Research On Intelligent Vehicle Path Planning And Tracking Control Based On High-precision Maps

The computer and communication technologies have advanced continuously,and this has accelerated the growth of the automotive sector.Automobiles have greatly facilitated people’s daily lives,but they have also contributed to issues like traffic congestion,pollu tion,and safety on the road.Autonomous driving technology is getting more and more at tention as one of the major technologies to increase the safety of driving a car.Three primary components make up automatic driving technology: positioning and e nvironment perception,intelligent planning and decision-making,and control execution.T he primary work covered in this paper includes research on high-precision maps,path pla nning,and path tracking control for self-driving cars.(1)First,a suitable to intelligent vehicles lane level high accuracy map is created.T he combined RTK-GPS/INS navigation and positioning system is finally adopted as the h igh-precision positioning system in this paper,and the differential service of Thousand Se ek Location is used to make the positioning accuracy reach centimeter level.This paper explains the composition,coordinate conversion,and operating principle of GPS positioni ng system and inertial navigation system in order to meet the demand of autonomous ve hicles for positioning accuracy.A lane level road network model was chosen,and data g athering on the campus roadways was used to construct a lane level high precision map network model with lane centerline nodes.(2)Secondly,the map data of the experimental collection area was abstracted into a directed graph with 64 vertices and 78 edges,using the length of the lanes as the weight s of the edges,and the abstracted directed graph data was stored in the form of an adjac ency matrix by writing the appropriate MATLAB code.This was done to express the rel ationship between the road topology and the relationship between the map data and the d ata abstraction.The east and west gates of the campus were chosen as the starting and e nding locations of the path planning,respectively,and the shortest path from the east gat e of the campus to the west gate was produced by using Dijkstra’s algorithm as the short est path planning algorithm in this work.(3)Following the establishment of the vehicle dynamics model and tire model,a pat h tracking controller based on the model predictive control algorithm was designed.A joi nt simulation platform of Carsim and Simulink was built,and the tracking algorithm’s via bility was assessed by choosing a double-shifted trajectory.The results revealed that the algorithm was capable of achieving the goal of tracking control with a minimal tracking error.(4)Finally,a real-world experiment was used to confirm the viability of the high-precis ion map and path tracking control method described in this article.The experimental find ings demonstrate that,with a maximum tracking error of 0.35 meters,the path tracking c ontroller enables the controlled vehicle to accurately track the reference lane centerline tr ajectory of the chosen test section.