Unilateral Telemanipulation System For Operator-Centered Training And Control In Industrial Robotics

Unilateral Telemanipulation System For Operator-Centered Training And Control In Industrial Robotics Mattia Tafuri, Davide Chiaravalli, Mona Saber Gharamaleki, Claudio Melchiorri, Gianluca Palli and Roberto Meattini Abstract—Teleoperation systems extend human capabilities in complex or unsafe environments. This paper presents a unilateral telemanipulation framework that combines direct and rate control modes for intuitive human–robot interaction. Direct control provides accurate mapping of haptic device motion to robot motion, while rate control enables efficient large-scale movements. Haptic feedback signals the current control mode and the velocity magnitude, improving transparency and usability. A dedicated Graphical User–Machine Interface (GUMI) offers real-time visualization of control states, stylus position, and synchronized video streams. The framework was implemented both in simulation (MuJoCo) and on a Franka Emika Panda manipulator, and validated on a wiring task with deformable linear objects. Results confirm that users can switch seamlessly between modes, leveraging precision in direct control and efficiency in rate control. The system demonstrates robustness, flexibility, and potential for industrial applications, especially in training and programming by demonstration. Index Terms—Unilateral teleoperation, industrial robotics, hybrid motion mapping, haptic interface. I. INTRODUCTION Teleoperation systems enable human operators to control robots in hazardous or unstructured environments that require expert supervision [1], [2]. This paper presents a teleoperation framework for intuitive robotic control in industrial settings, based on a unilateral architecture with a haptic device. The system supports both direct (device position →robot position) and rate (device position →robot velocity) control, improving flexibility and addressing traditional limitations [3], [4]. Haptic forces provide intuitive feedback on the activation of rate control mode and the intensity of the robot’s velocity commands, enhancing user experience [5]. A Graphical User-Machine Interface (GUMI) provides real-time feedback on inputs and system status. The framework includes both a MuJoCo simulation and a real-world robotic setup for training and deployment [6], [7]. Validation on a wiring task with deformable linear objects (DLOs) confirms the effectiveness and usability of the hybrid control scheme [8], [9], showing potential for remote manipulation and programming by demonstration in expertdriven scenarios [10], [11]. Authors are with the Department of Electrical, Electronic and Information Engineering (DEI), University of Bologna, 40136 Bologna, Italy (corresponding author’a email: [email protected]). This work was partially supported by European Commission’s Horizon Europe Framework Programme with the project IntelliMan under Grant 101070136, by MUR with the project Sustainable Mobility Center under Grant CN00000023-CUP J33C22001120001, and by MICS (Made in Italy – Circular and Sustainable) Extended Partnership and received funding from the European Union Next-GenerationEU (PIANO NAZIONALE DI RIPRESA E RESILIENZA (PNRR) – MISSIONE 4 COMPONENTE 2, INVESTIMENTO 1.3 – D.D. 1551.11-10-2022, PE00000004). (a) Reference frames. Leader side: Geomagic Touch haptic device. Follower side: Franka Emika Panda manipulator. (b) Graphical User Machine Interface. Fig. 1: Overview of the proposed teleoperation system. II. MATERIALS AND METHODS A. Leader-to-Follower Motion Mapping During the teleoperation of the manipulator via the haptic interface, the user can activate or deactivate the motion mapping between the leader device (Geomagic Touch) and the follower robot (Robot Franka Emika) by pressing the button on the stylus.To describe the leader-to-follower motion mapping approach adopted in this work, let us refer to Fig.1a. When the mapping is active, the system stores an initial reference frame that remains unchanged until the mapping is deactivated and then reactivated. B. Motion Control Modalities The motion control is performed through two main modalities: Direct Control: This mode is enforced when the position of the stylus satisfies condition ||SpT|| < r1, which means that it is within the radius sphere r1. In this case, translation and orientation of the leader device are directly mapped to the follower robot’s end-effector xEE =RpEE RγEE,where pdes and γdes are the position vector and orientation representation, 2025 I-RIM Conference October 17-19, Rome, Italy ISBN: 9788894580570 10.5281/zenodo.17629836 201 respectively, corresponding to the translation and rotation components of the homogeneous transformation matrix Tdes obtained as Tdes =TorientTinit STT. Rate Control: If the stylus position satisfies ||SpT|| ≥ r1, a velocity control paradigm is applied. Here, the haptic