Automatic Parking Path Planning Algorithm And CarSim Simulation Research

In the context of the "new four modernizations" of automobiles,the development of driverless technology determines the prospects of the automobile industry.Intelligent parking technology is based on practical problems and solves the contradiction between the large number of vehicles nowadays and the low availability of parking spaces.Automatic parking is more focused on solving the problems of difficult parking and low parking safety.Compared with other intelligent technologies,parking technology requires a single site,fewer car modifications,lower test speed and higher safety,so It will definitely enter R & D earlier than other technologies.Due to the complex parking environment,a single parking route will make it difficult to solve the planned route and increase the cost of parking.In order to solve this problem,the research work of this paper is to plan a parking route that meets the environmental requirements and vehicle motion requirements from the surrounding environment and vehicle body parameters,and use the corresponding path tracking algorithm to verify its tracking through simulation technology.The main contents are as follows:First,based on the principle of sensor ranging,the parking space recognition algorithm is designed.It mainly solves the distinction between idle parking spaces and other vehicles and real-time monitoring of idle parking spaces to eliminate the existence of obstacles.Secondly,starting from the environmental parameters and body parameters,the establishment of constraint conditions,based on the arc tangent function as the planning route,optimize the unknown parameters of the function,thereby establishing a constrained optimization problem.And use genetic algorithm to solve it.By analyzing and comparing the results,the optimal route is selected.Then,the horizontal motion control adopts pure tracking control algorithm,and the longitudinal motion control adopts PID-table look-up.The algorithm is built by MATLAB / Simulink,and the tracking performance is studied from the perspective of vertical control and horizontal control respectively.Finally,with the help of the powerful functions of CarSim software,this article simulates and validates the entire parking algorithm.Setting its sensor parameters,motor characteristic curve parameters,simulation interface parameters and setting up parking environment as the premise of simulation.The joint simulation of MATLAB / Simulink and CarSim verified the rationality of the algorithm.The simulation route is compared with the expected route,and an error analysis is performed to verify the tracking ability of the tracking algorithm.The results show that the planning algorithm can adapt to different environments and design a parking route that matches the body parameters and environmental parameters.The pure tracking algorithm can better track the route for parking and storage.