Robust switching adaptive tracking control for uncertain high-order fully actuated systems based on fully actuated system approaches

Abstract

This paper investigates the switching adaptive tracking control problem for high-order fully actuated (HOFA) systems with time-varying uncertainty. A type of special differentiable functions, named as pulse-like functions, are introduced firstly. Based on the function and the HOFA system approaches, this paper designs a new robust switching adaptive tracking (RSAT) controller, which guarantees that the state vector and its derivatives of the HOFA system are globally bounded, and the system tracking error vector converges into a preseted neighborhood of equilibrium. Most importantly, the preseted neighborhood is completely unrelated to the system initial values and the uncertainty. Furthermore, different from traditional switching adaptive control (SAC) methods, of which the controllers are often discontinuous, the RSAT controller designed in this paper is continuous (even differentiable), which is a key improvement to enhance dramatically the practicability of SAC. Finally, three simulation examples including single-link robot illustrate the effectiveness and the practicability of our proposed control method.