A Two-Level Algorithm for Scheduling Intelligent and Connected Vehicles at Unsignalized Intersections

With the advancement of autonomous driving technology and communication technology, the cooperative control of vehicles at intersections is gradually becoming possible. This paper proposes a two-level scheduling algorithm, in which the upper-level algorithm uses an improved particle swarm optimization (PSO) algorithm to constantly search for the optimal order of passing vehicles at the conflict points, so as to reduce total passing time, and the quadratic programming (QP) is used in the lower-level algorithm to plan smooth speed profiles that meet the comfort and collision avoidance requirements for vehicles. Finally, a comprehensive experiment is carried out to evaluate the performance of the algorithm proposed. The results show that the algorithm proposed can improve the efficiency of intersections while ensuring the driving comfort of vehicles on the premise of driving safety.