| In the 21st century,China’s economy presents an unprecedented development momentum.A large number of underground parking lots with large areas and numerous parking spaces appear in cities.However,the existing large parking lot layout is complex with similar structures so that it is difficult for people to orient themselves.Thus,it makes the car owners always cannot find their own parking space when returning to look for their car,which also brings the problem of "difficulty in finding the car ".In order to solve the above problems,this thesis focuses on the design of reverse car-searching system for large underground parking lots,car-searching path planning algorithm,license plate recognition method and reverse car-searching terminal software development to study based on the analysis and research of related technology development at home and abroad.The main research contents are as follows:The overall design scheme of reverse car-searching system for large underground parking lot is given.The design of vehicle parking position detection module and vehicle information acquisition module based on ultrasonic sensor.The parking space nodes in the parking lot are networked,and the collected vehicle monitoring image information and parking data are uploaded to the server using the wireless communication mode based on Zigbee technology.Using MySQL database to store data such as the parking space status and vehicle information.Aiming at the one-to-three monitoring mode between the monitoring equipment and the parking space in the system setting,a multi-license plate simultaneous recognition method based on the license plate interval symbol is proposed.Research the improved A*path planning algorithm,and the search scope constraints and redundant points are eliminated by means of direction constraint and path segmentation to complete the path optimization of car-searching system.Environment modeling and simulation were carried out for the algorithm,and the path planning results in different environments were compared.The algorithm was simulated and compared with the path planning results in different environments.The numbers of nodes of the improved A* algorithm is reduced by 60%-80% compared with the A* algorithm when searching.Furthermore,vector map modeling and development of parking lot are carried out based on ArcMap software.In this thesis,Java programming language is used to realize the PC client software of reverse car-searching system,providing a variety of car-searching query schemes,realizing the transplantation of path planning algorithm,adding reverse car-searching function on the basis of the existing parking lot management function to help users quickly find cars.The research work of the thesis has promotion function to the existing intelligent parking measurement and control management system.The path planning algorithm and multi-license plate recognition method studied in this paper have certain reference application value for other fields involved in path optimization and image location recognition,especially for the automatic control of static traffic management. |
