考虑行人影响的时延Petri网交叉口信号控制研究

Research on Signal Control of Single Intersection with Timed Petri Net Considering Pedestrian Influence

论文基于时延Petri网(Timed Petri Net,TdPN)建立了一种考虑行人影响的可变相序信号控制模型,以有效缓解城市交叉口交通拥堵、车辆通行效率低下的问题。首先使用时延Petri网描述行人过街行为及其与车流冲突的过程,并对可变相序下车流受行人影响的配时进行决策,提出了一种滚动优化算法。该算法结合每个信号周期的车流与行人信息,按照先决策相序后决策配时的步骤对系统进行滚动优化,最终获得当前周期的最佳信号控制方案。通过将此模型与其他未虑行人影响的可变相序模型[12]和固定相序时延Petri网模型综合对比验证发现,该模型能有效缓解交通拥堵、提高交叉口的通行效率。

In this paper,a variable phase sequence signal control model considering the influence of pedestrians is estab⁃lished based on timed Petri net to effectively alleviate the problems of traffic congestion and low vehicle passing efficiency at inter⁃sections.The timed Petri net is used to describe the pedestrian crossing behavior and the process of conflict with traffic flow,and a rolling optimization algorithm is proposed to decide the signal timing of traffic flow affected by pedestrians under the variable phase sequence.The algorithm combines the information of traffic flow and pedestrians in each signal cycle,carries out rolling optimiza⁃tion of the system according to the steps of determining the phase sequence first and then the timing.Finally,the best signal control scheme in the next cycle is obtained.By comparing this model with other variable phase sequence model[12]without considering the influence of pedestrians and fixed phase sequence model based on the timed Petri net,it is found that this model can effectively alle⁃viate traffic congestion and improve the traffic efficiency at intersections.