Hardware-in-the-loop testing of current cycle feedback ILC for stabilisation and tracking control of under-actuated visual servo system

This paper presents an iterative learning control (ILC) scheme augmented with the feedback control for solving the nonlinear stabilisation and tracking control problem of ball on plate system, which is a class of under-actuated visual servo system. To enhance the trajectory tracking performance and deal with the real-time challenges of ball on plate system including the nonlinearity, inter-axis coupling and uncertain dynamics, we present a feed-forward learning control scheme, which iteratively updates the control input from one trial to the next, integrated with the cascade control. The ILC update law is synthesised based on the current iteration tracking error (CITE), and the uniform convergence of the input control sequence is presented using the contraction mapping technique. From image processing standpoint, for detecting the foreground objects from a video stream, a background subtraction algorithm using frame difference technique is employed. The efficacy of the proposed scheme is tested on a laboratory scale ball on plate system using hardware-in-the-loop (HIL) testing. Experimental results substantiate that augmenting the learning control with the feedback control not only reduces the tracking error significantly but also enhances the robustness of the closed loop system against the poor lighting conditions.