Path Planning And Tracking Control Based On Passenger Car Automatic Parking

The automatic parking system can alleviate the driver’s parking troubles and improve the safety of parking.Major auto manufacturers are optimistic about the development of this field and increase funds to develop this product.Automatic parking technology has been greatly improved.This article focuses on the core technology of the automatic parking system-path planning and path tracking.Path planning is carried out for parallel parking and vertical parking scenarios,and a path tracking controller is designed,and the feasibility of the planned path is verified through software simulation.Control effect with path tracking controller.This paper first established a model of a certain passenger car in Simulink at low speed,set the same parameters to the Carsim model,and verified the feasibility of the established kinematics model.For the parallel parking scenario,the minimum parking space size and feasible parking area required for single-step parallel parking are studied based on the traditional path,and the collision avoidance constraints and vehicle kinematics constraints in the single-step parallel parking scenario are analyzed,using five times The polynomial function establishes a single-step parallel parking path optimization function.For the vertical parking scenario,the collision avoidance constraint conditions are analyzed,and the feasible parking area is derived.Based on the different parking areas where the vehicle is located,single-step vertical parking and multi-step vertical parking are divided.First,a single-step parking path is planned using the traditional path,and then the traditional path is optimized by a fourth-order polynomial function;secondly,for the complex multi-step vertical parking path,particle swarm optimization is used to optimize the path length and improve the safety of parking.Then,for path tracking,a pure tracking controller and a model predictive controller are designed to control the vehicle to track the parking path respectively,and the tracking effects are compared.By adjusting the preview distance of the pure tracking controller,the path tracking error is kept within a reasonable range,and the tracking error is reduced by adjusting the N_pvalue in the predicted time domain and the N_C value in the control time domain.Based on the comparison of the path tracking effects of the two controllers,model predictive control is finally used to track parallel parking paths and single-step and multi-step vertical parking paths.Finally,Carsim software and Simulink are used for joint simulation to verify the rationality of the planned path and the control effect of the path tracking controller.