Distributed Thermal Identification and Exploitation for Multiple Soaring UAVs

Depósito de Investigación de la Universidad de Sevilla https://idus.us.es/ This version of the book chapter has been accepted for publication, after peer review and is subject to Springer Nature’s AM terms of use, but is not the Version of Record and does not reflect post-acceptance improvements, or any corrections. The Version of Record is available online at: https://doi.org/10.1007/978-3-319-10807-0_17 Cobano, J.A., Alejo, D., Vera, S., Heredia, G., Sukkarieh, S., Ollero, A. (2014). Distributed Thermal Identification and Exploitation for Multiple Soaring UAVs. In: Spagnolo, P., Mazzeo, P., Distante, C. (eds) Human Behavior Understanding in Networked Sensing. Springer, Cham. https://doi.org/10.1007/978-3-319-10807- 0_17 Distributed thermal identification and exploitation for multple soaring UAVs J.A. Cobano, D. Alejo, S. Vera, G. Heredia, S. Sukkarieh and A. Ollero Abstract This chapter discusses the problem of cooperatively sensing the wind map of an area in order to efficiently harvest the wind energy with fixed-wing gliding UAVs, known as soaring. Moreover, a cooperative system with multiple gliding fixed-wing UAVs is presented for long endurance missions. This system is composed by three main blocks that include thermal detector, path planning and collision avoidance. The main advantage is its low computational load, making it suitable for real time applications. Extensive simulation results in several scenarios are given to test the complete system. In addition, real experiments have been carried out with real gliding aircrafts of the Robotics Vision and Control Group (GRVC) of the University of Sevilla in the Airfield of La Cartuja (Sevilla). The results of these experiments show the convenience of the proposed method. 1 Introduction The interest in using multiple Unmanned Aerial Vehicles (UAVs) has been increasing in recent years taking into account potential applications in different missions such as exploration, data collection, mapping, surveillance, coast control, fire detection, monitoring and many more. In all these cooperative missions, the UAVs should visit different places during their flight in order to explore the area. The common problem in this kind of missions is the short endurance of many vehicles. They have to land in order to refuel or recharge batteries. Extending the flight duration of multiple UAVs cooperating for long endurance missions arises as a critical issue in many applications. J.A. Cobano, D. Alejo, S. Vera, G. Heredia and A. Ollero University of Seville, Camino de los Descubrimientos s/n Seville (Spain), e-mail: [jcobano,dalejo,svera,guiller,aollero]@us.es, S. Sukkarieh The University of Sydney, Sydney NSW 2006 (Australia), e-mail: [email protected] 1 2 J.A. Cobano, D. Alejo, S. Vera, G. Heredia, S. Sukkarieh and A. Ollero New ideas such as the so-called autonomous soaring have been proposed to extend the flight endurance of a single glider aerial vehicle [1]. Soaring could be defined as flight in which a propulsion system is not used and favourable wind conditions are exploited to extend flight duration. This phenomenon was noticed in some birds by observing how they are able to fly without flapping their wings [2]. The clearest example is the Wandering Albatross that covers large areas with minimal energy consumption [3]. In 1885, I. Lancaster [4] published a work on soaring of birds. Thus, soaring flight became an important research area. There are two types of soaring, static and dynamic soaring. The first one uses rising air to gain energy and the second one uses wind gradients or distributions. Aerial vehicles should be capable of extracting energy from the atmosphere to gain altitude in order to stay aloft [5]. This energy can be extracted from different sources such as wind gusts over surfaces (such as the ocean), shear generated by flow around geographic obstacles, and meteorological shear from temperature inversions. However this work will consider energy that comes from static soaring. Static soaring depends on vertical movements of the atmosphere, also so-called thermals. Thermals are caused by convection in the lower atmosphere and could be exploited to increase the altitude of multiple UAVs. This chapter addresses long endurance cooperative missions with multiple glider aerial vehicles. The mission is given by a set of places defined by Point of Interest (PoIs) which should be visited. Existence of thermals in the environment is considered. Thus, the aerial vehicles should cooperatively visit the PoIs and each aerial vehicle should detect and identify the thermals present in the environment during the mission in order to exploit their energy and gain altitude, and so extend the flight duration. Cooperative missions with multiple vehicles allow for faster detection and more efficient exploitation of the thermals, since each vehicle transmits to the rest of the