| This paper addresses the shortcomings of the traditional collision time algorithm in terms of collision time threshold and collision time calculation.In the collision time threshold,the traditional collision warning algorithm adopts a fixed collision time threshold,which will lead to premature warning and false alarm when the vehicle speed is very small,and will lead to late alarm when the vehicle speed is very large,resulting in missed alarm and the driver is too late to take action.The improved collision time algorithm uses an adaptive dynamic collision time threshold,which is related to the self-vehicle speed and takes into account the driver’s reaction time.In the collision time calculation,the traditional collision time algorithm only considers the vehicle speed and does not consider the impact of acceleration on the collision time calculation.The improved collision time warning algorithm integrates the vehicle speed and acceleration when calculating the collision time.And in the original straight road scene and curved road scene,the straight road and curved road mixed scene is considered.The simulation test scenarios are built through the panosim simulation platform,and the simulation experiments are carried out in these three scenarios.The simulation results show that the early warning effect of using the dynamic collision time threshold is better than the fixed collision time threshold in the three scenarios of straight road,curve and straight curve.When the vehicle speed is low,the dynamic collision time threshold is small,and the actual distance during early warning is small,which will not cause false positives;When the vehicle speed is high,the threshold is also large,and the actual distance during early warning is also large,which will not cause missed warning.Therefore,the use of dynamic collision time threshold can improve the effectiveness of early warning algorithm.In addition,the early warning accuracy considering acceleration and not considering acceleration in three cases of low speed,medium speed and high speed are tested respectively.The results show that considering acceleration can improve the early warning accuracy by 8.89% on average in the straight road scenario;In the curve scene,considering the average acceleration,the accuracy of early warning can be improved by 6.67%;In the straight bend hybrid scenario,considering the average acceleration,the accuracy of early warning can be improved by 6.67%.In order to further verify the effectiveness of this algorithm,a comparative experiment is carried out with the traditional fixed TTC algorithm.The experimental results show that compared with the fixed TTC algorithm,the improved algorithm proposed in this paper can effectively improve the accuracy of early warning.Based on the above warning algorithm,this paper implements the V2 X technology-based vehicle anti-collision warning algorithm.Firstly,the requirement analysis of the V2 X technology-based vehicle anti-collision warning system is carried out to determine the main functions of the warning system,then the overall architecture diagram of the warning system is designed according to the functional requirements,and the hardware platform and software platform of the warning system are designed respectively.According to the early warning system software platform design,the software platform data acquisition module,data sending and receiving module,early warning algorithm module and display module were implemented,and the data sending and receiving module was tested,and the test results showed that the early warning system data sending and receiving capability was normal.On this basis,the vehicle anti-collision warning system was tested in real vehicles under two conditions of other vehicle stationary and other vehicle uniform speed.The test results showed that the average correct warning rate of the warning system was 91.11%,the false alarm rate was 2.22% and the missed alarm rate was 6.66%,which verified the effectiveness of the warning algorithm. |
PDF Full Text Request