Research On Path Planning And Tracking Control Strategy For Unmanned Autonomous Valet Parking

In recent years,the number of car ownership in the world has increased year by year.Excessive car ownership has brought about a series of problems in energy,environment,safety and transportation.Unmanned vehicles rely on advanced high-precision environment sensing systems and path decision planning systems.And the body motion control system can significantly improve the safety and efficiency of the vehicle.The Autonomous valet parking technology is an unmanned L4 level technology in which the vehicle independently searches for a parking space and performs parking,and is a key technology for solving the parking problem.This paper relies on the National Key R&D Program of China " Development and application of new multifunctional intelligent vehicle terminal " to study the path planning algorithm and path tracking algorithm of unmanned vehicles in the environment of Autonomous valet parking.The main research contents include:(1)Establish vehicle dynamics and kinematics models,and build a road for simulation and experimental environment of Autonomous valet parking.The application environment of Autonomous valet parking has the characteristics of low driving speed and complex driving environment.This paper establishes a vehicle dynamics model to accurately obtain vehicle related parameters,and establishes a kinematic model to accurately express the precise position of the vehicle in space.At the same time,the vehicle pose correction model was established to calculate the vehicle lateral position deviation.Finally,the design and selection of the Autonomous valet parking simulation environment and the real vehicle experimental environment were laid,which laid the foundation for the subsequent research work.(2)Research on Autonomous valet parking path planning algorithm and smoothing algorithm.Considering the kinematics limitation of unmanned vehicles and the application scene characteristics of independent valet parking technology,this paper studies the algorithm of Autonomous valet parking path based on rapidly exploring random tree algorithm,aiming to maximize its probability completeness and quickly generate feasible paths under complex roads.Aiming at the problem that the original algorithm has poor real-time performance and low path quality,this paper uses the target bias strategy to extend the target point as a random point with a certain probability,and uses the vehicle corner range to limit its sampling area,which significantly improves the algorithm planning speed.Finally,a rapidly exploring random treeconnect algorithm is proposed for path planning.The results show that the algorithm has a significant effect on the expansion speed.Because the path generated by the fast search random tree algorithm has a polyline,the B-spline curve algorithm and the local weighted regression scatter-smoothing algorithm are used to smooth the planning path.The results show that the former can get better smoothing effect in the local area,but the overall smoothing effect of the path is not good;the latter can obtain a more ideal vehicle travel path curve.(3)Research on the path tracking algorithm of Autonomous valet parking.In order to enable the vehicle to accurately track the planned path,this paper controls the vehicle wheel angle based on the model predictive control strategy.In this paper,the vehicle two-degree-offreedom model is used as the control algorithm to predict the model.At the same time,the vehicle preview model and the pure tracking model are used to obtain the lateral deviation between the current position and the planned path of the vehicle during the vehicle tracking planning path,which is used as the input of the model predictive control algorithm.The ideal vehicle wheel angle is calculated to accurately control the vehicle tracking planning path.(4)Research on simulation and real vehicle experiment of Autonomous valet parking path planning and tracking control.In this paper,a C-class sedan is selected as the research object.The effectiveness of the Autonomous valet parking path planning algorithm is verified by simulation and real vehicle experiments.The simulation experiment was carried out in the MATLAB environment,and the real vehicle experiment was completed in the Chang’an University Automobile Comprehensive Experimental Field.Based on the simulation scenario of Autonomous valet parking established in Chapter 2,the driver’s driving path and algorithm planning path of different scenes are compared.At the same time,the simulation experiment of the path tracking control algorithm is used to verify the vehicle’s ability to track the desired trajectory.