[4F2-GS-10m] AI応用：移動と交通

[4F2-GS-10m-03] Proposal of a Traffic Signal Split Control System Considering Information of Pedestrian Puddles

〇Yuji Yamada¹, Fumito Ihara², Ken Hayashi², Ryu Fujimori¹, Satoshi Suga¹, Satoshi Kurihara² (1. Graduate School of Science and Technology, Keio University, 2. Faculty of Science and Technology, Keio University)

Keywords:AI, Traffic signal control, Dynamic control

Traffic congestion is regarded as a serious social problem that causes environmental problems and economic losses, and decreasing of traffic congestion in urban areas is urgent theme. One approach to relieving traffic congestion is adaptive setting of traffic signal control parameters. On the other hand, up to now only vehicle information is utilized by the traffic signal control systems. So, in this study, we proposed a new method of traffic signal split control algorithm that takes into account pedestrian information as well as vehicle information. In order to evaluate the performance of the proposed method, we used the traffic simulation software SUMO to reproduce the intersection, vehicle flow, and pedestrian flow. As a result, it was confirmed that the number of people waiting for pedestrians was reduced by the method that considered pedestrian information.