teams the location of the thermals it has identified. The guidance and control of an autonomous soaring UAV is not addressed in this work, but we use the approach in [6]. A path planner is also needed to efficiently carry out the cooperative mission. The planner has to consider constraints such as the energy available and localization of the thermals which will influence the computation of collisionfree trajectories. Moreover, collision avoidance is a critically important aspect in applications with multiple UAVs to successfully perform the mission. Therefore, a collision avoidance block should be implemented to ensure the fulfillment of the mission. In summary, the objectives of the chapter are: 1. Explore all the PoIs without landing and decrease the total time of the mission. 2. Identify the presence of thermals in the environment in order to exploit them and extend the flight duration. 3. Compute safe trajectories to perform the mission. A new system made up different blocks is developed in order to meet the objectives. The Path Planner block considers a method to assign each PoI or Thermal Point (TP) to a vehicle. A Thermal Detector block is added to detect and identify thermals during the mission. The Conflict Detection and Resolution block is based on the RRT* (Optimal Rapidly-exploring Random Trees) planning algorithm. Stud- Distributed thermal identification and exploitation for multple soaring UAVs 3 ies with many simulations show the performance and advantages of the developed system. Experiments have been carried out in the airfield of La Cartuja (Seville, Spain) with the gliding fixed-wing UAV shown in Figure 1 in order to show the reliability of the system. Fig. 1 Gliding fixed-wing UAV used in the experiments. The chapter is organized into nine sections. The state of the art is presented in Section 2. The developed system is described in Section 3. Section 4 presents the thermal model, the features of the environment considered and the algorithm to detect and identify the thermals. The path planner is explained in Section 6 and the conflict detection and resolution algorithm is described in Section 6. The simulations and experiments performed are shown in Sections 7 and 8, respectively. Finally the conclusions are detailed in Section 9. 2 State of the art First studies on soaring were based on the flight patterns of birds [2]. Soaring UAVs capable of extracting energy from the atmosphere to gain altitude in order to stay aloft have been presented in [5]. The works presented in [7][1] and [8] show the first analysis on autonomous thermal soaring. Other results with an autonomous soaring controller are reported in [9]. The static soaring problem is applied to the Vehicle Routing Problem with Time Windows (VRPTW) in [10]. It develops an exact solution method including preprocessing, route optimization and route validation . The total flight time minimization is achieved and a considerable increase in the level of autonomy of a soaring UAV is attained. 4 J.A. Cobano, D. Alejo, S. Vera, G. Heredia, S. Sukkarieh and A. Ollero Detection and identification of thermals have also been addressed in the literature [11][12]. Models of thermals should be considered in the studies and several of them are presented in [6] and [13]. Energy-constrainted motion planning is done by considering the problem of a gliding UAV searching for a ground target while simultaneously collecting energy from known thermal energy sources [14]. Path planning is also addressed in other studies. A graph-based method for planning energy-efficient trajectories over a set of waypoints is presented in [15]. A method to generate a spatio-temporal map of the wind and a path planning to generate energy-gain paths based only on local observations of the wind is presented in [16]. The trajectory generation for autonomous soaring is also addressed in [17] and[18]. The coordination of multiple UAVs in order to perform a long endurance mission considering the detection of thermals and the computation of collision-free trajectories has been studied in [13]. UAVs communicate to each other the location of potential thermals in the area but the simulation shown only considers one UAV. [19] describes a new algorithm for maximizing the flight duration of a group of UAVs using thermals located in the area. A simultaneous perturbation stochastic approximation method (SPSA) is used to detect the center of the thermal and the method treats the thermal center drift effectively. It is assumed that there exists a path planning algorithm that keeps the vehicles from colliding with each other. This work considers small unmanned powered glider, so the propulsion system or soaring can be used. [20] presents a method for distributed mapping of the wind field. The map is discretized and a Kalman filter is used to estimate the vertical wind speed and associated covariance in each cell. Flocks of small UAVs are considered to maximize the endurance. [21] investigates the possible benefits of using a cooperating team of small UAVs to increase the probability of finding thermal lift. A collision-free trajectory planning algorithm is not implemented to optimize the search of thermals in [20] and [21]. These works and [10] consider lifetime and drift of the thermals. Moreover, The proposed system allows applications in real time because of its low computational needs. 