机器人系统的加权快速终端滑模主动容错控制

Weighted Fast Terminal Sliding Mode Active Fault-tolerant Control for Robotic System

针对受执行器故障的非线性机器人系统,提出一种加权快速终端滑模主动容错控制方法。首先利用观测器估计机器人系统中的执行器故障信息,并通过自适应律对故障未知的界进行估计。然后根据机器人各关节的加权位置误差进一步设计快速终端滑模控制器对获得的故障信息做出补偿,从而实现有限时间主动容错控制。通过李雅普诺夫函数法证明了闭环系统的稳定性,并采用两关节机器人验证了方案的有效性。该方案可以通过对不同关节位置误差权重值的分配来相应地补偿故障的影响,能够在有限时间内使得机器人位置跟踪误差快速收敛且跟踪精度得到提高。

In this study,we propose a weighted fast terminal sliding mode active fault-tolerant control for nonlinear robotic systems subject to actuator faults.Firstly,the actuator faults in the robotic system are estimated using an observer,and the unknown boundary of the actuator faults is estimated by an adaptive law.In order to compensate for the estimated actuator faults,a fast terminal sliding mode controller is designed according to the weighted position error of each joint.Consequently,the finite-time active fault-tolerant control is achieved.The Lyapunov function method is used to prove the stability of the closed-loop system,and a two-joint robot is used to verify the effectiveness of the proposed scheme.Our experimental results showed that the proposed scheme can compensate for the actuator faults by assigning weight values to different joint position errors.Thus,the position tracking error of the robot can converge fast in a finite time,and the tracking accuracy can be improved.