stylus position is mapped to a velocity command for the remote manipulator. A force feedback is applied to the stylus to inform the user that the system is in rate control mode and to indicate the magnitude of the velocity command. The reference velocity is computed as vr=            03×1,if ||SpT|| < r1 Vmax ∥SpT∥−r1 r2−r1 SpT ∥SpT∥,if r1≤ ||SpT|| ≤ r2 Vmax SpT ∥SpT∥,if ||SpT|| > r2 (1) where Vmax ∈R+is a proper maximum velocity value. C. Graphical User Machine Interface To enhance user experience, a Graphical User Machine Interface (GUMI) (Fig. 1b) has been developed. The interface provides a real-time 3D rendering of the haptic workspace, showing the current position of the stylus relative to the two spheres. When in rate control, a force vector is displayed, pointing towards the sphere’s center. Additionally, labels provide real-time feedback on the current control mode. On the right side of the interface, two synchronized camera feeds provide real-time visual feedback of the remote environment. The system also includes a homing procedure that resets the orientation of the robot’s end-effector to a predefined pose, which can be activated directly through the GUMI. III. RESULTS AND CONCLUSIONS The user performed a wiring task involving a cube and a flexible cable, starting from the workspace center. The user initiated control by pressing a button on the haptic device’s stylus, moved the end-effector toward the cube, and adjusted its orientation to grasp the object. After lifting the end-effector, the user routed the cable through two guiding channels and inserted the object into its target location, completing the task with a homing procedure.The same setup was also developed in a MuJoCo-based simulation environment, allowing users to familiarize themselves with the system before operating the real robot. Figure 2 shows translation data for both the haptic device and end-effector. Stylus movements reveal a nearly spherical region under direct control, with red segments indicating zaxis excursions triggering rate control. End-effector data show the use of rate control mainly during the routing phase, while position control dominated phases requiring precision, such as the grasping and insertion of the cube. These results demonstrate the effectiveness of the proposed unilateral teleoperation framework, integrating direct and rate control guided by task demands, supported by haptic feedback and an intuitive graphical interface. The system proved robust, flexible, and Fig. 2: Haptic device (top) and end-effector (bottom) linear trajectories during the execution of the task. suitable for industrial tasks with deformable objects, with future work focusing on adaptive control, enhanced force feedback, and validation in more complex scenarios. REFERENCES [1] G. Niemeyer, C. Preusche, S. Stramigioli, and D. Lee, “Telerobotics,” Springer handbook of robotics, pp. 1085–1108, 2016. [2] L. Basa˜ nez and R. Su´ arez, “Teleoperation,” in Springer Handbook of Automation. Springer, 2009, pp. 449–468. [3] S. E. Salcudean, M. Zhu, W.-H. Zhu, and K. Hashtrudi-Zaad, “Transparent bilateral teleoperation under position and rate control,” The International Journal of Robotics Research, vol. 19, no. 12, pp. 1185– 1202, 2000. [4] H. V. Quang and J.-H. Ryu, “Implementation of time domain passivity approach on rate-mode bilateral teleoperation,” in 2012 IEEE International Conference on Automation Science and Engineering (CASE). IEEE, 2012, pp. 846–850. [5] Z. Jiang, F. Ni, D. Yang, C. Li, F. Yang, and H. Liu, “A hybrid mapping method with position and stiffness for manipulator teleoperation,” Applied Sciences, vol. 9, no. 23, p. 5005, 2019. [6] L. Dominjon, A. L´ ecuyer, J.-M. Burkhardt, and S. Richir, “A comparison of three techniques to interact in large virtual environments using haptic devices with limited workspace,” in Computer Graphics International Conference. Springer, 2006, pp. 288–299. [7] W. Bu, G. Liu, and C. Liu, “Rate-position-point hybrid control mode for teleoperation with force feedback,” in 2016 International Conference on Advanced Robotics and Mechatronics (ICARM). IEEE, 2016, pp. 420–425. [8] S. E. Salcudean, N. Wong, and R. L. Hollis, “Design and control of a force-reflecting teleoperation system with magnetically levitated master and wrist,” IEEE Transactions on Robotics and Automation, vol. 11, no. 6, pp. 844–858, 1995. [9] J. Barrio, M. Ferre, F. Su´ arez-Ruiz, and R. Aracil, “A remote handling rate-position controller for telemanipulating in a large workspace,” Fusion Engineering and Design, vol. 89, no. 1, pp. 25–28, 2014. [10] R. Meattini, A. Amer´ ı, A. Bernardini, J. Gonzalez-Huarte, A. Ibarguren, C. Melchiorri, and G. Palli, “Neuromuscular interfacing for advancing kinesthetic and teleoperated programming by demonstration of collaborative robots,” IEEE Transactions on Haptics, 2024. [11] S. Calinon, Robot programming by demonstration. EPFL Press, 2009. 202