Modelling And Control Of Aerial Manipulators Considering Aerodynamic Effects

DEPARTAMENTO DE INGENIER´ IA DE SISTEMAS Y AUTOM´ ATICA ESCUELA SUPERIOR DE INGENIER´ IA UNIVERSIDAD DE SEVILLA MODELLING AND CONTROL OF AERIAL MANIPULATORS CONSIDERING AERODYNAMIC EFFECTS por Pedro Jes´us S´anchez Cuevas Graduado en Ingenier´ıa Aeroespacial / Master en Ingenier´ıa Aeron´autica PROPUESTA DE TESIS DOCTORAL PARA LA OBTENCI ´ ON DEL T´ ITULO DE DOCTOR POR LA UNIVERSIDAD DE SEVILLA SEVILLA, 2020 Directores Dr.Ing. Guillermo Heredia Benot, Catedr´atico Dr.Ing. An´ıbal Ollero Baturone, Catedr´atico Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 ii Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 UNIVERSIDAD DE SEVILLA Memoria para optar al grado de Doctor por la Universidad de Sevilla Autor: Pedro Jes´us S´anchez Cuevas T´ıtulo: MODELLING AND CONTROL OF AERIAL MANIPULATORS CONSIDERING AERODYNAMIC EFFECTS Departamento: Departamento de Ingenier´ıa de Sistemas y Autom´atica V◦B◦Director: Guillermo Heredia Benot V◦B◦Director: An´ıbal Ollero Baturone El autor: Pedro Jes´us S´anchez Cuevas iii Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 iv Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 A mi familia, A mis profesores, A ella. v Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 vi Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 Agradecimientos Agradezco al grupo de rob´otica, visi´on y control de la Universidad de Sevilla la confianza depositada en mi persona para la realizaci´on de esta tesis. Me gustar´ıa hacer especialmente responsable y agradecer esta oportunidad a An´ıbal Ollero. ´ El fue el principio de la aventura. Tambi´en quiero citar la ayuda y profesionalidad de Guillermo Heredia Benot, que ha sido la soluci´on en los momentos m´as complicados, siempre manteniendo la calma y acertando en sus decisiones y consejos. Agradezco a mis directores, su confianza, su responsabilidad y las oportunidades que me han dado para crecer y dar los primeros pasos en este apasionante mundo de investigadores. Me dirijo tambi´en a mis compa˜neros del laboratorio, nada hubiera sido posible sin la atenci´on y ayuda que diariamente me han dedicado. Siempre me han prestado el consejo de la experiencia y gran parte de su talento resolutivo. Gracias especialmente a V´ıctor, Antonio, Fran, Arturo, Manu, Alejandro y Rafa. Lo que me habeis hecho vivir, es algo que siempre ir´a conmigo. Por ´ultimo, agradecer el apoyo de mi familia, que desde la distancia estaban cerca en los momentos mas duros de este camino, que me lo dieron todo sin preguntar y, por supuesto, sin esperar nada a cambio. Gracias a Rebeca, mi compa˜nera de viaje, mi apoyo en la batalla. Me has hecho crecer como persona y me has ense˜nado a confiar en mi mismo. Eres parte de esta victoria. Muchas Gracias. Pedro Jes´us S´anchez Cuevas Grupo de Rob´otica Visi´on y Control, Sevilla, 2020 vii Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 Este trabajo ha sido financiado a trav´es del programa de formaci´on del profesorado universitario (FPU) y por el proyecto nacional ARTIC (RTI2018-102224-B-I00) financiados por el Ministerio de Econom´ıa, Industria y Competitiviad y por los proyectos europeos H2020 AERIAL-CORE (H2020-2019-871479), RESIST (H2020-MG-2017- 769066) y HYFLIERS (H2020-ICT-25-2017-779411) financiados por la Comisi´on Europea. This work has been supported by the FPU Program and by the national ARTIC (RTI2018-102224-B-I00) project funded by of the Spanish Ministerio de Educaci´on, Cultura y Deporte and by the H2020 AERIAL-CORE (H2020-2019-871479), RESIST (H2020-MG-2017-769066) and HYFLIERS (H2020-ICT-25-2017-779411) projects funded by the European Commission. viii Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 Resumen El inter´es de la comunidad investigadora en los veh´ıculos a´ereos no tripulados (UAV) ha crecido significativamente en la ´ultima d´ecada. De hecho, este tema se est´a volviendo especialmente atractivo desde el punto de vista empresarial debido a su variada gama de aplicaciones. A pesar de que en las aplicaciones comerciales actuales, en las que los drones se utilizan para realizar tareas pasivas, como filmaci´on, vigilancia, inspecci´on visual, monitorizaci´on o teledetecci´on, la comunidad cient´ıfica est´a yendo m´as all´a utilizando estos drones como robots a´ereos que pueden interactuar f´ısicamente con el entorno. En esa interacci´on, el robot a´ereo est´a trabajando muy cerca de varios obst´aculos u objetos, por lo que la caracterizaci´on aerodin´amica de los fen´omenos que surgen al volar cerca de esos obst´aculos es muy importante para asegurar que la operac´ıon de manipulaci´on a´erea se pueda realizar de forma segura y eficaz. En esta tesis se aborda este problema espec´ıfico, que surge cuando los robots a´ereos que vuelan cerca de obst´aculos, mediante el desarrollo de nuevos modelos e implementaci´on de t´ecnicas de control adaptadas para garantizar la seguridad y efectividad durante la operaci´on en escenarios reales. A lo largo de esta tesis, el autor propone el estudio de dichos efectos aerodin´amicos a trav´es de la generalizaci´on del m´etodo cl´asico basado en la aerodin´amica potencial linealizada y estudios experimentales. Tambi´en trata dos casos de uso diferentes en los que se ha tenido en cuenta el efecto aerodin´amico desde el punto de vista del control. Por ´ultimo, da un giro a la forma de modelar el efecto aerodin´amico yendo m´as all´a del m´etodo cl´asicos, estudiando c´omo surgen diferentes efectos aerodin´amicos cuando el robot se acerca a diferentes obst´aculos desde un punto de vista bidimensional. ix Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 xvi Contents B Linearized potential aerodynamics 135 C Aerodynamic planning 139 C.1 The aerial long-reach manipulator for multidirectional inspection . . . 140 C.2 Motion planner with dynamics and aerodynamics awareness . . . . . 141 C.3 Application scenario . . . . . . . . . . . . . . . . . . . . . . . . . . . 142 C.4 Simulationresults............................. 144 References 145 Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 List of Figures 1.1 Aerodynamicforces............................ 5 1.2 Blade element theory: Forces and velocities in a generic blade element 6 1.3 Timeline of the contributions of the thesis . . . . . . . . . . . . . . . 13 1.4 Schemeofthethesis ........................... 19 2.1 Scheme of the aerodynamic problem modeled with the images method for one rotor: va represents the induced velocity by the rotor and vai is the induced velocity in the rotor by the image rotor. . . . . . . . . . . 25 2.2 Axial Velocity and streamlines of PFI model. . . . . . . . . . . . . . . 28 2.3 Illustration of the induced velocities applying the method of images in a general quadrotor. viim is the influence of the image rotor placed in i ontherotor1............................... 29 2.4 Generic multirotor schematics. . . . . . . . . . . . . . . . . . . . . . . 30 2.5 Test stand used for the characterization of the aerodynamic effects. a), b), c) and d) shows different setups of the test-bench. e) illustrates the variables involved in the measurement process . . . . . . . . . . . . . 33 2.6 Test stand and multiple configurations. . . . . . . . . . . . . . . . . . 34 2.7 Cases of use of the test stand. . . . . . . . . . . . . . . . . . . . . . . 35 2.8 GUI implemented to the test stand. . . . . . . . . . . . . . . . . . . . 35 2.9 Ground effect in a single rotor. Thrust increment as a function of the normalized distance to the ground. Mean values and standard deviation (errorbar) in color, curve approximation in black. . . . . . . . . . . . 37 2.10 Ground effect in a full multirotor. Experimental and PFI comparison. 39 xvii Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 xviii List of Figures 2.11 Conceptual CFD simulation of the velocity field approach for a multirotor flying close to the ground. Blue error bar represents the experimental results, magenta dashed line is the theoretical results using the generalisation of the PFI model, red error bar shows the experimental results obtained for a single rotor and black dashed line is the PFI theory for asinglerotor ............................... 39 2.12 Ground effect in a full multirotor. Green errorbar is the experimental results for a full multirotor, blue dashed line shows the proposed PFI+Geometrical term model, red errorbar and black dashed line, represent, respectively, the experimental and theoretical results for a single rotor..................................... 41 2.13 Experimental results of the ceiling effect in a single rotor. It shows that the thrust is increased when it works close to the ceiling. . . . . . . . 43 2.14 Changes in RPM for a single rotor working in ceiling effect conditions with the same PWM signal commanded. It shows how the ceiling effect produces a significant increment in the rotational speed. . . . . . . . 44 2.15 Ceiling effect in a single rotor. Experimental results and fitted model obtained using the least squares minimum . . . . . . . . . . . . . . . 45 2.16 Comparison between the ceiling effect in a quadrotor (in blue errorbar) and a single rotor (in red errobar). Black dashed line is the experimental fittedmodelof2.18............................ 46 2.17 Wall effect results. The presence of the wall does not clearly disturb therotorperformance .......................... 48 2.18 Ground tube effect (left) and ceiling tube effect (right) experimental results.It clearly shows that in this case, the model does not only depend on the vertical distance but also on the horizontal one. This will be explained in the 2D approach presented in the Chapter 5 . . . . . . . 49 2.19 Evolution of the tube effect with respect to time when the rotor moves above a pipe. The effect of the pipe is not constant due to the problem does not have symmetry of revolution. . . . . . . . . . . . . . . . . . 50 Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 List of Figures xix 2.20 Simple scheme of a tilt rotor in ground effect (left) and experimental results (right). A tilting rotor is less disturbed by the ground because the rotor wake has more space to be developed. . . . . . . . . . . . . 51 2.21 Ground and Vertical Wall Effect. Fountain effect arises due because the reverted flow that goes up along the wall affect to the rotor efficiency. 52 2.22 Ground + Vertical Wall Effect. CFD Approach to velocity field. This result clearly shows how the flow is reverted when reach the ground. . 53 2.23 Tandem rotor in ground effect . . . . . . . . . . . . . . . . . . . . . . 54 3.1 Partial ground effect close to the ground. . . . . . . . . . . . . . . . . 58 3.2 Partial ground effect in a typical aerial manipulation application. . . 59 3.3 (a) Case 1: flying over an obstacle under the influence of the ground effect. (b) Case 2: hovering at a point with only one rotor under the influence of ground effect. . . . . . . . . . . . . . . . . . . . . . . . . 63 3.4 PID controller: Flying over obstacle. Red line shows how the ground effect decreases the speed of the multirotor when it approaches to the obstacle. Green line shows how with a linear controller the multirotor is not able to overcome the obstacle if it is flying very close to it . . . 65 3.5 PID controller: Hovering in partial ground effect. In red and green, the controller is not able to achieve the references due to the ground effect. Actually, green line shows how the multirotor maintains oscillating trying to overcome the obstacle without success . . . . . . . . . . . . 66 3.6 PID controller, velocity of 1 m/s and z/R = 2. Hovering in partial ground effect. In this simulation, the robot is commanded to fly faster to avoid the ground effect thanks to the inertia. Although in b) the result is better, in a) the robot oscillates without reaching the reference. 66 3.7 PID controller - Hovering in partial ground effect. Simulation with sensor noise shows that oscillations are worse in a realistic scenario. . 67 3.8 Estimated torque controller. . . . . . . . . . . . . . . . . . . . . . . . 68 Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 xx List of Figures 3.9 Estimated torque controller: Flying over obstacle. The estimated torque controller improves the results compared with the linear controller, however, when the robot is very close to the obstacle (green line) the disturbance is again significant. . . . . . . . . . . . . . . . . . . . . . 69 3.10 Estimated torque controller: Hovering in partial ground effect. The estimated torque controller improves the results compared with the linear controller, however, when the robot is very close to the obstacle (green line) the steady state error is important. . . . . . . . . . . . . 69 3.11 Control scheme of feed-forward controller. . . . . . . . . . . . . . . . 70 3.12 Feed-forward controller: Flying over obstacle. Feed-forward controller is able to reach the reference and compensate the aerodynamic disturbance even in the closest case (green line) . . . . . . . . . . . . . . . . . . . 71 3.13 Feed-forward controller: Hovering in partial ground effect. Feed-forward controller is able to reach the reference and compensate the aerodynamic disturbance even in the closest case (green line) . . . . . . . . . . . . 72 3.14 Simulation of feed-forward controller using maps with offset errors: left - 5 cm error. right - 15 cm error. It shows how an error in the estimation of the aerodynamic disturbance can lead to the instability. . . . . . . 72 3.15 Flying over obstacle -Comparison of controllers. . . . . . . . . . . . . 73 3.16 Hovering in partial ground effect -Comparison of controllers. . . . . . 74 3.17 Experimental tests with multirotors in the testbench: a) the PQUAD, b) the PQUAD with a ground plane . . . . . . . . . . . . . . . . . . . 75 3.18 Experiments in the testbench, influence on the pitch angle of placing a ground plane under one of the rotors (green background). a) Effect in the PQUAD multirotor. b) Effect in the AMUSE multirotor . . . . . 75 3.19 Experiments in the testbench, variations of the quadrotor pitch angle when placing a ground plane under one of the rotors (green background): a) with the standard PID controller (not compensating), b) when the compensation is implemented (feed-forward controller) . . . . . . . . 76 3.20 Experiments in the testbench, variations of the quadrotor pitch angle when placing a ground plane under one of the rotors (green background) 77 Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 List of Figures xxi 3.21 Partial ground effect tests with multirotor flying outdoors: a) multirotor in free flight; b) multirotor flying under partial ground effect (the rotor at the right with a ground plane close to it) . . . . . . . . . . . . . . 78 3.22 Figure 30 Partial ground effect test with the multirotor in flight . . . 78 4.1 First prototype of the bridge inspection platform . . . . . . . . . . . 85 4.2 First prototype of the bridge inspection platform during the experiments intheteststand.............................. 86 4.3 Ceiling effect in a multirotor with fairings compared with the experimental curve for an isolated rotor . . . . . . . . . . . . . . . . . . . . 87 4.4 Geometry of the refined prototype for bridge inspection and mesh for CFDdetail ................................ 88 4.5 Velocity field when the multirotor is: a) in contact with the ceiling; b) flying 10 cm below the ceiling. . . . . . . . . . . . . . . . . . . . . . . 