Design Of Lane Changing System For Intelligent Vehicle On Expressway

Expressway has good road conditions,low vehicle density and one-way driving of vehicles on the same side,which is one of the most suitable scenarios for realizing intelligent driving.Therefore,intelligent driving technology on expressway has attracted extensive attention from scholars at home and abroad.Lane-changing behavior of vehicles is a common driving behavior in expressway conditions,and it is also the core technology to realize intelligent driving.It is necessary to conduct in-depth research on the autonomous lane-changing algorithm of vehicles.The research content of this paper is divided into four parts:The first part: Based on the NGSIM traffic data set,the driving data of lane-changing vehicles meeting the preset conditions are extracted.According to the judgment conditions of the starting and ending points of lane change,the complete lane change trajectory is extracted.The extracted lane change track is smoothed to reduce the data error caused by the camera collection,and then the lane change characteristic parameters are statistically analyzed based on the track data after smoothing.In the second part,the fuzzy sets and sub-fuzzy sets of each parameter response are designed with the relative distances between the main vehicle and the car in front of the current lane,the car in front of the target lane and the car behind the target lane as fuzzy inputs.In order to make the lane change decision conform to the dynamic driving environment of expressway,the change of relative distance and relative speed between the main vehicle and the surrounding vehicles was considered in the design of fuzzy input port,and the scaling factor was added.The fuzzy rule base is designed according to the experience,and the lane change decision threshold is set to defuzzize the fuzzy decision output value,and the binary decision system with lane change/no lane change is generated.The fuzzy decision is verified with NGSIM data set,and the accuracy is 85%.The static lane changing trajectory is designed based on the quintic polynomial,and the lane changing process is verified by the co-simulation platform.The third part: In order to avoid the collision risk in the process of dynamic lane changing,the potential field force function of obstacle vehicle(object)and road boundary is designed based on virtual force field method.A vehicle prediction model including dynamics and kinematics was established,and a dynamic obstacle avoidance trajectory planning and tracking integrated controller for lane change process was designed based on the model predictive control principle,which took no collision between the main vehicle and the surrounding vehicles,followed the desired trajectory and ensured the stability of the vehicle as the optimization objectives.Part Ⅳ: Build the software simulation platform(Simulink/Prescan),design the corresponding lane change test conditions,and verify.The verification results show that the vehicle can accurately judge the timing of lane change,and when the main vehicle meets the obstacle vehicle(object)in the process of lane change,it can safely,timely and quickly avoid the obstacle vehicle(object)to complete the lane change operation.