Event-Triggered Learning Robust Tracking Control of Robotic Systems With Unknown Uncertainties

Zhinan Peng, Weilong Yan, Rui Huang, Hong Cheng, Kaibo Shi, Bijoy Kumar Ghosh

Abstract: In this brief, a tracking control problem for robotic systems with unknown uncertainties is addressed by using an event-triggered adaptive dynamic programming (ADP) method. First, the tracking control of a $n$ -degree of freedom (DOF) robotic system is transformed to the optimal control of an auxiliary system such that the robust control design of the original system is feasible based on the ADP framework. To reduce the computational burden, an event-triggering mechanism is introduced. The cost function and the optimal control are approximated by a critic neural network (NN), where new weight updating laws are designed to relax the persistence of excitation condition and the requirement of initial stabilizing control. In addition, the stability analysis is rigorously given to prove that the closed-loop system is asymptotically stable while the NNs’ weight approximation error is uniformly ultimately bounded. Finally, a simulation case based on a 2-DOF robotic manipulator is given to verify the effectiveness of the designed control methods.