Model-free iterative learning of time-optimal point-to-point motions for LTI systems

This paper presents a model-free iterative learning control algorithm, which generates time-optimal point-to-point motions for linear time-invariant systems. The proposed optimization-based algorithm consists of two levels. At the first level, a bisection algorithm determines the fastest possible point-to-point motion, i.e. the motion time is minimized, subject to actuator limitations. At the second level, an iterative learning control algorithm for point-to-point motions learns the system input that results in a point-to-point motion, with the minimal motion time obtained by the bisection algorithm at the first level. Simulation results show that the proposed model-free method is able to learn the time-optimal system input for a given point-to-point motion problem in the presence of measurement noise and repeating disturbances.