3 Overview of the system This section describes the proposed system for long endurance missions with multiple gliding fixed-wing UAVs. It assigns the PoI to the UAVs and computes the collision-free trajectory for each UAV if a collision is detected. During the flight, each UAV should detect and identify unknown thermals. Figure 2 shows the block diagram: Distributed thermal identification and exploitation for multple soaring UAVs 5 Fig. 2 Block diagram of the system. 3.1 Local Path Planner The Local Path Planner block (PP) is responsible for generating the flight plan of the UAV taking into account the knowledge of the wind map, i.e. detected thermals so far, and the PoIs to be visited. In order to do this, it should communicate with the Thermal Manager block and the Mission Manager block. This block uses a path planning algorithm called Bounded Recursive Heuristic Search method (BRHS) [22] which is based on a Depth-First Search algorithm (DFS). The basic behaviour of this block is the following: it periodically generates a new flight plan taking into account the current location of the UAV (from the Autopilot), position of the remaining PoIs and Thermal Points (TPs), and minimum flying altitude of the UAV. The flight plan is computed every second and the execution time is below one millisecond. When a new flight plan is generated and is significantly different of the current flight plan, PP transmits the new flight plan to the Autopilot. The generated flight plan can be defined by one of the following four alternatives: 1. One waypoint: If the altitude of the UAV goes below a minimum flying altitude, the UAV is commanded to go Home for landing. 2. Two waypoints: Current location and one PoI. UAV does not need a thermal to gain altitude. 3. Three waypoints: Current location, entry point and exit point of the thermal. The UAV can not reach a PoI and should access to a thermal first. 6 J.A. Cobano, D. Alejo, S. Vera, G. Heredia, S. Sukkarieh and A. Ollero 4. Four waypoints: Current location, entry point and exit point of the thermal, and a PoI. UAV could visit a PoI after gaining altitude in a thermal. 3.2 Autopilot Each UAV should be eqquiped with an Autopilot to be capable of following 3D flight plans. It is also responsible for estimating the state of the UAV and of providing the other blocks with this information. In the experimental platforms currently developed in the Robotics, Vision and Control Group of the University of Seville we have installed an Ardupilot Mega 2.5 of the company 3DRobotics [23]. This is an open-source autopilot that is easily configured and gives good performance. The experimental setup is shown in Figure 3. Fig. 3 Experimental setup of one of the gliding fixed-wing aircraft UAV. Distributed thermal identification and exploitation for multple soaring UAVs 7 3.3 Thermal Detector The thermal detector (TD) is responsible for detecting new thermals in the environment from changes of energy of the UAV. It is constantly monitoring the UAV state in order to check for unexpected ascensions. Each UAV has on-board its own TD. Whenever an unexpected ascension occurs and it meets some requirements, a new thermal is added to the system and new points are added and sent to the Mission Manager in order to actively sense the characteristics of the thermal. The details of these procedures are given in section 4. The characteristics of the new thermal are also sent to the Thermal Manager block. 3.4 Mission Manager The Mission Manager (MM) block stores the list of remaining PoI and TPs to be visited by the UAVs. It is also responsible for assigning and reassigning these waypoints to the UAVs as requested by the PP modules. In the proposed system, two types of points to visit are distinguished: PoIs and TPs. Both could be considered as an exploration process: PoI to explore some places, and TPs to explore potential thermals. PoIs are set by the operator and TPs are generated when the location of a potential thermal is received from TD block. The TPs are computed to provide a better estimation of the center of a detected thermal. Initially the list shows all the PoI defined in the environment. The list is updated every time one PoI is visited or a thermal is detected. The Local Path Planner (PP) of each UAV proposes visiting a PoI or TP when generating its flight plan. MM should assign to a UAV a PoI or TP if its estimated time of arrival (ETA) to it is the lowest one so far. In order to prevent oscillatory behaviours, whenever an UAViproposes visiting a PoIkor TPkthat