Predefined-Time Reliable Control for Robotic Systems With Prescribed Performance
Jianxing Liu, Yizhuo Sun, Zhuang Liu, Yabin Gao, Ligang Wu, José I. Leon, Leopoldo G. Franquelo
Abstract
This article presents a predefined-time reliable control strategy with prescribed performance for tracking control of robotic systems with actuator faults and parametric uncertainties. A compensation mechanism based on adaptive neural network is proposed for system uncertainties and nonlinear actuator faults, by which the control strategy does not need prior precise knowledge of uncertainties. A prescribed performance function (PPF) is introduced to improve the tracking performance. A predefined-time nonsingular terminal sliding mode control (NTSMC) strategy is proposed to realize the practical predefined-time convergence of the tracking errors. In addition, combined with the PPF, the predefined-time NTSMC is developed to ensure that actuator faults are tolerated. Meanwhile, the tracking errors always remain within prescribed bounds and converge to the equilibrium within the predefined time. Experiments verify the effectiveness and advantage of the proposed control strategy.