Actuator Saturation Limiter Under Adaptive Iterative Learning Control Study

Uncertain nonlinear systems with actuator saturation are widely existed in practice engineering. Dynamic performance and stability of the systems are influenced by actuator saturation, so it is important to research the control algorithms with actuator saturation.Based on Lyapunov stability theory, nonlinear systems with uncertain parameters and actuator saturation are discussed; adaptive iterative learning control (AILC) strategies with actuator saturation are designed accordingly. Moreover, theoretical analysis, numerical simulations and application researches are given.Firstly, for a class of nonlinear systems with actuator saturation, unknown disturbance and time-varying parameters, a robust adaptive iterative learning control (RAILC) algorithm with actuator saturation is proposed. The unknown parameters are estimated in time-domain and the disturbance is inhibited by robust control; for a class of nonlinear systems with actuator saturation, unknown time-invariant and time-varying parameters, a composite adaptive iterative learning control (CAILC) algorithm with actuator saturation is proposed. Time-invariant and time-varying parameters are estimated in time-domain and iteration-domain respectively in the algorithm, which makes full use of the time-domain and the iteration-domain information, the convergence speed can be hurried up.Secondly, both of the algorithms are promoted to the second order uncertain nonlinear systems with actuator saturation by means of backstepping technology and anti-windup compensators. Based on Lyapunov stability theory, the convergence analysis of the two approaches under identical initial condition and alignment condition are carried out. The effectiveness of these algorithms is demonstrated by simulations.Finally, the applications of the above two algorithms in parametric uncertain robot manipulators with unknown disturbance and actuators saturation are carried out. The dynamic model of the robot manipulator system with two degrees of freedom and actuator saturation is given. The boundedness of all internal signals and the convergence along the iteration horizon of output tracking error are proved. The effectiveness of these algorithms is demonstrated by simulations.