Research On Automatic Parking System Based On Multi-sensor Information Fusion

As an important part of advanced driver assistance system,automatic parking can greatly reduce the burden of parking,improve parking safety and shorten parking time.The existing automatic parking system is still unable to identify irregular parking spaces,and the parking accuracy is not high,and the parking is not smooth enough,resulting in the experience of the driver is not comfortable.In view of these problems,this paper has carried out a relevant research.Firstly,the automatic parking system is analyzed,and the constitution and principle of the automatic parking system are introduced.Then the work flow of the automatic parking system is explained in detail,and the possible interrupt situation and the interrupt processing are expounded.Finally,the hardware configuration of the automatic parking system is introduced,and the configuration parameters of some related sensors are given.Secondly,aiming at the problem that the automatic parking system needs to correctly identify the parking space under multiple working conditions,this paper proposes a parking space recognition method according to the fusion of distance and visual information based on the camera and ultrasonic sensor.Firstly,the parking space is identified by the camera and ultrasonic sensor.If the identification is successful,the parking space type is output.Otherwise,the information of the camera,ultrasonic sensor and wheel speed sensor is fused to obtain the vehicle attitude characteristic parameters,and the corresponding parking space type is output by the fuzzy reasoning method.Then,the simulation model of parking space identification under multiple working conditions is built on MATLAB,and the simulation analysis is carried out for different parking space scenarios.The simulation results show the rationality and reliability of the method.Then the problems of curvature discontinuity and wheel steering in place in path planning of automatic parking system are addressed.A parallel parking path planning scheme based on linear migration and sine function is proposed,and a fractional order active disturbance rejection tracking controller is proposed.Firstly,a parking path planning scheme based on the combination of liner migration and sine function is proposed,which has the advantages of continuous curvature and simple calculation.Secondly,in order to improve the reliability of parking,and EPS dynamics model is established,and angle tracking is realized by the designed fraction active disturbance rejection controller,so as to realize the path tracking control.The feasibility of this scheme is verified by simulation.Finally,a passenger car is used as the experimental platform to carry out real vehicle experiments on the parking space identification method,trajectory planning and tracking control respectively.By identifying parking spaces in different scenarios,the success rate of identifying parking spaces reaches 90% or above,which verifies its reliability and effectiveness.Then through the experiment of automatic parking,the success rate of parking reaches 90%,which verifies its feasibility.