Improving wall-following robot performance using PID-PSO controller

Abstract

A wall-following robot is one of the main issues in autonomous mobile robot behavior. However, a wall-following robot needs a robust controller to perform specific tasks accurately. This paper presents an optimization method termed Particle Swarm Optimization (PSO). It was used to automatically produce necessary parameters of the PID controller; henceforth, it was termed as PID-PSO Controller. A new technique of PSO was introduced to enhance the ability of a PID controller to maintain the linear velocity of a mobile robot. The PID-PSO controller was applied to a wheeled wall-following robot. A number of experiments were carried out, and the simulated results were adopted and performed in real applications. Based on several experimental results it can be obtained that the accumulative errors the robot use PID controllers tuned manually, tuned by GA and tuned by PSO are 0.7866, 0.78543 and 0.74619, respectively. Also, the convergence process of PID parameters using the proposed PSO is faster and more optimal than GA. Therefore, it can be said that the proposed system can improve the performance of wall-following robots by decreasing the accumulative error of up to 9%.