Robust Precision Motion Control for a Pneumatic Flexible Manipulator With Joint Coupling and Grasped Loads Based on a Nonlinear ESO and a Sliding Mode TD

In this paper, a motion control strategy is proposed for a pneumatic flexible manipulator with joint decoupling and a grasped load. A nonlinear extended state observer (ESO) is employed to estimate grasped load disturbances and joint coupling in the pneumatic flexible manipulator. An adaptive tornambe controller (TC) is proposed to provide further compensation and achieve robust precision motion control. A sliding mode tracking differentiator (TD) is designed to arrange a rapid transition process for the pneumatic flexible manipulator. Moreover, stability analyses of the sliding mode TD, the nonlinear ESO and the adaptive TC are conducted by Lyapunov methods. Experimental results are given to illustrate the feasibility of the proposed motion control strategy for the pneumatic flexible manipulator. Note to Practitioners—The motivation of this paper is to address the challenges faced by a pneumatic flexible manipulator with joint coupling and a grasping load during motion. Pneumatic flexible manipulators have strong adaptability and high safety, but they face three major challenges: 1. Pneumatic flexible manipulators exhibit high nonlinearity because of the hysteresis and creep characteristics of pneumatic artificial muscles (PAMs), which brings challenges for control performance enhancement. 2. There exists a complex coupling relationship between multiple PAMs in the pneumatic flexible manipulator, which causes difficulty in decoupling by model analysis. 3. Owing to the flexibility of the joints, the adverse effects of grasping loads on the motion control are not negligible. To address the above major challenges, a motion control strategy is proposed combined with an adaptive TC, a nonlinear ESO and a sliding mode TD in this paper. Notably, an adaptive control law is designed with TC to reduce the impact of the varying disturbances on the system control performance. In contrast to more complex decoupling techniques, the proposed adaptive TC is less computationally intensive and easier to implement in practical applications. Comparative experiment results show the feasibility and superiority of the motion control strategy. Future research will concentrate on improving the motion control strategy of the pneumatic flexible manipulators to decrease tracking errors.