Research On Modeling Simulation And Algorithm For Autonomous Lane Changing Of Intelligent Driving Vehicle

With the development of network communication,sensing,artificial intelligence and other technologies,the automobile industry has undergone a great transformation,and intelligent driving technology has attracted extensive attention at home and abroad.Mature intelligent driving technology makes driving rules more standardized,can better support intelligent transportation systems,effectively solve traffic congestion problems,improve travel efficiency,and can avoid traffic accidents caused by driver fatigue and insufficient technology.Lanechanging condition is one of the most common operating conditions during vehicle driving,and the phenomenon of traffic accidents and road jams caused by lane changing is very common.According to statistics,traffic accidents caused by lane changing accounted for 27% of the total traffic accidents.The development of autonomous lane changing strategy based on intelligent driving technology can be used to improve driving safety,driving efficiency and ride comfort,and has become one of the hot issues in the research of intelligent driving technology.Therefore,this paper has carried out research on this issue from the following aspects.(1)The autonomous lane change behavior decision is studied.Firstly,the reasons of lane change and the safety of lane change are analyzed.Secondly,an autonomous lane changing decision model is established based on the rule-based decision algorithm.According to the information of the sensing module,whether the road conditions,lane changing requirements and lane changing safety meet the lane changing requirements is judged.Finally,based on Prescan and Simulink co-simulation,the reliability of the road condition judgment,lane change demand judgment,lane change safety judgment model and the feasibility analysis of the overall lane change behavior decision-making model are respectively analyzed.The simulation results show that the autonomous lane changing behavior decision model can meet the lane changing demand and ensure the safety of lane changing under the urban standard road.(2)The method of autonomous lane change trajectory planning is studied.First,the performance of common lane change trajectory model is analyzed,and its comfort,ride comfort and plasticity are compared.Then,the heptic polynomial lane change trajectory is selected as the fitting way of lane change trajectory planning.Lane change trajectory set was generated in Frenet coordinate system based on sampling algorithm.Lane change time,lateral acceleration change rate and lateral position at the end of lane change were taken as evaluation parameters,and the optimal lane change trajectory was selected by loss function algorithm.The results show that the optimized lane change trajectory,considering the influence of multiple parameters,improves the comprehensive performance of lane change trajectory.(3)The algorithm of trajectory tracking control is studied.Firstly,the kinematics and dynamics models of the vehicles were established respectively,and the real vehicle experiments were carried out,and the vehicle dynamics model with the similar motion state was selected.Aiming at the large tracking error caused by the increasing curvature,the preview distance control based on curvature is adopted.An optimization scheme based on pure tracking algorithm is proposed,and its simulation and real vehicle test are carried out.The results show that the optimized trajectory tracking control algorithm can improve the tracking accuracy of the path,while taking into account the steering frequency and vehicle stability.(4)The autonomous lane change control strategy was modeled based on Prescan and Simulink.Two different working conditions were designed to simulate and analyze the autonomous lane change control algorithm,which verified the effectiveness and reliability of the autonomous lane change control strategy model established in this paper,and the motion parameters of intelligent vehicles in the process of lane change were analyzed and evaluated.The simulation results show that the autonomous lane changing control strategy established in this paper can complete lane changing on urban standardized roads,and ensure the comfort,ride comfort and safety of lane changing process.In this paper,the decision-making subsystem,the planning subsystem and the trajectory tracking subsystem of the autonomous lane change of the intelligent vehicle are constructed respectively,and their functions are verified by the co-simulation of Prescan and Simulink.The finite-state machine model is adopted in the decision system to realize the autonomous lane change decision of intelligent vehicles.The planning subsystem uses the seventh degree polynomial curve as the fitting method,and the optimal lane change trajectory is generated by sampling iteration based on Frenet coordinate system.The trajectory tracking subsystem adopts the tracking control model based on preview,realizes the lateral control of the intelligent vehicle through the optimization process,and realizes the longitudinal control of the intelligent vehicle through the simple PID algorithm.