Indirect sliding mode control based on gray-box identification method for pneumatic artificial muscle

Abstract We present an indirect robust nonlinear controller for position-tracking control of a pneumatic artificial muscle (PAMs) testing system. The system modeling is reviewed, for which the existence of uncertain, unknown, and nonlinear terms in the internal dynamics is presented. From the obtained results, an online identification method is proposed for estimation of the internal functions with learning rules designed via a Lyapunov derivative function. A robust nonlinear controller is then designed based on the approximated functions to satisfy the control objective under the sliding mode technique. Appropriate selection of the smooth robust gain and the sliding surface ensures convergence of the tracking error to a desired level of performance. Stability of the closed-loop system is proven through another Lyapunov function. The proposed approach is verified and compared with a conventional proportional–integral–differential (PID) controller, adaptive recurrent neural network (ARNN) controller, and robust nonlinear controller in a real-time system with three different kinds of trajectories and loading. From the comparative experimental results, the effectiveness of the proposed method is confirmed with respect to transient response, steady-state behavior, and loading effect.