Analysis and review of state-of-the-art automatic parking assist system

The emerging advanced technology, automatic parking assist, makes it possible to free drivers' hands from bored parking maneuvers and offers the possibility of reducing driver stress. Toyota Prius hybrid vehicle unveiled in 2003 is the first automobile equipped with automatic parking assist system(APAS), the research and development in APAS has become a growing trend spread from automobile enterprises to vehicle research centers of institutions and universities over ten years. APAS consists of four parts - environment recognition module, path planning generation module, path tracking controller subsystem and improved actuators like active steering/breaking subsystem. This paper mainly dealt with the new progress of APAS, on the basis of brief introduction of system with nonholonomic constraints. Firstly, the basic structures and operating principles of APAS were elaborated, and meanwhile, two methodologies applied for analyzing APAS from the vehicle engineers and robot scientists were taken into consideration respectively. Then, the products and hot topics of APAS were categorized and demonstrated, in terms of environment recognition (including vehicular position perception and obstacles detection using ultrasonic radars and vision-based sensors), path planning generation algorithms involved with continuous-curvature path using clothoid curves, different trajectory tracing control strategies based on fuzzy logic controller and fractional order controller, so as to present a detailed state-of-the-art research situation. Last but not the least, based on the current research situation, some urgent problems and development trends were discussed.