Presented at International Joint Conference on Artificial Intelligence

The study on “Cooperative Control of Multiple Intersections” by Tatsuya Iwase, assigned in Toyota Motor Europe, was conducted in collaboration with the University of Southampton and the University of Queensland. This study was selected for presentation at the International Joint Conference on Artificial Intelligence.

With the communication between intersections and cars, it is possible to control the situation so that, even without signals, collisions do not occur. This promises to improve traffic efficiency. In contrast to the previous studies that have focused on single intersections due to the computational complexity, this study clarifies that the localized control of a single intersection may cause inefficient traffic when applied to a network of multiple intersections. To address this issue, we used the game theory to develop an algorithm that adjusts multiple intersections simultaneously so that vehicles are delayed in passing through the next intersection in the order in which they are most likely to collide. Our findings revealed a 29% reduction in traffic delays compared to methods that managed a single intersection. These results are expected to contribute to improved traffic efficiency and CO₂ reductions in smart cities.

Title: A Polynomial-time Decentralised Algorithm for Coordinated Management of Multiple Intersections
Authors: Iwase, T., Stein, S., Gerding, EH., Chapman, A.
Appers in: Thirty-first International Joint Conference on Artificial Intelligence
Presented: July 27, 2022
https://doi.org/10.24963/ijcai.2022/534