The design of an autonomous parallel parking neuro-fuzzy controller for a car-like mobile robot

Abstract

The paper presents an autonomous parallel parking neuro-fuzzy controller for a car-like mobile robot. Three ultrasonic sensors are mounted in the front left corner of the car to obtain the parking environmental information and a compass sensor is mounted in the centre position of the car to measure the posture of the car. Through the proposed mechanism, one of the three ultrasonic sensors and the compass sensor are used to decide on the turning angle. Fifth-order polynomial reference paths starting from three different points to the same goal point have been used to generate the training data set. The fuzzy controller has been identified by subtractive clustering algorithm and trained by the adaptive neuro-fuzzy inference systems. The simulation results show that the proposed approach can successfully determine the motion direction at each sampling time without knowing the parking lot dimension, based on the direct sensors readings.