Research On Lane Keeping Control In Park Supported By High Definition Map

Intelligent automobile can effectively solve the problems of traffic safety and traffic congestion,which has expensive development and application prospects.Intelligent automobile is the hotspot and difficulty in vehicle research at present.With the rapid development of control theory and technology,more and more advanced intelligent control methods and control theories are widely used in the field of automatic driving of intelligent vehicle.Therefore,how to determine the optimal control methods according to different road driving environments and complex driving conditions has become a hot research subject.In this paper,based on the application scenario of driverless vehicle landing on the ground quickly to realize automatic driving-the enclosed environment in the park,taking intelligent automobile in the park as the research object,carrying on global static path planning based on the park’s map for highly automated driving electronic navigation,the control algorithm and control strategy of lanekeeping system is developed to adapt to the traffic environment in the park,to realize demonstration operation of low-speed automatic driving in the park,and the low-speed automatic driving indication in the park is realized,the effectiveness and robustness of the control algorithm are validated by union simulation and real vehicle test.The main research contents of this paper are as follows:First of all,this paper aims at intelligent automobile generally run at medium and low speeds in the park environment.In addition,the park has the characteristics of multiple deceleration belts,speed limit areas and simple road structure.The high-precision positioning model of vehicle is established by the integrated positioning and navigation fusion system of RTK-GPS(Real Time Kinematic Global Position System)and IMU(Inertial Measurement Unit),providing location information for the global static path planning and the control of lane keeping.Based on the lane-level highly automated driving map in the park,to extract images the road network data format in the highly automated driving electronic map,further improvement of road network definition file in abstract map to make it fully meet the needs of driverless vehicle navigation system and observe traffic regulations.Application of A* algorithm to carry out static lane-level global path planning based on graph search theory.Planning the optimal driving path from the starting point to the target point to make it observe traffic regulations,and validates its planning algorithm by simulation.Then,this paper determines the best reference path for vehicle travel after completing the static lane-level global path planning.This paper establishes the vehicle dynamics model and the motion control strategy of the lane-keeping system,so that the vehicle can follow the planned reference path accurately.Using RTK-GPS and IMU inertial navigation integrated equipment to acquire accurate real-time positioning and vehicle motion state estimation information during lane keeping.According to the lateral deviation and heading angle deviation between vehicle posture and desired path,the lane keeping system controller achieves the desired path tracking by eliminating the current error.Lane keeping system controller using optimal preview lateral acceleration model of lateral control algorithm and the iterative learning control of longitudinal control algorithm,the speed and direction of vehicle are separately controlled during lane keeping controlling process,to improve the applicability and robustness of the controller.Finally,this paper builds a software-in-the-loop test simulation platform and vehicle driving scenarios corresponding to various working conditions in the park on the joint simulation platform of CarSim &MATLAB/Simulink,and system verification of lane keeping system control algorithm and control strategy in the park through joint simulation.The construction of the Haval H7 wire-controlled real vehicle test platform was completed,and the rapid control prototype MicroAutoBox was used as the controller to integrate the lane keeping system control algorithm proposed in this paper on the real vehicle platform,and the real vehicle verification was carried out for the working conditions in the park.The experimental results of the joint simulation of MATLAB/Simulink & CarSim and the remote control real vehicle platform show the effectiveness and feasibility of the proposed algorithm.