Development of the Neural Network Steering Controller for Outdoor Autonomous Mobile Robot Using Magnetic Fields

Abstract

This paper describes a lateral guidance system for an outdoor autonomous mobile robot, using a neural network to model the magnetic fields from road embedded magnets. The analytical magnetic fields model was compared with measured data and found to have less than 1 % difference. The neural network was also used to learn the steering behaviour of the mobile robot relative to the magnetic field values (Bx, By, Bz). A computer simulation of the mobile robot (including robot dynamics and steering) was used to verify the steering performance of the robot controller using the neural network. Good results were obtained. Also, the real outdoor autonomous mobile robot using neural network controller verified good results.