Design of an image-based fuzzy controller for autonomous parking of four-wheeled mobile robots

Abstract

This paper develops an automatic parking system for a car-like mobile robot using image-based control, where the robot equipped with a single camera detects a rectangular parking frame drawn on the floor. The desired target line to be followed by the robot is produced by using Hough transform from a captured image, together with thresholding and Canny edge detector. The fuzzy controller is designed with a reasoning mechanism composed of two inputs, which are the gradient and intercept of the target line, and one output that is the steering angle of the robot. The effectiveness of the proposed method is demonstrated through some experimental results with an actual mobile robot.