Research On Tracking Control Of Intelligent Vehicle Lane Change

In recent years,the number of motor vehicles and passenger cars across the country has been increasing rapidly,posing difficult challenges to road traffic safety and travel efficiency.Vehicle lane changing is one of the most frequent daily driving behaviors.Generally,due to the driver’s improper judgment and operation of the driving environment,traffic congestion or serious road safety accidents occur.One of the goals of autonomous driving technology is to reduce traffic accidents caused by human factors.At present,the focus is on the research of environmental perception,driving decision-making,trajectory planning and tracking,etc.,in order to realize autonomous driving as soon as possible.This paper explores the problem of trajectory planning and tracking in the process of changing lanes,in order to reduce a series of accidents caused by changing lanes,improve traffic efficiency,and achieve efficient and collision-free lane changing.This paper first briefly describes the research background and research significance of lane-changing related technologies.The intelligent vehicle lane-changing trajectory tracking control mainly includes driving decision-making,trajectory planning and tracking,and introduces the domestic and foreign research status.For lane-changing trajectory planning and decision-making,the process of completing lane-changing driving is firstly divided,and the reasons for the intention of lane-changing driving are analyzed.Based on the consideration of lane change safety,a simplified one-way two-lane lane change scenario is established according to the actual road conditions,and the minimum safety of the vehicle and the preceding vehicle in the current driving lane,the preceding vehicle in the expected lane and the rear vehicle in the expected lane are deduced Lane-changing distance,and based on this to determine whether the current moment meets the conditions for safe lane-changing driving.Secondly,the advantages and disadvantages of five kinds of geometric functions commonly used in lane changing research are compared and analyzed,and a quintic polynomial is selected to generate the desired lane changing trajectory,which has the advantages of small calculation amount for deriving unknown parameters,continuous velocity,acceleration and trajectory curvature.It shows that the lanechanging trajectory generated by the algorithm is continuous and smooth.For lane-changing trajectory tracking control,in order to ensure the real-time performance of the calculation,the linear time-varying model predictive control principle is selected to establish a variable predictive time domain tracking controller.Based on the three-degree-of-freedom dynamic model established in Chapter 2,the prediction model is designed,and the optimization problem and objective function satisfying multiple constraints are solved by rolling,and the control quantity of each cycle is calculated to complete the tracking of the desired trajectory.Considering that the prediction time domain has a great influence on the tracking performance of the controller,the corresponding relationship between different prediction time domain values,different vehicle speeds and the tracking performance of the controller is obtained through experimental analysis,so a variable prediction time domain tracking controller is established.Finally,a co-simulation platform is built to verify the effectiveness of the trajectory planning and tracking control algorithms,and simulation experiments are carried out on the road with two attachment conditions and under three different vehicle speeds.It can be seen from the results that the variable prediction time domain tracking controller has good robustness and applicability in the driving environment of three vehicle speeds and two adhesion conditions.