كنترل امپدانس متغير سيستم رباتيك از راه دور دوطرفه با استفاده از يادگيري تقويتي

This thesis focuses on the study, design, comprehensive analysis, an‎d implementation of a nonlinear bilateral tele-robotic system in a leader–follower configuration. The primary objective of this research is to achieve position an‎d force tracking in order to reach high-precision transparency in bilateral tele-robotic systems. To this end, robust control is employed to ensure accurate position tracking. The novelty of this work lies in the application of variable impedance control to accomplish force tracking in the nonlinear bilateral robotic structure, where the reference trajectory of the leader robot is determined based on external force errors through this controller. In the first part, impedance parameters are estimated using the Gradient-based method, an‎d the extended Kalman filter is utilized to estimate the force impedance components. In the second part, a deep reinforcement learning algorithm is applied to enhance the transparency level of the system. Furthermore, the stability of the system is established using the input-to-state stability approach. The results of simulations an‎d experimental implementations demonstrate the effectiveness, robustness, an‎d reliability of the proposed method under various conditions. This system is particularly suitable for advanced applications in tele-robotic control, remote surgery, remote industrial operations, rehabilitation, an‎d other fields that deman‎d precise control an‎d accurate force feedback.

6

143505