Research On Autonomous Lane Changing Method Of Intelligent Vehicle

With the development and application of intelligent vehicle-related technology, people’s awareness of the traditional car and driving way gradually changed. Finally, the "driver-vehicle- environment" closed-loop system will be changed to "vehicle- environment" system, because the development of Intelligent Vehicle. In other words, in order to avoid accidents or traffic jams, which caused by driver fatigue, poor driving habits, lack of driving experience and other factors, intelligent vehicle use environmental perception, intelligent control and electronic information technology to complete traffic environments common driving operation by itself. Lane change is one of the most common driving behavior, this behavior sometimes lead to accidents and traffic congestion. Compared with the operation of the driver’s lane change, vehicle automatic lane changing operation more efficient,reasonable, and security, this is the purposes of research on key technology of autonomous lane changing.This paper researches intelligent vehicle automatic lane changing in the high-speed and dynamic traffic environment. The autonomous driving operation of intelligent vehicle can be intelligent and reliable just like a skilled driver in the complex driving environment, what may take some time. In the actual traffic environment, the lane change process takes a relatively short time, in this process, speed of the vehicle remains stable, therefore, during the study, without considering vehicle longitudinal control.Firstly, this paper preliminary studies on lane change decision of intelligent vehicle.Through the analysis of characteristics of the driver’s lane change operation, the paper introduces the concept of desired space and desired vehicle speed, then,to consider security and personalization for a preliminary design of automatic lane-changing decision module.About the lane-changing trajectory planning, most algorithms are described by polynomial lane change trajectory, but basically the starting point and destination point determined,don’t take into account the changes of traffic environment in the planning process. Path planning originated in the field of robotics, but the car’s larger size, traditional trajectory planning method vehicle simplified as particle, only consider translational not consider the rotation, will bring relatively large error, meanwhile without considering vehicle dynamics will bring big error, because the vehicle speed is relatively high. This paper analyzes the characteristics of lane- changing trajectory, uses polynomial to generate the lane change trajectory clusters, then considers the vehicle security, stability, legitimacy, comfort,economy and other performance indicators, selected the best lane change trajectory. A trajectory family is generated based on a vehicle steady-state kinetic model, which is used to ensure its safety against the traffic obstacles. Finally a trajectory is selected based on optimization among multiple objectives on driving safety, efficiency and performance.About the lane-changing trajectory tracking control, some conventional control method has a high dependence on the environment and parameters, when the environment have a greater dynamic change, the tracking control methods are not well adapted; there are kinematic constraints and constraints of implementing agencies, when the vehicle is traveling, when the speed of vehicle is high, we also need to consider dynamic constraints,there are also some physical constraints such as road surface friction, maximum lateral displacement. Conventional control algorithm is difficult to deal with these constraints. This article uses the model predictive control which can predict the future trajectory and handling multi-objective constraints, to establish a reasonable lane-changing process constraints,controller design, implement lane-changing operation of the vehicle. Finally controller is designed, implement lane-changing operation of the vehicle.At the end of the article, the virtual simulation platform(Panosim) is used to build Simulation scenarios which is developed by our laboratory, to verify the validity and reliability of the autonomous lane-changing methods studied in this paper.