Research On Track Planning And Tracking Control Of Unmanned Vehicles Changing Lanes

With the development and application of related technologies for automobile intelligence,people’s understanding of traditional automobiles and driving methods have gradually changed.The ultimate goal of intelligent driving vehicles is to turn the closed-loop system of "peoplevehicle-environment" into a "vehicle-environment" system.In recent years,the technology of automobile intelligence has attracted wide attention,and the research work carried out around it has become more in-depth.Studies have shown that most of the causes of car accidents are due to improper operation of the driver.When driving on the road,changing lanes is one of the most common driving behaviors,and it is also one of the most likely behaviors to cause accidents or road jams.The purpose of the research on the key technology of autonomous vehicle lane changing is to make the lane changing operation performed by the vehicle autonomously more accurate,reasonable,safer and more comfortable than the driver’s operation.The research content of this article is as follows:(1)Analysis of lane changing behavior strategy.First,analyze the behavioral factors that affect lane changing.The main significant factors include road characteristics,vehicle characteristics,traffic flow characteristics,and driver characteristics.Secondly,analyze the driving behavior characteristics of human drivers in actual traffic environments.The road scene is used to expand,considering the collision situation of the body attitude,and deriving the safe collision distance model.(2)Design of lane-changing traj ectory algorithm.In view of the current autonomous lanechanging trajectory planning algorithms,most of them are static trajectory planning or the problem of setting up known objects.This paper designs a dynamic trajectory planning algorithm based on adaptive potential field.The algorithm is first based on the potential field method.The car is the center of the drivable area,and an adaptive factor is added to the traditional potential field function,so that it can automatically change the distribution of the potential field with the change of the relative speed of the two cars.According to the current state of the vehicle,a reasonable plan is made.For a lane-changing trajectory,the algorithm executes the above steps in each cycle,so the lane-changing trajectory can be dynamically planned.(3)Tracking control of lane change.This paper uses the model predictive control algorithm,which has the ability to predict the future trajectory and deal with multi-object constraints,reasonably establish the constraints in the lane change process,design the controller,and realize the lane change trajectory tracking of the unmanned vehicle operating.(4)Time domain influence of lane change prediction.This article takes into account the influence of the prediction time domain on the performance of track-changing trajectory tracking,and explains the situation of trajectory tracking in different prediction time domains.If the prediction time domain value is too small or large,the tracking error of the lane-changing trajectory is different.It may even lead to the failure of the unmanned vehicle to change lanes,so it is necessary to select a suitable prediction time domain for the unmanned vehicle to change lanes.(5)Verification and evaluation of lane-changing algorithm.In order to verify the effectiveness and accuracy of the designed driverless car lane-changing algorithm,this paper verifies the lane-changing algorithm with the vehicle model established in CarSim and Matlab/Simulink simulation platform,which proves the rationality and effectiveness of the algorithm.