论文部分内容阅读
通过对比闯红灯预测算法的性能,选择出用于区分车辆走、停行为的车辆轨迹特征参数。为实现对交叉口车辆闯红灯行为预测,根据闯红灯车辆为在红灯亮起后通过停止线的车辆这一定义,提出闯红灯预测算法的两个主要功能:估计车辆到达停止线的时刻;预测车辆到达停止线后的走、停行为。研究结果表明:在速度、加速度、行驶时间和车头时距4个物理量中,车辆的速度对闯红灯行为的预测性能最好;若用速度分别与其他3个物理量联合对车辆走、停行为进行区分,性能有所提高,但并不显著;速度-车头时距组合比以往常用的速度-加速度组合进行闯红灯行为预测的性能更好。
By comparing the performance of running a red light forecasting algorithm, a vehicle trajectory characteristic parameter for distinguishing between walking and stopping of a vehicle is selected. In order to achieve the prediction of vehicle red light running at intersections, two main functions of the red light prediction algorithm are proposed according to the definition that the red light vehicle passes the stop line after the red light is on: estimating the moment when the vehicle reaches the stop line; Stop after the line to go, stop for. The results show that the vehicle speed is the best one to predict the behavior of running red light among the four physical quantities of speed, acceleration, driving time and headway; if the speed is separately combined with the other three physical quantities, , The performance has been improved, but not significant. The speed-headway combination performs better than the conventional speed-acceleration combination in predicting red light behavior.