has been already assigned to a UAVj, the ETA of UAVi not only should to be lower than the ETA of UAVjbut also should decrease this time with a given margin, tvisit . Otherwise PoIkor TPkcontinues being assigned to UAVj. 3.5 Thermal Manager The Thermal Manager (TM) block stores the information on the thermals and manages the access to them. It communicates to each Local Path Planner the existing thermals in the space and the temporal constraints to access a thermal. Whenever an UAV needs to gain energy, it should request access to any thermal to the TM block. This block checks whether the flight plan proposed by each UAV to gain altitude is safe or not. A flight plan to access to a thermal is safe when the vertical separation between two UAVs within the thermal is larger than a safety margin, dsa f ety. The outputs are temporal constraints to access to the thermal. If 8 J.A. Cobano, D. Alejo, S. Vera, G. Heredia, S. Sukkarieh and A. Ollero a flight plan is not safe, TM sends the temporal constraints to meet the vertical separation to the corresponding UAV. 3.6 Collision Detection and Resolution block The Collision Detection and Resolution block (CDR) is responsible for ensuring collision-free trajectories between UAVs in the system outside the thermals. Note that the Thermal Manager block arbitres the access of UAVs to the thermals, so the collisions inside thermals should not occur. This module can be divided into two different blocks: the detection and the resolution blocks. The first block, takes as inputs the state of the UAVs in the systems and their current flight plans in order to detect conflicts between their trajectories. The second block is activated whenever a conflict is detected and will modify the flight plans of involved UAVs in order to prevent potential collisions. 4 Thermal detector This section describes how the wind map of the environment is generated and the potential thermals are detected. The parameters that define a thermal are: center of the thermal (C), vertical wind velocity (w), radius (R), maximum altitude (A) and drift of the thermal, (Vdri f t ) . The environment is divided into uniform two-dimensional cells. A value of the vertical wind velocity is assigned in each cell. Each UAV will estimate the vertical wind velocity of the center of the thermal from the changes of energy. The UAV speed is constant, so only changes of altitude are taken into account. 4.1 Thermal model The model of the thermal used is based on the model presented in [11]. Some modifications are performed to compute a more realistic initial wind map: • A spatially uncorrelated zero-mean Gaussian noise is added to the wind field. • The lifetime of the thermal is considered. Thus, the vertical wind velocity distribution that defines the thermal decreases with respect to time. • The drift of the thermal is considered. Thus, center of the thermal is in relative movement to the ground. In this work, a test set has been generated to validate the proposed system. An algorithm to randomly generate the wind maps has been implemented. The inputs are: number of thermals at the start, lifetime of each thermal, drift of each thermal, zero- Distributed thermal identification and exploitation for multple soaring UAVs 15 Fig. 6 UAV trajectories computed from the RRT* planning algorithm to solve a collision (instant tc): solid line is the solution trajectory and dashed line is the initial trajectory. Next a study to analyze the behaviour of the system with several cooperative UAVs is presented. A wind map generated randomly is considered. Initially, twelve thermals are created (thermals 1-12). Five thermals more are created during the mission in different times (thermals N1-N5) (see Figure 7). Each UAV does not have any information about their location or strength, so the thermals are unknown to the system. Ten simulations are performed in each case by changing the initial position of each UAV. Table 2 shows the results obtained by considering different number of UAVs. The time of the mission and the number of thermals detected is shown. Figure 7 presents one of the simulations to show how the exploration is carried out with three UAVs. It is important to highlight the main advantages of the proposed system. It ensures the safety of the system. Each UAVs takes into account the drift estimated to exploit the thermals detected. The mission time is reduced by using cooperative UAVs and more thermal can be detected. Finally, the flight duration is extended to carry out the mission while thermals exist in the environment. 16 J.A. Cobano, D. Alejo, S. Vera, G. Heredia, S. Sukkarieh and A. Ollero Table 2 Detection and identification of thermals considering ten simulations. UAVs Elapsed Time (s) Thermals detected 1 1345.31 ±57.20 4.63 ±1.51 2 759.92 ±42.02 4.81 ±1.68 3 498.844 ±36.01 5.75 ±1.92 4 466.24 ±44.03 6.89 ±0.84 5 344.25 ±31.63 5.73 ±1.34 Fig. 7 Mission with three UAVs: UAV trajectories to pass through fifty PoI (black points). Wind map considered corresponding with t=450 seconds. In order to show how the detection of thermal is done, Figure 8 shows the evolution of the altitude in a simulation with an UAV. It identifies seven thermals (thermals 4,1,N3,7,5,10 and 12 of the Figure 7) and it exploits the sixth thermal by gainning approximately 140 meters. 