91 4.6 CTvs z/R fordifferentRPMs...................... 92 4.7 Kpvs z/R fordifferentRPMs...................... 92 4.8 CFD Experimental Results Comparison. Blue error bar and red errorbar are experimental results for quadrotor and multirotor respectively. Black braked line is the model proposed in 2.5.2 for the single rotor in ceiling effect. Yellow braked line is the model used in this work for the aerial platform designed (4.8) . . . . . . . . . . . . . . . . . . . . . . . . . . 93 4.9 Response of a multirotor in ceiling effect and safety flight envelope. Red area represents the very close to the bridge area in which flight forwards commands are not allowed. In yellow area, the ceiling effect gradient will be lower than a 5% between the rotors to ensure the stability of theplatform. ............................... 97 4.10 System overview. a) Aerial Platform. b) Reflector Prism. c) GCS. d) Robotictotalstation ........................... 102 4.11 Hardware/software architecture. WiFi communication is represented by dashed arrows and wired communication by the bold arrows. . . . 103 4.12 Navas bridge and experiment overview . . . . . . . . . . . . . . . . . 104 Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 xxii List of Figures 4.13 Altitude telemetry results during in manual flight. Red zone represents the dangerous area in which the aircraft critically loses its altitude and it could generate a dangerous flight condition . . . . . . . . . . . . . 105 4.14 Altitude nonlinear controller results. Altitude vs altitude reference control, thrust and mean value during the flight and contact (red dashed), and estimation of the ceiling effect factor ( Kc ) during the flight 107 4.15 PID vs Nonlinear z-velocity and z-position controller comparison. Green backgrounds represent the contact condition; red background is the dangerous zone where it can be observed an overshooting. It is shown that the nonlinear controller is more than 4 times faster moving away fromtheceiling............................... 108 4.16 Position controller results during an autonomous mission . . . . . . . 109 4.17 Attitude controller results during an autonomous mission . . . . . . . 110 5.1 Characterisation of aerodynamic effects: ground, ceiling and wall effects. Experimental results (red dots and blue errorbars) and theoretical models(blackcurves). .......................... 116 5.2 Test configuration and nomenclature . . . . . . . . . . . . . . . . . . 117 5.3 Ground and ceiling effect with a flat surface - experimental results . . 119 5.4 Flyable area detail and forbidden zones. Note that, in both cases we have symmetrical transition effects. . . . . . . . . . . . . . . . . . . . 120 5.5 Sample of aerodynamic effect close to a rectangular obstacle (left) and apipe(right) ............................... 121 5.6 Obstacle overlapping and shadows conditions. . . . . . . . . . . . . . 121 5.7 Aerodynamic effect map with rectangular obstacles . . . . . . . . . . 122 5.8 Aerodynamic effect map with tubular obstacles . . . . . . . . . . . . 122 5.9 Aerodynamic effect map with rectangular and tubular obstacles . . . 123 6.1 Aerodynamic effect in the robotic arm . . . . . . . . . . . . . . . . . 128 6.2 Aerodynamic effect inside of a tunnel . . . . . . . . . . . . . . . . . . 129 Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 List of Figures xxiii C.1 Aerial Robotic System for Long-Reach Manipulation (ARS-LRM). Geometry and mass distribution. . . . . . . . . . . . . . . . . . . . . . . 140 C.2 Block diagrams of the UAV controller (left) and the arm controller (right). 141 C.3 Operation basis of the MP-ARM-ADA algorithm. . . . . . . . . . . . 142 C.4 Application scenario given by a bridge inspection task. . . . . . . . . 143 C.5 3D map of aerodynamic effects for the bridge scenario. . . . . . . . . 143 C.6 Snapshot diagram corresponding to the closed-loop execution (dark blue) of the trajectory planned with the MP-ARM algorithm (light blue). 144 C.7 Snapshot diagram corresponding to the closed-loop execution (dark orange) of the trajectory planned with the MP-ARM-ADA algorithm (lightorange)................................ 144 Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 xxiv List of Figures Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 Chapter 1 Introduction Design is not how it looks like and feels like. Design is how it works Steve Jobs (Apple co-founder and former CEO) The interest of the research community in Unmanned Aerial Vehicles (UAVs) [ 1 ] has significantly grown in the last decade. Actually, this topic is becoming specially attractive for industrial companies due to its varied application range. There are a lot of companies,such as Amazon or DHL, which consider this technology as a clear business opportunity, and which, respectively propose [ 2 ] and [ 3 ] the use of drone with cargo capabilities as the future delivery system. They literally say that ““Unmanned aerial vehicles (UAVs) or drones can be used to deliver goods in the first and last mile as well for intralogistics and surveillance operations. ”” Nowadays, most of the commercial applications are coming from the image and remotely data acquisition field, where UAVs are mainly presented as flying cameras that can be used for filming, surveillance, or photogrametric applications [ 4 ], [ 5 ] among others. In those ones, UAVs stand out for being significantly cheaper and safer than a manned helicopter. On the one hand, in the field of visual monitoring and inspection, it is possible to find some specific samples in agricultural applications, where the WinterGreen 1 Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 8Introduction 1.1.2 Aerodynamic effects close to obstacles: A challenge in aerial manipulation Aerial manipulation operations usually involve flying close to objects, structures and other obstacles. Tasks, such as, grasping or manipulating objects that are placed on the ground, over surfaces, close to walls, under surfaces or even passing very close to a surface to grasp and object while flying, require that the multirotor operates close to the obstacle while performing the manipulation operation. However, as it is well-known by the aerodynamic community, flying in an constrained environment disturbs the free development of the streamlines of a fluid producing changes that affect the complete fluid field. In the case of the aerial manipulation, since the aerial manipulators are commonly rotorcrafts, the wake produced by the movement of the propeller or propellers is distorted by the environment, leading changes in the aerodynamic problem upwards and downwards, and, consequently, it changes the performance of the rotors in terms of thrust and drag. Then, aerodynamic effects changes the performance of the aircraft in the proximity of the obstacles, which is the nominal operation point of the aerial manipulator. In fact, in these situations, the needs of accuracy imposed to the manipulator and the aerial platform are the greatest ones because it is usually when the manipulation task is taking part. Thus, modelling and controlling this kind of aerodynamic effect is becoming essential to guarantee that the manipulation operation can be successfully performed at the same time that the aerial platform is working under safety conditions. Traditionally, the most common of these aerodynamic effects is the one produced by flying close to the ground. This phenomenon, usually known in the literature as ground effect, is more pronounced in rotorcrafts operating in hover and low speed. For rotor-craft hovering close to the ground, the rotor wake must rapidly expand as it approaches the surface, transitioning from the almost vertical down-wash to radial out-wash parallel to the ground. This alters the velocity of the slipstream and the induced velocity, which then affects the rotor thrust and power. The ground effect in helicopters has been well researched in literature [ 49 ], [ 50 ], [ 51 ] and has been studied for the take-off, landing, and hovering close to the ground [ 52 ], [ 53 ], [ 54 ]. However, for Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 1.1 Aerodynamics: A key aspect in a flying robot 9 multirotors, it had not received much attention at the start of this work, although these platforms were being increasingly used in multiple applications. The influence of ground effect in helicopters has been studied through the use of an underlying aerodynamic model or empirically. A classical analytical model for the ground effect is provided by [ 49 ], which uses the assumptions of the linearised potential aerodynamics placing a single source to model the rotor airflow and the method of images to account for the ground effect. Other authors have provided empirical expressions for the rotor thrust increment in ground effect for large [ 51 ] and small UAV helicopters [ 55 ]. However, the model of [ 49 ] has been, traditionally, the most used in robotics because it has a simple analytical form, which experimentally addresses the relationship between the rotor thrust in ground effect and the rotor height over the ground surface with enough accuracy As it was beforementioned, the ground effect in multirotors has received much less attention. Several papers deal with the aerodynamic models of multirotors [ 56 ], which are used for navigation [ 57 ] or even for power control [ 58 ]. Disturbance observers have become popular in the last years for estimating external wrench in multirotors [ 59 ], [ 60 ]. In most cases, it is assumed that the external disturbance source is unique and mainly contact forces [ 61 ], [ 62 ] or wind. In [ 63 ], the simultaneous online estimation of aerodynamic and contact forces is studied to discriminate between them and compute the wind velocity by means of model inversion. However, disturbance observers have not been used for the ground effect estimation. The influence of the ground effect has been considered in the development of controllers for low-altitude flight in [ 64 ] using an adaptive controller and a height estimator that works well in the experiments, although false measurement from the ultrasonic sensor can destabilise the system. A take-off and landing controller that uses an ant colony filter for estimation of the ground effect is presented in [ 65 ]. The algorithm is tested in simulation with PID and sliding modes controllers, and in both cases the consideration of the ground effect improves significantly the controller performance. In [ 66 ], a PID landing controller with a ground effect robust compensator has been presented. The experimental tests show that accounting for the ground effect improves the controller behaviour. Also, the ground effect estimation using a Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 10 Introduction vision sensor to estimate distance to obstacles and learning from previous flights has also been proposed [ 67 ]. A dynamic controller for rotorcrafts landing and hovering in ground effect using feedback control based on flow field estimation has also been developed and tested in simulation [68]. However, in almost all cases, the model of the ground effect in [49] is assumed or evaluated for a single rotor. The only experimental results that have been reported, making experiments with a small quadrotor flying in hover over the ground at different heights [ 69 ] and using a test bench [ 70 ], suggest that the ground effect in multirotors may be larger than predicted by [ 49 ], although the issue had not been further analysed at the beginning of this work. Although, the ground effect has been the most studied aerodynamic effect, a similar effect is caused by flat surfaces placed above the rotors, known as the ceiling effect. This phenomenon induces an additional thrust on the robot, pushing it towards the ceiling. Thus, this fact can be dangerous for standard multirotors because propellers and engines may collide with the ceiling and break, or destabilise the platform. Ceiling effect is relevant in applications in which the aerial manipulators fly close to an obstacle from the underside. For instance, in [ 71 ], authors proposed using an aerial manipulator with the arm attached on top of the multirotor body above the rotors for conducting the inspection by the contact of bridge elements without any human intervention, since the arm can be used to hold the ultrasound sensor that should be in contact with the bridge while the multirotor is hovering [72]. Although the ground and ceiling effects are the most significant aerodynamic ones, there are others that could arise in the current application scenery of the aerial manipulators. This is the case of flying, for example in narrow environments, such us tunnels, mines or cluttered and close environments, like tanks or vessels. In other cases, it is necessary to fly in complex environments, such us chemical plants or refineries to carry out inspection and maintenance operations. All these situations produce different aerodynamic disturbances that could be decisive to guarantee the reliability and suitability of using UAVs in real contact-based operations. Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 1.2 Scope of this work 11 1.2 Scope of this work This Ph.D falls within the framework of three research projects funded by the European Commission. The first project is entitled “AErial RObotic system integrating multiple ARMS and advanced manipulation capabilities for inspection and maintenance” (AEROARMS) [ 73 ] proposes the development of the first aerial robotic system with multiple arms and advanced manipulation capabilities to be applied in industrial inspection and maintenance (I&M). The second one “AErial RObotic System for In-Depth Bridge Inspection by Contact” (AEROBI) [ 33 ] intends to develop a robotic system to inspect the bridges and assess their status. The third project is the “AERIAL COgnitive integrated multi-task Robotic system with Extended operation range and safety” (AERIAL CORE) [ 36 ] project which is about infrastructure inspection of large infrastructure. Thus, the thesis is focused on the modelling and characterisation of the aerodynamic effects that arise in aerial manipulators when flying in the proximity of obstacles and their applications in the context of the aforementioned projects, developing control and planning techniques which take the previous models and characterisations into account. Moreover, the work of this thesis has been done in order to take the greater range of possibilities that can arise in that aerial robotic manipulation applications into consideration. The workflow of the research includes the following tasks: • Study of aerodynamic effects of flying close to obstacles. This involves studying the effect on the performance of UAVs and the deficiencies of current control techniques when flying near obstacles. In particular, these aspects have been addressed when the UAV flies in: –The proximity to flat surfaces: floor, ceiling, etc. – Industrial environments similar to those of the AEROARMS project: flights of the UAV near pipes, studying the flow around pipes... – Environments close to bridges or infrastructures similar to those of the AEROBI project: UAV flight near the ceiling and pillars. Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 12 Introduction • Study of the thrust generation in conventional rotors of a multirotor: The study of the characteristics of the typical motorisation of this type of UAVs. In particular, it focuses on characterising and modelling its optimal regime, behaviour in different conditions and in the proximity to obstacles. • Advanced study of the influence of flying in the proximity to surfaces: This task consists of modelling, the static study and the study in dynamic conditions of the different disturbances. Disturbances have been studied in different UAVs, with different sizes and configurations. The final objective of this part is the development of aerodynamic effects maps. • Estimation of aerodynamic effects to be considered in the controller: This part of the thesis consists of the estimation of aerodynamic effects in order to take them into account in the UAV controller. It is the integration of aerodynamic effects maps of the previous section and the models effect in the environment in the controllers. The ultimate goal is to estimate external forces and moments only with navigation sensors. • Multirotor control techniques with aerodynamic effects: In the study of multirotor control techniques, the classic linear and non-linear control techniques (backstepping, passivity, etc.) and their behaviour have been discussed before the aerodynamic disturbances. • The applications of the models in planning techniques which take the aerodynamic behaviour when flying close to obstacles into account, in order to generate more efficient and safer trajectories. • Experimental validation of aerodynamic controllers: The objective of this phase is the validation with real flights of the aerodynamic control techniques developed in the previous phases. These experiments have been carried out in the scenarios of the AEROARMS and AEROBI projects. Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 1.3 Contribution of the work 13 1.3 Contribution of the work The main contribution of this work consists of generating a scientific background of the aerodynamic effects that can affect aerial manipulators, as well as the consideration of them in pursuit of safety and effectiveness while carrying out the manipulation task in the framework of the AEROARMS and AEROBI projects. In the following, it is presented a timeline (Figure 1.3) with the different contributions and works developed. Although the study of the behaviour of the flow has been widely studied, particularly, in ground effect from different points of view, the scheme only shows previous studies that have been widely extended along the robotic community [49] [51]. 1955 - 1976 2013-2017 2002 2017 Ground effect on a helicopter [49] Cheeseman and W. E Bennet [51] J. S. Hayden Dual-Rotor Interference [74] D. A. Griffiths and G. Leishman. First Approach Ground Effect in Multirotors [69]C. Powers, D. Mellinger, A. Kushleyev, B. Kothmann, and V. Kumar, [70]I. Sharf et al. [75] D. Del Cont Bernard, F. Riccardi, M. Giurato, and M. Lovera First Approach Ceiling Effect in Multirotors [78] Conference Paper First Approach to Aerodynamic Effects close to obstacles [76] Conference Paper Influence of Ground Effect in Multirotor Control [79] Journal Paper Starting Point 2019 Multirotor Aerodynamic Effects in Aerial Manipulation [81] Book Chapter Robotic System for bridge inspection using UAVs [77] Journal Paper First Aerodynamic Map proposal [82] Conference Paper Influence of Ceiling Effect in Multirotor Control [80] Journal Paper 2020 Aerodynamic awareness planning [84] Journal Paper Control of Aerial Robotic Manipulators [83] Book Chapter Figure 1.3: Timeline of the contributions of the thesis As it was aforementioned, at the beginning of this research in 2015, there was not a deep background in the characterisation of aerodynamic effects in multirotors. In fact, there were only first approaches to them that were inherited from the classical helicopter aerodynamic ground effect and a few results applied to tandem rotors [ 74 ] and multirotor [ 69 ] [ 70 ] [ 75 ]. Moreover, the knowledge about other aerodynamic effects different from the ground effect was restricted to the ceiling effect [ 69 ]. In fact, the approach of studying these aerodynamic effects in aerial manipulators was practically non-existent. Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 14 Introduction This work has contributed to this research community being one of the first that analysed the aerodynamic effects when a rotor works close to different kinds of obstacles in a general approach, studying not only the ground and ceiling effect but also the effect of flying close to a wall or a corner and the combined effect of two rotors working together [ 76 ], putting an special interest in the ceiling effect in the following publications [ 77 ], [ 78 ]. It was also the first research that considered the interaction of the flow-field with the quadrotor body through a combined experimental and the CFD analysis to adapt the previous theoretical results of helicopters to multirotors including also the use of model-based controllers to predict the behaviour of the multirotor in presence of these obstacles [ 79 ], [ 80 ]. The author also proposed a study of the generalisation of the classical method based on the linearised potential aerodynamics and published a chapter about the aerodynamic effects that are suitable to appear in aerial manipulators in [ 81 ]. Lastly, the contribution of [ 82 ] and [ 83 ] open a whole world of possibilities with the proposal of developing an aerodynamic effect map which could be applied to more complex control strategies or in an aerodynamic awareness planning algorithm [84] and specific aerodynamic awareness controllers. Extra contributions In addition to the work briefly described above, the author of this thesis has contributed to different publications about aerial manipulation [ 29 ], [ 23 ], to a new concept of spherical drone with autonomous indoor localisation [ 85 ], to propose a cybersecurity solution in the UAV architecture to be deployed in post-disaster situations [ 86 ] and to design a new concept of robot for inspection in oil and gas industries and aerial physical interaction with in grabbing conditions with a compliant dual-arm aerial manipulator. Details of those papers are included on the list of publication in Section 1.4. Moreover, the author is decision pending of one journal paper about multirotor autopilot system for fast prototyping solutions, two journal publications about the results obtained in the MBZIRC competition and a IEEE Transaction on Industrial Electronics which presents a fully autonomous operation of an aerial robot. Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 1.4 List of publications 15 1.4 List of publications 1.4.1 Published papers Journal Papers • Pedro J. Sanchez-Cuevas, G. Heredia, and A. Ollero, Characterization of the Aerodynamic Ground Effect and Its Influence in Multirotor Control. International Journal of Aerospace Engineering, vol. 2017, 2017. • Pedro J Sanchez-Cuevas, Pablo Ramon-Soria, Bego˜na Arrue, Anibal Ollero, and Guillermo Heredia. Robotic system for inspection by contact of bridge beams using uavs. Sensors, 19(2):305, 2019. • Antonio E Jimenez-Cano, Pedro J Sanchez-Cuevas, Pedro Grau, Anibal Ollero, and Guillermo Heredia. Contact-based bridge inspection multirotors: Design, modeling, and control considering the ceiling effect. IEEE Robotics and Automation Letters, 4(4):3561-3568, 2019. • Salua Hamaza, Ioannis Georgilas, Manuel J Fernandez, Pedro J Sanchez-Cuevas, Thomas Richardson, Guillermo Heredia, and Anibal Ollero. Sensor installation and retrieval operations using an unmanned aerial manipulator. IEEE Robotics and Automation Letters, 2019. • Pedro J. Sanchez-Cuevas, Antonio Gonzalez-Morgado, Nicol´as Corte´s, Diego Benjumea-Gayango, Antonio E. Jimenez-Cano, Anibal Ollero and Guillermo Heredia. Fully-Actuated Aerial Manipulator for Infrastructure Contact Inspection: Design, Modeling, Localization, and Control. Sensors, 20(17), 4708, 2020 • Alvaro Caballero, Pedro J. Sanchez-Cuevas, Manuel Bejar, Guillermo Heredia and Anibal Ollero. An aerodynamic extension for motion planning with dynamics awareness in aerial long-reach manipulators. International Journal of Aerospace Engineering, vol. 2020, 2020. Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 16 Introduction • Alejandro Suarez, Alvaro Caballero, Ambar Garofano, Pedro J. Sanchez-Cuevas, Guillermo Heredia and Anibal Ollero, Aerial Manipulator With Rolling Base for Inspection of Pipe Arrays. IEEE Access, vol. 8, pp. 162516-162532, 2020. Conference Papers • Pedro J. Sanchez-Cuevas, G Heredia, and A Ollero. Multirotor uas for bridge inspection by contact using the ceiling effect. In 2017 International Conference on Unmanned Aircraft Systems (ICUAS), pages 767-774. IEEE, 2017. • Pedro J. Sanchez-Cuevas, Guillermo Heredia, and Anibal Ollero. Experimental approach to the aerodynamic effects produced in multirotors flying close to obstacles. In Iberian Robotics Conference, pages 742-752. Springer, 2017. • Pedro J. Sanchez-Cuevas, Victor Mart´ın, Guillermo Heredia and An´ıbal Ollero Aerodynamic effects in multirotors flying close to obstacles: Modelling and mapping. In Iberian Robotics conference, pages 63-74. Springer, 2019. • Alejandro. Su´arez, Pedro J. Sanchez-Cuevas, Manuel Fernandez, Manuel Perez, Guillermo Heredia, and Anibal Ollero (2018, October). Lightweight and compliant long reach aerial manipulator for inspection operations. In 2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), pp. 6746- 6752, IEEE, 2018. • Agustin Ramos, Pedro J. Sanchez-Cuevas, Guillermo Heredia and Anibal Ollero. Spherical fully covered uav with autonomous indoor localization. In Iberian Robotics conference, pages 355-367. Springer, 2019. • Manuel Fernandez, Pedro J. Sanchez-Cuevas, Guillermo Heredia and Anibal Ollero. Securing UAV communications using ROS with custom ECIES-based method. International Workshop on Research, Education and Development on Unmanned Aerial Systems. RED-UAS 2019, Cranfield. • Abraham L´opez, Pedro J. Sanchez-Cuevas, Alejandro Suarez, Ambar Soldado, Anibal Ollero, and Guillermo Heredia. MHYRO: Modular HYbrid RObot for Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 1.4 List of publications 17 contact inspection and maintenance in oil&gas plants. In 2020 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), IEEE, 2020. Book Chapters • Pedro Sanchez-Cuevas, Guillermo Heredia, and Anibal Ollero. Multirotor aerodynamic effects in aerial manipulation. In Aerial Robotic Manipulation, pages 67-82. Springer, 2019. • Guillermo Heredia, Pedro J. Sanchez-Cuevas, Control of Aerial Robotic Manipulators. In Encyclopedia of Robotics. Springer, 2020. 1.4.2 Submitted Papers Journals • Carlos R. de Cos, Manuel J. Fernandez, Pedro J. Sanchez-Cuevas, Jose A. Acosta and Anibal Ollero. MAS+: A High-Level Modular Autopilot Solution for Fast Prototyping of Unmanned Aerial Systems. IEEE Access. • Fran Real, Angel Casta˜no, Jesus Capitan, Pedro J. Sanchez-Cuevas, Manuel Fernandez, Manuel Villar, and Anibal Ollero. Autonomous Fire-fighting with Heterogeneous Team of Unmanned Aerial Vehicles. Field Robotics • Fran Real, Angel Casta˜no, Jesus Capitan, Pedro J. Sanchez-Cuevas, Manuel Fernandez, Honorio Romero, and Anibal Ollero. A Team of Multiple Unmanned Aerial Vehicles for Cooperative Construction. IEEE Access • Pedro J. Sanchez-Cuevas, Alejandro Suarez, Pablo Ramon-Soria, Manuel Fernandez, Rafael Salmoral, Guillermo Heredia, Bego˜na C. Arrue, and Anibal Ollero. Compliant Dual Arm Aerial Manipulator to Grasping and Installation of Inspection Tool in Industrial Environments. IEEE Transactions on Industrial Electronics Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 24 Modelling where M is the inertia matrix, the Coriolis and centrifugal terms are represented by C and the gravitational force terms are described by G , F is the vector of forces and torques generated by the actuators. It can be decomposed into the thrust forces FQ and moments, τQ generated by the rotors, and the torque of each link of the manipulator, τM . External forces and moments are represented by FE , which include the contact forces and moments of the end effector. However, in case of a flight under the influence of an aerodynamic effect, the generalised vector force changes because the forces change due to the relative position between the multirotor and the different obstacles of the environment, so the generalised vector is a function of the state of the multirotor: F = F(˙ ξ) . Thus, the dynamic model is rewritten as: M(ξ)¨ ξ+C(ξ,˙ ξ)˙ ξ+G(ξ) = F(ξ)+FE(2.2) Then, it is clear that considering this aerodynamic effect affects the dynamics of the full system, and it is necessary to take it into account in such applications that require operating close to obstacles, like a ground or ceiling in a accurate way. 2.3 Analytical Modelling based on potential flow 2.3.1 Classical method The classic method used to approach the aerodynamic effect presented in this chapter was introduced by [ 90 ] and [ 49 ] using the assumptions of the linearised potential aerodynamics (Appendix B). In the first theoretical study, Zbrozek [ 90 ] proposed to model the rotor as a sink of constant strength, s = Avi/ 4 π , where R is the rotor radius and vi is the velocity induced by this rotor. However, this expression was reinterpreted by Cheeseman in [ 49 ] proposing changing the sink by a source due to the flow pattern beneath a lifting rotor resembles a source rater than a sink. Then, the expression of the potential flow is: Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 2.3 Analytical Modelling based on potential flow 25 φ=−s p(x−x0)2+ (y−y0)2+ (z−z0)2(2.3) In order to model the effect of a surface like a ground, this classical approach used the method of the images placing a “virtual rotor” on the opposite side of the surface and at the same distance and with the same strength. This method uses the symmetry conditions to create a “virtual surface” (like the ground) in which the normal velocity is zero and the impenetrability condition is ensured, which is n·∇φ = 0, where n vector normal to the ground surface, · is the scalar product and ∇φ represents the velocity vector derived from the potential equation (2.3) according to the assumptions of the linearised potential aerodynamic. In this way, using equatoin (2.3) and images method, the potential equation of the problem illustrated in Figure 2.3.1, this can be written as (2.4). This expression models a single rotor, placing one rotor in x = 0, y = 0 and z = h and the image rotor in x= 0, y= 0 and z=−h, in this way, the distance to the ground is assumed as h. Figure 2.1: Scheme of the aerodynamic problem modeled with the images method for one rotor: va represents the induced velocity by the rotor and vai is the induced velocity in the rotor by the image rotor. φ=−s px2+y2+ (z−h)2+−s px2+y2+ (z+h)2(2.4) Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 26 Modelling Assuming that the power rotor at zero air speed could be written as P = Tvi , where Tis the thrust and viis the induced velocity by the rotor. Therefore, at constant power, TOGEviOGE =TIGEviIGE , we can write: TIGE TOGE =viOGE viIGE (2.5) where the subindex IGE and OGE refer to conditions inside and outside the ground effect. If viOGE ( x = 0 , y = 0 , z = h ) = va and viOGE ( x = 0 , y = 0 , z = h ) = va−δvai , where δvai is the velocity induced by the image rotor (located in z = −h ) in the position of the rotor (x= 0, y = 0, z =h), according to the Figure 2.3.1. Then, expression (2.5) can be written now as follows: TIGE TOGE =1 1−δvai vi (2.6) Then, δvai can be obtained following the aerodynamics assumptions which say that the axial velocity field can be obtained partially deriving (2.4) with respect to the axial direction, va ( x, y, z ) = ∂φ ∂y and substituting the values of x = 0 , y = 0 , z = h . Deriving, it is possible to obtain the expression (2.7) which shows the axial velocity induced by the mirror-image rotor called as vai(x, y, z). vai(x, y, z) = ∂φ ∂z =s(z+h) p(x2+y2+ (z+h)2)3(2.7) substituting in 2.7, the values of x = 0 , y = 0 , z = h and taking into account that s=Avi/4π, it is concluded that: δvai=R2vi 16h2(2.8) Therefore, generalizing for any h=z, TIGE TOGE =1 1−1 16(R z)2(2.9) Although other authors have presented empirical expressions for the increment of rotor thrust in ground effect which show similar results [ 51 ] [ 55 ], the ”Potential Flow Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 2.3 Analytical Modelling based on potential flow 27 with the method of Images” (PFI) model presented in [ 49 ] and summarised in this chapter have been widely used because it has the simplest analytical form and has been experimentally shown to accurately capture the relationship between the rotor thrust in ground effect and the rotor height over the ground surface. To obtain a visual description of the flow field represented by the potential flow, the expressions of the velocities: vx(x, y, z) = ∂φ ∂x =sx p(x2+y2+ (z+h)2)3+sx p(x2+y2+ (z−h)2)3(2.10) vy(x, y, z) = ∂φ ∂y =sy p(x2+y2+ (z+h)2)3+sy p(x2+y2+ (z−h)2)3(2.11) vz(x, y, z) = ∂φ ∂y =s(z+h) p(x2+y2+ (z+h)2)3+s(z−h) p(x2+y2+ (z−h)2)3(2.12) has been obtained and represented in the Figure 2.3.1, in such a way that v= [vx(x, y, z)vy(x, y, z)vz(x, y, z)]0 Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 28 Modelling Figure 2.2: Axial Velocity and streamlines of PFI model. Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 2.3 Analytical Modelling based on potential flow 29 2.3.2 PFI generalization to Multirotors The formulation applied to the case of a single rotor in ground effect presented before can be extended to the case of a complete multirotor with n rotors. In this case, each rotor is replaced by a source like the presented in Figure 2.3 of the Section 2.3 and the ground by an image source of equal strength placed below this rotor at a distance equal to twice the height of the rotor. However, it is important to remark that all image sources induce changes in all rotors so it is necessary to rewrite the expression of the vIGE to take the axial velocity induced by the other rotors into account. For example, in the case of a quadrotor with all rotors equally spaced (see Figure 2.3), the expression for vIGE can be rewritten as (2.13): Figure 2.3: Illustration of the induced velocities applying the method of images in a general quadrotor. viim is the influence of the image rotor placed in ion the rotor 1 vIGE =vOGE −δv11imvOGE −δv21imvOGE −δv31imvOGE −δv41imvOGE (2.13) where vijim is the influence of the image rotor placed in i on the rotor j . All the rotors are influenced equally and in the case of the assumption that the power remains constant TIGEvIGE = TOGEvOGE for the complete multirotor, the following expression can be obtained: Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 30 Modelling TIGE TOGE =1 1−(R 4z)2−R2h √(d2+4h2)3−R2 2 h √(2d2+4h2)3 (2.14) Then, in a general way in which we have a multirotor composed of N rotors which are not at the same distance (Figure 2.4). It is possible to express the theoretical ground effect using the expression (2.15): TIGE TOGE =1 1−(R 4z)2−R2 2PN i=2 h √(d2 i+4h2)3 (2.15) where di is the relative distance between the i rotor and the reference rotor with i = 1. In this way, it is possible to obtain the value of TIGE TOGE for the rotor i = 1, which can be generalised to the full multirotor because it is dimensionless. Figure 2.4: Generic multirotor schematics. This result follows the basis used and widely tested to approach the aerodynamic ground effect in a single rotor, however, it is not enough to demonstrate if it is valid or not. Then, this thesis follows a detailed experimental part in which this theoretical model is compared with those previous results, especially with the result of the expression (2.14) to validate the generalisation of the theoretical approach. Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 2.4 Experimental modelling of aerodynamic effects - Methods and Materials 31 2.4 Experimental modelling of aerodynamic effects - Methods and Materials As it was aforementioned, it has been considered necessary to be able to assess the aerodynamic ground effect of a multirotor in a realistic setup which allows carrying out experiments with multirotors in multiple configurations. In order to aim this goal, a test stand was built for an experimental motor/rotor aerodynamic characterisation close to surfaces. This is presented in Section 2.4.1. Moreover, an experimental procedure has also been defined to properly compare the results obtained through the experimental characterisation (Section 2.4.1) ensuring that all the experiments are accomplished in a controlled which permits the comparison of the experimental results in a correct way. 2.4.1 Test stand This section describes the test stand developed in this thesis, which will be used later to validate the theoretical results and model different aerodynamic effects in multirotors. To sum up, the test stand is intended to measure the thrust of the propellers in the presence of different types of surfaces, such as ground, ceiling, wall or circular surfaces. Then, the test stand should address two kind of requirements. On the one hand, the physical requirements are: •Asses aerodynamic effects on multiple types of surfaces and obstacles • Accomplish each experiment with the possibility of moving the distance between the rotors and the obstacle •Be able to change the relative angle between the rotor and the obstacle • The possibility of assessing the aerodynamic effect in different kinds of surfaces, flat, rough, grid... Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 32 Modelling • Capacity to assess more than one rotor at the same time (tandem rotor or coaxial configuration) •Be able to assess a complete small size multirotor On the other hand, the functional requirements are: • It is necessary to obtain the data of the thrust developed by the rotor, rotors or the complete multirotor. •It is necessary to log the rpm of the rotor • It is necessary to be able to measure and control the pwm commanded to the rotors The result that addresses those requirements is the test stand presented in Figure 2.5. The test-bench consists of an isostatic L-shaped structure that integrates a load cell in its lower endpoint and the rotor, whose aerodynamics will be characterised in the distal end of the vertical part (see Figure 2.5.E). Additionally, a cylindrical joint, γ , allowing free rotation of the structure, has been used to avoid any resultant force in the holder between the ground basis and the corner intersection of the L-shaped structure. An analysis of the force balance in this structure allows to calculate the rotor thrust F3 through the measurement of the reaction force Flc applied to the load cell. The resultant equation of equilibrium is the following: F3=Flc dh dv (2.16) Where dh and dv are the dimensions depicted in Fig 7.b. Regarding the electronics embedded in the structure, the test-bench integrates an Arduino Mega 2560 as a data acquisition unit and it includes sensors to measure the rotor speed as well as a load cell to collect the measurements of the reaction force, Flc. Moreover, the PWM signal commanded to the rotor is monitored at the output using an interruption-based algorithm which runs in the board. The Arduino Mega 2560 is connected through Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 2.4 Experimental modelling of aerodynamic effects - Methods and Materials 33 the serial port to a PC with MATLAB that offers a graphical interface to process, displays and saves the data collected by the sensors. Moreover, the test stand has a bearing system that allows tests with different distances and inclination angles of the rotor plane with respect to the obstacle surface as it is shown in Figure 2.5.A/B/C/D. Moreover, changing the obstacle or surface behind, it is clear that the test stand can be used to test several obstacle conditions but, in this case, we are interested in the ceiling effect results. Figure 2.5: Test stand used for the characterization of the aerodynamic effects. a), b), c) and d) shows different setups of the test-bench. e) illustrates the variables involved in the measurement process Figure 2.6 shows how this device is also able to mount different rotor configurations and even a complete multirotor. Experimental procedure The experimental procedure is based on the measurements of the rotor thrust provided by the test-bench. These measurements will be collected in situations where an element is located close to the rotor (from above, below or laterally, depending on the particular phenomenon under study, that is, ground effect, ceiling effect or wall effect as it is stated in Figure 2.7). The PWM input signal that governs the rotor speed will be the same in all tests, which, in turn, implies that the desired rotor thrust will Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 40 Modelling the force which arises due to the presence of the multirotor frame (see F in Figure 2.11). The force, F has been modeled including an additional term in the equation provided by the PFI model. This additional term uses the velocity of the air at the central point of the body and adjusts its influence with an empirical coefficient Kb , which takes a value close to 2 in the experiments which are performed and it is strictly dependent on the geometry of the multirotor. Finally, superimposing this effect, due to the collision of the flow with the PFI model, for a complete multirotor the following expression can be obtained: TIGE TOGE =1 1−(R/4z)2−R2(z/ √(d2+4z2)3)−(R2/2)(z/ √(2d2+4z2)3)−2R2(z/ √(b2+4z2)3)Kb (2.17) This last expression is presented in blue dashed line in Figure 2.12. 2.5.2 Ceiling Effect in Multirotors In some emerging applications, aerial vehicles need to physically interact with the environment from the underside, as in bridge deck or beams inspections by contact operations. In those cases, the so-called “aerodynamic ceiling effect”, introduced by the author in [ 78 ], will arise affecting the multirotor just like the ground effect does when multirotors work close to a surface from the upper side. This made it essential to study this problem in detail, which, furthermore, had not received much attention in literature. Then, this section follows the same approach as the previous one, in which the effect of an isolated rotor is analysed and then, the same test is carried out with a complete multirotor to assess its behaviour. Ceiling effect - Single rotor experiments First of all, it is necessary to analyse the ceiling effect for a single rotor in order to get a first approach to the problem using the test stand presented in Section 2.4.1 Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 2.5 Experimental modelling of aerodynamic effects - Results 41 z/R 012345678 TIGE/TOGE 0.9 1 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 Ground Effect - Complete Quadrotor Figure 2.12: Ground effect in a full multirotor. Green errorbar is the experimental results for a full multirotor, blue dashed line shows the proposed PFI+Geometrical term model, red errorbar and black dashed line, represent, respectively, the experimental and theoretical results for a single rotor. Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 42 Modelling following the setup presented in Figure 2.7 for the ceiling effect. Next, the results obtained in these experiments are analyzed. In this way, Figure 2.13 shows the results obtained in these ceiling effect experiments for the rotor thrust. T is the thrust generated by the rotor, z is the distance to the surface and R the radius of the rotor (in this case 9 . 4 inches with DJI 2312 rotor). The acronym “ICE” means “In Ceiling Effect” and “OCE” means “Out Ceiling Effect”. The mean value and the standard deviation of the thrust for each normalised distance z/R are shown in Figure 2.13 and again represented in Table 2.2. The results clearly show that the thrust increases significantly when the rotor approaches the ceiling, i. e., when a multirotor is flying close to the ceiling, the thrust of the rotors will also increase and it will push the multirotor even closer to the ceiling. It is important to consider this effect in order to avoid a crash between the multirotor and the ceiling surface. Moreover, the effect is stronger than the ground effect. In order to find an explanation of the abrupt increase in the thrust, the author looked into other data collected during the experiment. In this case, these data also show that the aerodynamic ceiling effect affects the rotational speed of the rotor significantly. Figure 2.14 shows the changes in the RPM as the rotor approaches the surface meanwhile the PWM signal commanded to the rotors remains constant. This is consistent with the large increase in thrust of the Figure 2.13 because the RPM appears squared in the classical thrust expression. Then, this increase of the RPM is produced by a vacuum effect which decreases the propeller drag when it is very close to the obstacle. Therefore, then the rotor can rotate faster generating this thrust increase joined to the one created due to the change of pressures around the propeller. Table 2.2: Standard deviation for single rotor in ceiling effect experiments z/R 0.20 0.40 0.80 1.20 1.60 2.00 2.40 σ0.0031 0.0033 0.0037 0.0038 0.0037 0.0022 0.0020 Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 2.5 Experimental modelling of aerodynamic effects - Results 43 0 0.5 1 1.5 2 2.5 3 3.5 4 z/R 0.8 1 1.2 1.4 1.6 1.8 2 2.2 2.4 TICE/TOCE Ceiling Effect Single Rotor Figure 2.13: Experimental results of the ceiling effect in a single rotor. It shows that the thrust is increased when it works close to the ceiling. Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 44 Modelling 0 0.5 1 1.5 2 2.5 z/R 5600 5700 5800 5900 6000 6100 6200 6300 6400 RPM Rpm in Single Rotor Ceiling Effect Figure 2.14: Changes in RPM for a single rotor working in ceiling effect conditions with the same PWM signal commanded. It shows how the ceiling effect produces a significant increment in the rotational speed. Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 2.5 Experimental modelling of aerodynamic effects - Results 45 Since this effect cannot be modeled using the PFI so a least squares minimum method has been used to fit the experimental curve as follows: TICE TOCE =1 1−1 K1(R z+K2)2(2.18) Where the least squares minimum values obtained are K1 = 6 . 924 and K2 = 0.03782(m). This is the black braked line represented in Figure 2.15. 