Autonomous vehicle sequencing problem for a multi-intersection network: A genetic algorithm approach

This paper addresses a vehicle sequencing problem at multiple intersections under the framework of Autonomous Intersection Management (AIM). In the context of AIM, there is no more traffic signals. Autonomous vehicles are considered as independent individuals and the traffic control aims at deciding an efficient vehicle passing sequence. Since there are considerable vehicle passing combinations, how to find an efficient vehicle passing sequence in a short time becomes a big challenge. In this paper, we present a genetic algorithm based on these basic groups is designed to find an optimal or near-optimal vehicle passing sequence. Computational experiments and simulation results show that the traffic condition can be dramatically improved by applying our algorithm.