8 Experimental results Experiments have been performed with two gliding fixed-wing UAVs in the airfield of La Cartuja (Seville) in order to test the behaviour of the whole system in real time. One of them is a real gliding fixed-wing UAV and the other is simulated. The thermals are emulated; that is, when the real UAV accesses a simulated thermal it Distributed thermal identification and exploitation for multple soaring UAVs 17 Fig. 8 Vertical profile of the UAV flight. Seven thermals are identified and the UAV passes through each thermal twice to estimate its parameters. simulates the gaining of energy by using the propulsion system to gain altitude. The rest of the flight is carried out without propulsion. The following variables of the UAV were set: gliding angle (0.11rad), airspeed (13.89m/s) and climb rate (1.9m/s) within a thermal. Figure 9 shows a potential collision and the solution trajectory of the real gliding fixed-wing UAV in 2D. Both UAVs fly to the PoI2 but real UAV changes its trajectory to avoid the collision, so it passes through PoI1. Finally, Figure 10 presents the trajectories in the airfield of La Cartuja (Seville). 9 Conclusions The chapter considers long endurance cooperative missions with multiple gliding fixed-wing UAVs. The goal is to explore an environment as much as possible without landing. Multiple UAVs can be used in order to decrease the time to perform the mission and increasing the probability of detecting unknown thermals. A new system has been proposed to extend the flight duration by harvesting energy that comes from thermals. 18 J.A. Cobano, D. Alejo, S. Vera, G. Heredia, S. Sukkarieh and A. Ollero Fig. 9 Real and simulated flight in 2D to explore the environment in the airfield of La Cartuja (Seville). A potential collision is detected (a) and real UAV avoids it (b). Therfore, the real UAV passes through PoI1 and PoI3, and the simulated UAV passes through PoI2. An themal detection algorithm is implemented to identify unknown thermals and exploit them. First, a potential thermal is detected from the changes of altitude of a UAV. Then an algorithm computes two extra waypoints to ensure that a UAV will pass through the center of the thermal with a perpendicular direction to the first one. The thermal parameters are estimated after the second pass and these parameters are sent to the TM block. These thermals are considered as a shared resource of the system. A distributed path planning algorithm (BFRS) that automatically guides the UAVs in the system to visit both the PoIs and the TPs in the system has also been implemented. This algorithm will ask the TM block each time an UAV needs to enter in a thermal in order to gain energy. Moreover, a collision-free trajectory planning algorithm based on the RRT* has been implemented to solve the collisions detected between UAVs. The RRT* planning algorithm presented in [28] has been adapted to this problem by considering the energy. This algorithm is an important contribution with respect to the works presented on multiple UAV in autonomous soaring [13] [19] [20] [21]. Simulations demonstrate the utility of the system by adapting to changes in the environment while maximizing the endurance of the UAVs in the system. Experiments that have been carried out on a real platform shows that it can be applied for real time applications because of its low computational needs. This latter presents an important contribution with respect to [20] and [21]. Distributed thermal identification and exploitation for multple soaring UAVs 19 Fig. 10 UAV Trajectories and location of thermal represented in the airfield of La Cartuja (Seville): real gliding fixed-wing UAV (blue), simulated gliding fixed-wing UAV (red) and thermal (black cylinder). The real UAV trajectory described inside of the thermal is shown (blue circles). Acknowledgment This work was supported by the European Commission FP7 ICT Programme under the International Research Exchange Network ”Multi-UAV Cooperation for long endurance applications ” (SI-0971/2012) and and the CLEAR Project (DPI2011- 28937-C02-01) funded by the Ministerio de Ciencia e Innovacion of the Spanish Government. David Alejo is granted with a FPU Spanish fellowship from the Ministerio de Educaci´ on, Cultura y Deporte. References 1. M. J. Allen, “Guidance and control of an autonomous soaring uav,” NASA TM-214611, February 2007. 2. L. Rayleigh, “The soaring of birds,” Nature, no. 27, pp. 534–535, 1883. 20 J.A. Cobano, D. Alejo, S. Vera, G. Heredia, S. Sukkarieh and A. Ollero 3. G. T. M. J. S. S. A. Weimerskirch, H. and D. P. Costa, “Fast and fuel efficient optimal use of wind by flying albatrosses.” in Proceedings of the Royal Society of London - Biological Sciences, vol. 267, 2000, pp. 1869–1874. 