0 0.5 1 1.5 2 2.5 3 3.5 4 z/R 0.8 1 1.2 1.4 1.6 1.8 2 2.2 2.4 TICE/TOCE Thrust in Ceiling Effect - Single Rotor Experimental model Experimental results Figure 2.15: Ceiling effect in a single rotor. Experimental results and fitted model obtained using the least squares minimum Ceiling effect - Full multirotor experiments The same phenomenon which appears for a multirotor operating in ground effect appears when a multirotor is flying close to the ceiling. The flow field generated around each rotor is also affected by the other rotors and the multirotor frame. Thus, Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 46 Modelling the ceiling effect for a full multirotor produces a different effect than adding the effects of each individual rotor due to the coupling between the flow of the propellers. 0.5 1 1.5 2 2.5 3 3.5 4 1 1.2 1.4 1.6 1.8 2 2.2 2.4 Ceiling Effect Experimental model Experimental errorbar - Multirotor complete Experimental errorbar - Single rotor Figure 2.16: Comparison between the ceiling effect in a quadrotor (in blue errorbar) and a single rotor (in red errobar). Black dashed line is the experimental fitted model of 2.18 These results can be interpreted from different points of view. On the one hand, it is possible to develop more thrust for the same power if the multirotor is subjected to the aerodynamic ceiling effect. On the other hand, the multirotor can increase its maximum flight time because the rotor decreases its energy consumption. It is a great benefit which can be used in several UAV applications. Moreover, researches like [ 78 ] which is also presented in the Chapter 4 of this thesis, shows how to take advantage of the ceiling effect designing a platform which can maintain a contact with the ceiling. This solution helps to minimise the two typical problems which appear in the applications of multirotors. Firstly, as it was mentioned above, the ceiling effect increases the maximum flight time and secondly, the ceiling Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 2.5 Experimental modelling of aerodynamic effects - Results 47 effect solves almost completely the problem of the stability during the operation and the motion-blur because the platform is stick to the ceiling during the application task and has constrained all the displacements. The last problem arises especially in high precision applications, such as aerial manipulation close to a surface, which generates aerodynamic effects or inspection by contact applications. 2.5.3 Wall effect Since some of the typical applications require that the multirotor flies close to obstacles, like walls, for example, during the installation of a sensor in a vertical surface, this situation has also been studied. However, the classical assumptions in helicopter theory suggest that the airflow is almost perpendicular to the rotor plane. So, according to this hypothesis, the expected results in the wall effect experiment is that the wall has no influence on the rotor thrust. After accomplishing those experiments, the results of Figure 2.17 were obtained. These clearly show that the wall effect is negligible because there is not a clear effect in rotor thrust when the rotor operates close to the wall. The changes obtained in Figure 2.17 are very light compared to the ground effect which increases the thrust almost 25% and the ceiling effect which increasez approximately 50-100% the rotor thrust. So, we can conclude that the wall effect is negligible. Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 48 Modelling 1 1.5 2 2.5 x/R 0.94 0.96 0.98 1 1.02 1.04 1.06 TIWE/TOWE Wall effect Figure 2.17: Wall effect results. The presence of the wall does not clearly disturb the rotor performance 2.5.4 Other effects Following the structure of this section, a first approach to different aerodynamic effects which can appear in typical application scenarios are shown. However, in this case, the study is not as detailed as the previous one because, they are out of the scope of the main cases of use of the project in which this thesis was developed. Then, these cases of use are studied as a first approach and are treated as a future work in this research field. For example, in the case of the HYFLIERS project [ 34 ], which aim to carry out inspection and maintenance tasks in an industrial environment, like refineries or oil & gas industries, the multirotor should fly between tubes and pipes. So, in order to grant a good accuracy during the operation, it is necessary to model the effect of flying close to this kind of obstacles. Another interesting case to be studied is how this aerodynamic effect can affect the new fully-actuated platforms with tilted propellers. Finally, combined effects, like flying in corners or in fully closed environments are considered interesting for future applications. Pipe effect Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 2.5 Experimental modelling of aerodynamic effects - Results 49 In this case, which will be better analysed in chapter 5, the aerodynamic effect does not only depend on the distance to the obstacle, but it also depends on the relation between the size of the rotor and the tube, and the transverse coordinate in which the rotor is placed. Figure 2.18 shows preliminary results of this effects. Figure 2.18: Ground tube effect (left) and ceiling tube effect (right) experimental results.It clearly shows that in this case, the model does not only depend on the vertical distance but also on the horizontal one. This will be explained in the 2D approach presented in the Chapter 5 Furthermore, it can be observed that this problem does not have revolution symmetry, so the thrust will depend on the position of the propeller with respect to the pipe and it will be as a function of time in Figure 2.19. However, this effect is very fast and it is produced at a high frequency being negligible for robotic manipulation tasks. Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 56 Modelling Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 Chapter 3 Applications for aerial manipulation platforms A journey of a thousand miles begins with a single step. Lao Tzu 3.1 Introduction This chapter analyses the influence of flying close to surfaces on multirotor control. It also presents the partial ground effect which is a special phenomenon that only appears in multirotors when flying close to surfaces and it can be defined as situations in which only some of the rotors (but not all) are under the ground effect. Along this chapter, the control of the partial ground effect is studied in detail because it is the most related one to the practice of aerial manipulation. The chapter uses the previous results of the ground effect in Section 2.5.1 and it includes a comparative study of the results obtained in simulation tests of different control strategies of a quadrotor model under the ground effect in two different scenarios. The content of this chapter is based on the following publication: 57 Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 58 Applications for aerial manipulation platforms • Pedro J. Sanchez-Cuevas, G. Heredia, and A. Ollero, Characterization of the Aerodynamic Ground Effect and Its Influence in Multirotor Control. International Journal of Aerospace Engineering, vol. 2017, 2017. • Pedro J. Sanchez-Cuevas, Guillermo Heredia, and Anibal Ollero. Experimental approach to the aerodynamic effects produced in multirotors flying close to obstacles. In Iberian Robotics Conference, pages 742-752. Springer, 2017. 3.2 Problem Statement Aerial manipulation applications usually require that the multirotor approaches different obstacles to grasp or manipulate something. However, this approaching maneuvers could place the multirotor in asymmetrical conditions with respect to the ground, giving place to a new phenomenon called partial ground effect, which is unique to multirotors, and which comes from the way the rotor thrust forces vary with the distance to the ground. Figure 3.1: Partial ground effect close to the ground. Since the increment in rotor thrust depends on the distance of each rotor to the ground and each rotor has a different distance ( z1 and z3 ) a disturbance moment Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 3.2 Problem Statement 59 Mge2 will be generated that tries to rotate the multirotor opposing the disturbance, and thus, it is a stabilizing moment. In practice, however, it has been found that for typical attitude disturbances of about 5 − 10 ◦ this effect, although stabilising, is negligible unless the multirotor is very close to the ground, and then it is considered not significant for a controller development. This effect is illustrated in Figure 3.1 and Figure 3.2. The first one shows a multirotor hovering in ground effect that experiments an attitude perturbation. Since the increment in the rotor thrust depends on the distance of each rotor to the ground and these distances are different, a disturbance moment Mge2 will arise trying to rotate the multirotor opposing the disturbance, and, thus, this is a stabilising moment. In the second one, the aerial manipulation application, (grasping an object placed over a rectangular obstacle) the multirotor is flying in hover or at low speed in a situation in which only one or several of its rotors (but not all) are under the influence of the ground effect. It can be observed that rotor 3 experiments an increase in the thrust ∆ Tge1 that will generate a disturbing moment Mge1 , which tries to rotate the multirotor counter clockwise, and then moves apart from the object because of the tilting. Figure 3.2: Partial ground effect in a typical aerial manipulation application. Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 60 Applications for aerial manipulation platforms In this case, it is clear to derive that if one of the rotors is under the ground effect, its characterisation can be modeled as described in Section 2.5.1 and using the previously presented PFI model in Section 2.3. However, the cases where three of the rotors or complete multirotors are close to a horizontal surface under the ground effect will be different due to the coupling between the airflow of the rotors and the geometry of the multirotor. In this case, the used multirotor is the same as that which was previously studied in Chapter 2. Then, previous results will be used to take the results of this aerodynamic disturbance in the multirotor controller into consideration. 3.3 Multirotor dynamic model In this case, the dynamic model of the multirotor has been obtained using the Euler- Lagrange formulation [ 91 ] [ 88 ]. This model is basically obtained by considering the multirotor as a rigid body evolving in a 3D space and subject to one force (the total thrust generated by the rotors) and three moments generated by the differences in speed and thrust of the pairs of rotors. The dynamics of the electric motors are relatively fast and, therefore, they will be neglected as well as the flexibility of the blades. The generalised coordinates of the multirotor can be defined as: q= [p η γ]T(3.1) Assuming that the multirotor does not have articulated joints, γ = 0 , we define p = [ x y z ] T to denote the position of the center of mass of the multirotor relative to the inertial frame and η = [ φ θ ψ ] T to represent the three Euler angles (roll, pitch, and yaw) which are the orientation of a reference system attached to the multirotor body with respect to the inertial frame. If m is the total mass of the multirotor and J is the matrix of moments of inertia of the aerial vehicle in the inertial frame, the resultant Euler-Lagrange equations for the translations and rotational degrees of freedom are, respectively, Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 3.3 Multirotor dynamic model 61 m¨p+    0 0 mg     =RFp(3.2) J¨η+C1(η,˙η)˙η=τη(3.3) In 3.3, C1 ( η, ˙η ) is known as the Coriolis terms and it contains the gyroscopic and centrifugal effects; R is the rotation matrix between the body and inertial systems, and Fp = [0 0 T ] T and τη = ( τφ, τθ, τψ ) are the forces and torques generated by the rotors. Both translation and rotational dynamic equations can be expressed in a matrix form as: M(ξ)¨ ξ+C(ξ,˙ ξ)˙ ξ+G(ξ) = F(ξ) + FE(3.4) where F(ξ) are the generalised torques and forces, FE are the unknown forces (wind, turbulence...) and M , C and G are the mass, Coriolis and gravity matrices respectively. Traditionally, the thrust generated by each rotor and the reaction torque due to rotor drag are generally accepted to be proportional to the square of the angular velocity of the rotor ωiwhen flying on free air (Section 1.1.1): Ti=CTΩ2 i(3.5) Qi=CdΩ2 i(3.6) The thrust and drag coefficients CT and CT can be determined by static thrust tests. Then, for a quadrotor in which the distance of the geometric center to the axis of each rotor is d , there is a direct correspondence between the four generalised moments and the angular velocities of the rotors:       T τφ τθ τψ       =      CTCTCTCT 0CTd0−CTd −CTd0CTd0 −CTCT−CTCT       =      Ω2 1 Ω2 2 Ω2 3 Ω2 4       (3.7) Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 62 Applications for aerial manipulation platforms However, when the multirotor is flying close to the ground or other horizontal surfaces the rotor thrust (3.5) and drag model (3.6) are no longer valid. To account for the ground effect and partial ground effects in the simulator the thrust model is modified as: Ti=CTΩ2 ifGE(zr,i) (3.8) where fGE ( zr ) is the ground effect factor that accounts for the increment in thrust due to the ground effect and partial ground effect, which are obtained from the data in Section 2.5.1. This ground effect factor fGE ( zr ) depends on the relative distance of each rotor to the ground, zr . Furthermore, in the simulator it was also considered that the rotor thrust increment varies approximately linearly with the rotor area that is under the ground effect, as it has been reported with experiments with a helicopter rotor [92]. 3.4 Simulation results 3.4.1 Cases of study Two different cases have been tested in order to analyse the influence of the ground effect in multirotor control and to compare the different control approaches: • The multirotor flies over an obstacle (see Figure 3.3.a), beginning in point A and ending in point B. In the reference trajectory the y-coordinate remains constant, and the x-coordinate varies from x=0 (point A) to x=7 (point B). The obstacle is between x=3 and x=5. • multirotor is commanded to go from point A to point B and hover on B, where it has to perform a manipulation operation (see Figure 3.3.b). In point B only one rotor is over the obstacle, and then, under the influence of the partial ground effect. In the reference trajectory, the y-coordinate remains constant, and the x-coordinate varies from x=0 (point A) to x=2.8 (point B). The obstacle is between x=3 and x=5. Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 3.4 Simulation results 63 Three different control alternatives have been analysed in both scenarios: conventional control with cascaded linear controllers, estimated torque controller and control with rotor height estimation (feed-forward). The results obtained using the different control alternatives are shown below. A comparison between these alternatives will be shown at the end of this section. Figure 3.3: (a) Case 1: flying over an obstacle under the influence of the ground effect. (b) Case 2: hovering at a point with only one rotor under the influence of ground effect. 