Terminal Sliding Mode Adaptive Control for Robotic Manipulators

Abstract

A robust model-based adaptive control strategy is proposed for rigid manipulators subjected to unknown parameters and external disturbances. This method constructs firstly an estimated system to approximate manipulator's dynamic behavior by adaptive estimated law, and then additional part based on the terminal sliding mode is used to compensate the approximated errors and achieve robustness of the control system. Dead-zone is introduced to adaptive control law to prevent the drift of the estimated parameters. Because of its strong disturbance-rejected properties, the sliding mode control can also alleviate the drawback of dead-zone adaptive control. Simulation studies indicate a satisfactory control performance as well as strong robustness can be guaranteed by using the proposed controller.