Iterative Learning Control for AGV Drive Motor Based on Linear Extended State Observer

Abstract

In order to solve the problems of repetitive and non-repetitive interference in the workflow of Automated Guided Vehicle (AGV), Iterative Learning Control (ILC) combined with linear extended state observer (LESO) is utilized to improve the control accuracy of AGV drive motor. Considering the working conditions of AGV, the load characteristics of the drive motor are analyzed with which the mathematical model of motor system is established. Then the third-order extended state space equations of the system approximate model is obtained, in which LESO is designed to estimate the system states and the total disturbance. For the repeatability of AGV workflow, ILC is designed to improve the control accuracy. As the goods mass transported each time is not same, the LESO is utilized to estimate the non-repetitive load disturbance in real time and compensate the disturbance of the system to improve the position precision. The convergence of the combined algorithm is also verified. Simulation and experimental results show that the proposed iterative learning control strategy based on LESO can reduce the positioning error in AGV workflow and improve the system performance.

Optimization of servo accuracy of Y axis of dicing saw based on iterative learning control

“…Li, Poms et al (2014) applied the frequency domain filter to P-type ILC, which perfectly eliminated the external interference and nonlinear influence and ensured the stability of the whole control system. Jiang et al (2021) combined iterative learning with a linearly expanding observer (LESO) to compensate for system disturbances, using LESO to estimate non-repeating load disturbances in real time to improve positioning accuracy. Zhang (2022) proposed an iterative integral sliding mode (IISM) surface design that achieves robustness to irreducible disturbances and fast convergence.…”

“…Fuzzy self-tuning PID is based on the PID algorithm (Jiang, Zhu, and Zheng 2021). By calculating the current system error and error change rate, fuzzy rules are used for fuzzy inference, and fuzzy matrix is used for parameter tuning.…”

Hardware Development and Safety Control Strategy Design for a Mobile Rehabilitation Robot

“…Generally, linear actuators are used for load torque control to reduce the angular velocity as the load torque increases. Therefore, the relationship of the linear actuator between the motor speed and the load output force can be expressed by the following Equation [36]:…”

Advanced Autonomous Machines and Design Developments

“…Within the context of autonomous control vehicles, the authors of [5] proposed an Iterative Learning Control combined with linear extended state observer to improve the control accuracy of Automated Guided Vehicle (AGV) drive motor. In regard to medical devices for motion assistance, the authors of [6] design an elbow-assisting device based on a cable-driven parallel mechanism with design solutions.…”