A Practical Adaptive Sliding-Mode Control for Extended Trajectory-Tracking of Articulated Robot Manipulators

Abstract

This paper presents a practical adaptive sliding-mode control (PASMC) scheme and then applies it to an articulated robot manipulator. An adaptive law is newly-developed in the adaptive sliding-mode control scheme, which is designed to be based on the magnitude of the time-varying switching gains, the sliding variables, and their derivatives. It aims to provide fast and stable adaptation rate without the undesirable side effects while in operation. Besides, since it is only in need of one gain parameter, the proposed control algorithm offers outstanding advantage in producing convenience of use to practicing engineers while reducing the time-consuming tasks, including the tuning process of parameters. Next, the proposed PASMC scheme produces two-type pole-placement approach that can attain the tunable dominant poles depending on whether the sliding variables remain in the vicinity of the sliding manifold. It helps to improve the convergence rate without over-estimated switching gains near the equilibrium point. Finally, the proposed one uses one-sample delayed information to cancel out system uncertainties and nonlinearities so that it may provide simple and compact control structure. From these benefits, the proposed control scheme offers synergistic effects, including effectiveness and practicality. The system stability is proved to be uniformly ultimately bounded by Lyapunov stability. The performance of the proposed control scheme is illustrated through experiments on a real robot manipulator, which is compared with that of the existing control schemes.