Soft computing techniques in intelligent wall following control for a car-like mobile robot

In this paper, we present an intelligent wall following system that allows an autonomous car-like mobile robot to perform wall following tasks. Soft computing techniques were used to overcome the need of complex mathematical models that are traditionally used to solve autonomous vehicle related problems. A neuro-fuzzy controller is implemented for this purpose, which is designed for steering angle control, and is implemented with the use of two side sensors that estimate (1) the distance between the vehicle front side and the wall, and (2) the distance between the vehicle back side and the wall. These distances get then inputted to our steering angle controller, which as a result outputs the appropriate weal steering angle. The work in this paper is intended to show a detailed description of the neuro fuzzy controller that overcomes the performance of the previously implemented fuzzy logic based controller, and significantly reduces the trajectory tracking error, but at the expense of using more DSP resources causing a slower throughput time. This autonomous system was realized on an FPGA platform, which has the advantage of making the system response time much faster than software-based systems.