3.4.2 Conventional Control A standard control scheme with cascaded PID linear controllers in each channel has been considered to analyse the response of the system to a conventional controller which does not take the ground effect into account. This control scheme is widely used as a baseline in commercial autopilots, and it is a good reference for the comparison with other methods. The multirotor is moving forward along the x global axis and the yaw angle is maintained constant in the simulations, and so the relevant variables which are presented in the plots as a function of the time are the pitch angle, θ , and the global coordinate in the direction of motion, x. The results obtained in the simulation using the PID controller in the two experiments are shown in Figure 3.4 (Case 1-Flying over obstacle) and Figure 3.5 (Case Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 64 Applications for aerial manipulation platforms 2-Hovering in partial ground effect). In the figures, a dashed line has been included to mark the x-coordinate of the final destination of the multirotor (point B in both cases). Furthermore, the values of the x coordinates that are over the obstacle are highlighted with a green background. Several simulations are presented in each plot for comparison: “OGE”,which is the simulation of the multirotor flying high above the obstacle, so that no ground effect is present; and two simulations flying close to the obstacle (“IGE”) at two different heights normalised with the rotor radius R. Figure 3.4 shows the evolution of the x-coordinate and the pitch angle for the first case flying over an obstacle. It can be observed that the ground effect disturbance produces a torque which affects the system in attitude and position. This torque is more significant when the vehicle is closer to the obstacle. In fact, the results for z/R = 2 show how this torque pushed the vehicle away from the obstacle repeatedly and it cannot fly over it with this controller. In the second case (Figure 3.5), the multirotor tries to hover with only one rotor over the obstacle, but it is unable to do it in the x-reference, and it stays flying in a maintained oscillation going in and out of the ground effect. The simulations presented in Figure 3.4 and Figure 3.5 were performed with the quadrotor flying at a velocity of 0.2 m/s. In order to assess the behavior at different translation velocities, the simulations have been done at a maximum speed of 1 m/s. Figure 3.6.b shows the results of the hovering in a partial ground effect test case, and it can be seen that the behavior is similar to the one presented in Figure 3.4. Figure 3.6.a simulation of case 2, flying over an obstacle. In this case, although there is a strong perturbation in the pitch angle, the inertia of the multirotor moving at 1m/s allows it to surpass the obstacle unlike when flying at 0 . 2 m/s . Since moving at a slow speed is more restrictive and it is also more appropriate for aerial manipulation tasks, the rest of the simulations will be done with the multirotor flying at 0.2m/s. Other simulations have been done considering sensor noise. Figure 3.7 shows the simulation of the quadrotor in case 1, hovering in partial ground effect at a velocity of 1 m/s and z/R = 2. A normal distribution with a standard deviation of 0 . 015 rad/s has been supposed for the noise in the gyroscopes. It can be seen in Figure 3.7 that the presence of noise in the sensors amplifies the oscillations, although the qualitative Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 3.4 Simulation results 65 0 5 10 15 20 25 30 35 40 45 50 time (s) 0 2 4 6 8 x (m) x Obstacle Reference x OGE x IGE (z/R = 3.0) x IGE (z/R = 2.0) 0 5 10 15 20 25 30 35 40 45 50 time (s) -1 0 1 2 3 θ (º) θ θ OGE θ IGE (z/R = 3.0) θ IGE (z/R = 2.0) Figure 3.4: PID controller: Flying over obstacle. Red line shows how the ground effect decreases the speed of the multirotor when it approaches to the obstacle. Green line shows how with a linear controller the multirotor is not able to overcome the obstacle if it is flying very close to it Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 72 Applications for aerial manipulation platforms 15 20 25 30 35 40 time (s) 1.5 2 2.5 3 x (m) x Obstacle Reference x OGE x IGE (z/R = 3.0) x IGE (z/R = 2.0) 15 20 25 30 35 40 time (s) -1 0 1 2 3 θ (º) θ θ OGE θ IGE (z/R = 3.0) θ IGE (z/R = 2.0) Figure 3.13: Feed-forward controller: Hovering in partial ground effect. Feed-forward controller is able to reach the reference and compensate the aerodynamic disturbance even in the closest case (green line) Figure 3.14: Simulation of feed-forward controller using maps with offset errors: left - 5 cm error. right - 15 cm error. It shows how an error in the estimation of the aerodynamic disturbance can lead to the instability. Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 3.4 Simulation results 73 3.4.5 Comparison between different control alternatives Figure 3.15 (Flying over obstacle) and Figure 3.16 (Hovering in partial ground effect) presents a comparison of the simulations between the different multirotor controllers, flying at a relative distance of z/R = 2.0 from the obstacle. From the results in Figure 3.15 and Figure 3.16, it can be seen that the conventional PID-based controller that is implemented in many multirotors does not give acceptable results when flying close to obstacles or the ground, causing oscillations that may lead to dangerous situations, or even preventing the multirotor from reaching its destination because it gets trapped in a cycle at the beginning of the obstacle. With a detailed environment map the feed-forward controller will give better results, provided that the practical implementation issues are correctly addressed. Figure 3.15: Flying over obstacle -Comparison of controllers. Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 74 Applications for aerial manipulation platforms Figure 3.16: Hovering in partial ground effect -Comparison of controllers. 3.5 Experimental Results Several experiments have been conducted to experimentally assess the performance when the multirotor flies over an obstacle. A first set of experiments were conducted in a test-bench in which the multirotor is allowed to rotate freely around one of the axes (pitch angle) (Figure 3.17.a). The ground effect is introduced placing a large plate under one of the rotors at different distances (Figure 3.17.b), while the multirotor pitch angle is controlled by the cascaded PID controller. Figure 3.18 shows the results of the experiment, placing the ground plate at z/R = 2 with both the PQUAD multirotor and the AMUSE aerial manipulator. The evolution of the pitch angle for both cases is presented in Figure 3.18 marking with green background when the ground plate is placed below one of the rotors. It can be seen that when the (partial) ground effect is present, the torque disturbance induces an increase in the pitch angle that is difficult to correct by the PID controller. Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 3.5 Experimental Results 75 Figure 3.17: Experimental tests with multirotors in the testbench: a) the PQUAD, b) the PQUAD with a ground plane Figure 3.18: Experiments in the testbench, influence on the pitch angle of placing a ground plane under one of the rotors (green background). a) Effect in the PQUAD multirotor. b) Effect in the AMUSE multirotor Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 76 Applications for aerial manipulation platforms The next experiments were intended to demonstrate the viability of the control approaches presented in previous sections. A first test was performed implementing the feed-forward control scheme, in which the feed-forward term was activated by a switch in the radio of the safety pilot. It can be seen in Figure 3.19 when the feed-forward controller was not activated (Figure 3.19.a), the induced pitch perturbation is similar to the results in Figure 3.18. However, in Figure 3.19.b the feed-forward controller was tested using the switch in the radio and the controller could maintain the pitch angle stably with much smaller variations. Figure 3.19: Experiments in the testbench, variations of the quadrotor pitch angle when placing a ground plane under one of the rotors (green background): a) with the standard PID controller (not compensating), b) when the compensation is implemented (feed-forward controller) A second test was performed implementing the disturbance observer presented in Figure 3.8. Figure 3.20 shows the time evolution of the pitch angle during the experiment. As in previous experiments, the green background corresponds to the ground plate placed under one rotor. As can be seen in Figure 3.20, after placing the ground plate, the pitch angle begins to increase, behaving similarly to the first experiments in Figure 3.18, but, right after that, the disturbance observer is able to estimate the external wrench and the controller can effectively compensate it. The last step in the experimental tests was to reproduce how the ground effect affects during a normal flight. Figure 3.21 shows two images of the experiment; the Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 3.5 Experimental Results 77 Figure 3.20: Experiments in the testbench, variations of the quadrotor pitch angle when placing a ground plane under one of the rotors (green background) first one in free flight, and the second one when a ground plane is placed under one of the rotors. Figure 3.22 shows the evolution of the multirotor pitch angle and the reference during the experiment, in which the multirotor is being controlled with the standard PID cascaded controller, marking with green background when one of the rotors is under the ground effect (partial ground effect). It can be seen that initially the response is similar to the testbench experiments: the pitch angle suffers a perturbation caused by the torque generated by the partial ground effect (different thrust forces of the same rotors for the same input signal). In this case, there is a perturbation in the opposite direction when the partial ground effect disappears and that it was present in some of the testbench experiments but with smaller amplitude (see Figure 3.18.b and Figure 3.19.a). This can be due to several factors of flying outdoors, for example, wind gusts or turbulence. It can also be explained because the quadrotor is attached to the testbench at two points in opposite arms of the quadrotor, allowing it to rotate around the axis (see Figure 3.17), and the friction may dampen the oscillation. Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 78 Applications for aerial manipulation platforms Figure 3.21: Partial ground effect tests with multirotor flying outdoors: a) multirotor in free flight; b) multirotor flying under partial ground effect (the rotor at the right with a ground plane close to it) 254 256 258 260 262 264 266 268 Time (s) -6 -4 -2 0 2 4 6 θ (º) Obstacle θ (º) Ch2 Figure 3.22: Figure 30 Partial ground effect test with the multirotor in flight Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 3.6 Conclusions 79 3.6 Conclusions In this chapter, the impact of the aerodynamic ground effects on multirotor systems has been shown. The quantification of this effect is very relevant in applications involving flight very close to surfaces as required in aerial manipulation. This chapter has provided a precise characterisation of the ground effect in multirotors in different situations, which has been modelled similar to single rotor helicopters in many cases in literature. The ground effect has been shown to generate a stabilising moment in multirotors due to the differences in height over the ground of the different rotors. Moreover, the so-called partial ground effect, when only one rotor is over the surface, has been studied. The chapter has also shown which control strategies can be used to compensate this effect by simulations and experiments. Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 80 Applications for aerial manipulation platforms Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 Chapter 4 Applications in infrastructure inspection Efficiency is doing things right; effectiveness is doing the right things. Peter Drucker 4.1 Introduction This chapter analyses the influence of flying close to a ceiling surface from the underside during an inspection by contact operation. After presenting the problem, a specific aerodynamic study of the ceiling effect is presented. Then, based on these new results, it is presented and analysed a new concept of aerial platform that can take advantage of this aerodynamic effect while it is inspecting. The modelling and controller sections (Sections 4.5 and 4.6) reformulated the problem and present a nonlinear Lyapunov-based controller to deal with this aerodynamic disturbance. Finally, several experiments which are followed by the conclusions of this chapter are presented in a real scenario. The content of this chapter is based on the following publications: 81 Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 88 Applications in infrastructure inspection the aerodynamic ceiling effect on the quadrotor using Computational Fluid Dynamics (CFD) to optimise the performance of the aircraft during the inspection. These simulations have been carried out using the ANSYS environment. 4.4.1 Meshing Figure 4.4 shows the complete geometry of the proposed aircraft and the simulated mesh. The two symmetry planes are defined in ANSYS Fluent as symmetry boundary conditions. Therefore, the modeled domain is reduced to a portion of 1/4th, improving calculation times without any loss in Maccuracy. Figure 4.4: Geometry of the refined prototype for bridge inspection and mesh for CFD detail Different meshes have been done at different distances between the propeller and the ceiling surface, ranging from 4cm (at this distance, the quadrotor frame touches the ceiling) to 50cm. 4.4.2 Solver settings Firstly, the turbulence model used in this case was the “ k−ε Realizable” which is the most common model used in CFD to simulate mean flow characteristics for turbulent flow conditions. Secondly, the equations employed to solve the problem are the mass, momentum and energy conservation. Thirdly, the fluid is modelled as an ideal gas. Lastly, boundary conditions are defined as follows: the propeller is a wall with no-slip Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 4.4 Aerodynamics based design 89 condition which rotates around its axis. The frame and the top surface of the domain (ceiling) are considered as stationary walls with no-slip condition. The lateral and bottom limits of the area are both defined at constant pressure and temperature of 1 atm and 300K to emulate far-away flow field conditions. Both symmetry planes are established as symmetry boundary conditions. A transient simulation is solved to take the movement of the propeller into consideration. The time step is chosen to be equivalent to a 1/10th of a propeller complete revolution, so it varies between simulations where the angular speed of the propellers changes. For each time step, the software iterates until convergence is reached. This process continues for 10 propeller complete revolutions to ensure stable results. 