行人群体闯红灯行为决策模型

Pedestrian's decision model with non-complying colony

人车混行的现象在国内城市中十分普遍,而且在行人红灯期到达的行人往往倾向于违规穿越横道.经过较长时间在西安市实地观测,发现行人红灯期到达的行人选择违规穿越横道的几率随着需要等待绿灯的时间和人群规模的差异而有所不同.当需要等待的时间较长时,行人违规的几率就会变大.同样,当群体中的人数增多时,行人违规的几率也会变大.因此,结合蒙特卡罗仿真方法,提出了一种行人群体闯红灯行为随机决策仿真模型.在不同的流量条件下,对行人违规次数的实测值和仿真模拟值进行具有统计意义的显著性差异分析,验证了模型的有效性.

The phenomenon of mixed traffic exists all over the city in China. And pedestrians usually can cross streets successfully during pedestrian non-green phases. Based on the field research in Xi＇an, Shaanxi province, it is found that the proportion between pedestrian signal non-compliance and compliance are not uniform with different scale colony. And the ratio of traffic violator rises with increasing time for waiting the pedestrian＇s green signal. In this paper a simulation model is proposed to estimate the ratio of pedestrian signal noncompliance during non-green phase by Monte Carlo method. Finally, the proposed model has been verified by analyses of significant difference to the values of simulation and the measured value of the pedestrians violating times under different traffic flows.