Using Iterative Learning Control in a Walking Beam Motion System with a Flexible Coupled Robotic Mechanism

Abstract

The paper proposes a modification to enhance the performance of the control structure with the norm-optimal iterative learning control (ILC) algorithm for precise positioning of the gripper tip of a robotic mechanism that is flexibly coupled to an industrial walking beam motion system. For feedback systems involving controlled plants with unstable or poorly damped poles, it is more suitable to use a serial norm-optimal ILC controller, which may exhibit a tendency towards controller fragility and/or the generation of aggressive and excessive changes in control signals. To address this issue, adjustments to the implementation of the feedback controller are discussed, and modifications to the synthesis of the overall controller structure are proposed. This is done to prevent controller fragility and ensure achievable dynamics of the control variable without significantly impacting the attained system performance. Simulation verification is performed using the Matlab/Simulink software package by developing a nonlinear plant model in Simscape/Simulink, providing comprehensive insights into the system's behavior.