4.4.3 Methodology Previous results in Section 2.5.2 have shown that the ceiling effect does not only increase the thrust but it also changes the velocity of the rotors (RPM). This is due to the suction that arises when the rotors work very close to the ceiling, which decreases the air pressure around the propellers and the power consumed by the rotors. Most electronic speed controllers work sending PWM signals to the rotors and do not have closed-loop speed control. Therefore, they cannot compensate this effect. As a result, this problem is analysed as follows to properly model the ceiling effect: For a quadrotor in free flight conditions, the relations between thrust T , angular speed Ω, and power consumed P on each rotor have been assumed in the literature as: T(Ω) = CTΩ2(4.1) P(T) = KPT1.5(4.2) In these expressions, CT and KP are constants whose value only depend on the geometry of the propeller (diameter, pitch, airfoil distribution) and the ambient conditions (air density, relative humidity). However, if the platform flies close to an obstacle, these expressions are no longer valid. In the case of use of this work, these Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 90 Applications in infrastructure inspection variables also change with the distance to the ceiling when the aircraft flies close to the bridge. To model this ceiling effect, two assumptions can be made: Assumption 1: CT and KP do not remain constant but are functions of the distance from the propeller to the ceiling z . 2) CT and KP remain independent of the angular speed Ω. Assumption 2: The relations between T,Pand Ω remain unchanged. Both assumption 1 and assumption 2 will be validated using CFD in the next section. Then, the classical approach of Tand Pcan be reformulated as follows: TICE(z, Ω) = CTICE (z)Ω2(4.3) PICE(z, T) = KPICE (z)T1.5(4.4) Therefore, the ceiling effect is modelled through the functions CT ( zi ) and KP ( zi ) and then, it is assumed that the thrust can be modeled as: TICE(z, Ω) = Kc(z)CTOCE Ω2(4.5) 4.4.4 CFD Results Several CFD simulations were performed to validate the assumptions of the previous sections. These simulations consisted of assessing the behavior of a quadrotor in terms of thrust, T, and power, P. More than 30 different simulations for different distances and RPM were performed. The qualitative behavior of the flow field can be explained with the two simulations presented in Figure 4.5 In Figure 4.5.a the quadrotor is touching the ceiling and in Figure 4.5.b it is placed at 10cm of the ceiling. Then, when rotating, the propellers (in black in Figure 4.5) take the upstream air and accelerate it downwards. In the presence of the ceiling, the air movement is restricted, and a low-pressure bubble also appears between the rotor Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 4.4 Aerodynamics based design 91 and the ceiling. Thus, the air particles close to the ceiling and in the proximity of the propeller are attracted to the rotor area and accelerated down-wash. Far away from the propeller, the fluid remains unaltered (deep blue). Figure 4.5: Velocity field when the multirotor is: a) in contact with the ceiling; b) flying 10 cm below the ceiling. Assuming the distance to the ceiling as z = zj and the angular speeds as Ω = Ω k , the equations of the assumptions 1 and 2 become linear when taking logarithms: log Tjk = log CTICE,jk + 2 log Ωk(4.6) log Pjk = log KTICE,jk + 1.5 log Tjk (4.7) Thus, since the CFD simulations provide a complete solution to the flow field of the problem, it is possible to validate these expressions as a consequence the assumption 2. This has been done calculating the values of the exponents in the expressions of assumption 2 by means of the values of T , CT , P and KP provided by the CFD. The results are presented in the table below and validate the relations between T , P and Ω. Assumption 1 is validated through the results presented in Figure 4.6 and Figure 4.7 for different rotation velocities, where it can be clearly seen that the ceiling effect does not depend on the RPM. Last, Figure 4.7 shows the results of Kc ( z ) that appear when the rotors approach the ceiling assuming that the power is constant. This model, which is represented by the yellow dashed line in Figure 4.7, has been compared with previous results with Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 92 Applications in infrastructure inspection 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 z/R 1 1.05 1.1 1.15 1.2 1.25 1.3 1.35 1.4 CTICE /CTOCE Dependency of CTICE /CTOCE with and z/R =5600rpm =5900rpm =6200rpm =6500rpm =6800rpm Figure 4.6: CTvs z/R for different RPMs 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 z/R 0.7 0.75 0.8 0.85 0.9 0.95 1 1.05 KPICE /KPOCE Dependency of KPICE /KPOCE with and z/R =5600rpm =5900rpm =6200rpm =6500rpm =6800rpm Figure 4.7: Kpvs z/R for different RPMs Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 4.4 Aerodynamics based design 93 Model Exp. Mean value Exp. Standard deviation TICE(z, Ω) = CTICE (z)Ω2−Exp2 2.0487 0.0178 PICE(z, T ) = KPICE (z)T2−Exp1.5 1.4796 0.0482 other platforms [ 78 ], showing that the designed platform exploits the ceiling effect in a more effective way than in previous designs. 0.5 1 1.5 2 2.5 3 3.5 4 z/R 1 1.2 1.4 1.6 1.8 2 2.2 2.4 Kc(z) Ceiling Effect CE SR Curve Experimental errorbar - Quadrotor Experimental errorbar - Single rotor CFD Results Figure 4.8: CFD Experimental Results Comparison. Blue error bar and red errorbar are experimental results for quadrotor and multirotor respectively. Black braked line is the model proposed in 2.5.2 for the single rotor in ceiling effect. Yellow braked line is the model used in this work for the aerial platform designed (4.8) For this platform, KCis modeled as: KC(z) = 1 1−1 a1(R a2+z)2(4.8) Both, a1 = 0 . 081 and a2 = 1 . 026 are values obtained through least squares minimising the error with the CFD results. Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 94 Applications in infrastructure inspection 4.5 Dynamic model of the multirotor close to the ceiling The dynamic model of a quadrotor is widely known in bibliography [ 102 ]. Thus, defining a set of generalized coordinates as ξ = [ x y z φ θ ψ ] T , and considering the ceiling effect, the dynamic model of a quadrotor is described by the following equations. ¨ ξ=             0 0 −g ˙ θ˙ ψ(Iy−Iz Ix)−Jr Ix ˙ θΩ ˙ θ˙ ψ(Ix−Iz Iy)−Jr Iy ˙ θΩ ˙ φ˙ θ(Ix−Iy Iz)             +"1 mI30 0αI−1#F(ξ) + Fe(4.9) with α=    d0 0 0d0 0 0 1     ,I=    Ix0 0 0Iy0 0 0 Iz     , I3=    1 0 0 0 1 0 0 0 1     (4.10) Ω = Ω1−Ω2+ Ω3−Ω4(4.11) Here, g is the gravity, d is the distance between the geometric center of the quadrotor and the center of a rotor; Ii ( i = x, y, z ) are diagonal terms of the inertia matrix; m is the aerial vehicle total mass and Jr is the rotor moment of inertia. Vector F ( ξ ) encompasses actuator forces and torques exerted by the rotors. These forces and torques depend on the state of the multirotor due to the aerodynamic effects. Lastly, Fext includes external forces and torques which are not known or modelled, such as contact forces or wind disturbances. The mass matrix and the gravity vector are the same that have been widely presented in literature, but, in this case, and due to the flight conditions, the force vector acting on the vehicle is: Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 4.6 Control 95 F(ξ) =             (cφsθcψ+sφsψ)T(ξ) (cφsθsψ+sφcψ)T(ξ) cφcθT(ξ) τφ(ξ)) τθ(ξ) τψ(ξ)             (4.12) where cα = cos α , sα = sin ( α ) and ( φ, θ, ψ ) are the Euler angles. Defining the total thrust T ( ξ ), as T ( ξ ) = P4 i=1 Ti ( ξ ), and taking it into account that the platform is a quadrotor in cross configuration, there is a direct correspondence between the four generalised moments and the angular velocities of the rotors in which the vector F ( ξ ) follows 4.12, where the aerodynamic ceiling effect has been included. F(ξ) =       T τφ τθ τψ       =      Kc(zc1)cTKc(zc2)cTKc(zc3)cTKc(zc4)cT −Kc(zc1)dcTKc(zc2)dcTKc(zc3)dcT−Kc(zc4)dcT Kc(zc1)dcT−Kc(zc2)dcTKc(zc3)dcT−Kc(zc4)dcT cDcD−cD−cD             Ω2 1 Ω2 2 Ω2 3 Ω2 4       (4.13) where z(ci) is the distance from the i-rotor to the ceiling, defined as follows: zc1=zc−√2 2d(sin(θ)−cos(θ) sin(φ)) zc2=zc−√2 2d(−sin(θ) + cos(θ) sin(φ)) zc3=zc−√2 2d(sin(θ)−cos(θ) sin(φ)) zc2=zc−√2 2d(−sin(θ) + cos(θ) sin(φ)) (4.14) where zc is the distance from the center of gravity of the multirotor to the ceiling. 4.6 Control In this section, we present a nonlinear Lyapunov-based attitude, altitude and position controllers for bridge inspection flights. The main goal of the designed controller Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 96 Applications in infrastructure inspection is that the quadrotor can fly very close to the beams and carry out contact tasks with safety and accuracy. The ceiling effect is the most critical aerodynamic effect occurring during these inspection tasks. In the previous section, the ceiling effect has been modelled and introduced in the equations of the motion system. The relationship between the control inputs and the angular velocities of the rotors have been shown in equation 4.13. In this expression, the zci of each rotor has been calculated to obtain the ceiling effect for each one. Depending on the multirotor attitude, the ceiling effect is different in each rotor. As Figure 4.9 shows, if the aircraft has a pitch angle about ± 10 degrees (flight forward), the aerodynamic ceiling effect is approximately a 22% higher in one rotor than in the other when zc = 0 . 1 m . However, it decreases to 5% when the attitude angles are about ± 3 degrees. Furthermore, it decreases to 2% when zc = 0 . 25 m . This result implies that it is possible to make the following assumptions to increase the safety during the operation: 3.1: A forward flight will be done to a minimum distance of 0.25 m. to the ceiling surface. 3.2 The only operation allowed for distances less than 0.25 m will be the hover until reaching the contact with the surface. Taking into account the assumptions 3.1 and 3.2, expression (4.8) is linearised in intervals in which the difference between the aerodynamic effects is less than 5% and, therefore zc∼zc1∼zc2∼zc3∼zc4. Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 4.6 Control 97 Figure 4.9: Response of a multirotor in ceiling effect and safety flight envelope. Red area represents the very close to the bridge area in which flight forwards commands are not allowed. In yellow area, the ceiling effect gradient will be lower than a 5% between the rotors to ensure the stability of the platform. Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 104 Applications in infrastructure inspection Figure 4.12: Navas bridge and experiment overview Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 4.7 Experimental Results 105 4.7.3 Manual Flight Results The first experiments, which were carried out piloted by a human, show that the aircraft is very unstable at the end of the contact phase. The results in Figure 4.13 show that the pilot must significantly decrease the thrust signal, at least, up to a 40% less than the thrust hover to finish the contact task. This situation leads to conditions in which the aircraft critically loses its altitude and it could generate a dangerous flight condition. Then, it becomes it clear the need of implementing an automatic control strategy to fix this behavior. This is one of the main conclusions of this work. Figure 4.13: Altitude telemetry results during in manual flight. Red zone represents the dangerous area in which the aircraft critically loses its altitude and it could generate a dangerous flight condition Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 106 Applications in infrastructure inspection 4.7.4 Nonlinear Controller Results The next experiment was accomplished with the automatic control strategy proposed in Section 4.6. In this case, the results of Figure 4.14 and Figure 4.15 show the altitude controller result, as well as the values of the thrust, the aerodynamic ceiling effect estimation and the vertical velocity. These results reveal that the controller implemented allows the platform to fly in safe conditions when it approaches and moves away from the bridge surface. Moreover, Figure 4.14 also shows how the controller acts during the contact and the commanded thrust. It shows that, in spite of the oscillations due to the error in the state estimation, the mean value of thrust (represented with the red dashed line) decreases as a consequence of taking the ceiling effect in the nonlinear controller into consideration. It is important to remark that the most important benefits of this nonlinear control appear in the z-position and z-velocity controller. The result in Figure 4.15 shows the z-velocity controller results, in which, it can be observed that the nonlinear controller is faster and the unstick command is followed in a more efficient way. For instance, this result clearly shows that the nonlinear controller easily follows the reference with a small delay (0.25 s), while a standard PID controller needs to integrate and accumulate too much error for following the reference (4 s) because this one does not take the aerodynamic ceiling effect into account. On the other hand, Figure 4.15 also presents the performance of the z-position controller. These results show that the linear controller is not only slower, but it is also less safe due to the overshooting presented in the dangerous zone (red background) which does not appear with the proposed nonlinear controller. Moreover, the reason why both controllers show a small delay is due to the limits of the sensors used to estimate the position of the vehicle. The robotic total station provides a very high positioning accuracy ( ∼ 1 cm), but it has a delay and its updated rate is temporal changing and relatively low, limiting the performance of the nonlinear controller and causing a small delay in the response of the multirotor. The next section shows a complete telemetry set of the multirotor flying in an inspection of the bridge, where it is possible to assess the result of the controller along the flight. Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 4.7 Experimental Results 107 Figure 4.14: Altitude nonlinear controller results. Altitude vs altitude reference control, thrust and mean value during the flight and contact (red dashed), and estimation of the ceiling effect factor (Kc) during the flight Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 108 Applications in infrastructure inspection Figure 4.15: PID vs Nonlinear z-velocity and z-position controller comparison. Green backgrounds represent the contact condition; red background is the dangerous zone where it can be observed an overshooting. It is shown that the nonlinear controller is more than 4 times faster moving away from the ceiling. Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 4.7 Experimental Results 109 4.7.5 Autonomous Inspection The last experiments consisted of accomplishing a full inspection by contact in an autonomous mode. These results, which are presented in Figure 4.16 and Figure 4.17, show that the system is able to navigate between different contact points in safety conditions. In Figure 4.16, it can be seen that the proposed z-position controller allows the aircraft to move away from the ceiling without entering dangerous flight conditions or losing their altitude critically, as it happens in Figure 4.13 (see red area in Figure 4.16). Figure 4.16: Position controller results during an autonomous mission Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 110 Applications in infrastructure inspection Figure 4.17: Attitude controller results during an autonomous mission Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 4.8 Conclusions 111 4.8 Conclusions This chapter has presented the design, modeling and control of a multirotor which is able to fly very close to the ceiling surface and even establishes a full contact with it for carrying out the inspection of a bridge by contact. The design of the aerial platform was based on a detailed aerodynamic study of the ceiling effect performed with CFD. Thus, the aircraft can exploit the ceiling effect in its favour and do the inspection operation in a more effective way. This ceiling effect was also modeled through the results of the CFD and it was included in the dynamic model of the quadrotor reformulating the classical approach of these equations. Moreover, the controller scheme used to accomplish the contact and to fly close to the surface of the bridge has been described and implemented in a real platform. Lastly, the outdoor experiments in a real bridge have validated the proposed design, as well as the controller and the aerodynamic characterisation, even during a fully autonomous mission. The future work related to this research will be focused on the end user of this technology and the inspection application. The aim of these works will be: increase the applicability of this technology, increase its technology readiness level (TRL), and, turn it into a competitive option in the commercial exploitation Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 112 Applications in infrastructure inspection Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 Chapter 5 2D Aerodynamic Modelling: From the 1D aerodynamic effect to the 2D aerodynamic maps Una de las virtudes de un verdadero ingeniero es la eficiencia. Guang Tse 5.1 Introduction One of the most disruptive trends in UAV applications is the use of the previously presented aerial manipulator in tasks , such as contact inspection and sensor installation in inspection and maintenance (I&M) of infrastructures or industrial plants [ 103 ] [ 77 ]. These I&M tasks usually require that the aerial platform flies very close to different obstacles. For instance, the PILOTING [ 37 ] and HYFLIERS [ 34 ] projects are focused on aerial manipulation for outdoor applications in I&M in oil and gas plants. Both cases imply that the aerial platforms need to fly in the proximity or even maintain a contact with an infrastructure to carry out the I&M operations. Moreover, they need to do it without lacking accuracy or safety conditions during the operation. In those situations, which involve an UAV flying close to different structures, surfaces, or 113 Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 120 2D Aerodynamic Modelling 5.4.1 Assumptions This section is focused on defining the limits of the flying area and on establishing the assumptions when the rotor is working close to different obstacles. These assumptions are established in terms of defining the flyable area during the operation and of working out how to solve the problem in the points which are under the influence of more than one obstacle. Flyable area Figure 5.4 shows the flyable areas close to a flat/rectangular obstacle and a tubular one. This area establishes the limits of the map and envelopes the operation area of the aerial platform. Figure 5.4: Flyable area detail and forbidden zones. Note that, in both cases we have symmetrical transition effects. The map around these obstacles will be modelled as it is shown in Figure 5.5. The flat surface where the value of the non-dimensional thrust is 0 . 95 is out of the flying area but it has been defined with this value to avoid infinite elements in the plot. Obstacle overlapping The assumption to model the aerodynamic effect in the points of the map which are affected by more than one obstacle is that it is possible to apply the principle of superposition. However, it is assumed that the superposition method is no longer valid if one obstacle is in the shadow of another one. This is clearly explained in Figure 5.6, where the red area shows the influence area of the Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 5.4 Aerodynamic Mapping 121 Figure 5.5: Sample of aerodynamic effect close to a rectangular obstacle (left) and a pipe (right) obstacle (a) and the blue area shows the influence of the obstacle (b). In the section in purple, it is where the principle of superposition is applicable and the grey zone is a shadow area and it shows an area in which the aerodynamic effect of the obstacle (a) is considered blocked by the presence of the obstacle (b). Figure 5.6: Obstacle overlapping and shadows conditions. Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 122 2D Aerodynamic Modelling 5.4.2 Results Lastly, this section presents the results of different aerodynamic maps in which the changes of the thrust due to the influence of the obstacles are represented. Figures 5.7, 5.8 and 5.9 show the results in three different scenarios which can be used later to design different planning or control techniques. Figure 5.7: Aerodynamic effect map with rectangular obstacles Figure 5.8: Aerodynamic effect map with tubular obstacles These results evidence that this kind of aerodynamic map can open several research lines in which the aerodynamic disturbance does not take into account as a punctual problem but a environmental one. In this context, aerodynamic awareness planners or controllers could be used to navigate in a complex and narrow environment with Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 5.5 Conclusions 123 Figure 5.9: Aerodynamic effect map with rectangular and tubular obstacles several obstacles. For instance, in some cases, it could be convenient to move along a constrained environment if this help the multirotor to lift their weight thanks to the aerodynamic effects. That is similar to the case of the bridge presented in Chapter 4, where we use the ceiling to stabilise the platform and increase their maximum flight time, but in a general way in which we can navigate through the environment to exploit the aerodynamic effect in our benefit. 5.5 Conclusions This chapter has presented a new approach to the modelling of the aerodynamic effects that can arise during the operation of an UAV flying close to obstacles in oil and gas plants. This approach consists of creating an aerodynamic effects map which links the relative position of the vehicle with the aerodynamic effect that the environment produces in this point. The different aerodynamic effects have been independently studied and the different assumptions about the areas under the influence of more than one obstacle are also presented. The future work related to this research will be focused on the application of this map to design different control techniques or planning methods which take them into Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 124 2D Aerodynamic Modelling account the aerodynamic effects on the aerial vehicle to improve its behavior or to optimise the use of resources, like the power or time consumption during the operation. This kind of map has a direct application in the direction of the actual application of the UAV in the Inspection and Maintenance of oil and gas industries. Actually, the appendix C presents results of collaboration work with ´ Alvaro Caballero, (PhD Student of the University of Seville) in which this kind of aerodynamic maps is included in an aerodynamic awareness planning method [84]. Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 Chapter 6 Conclusions and Future Works Try not to become a man of success. Rather become a man of value. Albert Einstein 6.1 Conclusion Aerodynamic effects take part in most of the applications of aerial manipulation which are ranging from grasping an object to inspect a surface through establishing a physical interaction. Those effects are higher as the robot is closer to different elements of the environment, where, the needs of maintaining a high accuracy and safe conditions are essential. Those reasons motivated the work developed in this thesis. Throughout it, several studies about different aerodynamic effects have been presented, actually, different solutions to deal with it while flying have been proposed. At the beginning of this work, the background about this aerodynamic phenomenon in aerial manipulator was not widely extended (Section 1.3). In this way, this research started to look into a field with small background compared to other topics that have been widely studied by the research community. The published contributions have helped then to improve the knowledge about this topic. To sum up, this thesis has presented an analysis of the aerodynamic effects that can arise in typical aerial 125 Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 126 Conclusions and Future Works manipulation applications. In addition, it has thoroughly modelled the aerodynamic effect that appears in two cases of use in the framework of two European Projects, AEROARMS [ 73 ] and AEROBI [ 33 ]. In both cases, different control strategies have been presented and tested in real flights. After that, and following the research lines of new European Projects, like HYFLIERS, RESIST, AERIAL-CORE and PILOTING, the author intends to present a paradigm shift in terms of how to model those aerodynamic effects, changing the perspective of modelling those effects to a complexer one produced by the environment. In this way, it is possible now to generate aerodynamic maps which can be exploited in the controller or the planning (Appendix C) strategies to fly in a more efficient and safer way. Despite of the fact that the document has been specifically focused on the study of these aerodynamic effects, in this thesis, the author wants to remark the experimental part and the efforts dedicated to it. To sum up, this thesis will concluded with the following key aspects: • Aerodynamic effects affect significantly in those operations that require the robot flies very close to different obstacles. • Although it is possible to develop a general model that approximate different aerodynamic effects, it is extremely difficult to solve the complete problem because the specific geometry of the robot takes part in the aerodynamic effect. That means that this kind of disturbances will affect each platform in a different way. • Those effects can be use to improve the performance of an specific operation as Chapter 4 have shown with the ceiling effect. • The knowledge of the aerodynamic effect could help to exploit aerial robot in difficult to access and small environments. • This thesis has helped and contributed to advance and improve the knowledge about the aerodynamic effect in aerial manipulators. Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 6.2 Future Works 127 6.2 Future Works 6.2.1 Complete aerodynamic awareness while operating In the future, the main goal of the aerial manipulator working close to different obstacles will be the complete aerodynamic awareness while operating. For instance, a clear sample that has received less attention is the aerodynamic effects that can appear in the robotic arm by the presence of airflow developed by the propellers. However, this effect is not negligible, especially if the arm is compliant and is able to admit certain deformations while operating. Our experience in aerial manipulation applications pushed us to look into this aerodynamic effect as the disturbance does not only arise in the aerial platform, but it also distorts the robotic arm. The first effect is caused by the arm which acts like an obstacle and obstructs the development of the propeller wake changing the aerodynamic flow field and, consequently, the force and torques developed by the rotor. The second one is due to the forces that the fluid exerts against the robotic arm ,which can produce two different effects depending on the characteristics of the arm joints and the geometrical design design of it. In rigid joints which do not protect the servo actuators against impacts and overloads, the drag that appears due to the movement of the propeller could force the servo beyond its range causing a complete failure of the joint. This is one of the most important drawbacks of using rigid joints in aerial manipulation. However, in terms of the aerodynamic effect, a compliant joint is also affected in an important way. Thus, although the flexible link partially protects the actuator and solves this problem, the aerodynamic forces will deflect these joints causing a deformation and losing accuracy in the manipulation. In addition to this, it is important to remark that the geometrical design of the arm is relevant from the aerodynamic point of view. It is obvious that if the arm acts like a blunt body, the drag that it is going to support is much greater than being a streamlined body. Figure 6.1 illustrates this effect from an schematic point of view. FAarm and TAarm are the forces and torques that the arm is suffering due to the presence of the airflow impacting on it. FArotor and TArotor are the forces and torques that appears in the Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 128 Conclusions and Future Works Figure 6.1: Aerodynamic effect in the robotic arm rotors die to the disturbance produced by the arm. These forces and torques are usually considered negligible. However, the forces that act in the manipulator produce two important effects that must be taken into account to be able to carry out aerial manipulation operations with enough accuracy. Firstly, the position of the arm is going to depend on the airflow exerted by the rotors which ,in turn ,depend on the rotational velocity of the rotors ω , FAarm = f ( ω ) and MAarm = f ( ω ). Secondly, the force suffered by the arm will be transmitted to the multirotor, since they are in a common body, producing an undesirable disturbance. This effect as well as those presented above are a challenge in aerial manipulation that must be deeply studied to be able to develop robust and reliable aerial manipulators which can be included in the industries of the future. 6.2.2 Aerodynamic effect in closed spaces Several applications will require using multirotor in confined spaces, for instance, for inspecting vessels or tunnels among others. In those environment the aerodynamic effect could have an important impact in order to guarantee the stability of the robot while flying. Aerodynamic effects, then, are expected to be higher in those kind of scenarios where all the effect will be superimposed at the same time and additional aerodynamic effects due to the re-circulation, could accentuate the impact in the dynamic of the robot (see Figure 6.2). Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 6.2 Future Works 129 Figure 6.2: Aerodynamic effect inside of a tunnel Código seguro de Verificación : GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 | Puede verificar la integridad de este documento en la siguiente dirección : https://sede.administracionespublicas.gob.es/valida ÁMBITOPREFIJO CSV FECHA Y HORA DEL DOCUMENTO GEISER GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8 23/10/2020 11:12:36 Horario peninsular Nº registro DIRECCIÓN DE VALIDACIÓN O00008745e2000052783 https://sede.administracionespublicas.gob.es/valida GEISER-e6ae-26a5-3f4f-4e45-9b01-5e4a-e2e7-5ba8