A Method for Vehicle Control at T-Junctions for the Diffusion Period of Autonomous Vehicles

The number of people who must drive even as they age is increasing. Therefore, traffic accidents caused by elderly people are also increasing. Additionally, reductions in traffic jams and CO2 emissions are increasingly expected by societies. Car manufacturers are actively researching autonomous vehicles that will be introduced in the market in the near future. However, there are accidents that cannot be avoided by autonomous vehicles. Currently, there is a lack of legal structure for determining responsibility in such accidents. Moreover, even if all new vehicles are autonomous, a mixed environment of autonomous and non-autonomous vehicles is expected to exist for at least the typical life span of a vehicle, ten years. In this paper, in effort to reduce the potential for traffic jams, the authors develop a behavior model for autonomous vehicles, which is then simulated on a traffic simulation of a T-junction in a provincial city, measuring success by the reduction in the resulting length of the traffic jam. Finally, we propose a method for implementation of the supporting system and discuss the result.