Navigation and Control of Automated Guided Vehicle using Fuzzy Inference System and Neural Network Technique

Abstract

Automatic motion planning and navigation is the primary task of an Automated Guided Vehicle (AGV) or mobile robot. All such navigation systems consist of a data collection system, a decision making system and a hardware control system. Artificial Intelligence based decision making systems have become increasingly more successful as they are capable of handling large complex calculations and have a good performance under unpredictable and imprecise environments.
This research focuses on developing Fuzzy Logic and Neural Network based implementations for the navigation of an AGV by using heading angle and obstacle distances as inputs to generate the velocity and steering angle as output. The Gaussian, Triangular and Trapezoidal membership functions for the Fuzzy Inference System and the Feed forward back propagation were developed, modelled and simulated on MATLAB. The reserach presents an evaluation of the four different decision making systems and a study has been conducted to compare their performances.
The hardware control for an AGV should be robust and precise. For practical implementation a prototype, that functions via DC servo motors and a gear systems, was constructed and installed on a commercial vehicle.