Research On The Lane Change Warning Threshold Considering The Deceleration Of The Rear Vehicle

Lane change warning systems can remind drivers of dangerous lane change behavior,but different drivers have great differences in risk understanding and risk decision.In the same scene,some drivers will perform the lane change maneuver;however other drivers will terminate the lane change.The existing lane change warning systems have been promoted and popularized to a certain extent,but these systems have a high frequency of early warnings while driving,which may evoke conflicting emotions of the drivers.Meanwhile,too many warnings will cause driver distraction while normal driving.The rules and strategies of warning systems play a crucial role in driver acceptance of the lane change warning system.Accurately understanding drivers’risk decision in the lane change process is the key issue in the formulation of warning rules.In terms of kinematics,when the front vehicle is performing a lane change on the target lane,it will cause the rear vehicle on the target lane to decelerate,and the deceleration of the rear vehicle determines the safety of the lane change.If the rear car cannot be decelerated in time,there may be a lane change risk.The real driving experience shows that when the front vehicle is executing a lane change,it always causes the deceleration of the rear vehicle.If the absolute value of deceleration of the rear vehicle is small,the driver of the front vehicle will often continue to perform the lane change,which is a scenario and called game-style lane change.Therefore,in this paper,the deceleration of rear vehicle is used as the characterization parameter of the lane change safety.Based on the real road lane change test,the lane change warning thresholds which are suitable for the drivers in China are proposed.To solve the problems mentioned above,a small passenger car was used in this experiment;the car consists of two millimeter wave radars.Two kinds of lane change tests were carried out,which are real driving lane change test and extreme time lane change test.The real lane change behavior data of 30 subjects and the latest safe lane change behavior data of 12 subjects were collected respectively.Based on the analysis of the data,a safe lane changing model is established.The model is used to estimate the deceleration of the rear vehicle during the lane change at the extreme time.The warning threshold is set according to the deceleration distribution of the safe lane change,the front extreme time lane change,the failed lane change attempt,and the rear extreme time lane change.The main research contents and conclusions of this paper are as follows:1.Based on the real lane change data,investigating the lane change incentives,turn signal opening characteristics,and lane change trajectories.Among them,the proportion of the number of lane changes caused by the slow lead vehicle is 74.9%,which is the main cause of the lane change;the turn signal opening rate in the test is 79.33%,and the early turn signal opening rate is only 33.74%.Using the 7^th order polynomial to fit the lane change trajectory,the goodness of fit is close to 1,and the fitting effect is good.2.Based on the real lane changing process,analyzing the distribution of the model parameters such as the time to lane change,the time to line crossing,the acceleration of the test vehicle,the relative speed and relative distance between the test car and the rear car.With the consideration of the rear car’s velocity,based on the kinematics principle,a safe lane changing model is established,and the deceleration of the rear car in the real lane changing process is used to verify the validity of the model.3.Using the safe lane change model and the extreme time lane change data to calculate the deceleration of the rear vehicle,and calibrating the time to line crossing according to the distribution of the deceleration.The results show that the deceleration distribution is reasonable when the time to line crossing is 1.6s.4.According to the distribution characteristics of the rear car’s deceleration in the safe lane change,the front extreme time lane change,the failed lane change attempt and the rear extreme time lane change,the range of the warning threshold is divided.After the warning thresholds are verified and optimized,set-1.5m/s² as the first warning threshold and-3m/s² as the second warning threshold.The safe lane change data and the failed lane change attempt data were used to compare the different effects of the ISO standard warning rules and the warning thresholds in this paper.The results show that the accuracy of the ISO standard warning rules for the safe lane change data is low,which cause a certain false positive rate.The thresholds in this paper are better,and warning accuracy rate is high.