BLNN-based Adaptive Control for a Class of Spacecraft Proximity Systems with Output Constraints and Unmodeled Dynamics

Abstract

In this work, a novel broad learning neural network based adaptive control (BLNNAC) scheme is designed for a class of spacecraft proximity systems with external perturbations, unmodeled dynamics and symmetrical time-varying constraints. Firstly, for avoiding the unacceptable complexity caused by conventional Barrier Lyapunov function, the constrained output signals are transformed into unconstrained ones by several nonlinear functions. Secondly, by virtue of fast response and insensitivity to external disturbance, the integral sliding-mode control (ISMC) scheme is incorporated into the presented control scheme. Then, to suppress the adverse impact of the dynamic uncertainties, a novel broad learning neural network (BLNN) is established to ameliorate the approximation performance. Based on that BLNN scheme, a nonlinear disturbance observer (NDO) is designed to approach the time-varying disturbances. Furthermore, a fractional power function, as a particular term of control input, is designed to guarantee the fast convergence rate. It is shown that all the signals are bounded, and the transient-states of the output signals satisfy the constraint conditions constantly. Finally, simulation results illustrate the effectiveness and advantages of the presented scheme.