Design of ET-MPC-based Parameters Self-tuning Controller for Mobile Robot Based on Machine Learning

Abstract

In this paper, a model predictive control (MPC) algorithm framework based on machine learning (ML) is proposed for mobile robot system, which integrates event-triggered mechanism (ETM) and parameters self-tuning mechanism (PSM). Firstly, the kinematic model of the mobile robot is established, and the MPC-based path tracking controller is designed. Secondly, the PSM for MPC is designed based on the kernel extreme learning machine optimized by sparrow search algorithm (SSA-KELM). Then, two event-triggered mechanisms are designed to determine whether to perform optimization solution and parameters tuning, respectively. In addition, a state deviation compensation mechanism (SDCM) based on event-triggered MPC (ET-MPC) is designed. Finally, the theoretical result is applied to the actual mobile robot system. Compared to the basic MPC-based trajectory tracking controller, the controller designed in this study can obtain better control performance with less computing resources.