Design and Implementation of an Adaptive Control System for Industrial Automated Guided Vehicles Using Radial Basis Function Neural Network

Automated Guided Vehicles (AGVs) have been widely used in industrial factories and seaport due to their high efficiency and ability to move in recent years. Nevertheless, controlling AGVs to keep moving in the predefined (correct) direction is one of their most important challenges. In this paper, an automated guided control system is designed using the direct current of servomotor. In order to guide the automatic control robot in the predefined and rotational paths, the Proportional-Integral-Derivative (PID) control system with proportional, integral, and derivative parts is used. Adaptive adjustment of PID controller coefficients is difficult due to the nature of the used control system location. Hence in this study, an optimally set up of the coefficients of PID controller is used in control system of AGV by applying adaptive Radial Basis Function Neural Network (RBFNN). The simulation results indicate that the proposed controlling scheme can optimally increase the accuracy rather than traditional PID-based method. In addition, the proposed scheme is practically implemented as a prototype robot leading to provide an experimental validation to be used as an applicable solution in AGV control systems due to the low cost of design and optimal performance.