Research On Intelligent Lane Change Trajectory Planning For Typical Traffic Scenes

In recent years,with the rapid development of artificial intelligence,the automation technology has attracted more and more attention of the society and the scientific community.As the main means of transportation for people to travel,the number of cars is increasing rapidly,Therefore,unmanned driving technology in automation technology has also become a research hotspot in the automotive field.unmanned driving technology has a positive impact on the entire traffic system.On the one hand,it can replace Manual driving,improve driving safety,and reduce the occurrence of traffic accidents.On the other hand,it can also increase the effective utilization of energy,reduce the content of harmful emissions,and improve environmental quality.It can be seen that unmanned driving technology can effectively solve a series of existing traffic problems,so a large number of traffic researchers are committed to the research of unmanned driving technology.As one of the important technologies in unmanned driving,intelligent lane changing is more challenging than lane keeping technology.It is the main solution to improve the safety of lane changing,and has a positive role in promoting the further development of unmanned driving.According to the current research on intelligent lane change,it is mainly divided into three research topics: lane changing decision,lane changing trajectory planning and lane changing trajectory tracking.Intelligent driving is immature compared with other assisted driving,and the considerations in various aspects are not yet Comprehensive,especially in the planning of lane changing.A large number of studies reflect some common research defects.On the one hand,the basic scene for lane changing trajectory planning is too simple,or the algorithms considered when constructing the lane changing trajectory planning model are incomplete,on the other hand,because trajectory planning does not have real-time nature,The researched lane changing trajectory planning method can not adapt to the current complex and changeable traffic environment,and often loses its practical application value after some parameters change in the traffic environment.Therefore,this article will aim at the above defects and try to study safer,more comfortable and real-time a model for the lane changing trajectory planning.In order to improve part of the performance of intelligent lane changing,the simulation platform is established;then,Through the analysis of various lane changing trajectory planning models at present,it determines the basic lane changing trajectory planning model which is conducive to the study of this paper;Considering the relevant constraints comprehensively,it introduces anti-collision algorithm,anti-rollover algorithm,anti-side collision algorithm and objective optimization function into the lane changing trajectory planning model,and the influence of weather changes on each algorithm is analyzed.Finally,an intelligent lane changing planning trajectory model integrating real-time,safety,comfort and efficiency is constructed.In addition,according to the existing state of surrounding vehicles when changing lanes,this article sets up eight kinds of lane changing scenes,including no surrounding vehicles,the presence of a front vehicle on the target lane,the presence of a rear vehicle on the target lane,the presence of a rear vehicle and a front vehicle on the target lane,only a front vehicle on the original lane,a rear vehicle on the target lane and a front vehicle on the original lane,a front vehicle on the target lane and a front vehicle on the original lane,a rear vehicle and a front vehicle on the target lane and a front vehicle on the original lane,and the corresponding lane changing trajectory planning models are obtained for different lane changing scenes.In order to judge whether the lane changing trajectory planning model studied in this paper meets the expected performance requirements,the simulation platform is used for analysis;trajectory and speed are selected as the evaluation indicators of trajectory tracking;lateral and longitudinal acceleration are used as the evaluation indicators of comfort;The relationship between the optimal trajectory longitudinal displacement and the safe displacement interval is used as an evaluation indicators of efficiency;it is used to analyze the trackability,comfort and efficiency of the planned lane change trajectory,and in addition,the security of real-time lane change trajectory is analyzed by sample comparison method.The results show that the intelligent lane changing trajectory planning model designed in this paper has good tracking performance in various traffic scenes,and can respond to changes in the driving status parameters of surrounding vehicles in real-time,planning a new safe lane changing trajectory or returning to the original safe trajectory of driving in the lane.The peak of the acceleration curve and the violent degree of change show that the trajectory has good comfort;the ratio of the optimal longitudinal displacement of the lane change to the displacement interval of the safe lane change reflects the high efficiency of the track.Based on the performance analysis of various aspects of the trajectory,the lane changing trajectory planning model designed in this paper can provide useful reference value for the further development of intelligent lane changing.