Optimizing Lateral Motion Control of an Autonomous Ground Vehicle Using Modified Particle Swarm Optimization with Model Predictive Controller
Abhishek Thakur, Sudhanshu Mishra, Ankit Kumar, Subrat Kumar Swain, Subhendu Kumar Behera · 2024
This research presents a strategy to enhance the lateral movement control of Autonomous Ground Vehicles (AGVs) by combining a Modified Particle Swarm Optimization (MPSO) approach with a Model Predictive Controller (MPC). Concentrating on the AGV's position and orientation, this technique aims to boost agility and precision. We assess this method against algorithms like Particle Swarm Optimization (PSO), Genetic Algorithm (GA) and Batch Gradient Descent (BGD) analyzing their performance based on steering wheel angle, lateral position and tracking accuracy. Furthermore, the study utilizes sign and rank tests for an examination further confirming the efficacy of the MPSO based MPC approach. The results indicate that our method outperforms the algorithms across all metrics showcasing its potential to significantly enhance AGV side to side movement control. This progress shows promise in enhancing AGV performance, in driving scenarios, underscoring the vital importance of accurate and effective lateral motion control.