4. I. Lancaster, “The problem of the soaring bird,” The University of Chicago Press, 1885. 5. C. K. Patel and I. M. Kroo, “Theorical and experimental investigation of energy extraction from atmospheric turbulence,” in Proc. 26th Congress of International Council of the Aeronautical Sciences, Anchorage, Alaska, USA, 14-19 September 2008. 6. M. J. Allen, “Updraft model for development of autonomous soaring uninhabited air vehicles,” in 44 th AIAA Aerospace Sciences Meeting and Exhibit, AIAA 2006-1510, 2006, pp. 9–12. 7. J. Wharington, “Autonomous control of soaring aircraft by reinforcement learning,” Ph.D. dissertation, Royal Melbourne Institute of Technology, Melbourne, Australia, 1998. 8. M. Hazard, “Unscented kalman filter for thermal parameter identification,” in AIAA Region II Student Conference, AIAA, Washington, DC, USA, 2009. 9. D. J. Edwards, “Implementation details and flight test results of an autonomous soaring controller,” in AIAA Guidance, Navigation, and Control Conference, AIAA, Paper 2008-7244, Reston, VA, August 2008. 10. N. Kahveci, P. Ioannou, and D. Mirmirani, “Optimal static soaring of uavs using vehicle routing with time windows,” in AIAA Aerospace Sciences Meeting and Exhibit, AIAA-2007-158, AIAA, Washington, DC, USA, 2007. 11. M. J. Allen, “Autonomous soaring for improved endurance of a small uninhabited air vehicle,” in Proceedings of the 43rd Aerospace Sciences Meeting, AIAA, 2005. 12. D. Metzger and J. Hedrick, “Optimal flight paths for soaring flight,” in Proceedings of the 2nd International Symposium on the Technology and Science of Low Speed and Motorless Flight, 1974. 13. A. Klesh, P. Kabamba, and A. Girard, “Optimal cooperative thermalling of unmanned aerial vehicles,” Optimization and Cooperative Control Strategies, vol. 381, pp. 355–369, 2009. 14. J. Nguyen, N. Lawrance, R. Fitch, and S. Sukkarieh, “Energy-constrained motion planning for information gathering with autonomous aerial soaring,” in IEEE International Conference on Robotics and Automation ( ICRA2013), Karlsruhe, Germany, 6-10 May 2013, pp. 3825–3831. 15. A. Chakrabarty and J. W. Langelaan, “Energy-based long-range path planning for soaringcapable unmanned aerial vehicles,” Journal of Guidance, Control and Dynamics, vol. 34, no. 4, pp. 1002–1015, 2011. 16. N. R. J. Lawrance and S. Sukkarieh, “Path planning for autonomous soaring flight in dynamic wind fields,” in Proceedings of the IEEE International Conference on Robotics and Automation, ICRA2011, Shanghai, China, 2011. 17. W. B. Kagabo and J. R. Kolodziej, “Trajectory determination for energy efficient autonomous soaring,” in 2011 American Control Conference, San Francisco, CA, USA, June 29 - July 01 2011, pp. 4655–4660. 18. J. H. A. Clarke and W. H. Chen, “Trajectory generation for autonomous soaring uas,” International Journal of Automation and Computing, pp. 248–256, June 2012. 19. G. O. Antal C. and L. S., “Adaptive autonomous soaring of multiple uavs using simultaneous perturbation stochastic approximation,” in Proceedings of the 49th IEEE Conference on Decision and Control, Atlanta, GA, USA, 2010. 20. N. T. Depenbusch and J. W. Langelaan, “Coordinated mapping and exploration for autonomous soaring,” in AIAA Infotech@Aerospace Conference, St. Louis, Missouri, USA, Marchl 29-31 2011. 21. K. Andersson, I. Kaminer, K. Jones, V. Dobrokhodov, and D. Lee, “Cooperating uavs using thermal lift to extend endurance,” in AIAA Unmanned Unlimited Conference, Seatle, Washington, USA, April 6-9 2009. 22. J. A. Cobano, D. Alejo, S. Vera, G. Heredia, and A. Ollero, “Multiple gliding uav coordination for static soaring in real time applications,” in IEEE International Conference on Robotics and Automation ( ICRA2013), Karlsruhe, Germany, 6-10 May 2013, pp. 782–787. 23. 3DRobotics. (2014, apr) Apm multiplatform autopilot. [Online]. Available: http://ardupilot.com Distributed thermal identification and exploitation for multple soaring UAVs 21 24. S. Russell and P. Norvig, Artificial Intelligence. A modern approach., 2nd ed. Prentice-Hall, 2003. 25. S. J. Rasmussen, T. Shima, J. Mitchel, A. G. Sparks, and P. Chandler, “State-space search for improved autonomous uavs assignment algorithm,” in Proc. 43rd IEEE Conference on Decision and Control, Atlantis, Paradise Island, Bahamas, 2004, pp. 2911–2916. 26. S. Karaman and E. Frazzoli, “Sampling-based algorithms for optimal motion planning,” International Journal of Robotics Research, vol. 30, 2011. 27. T. W. McLain, R. W. Beard, and A. Beard, “Coordination variables, coordination functions, and cooperative timing missions,” 2003. 28. B. Akgun and M. Stilman, “Sampling heuristics for optimal motion planning in high dimensions.” in International Conference on Intelligent Robots and Systems (IROS2011), San Francisco, California (USA), 25-30 September 2011, pp. 2640 – 2645.