Adaptive repetitive learning control of robotic manipulators without the requirement for initial repositioning

摘要

This paper presents adaptive repetitive learning control for trajectory tracking of uncertain robotic manipulators. Through the introduction of a novel Lyapunov-like function, the proposed method only requires the system to start from where it stopped at the last cycle, and avoids the strict requirement for initial repositioning for all the cycles. In addition, it is more applicable, as it only requires the variables to be learned in an iteration-independent manner, rather than satisfying the periodicity requirement in a number of the conventional methods. With the adoption of fully saturated learning, all the signals in the closed loop are guaranteed to be bounded, and the iterative trajectories are proven to follow the profiles of desired trajectories over the entire operation interval. The effectiveness of the proposed method is shown through extensive numerical simulation results.