Parallel auto-parking of a model vehicle using a self-organizing fuzzy controller

Abstract The vehicle backward auto-parking process can be divided into three steps. First, the relative position between the Ego car and an available parking space should be detected using additional sensors, and the two-dimensional (2D) environmental map can be constructed. Then, the optimal parking trajectory is planned online based on this 2D environmental map. Finally, the auto-parking controller executes the real-time vehicle motion path estimation and the planned parking trajectory-tracking control. Here, multiple ultrasonic sensors were installed on a model vehicle to construct the 2D map of the area surrounding the vehicle. A sinusoidal function parking trajectory is planned and the corresponding state variable trajectories of the X and Y components and front-wheel steering angle are derived. A model-free intelligent self-organizing fuzzy controller is proposed to track the parking path by regulating the deviation of a selected state variable. The output command is the real-time front-wheel steering angle. This intelligent controller has a system-learning mechanism without expertise knowledge or a trial-and-error process. It is implemented on a one-sixth model vehicle for experimental study. The experimental results are compared with those of a non-linear feedback linearization controller.