An Iterative Learning Control Synthesis for Nonlinear Systems with Hard Input and Output Constraints

Engineered systems are always subjected to operational constraints that limit the range of feasible control input signals and their output signals. This paper proposes an iterative learning control (ILC) structure that can satisfy hard input and output constraints simultaneously for a class of nonlinear systems. This structure enables the decoupling between the design of feed-forward ILC and the output feedback. The role of feed-forward ILC is to track the desired trajectory under repetitive environment while the output feedback is added to handle output constraints with the help of a barrier Lyapunov function. The concept of virtual output constraints is proposed to ensure that the output constraints can be satisfied within the input limits by shifting and scaling the original barrier Lyapunov function. The proposed algorithm is able to ensure the perfect tracking performance and satisfaction of both input and output hard constraints. Simulation results are presented to demonstrate the effectiveness of the proposed method.