Research On Automatic Parking Route Decision Planning And Vehicle Control Algorithm

The rapid development of social science and technology has brought great changes and impacts to the automobile industry.The emerging industry represented by intelligent network vehicles has become a new growth pole in the automobile industry,and intelligent driving,as an important part of it,has been widely concerned by the public and favored by the capital in recent years.As the early development foundation of intelligent driving,the development and improvement of advanced driver assistance system plays a crucial role in guiding the development direction of intelligent driving.Automatic parking as a key technology,more and more domestic scientific research personnel and technology companies to study,therefore,exampled by a certain type of domestic vehicle as the research platform,in view of the automatic parking system program and path of the vehicle control algorithm analysis and exploration,the purpose is to improve the reliability and security of parking operation.The research and work contents of this paper on automatic parking system are as follows:Firstly,it introduces the method and principle of parking detection,understands the ultrasonic ranging technology,USES ultrasonic radar and wheel speed sensor as sensing elements,analyzes the starting point of parking space through the characteristics of radar ranging changes,and analyzes the size of parking space combining with the signal of wheel speed sensor.A parking space detection method based on multi-sensor fusion is proposed,and the validity of the method is verified by the parking space detection experiment.Secondly,the specific analysis and description of the vehicle parking process,and for parallel and vertical parking two different working conditions,respectively,the establishment of the vehicle parking kinematics model,the analysis of the influence of each position and each parameter of the car body on the parking process,through the simulation experiment to verify the correctness of the established model.On this basis,the minimum turning radius,minimum parking space and the starting area of vehicle parking are determined.Then,the control strategy of the automatic parking system is analyzed based on the fuzzy control theory,and the design idea of the parking controller is determined.Through the further analysis of the automatic parking system,the membership function of system related variables and the fuzzy rule set are determined,and the parking simulation model is built based on MATLAB/Simulink.Finally,for the overall simulation and real vehicle verification of the system,a simulation model was built based on Carsim and MATLAB/Simulink,and a joint simulation was conducted.The vehicle experimental platform was built around the Micro Autobox,a fast control prototype of dSPACE company,as a vehicle controller,and the parking control strategy was designed as the core for verification.The experimental results showed the effectiveness and reliability of the parking strategy.