Behavior Modeling of Bus Lane Utilizing by Connected Autonomous Vehicles in Heterogeneous Traffic Flow Based on Cooperative Game

The mixed traffic which composed of connected autonomous vehicles (CAVs) and human-driven vehicles (HVs) will remain the norm in urban roads. In order to fully utilize the advantages of CAV in enhancing traffic efficiency and reduce the negative impact of CAVs on human-drivers, this paper focuses on scenarios with low CAV permeability and innovatively constructs a lane change model of CAV utilizing bus lanes based on cooperative game theory from the perspective of dynamic bus priority right-of-way. The method proposed in this paper not only guarantees the priority of the dynamic bus right-of-way, but also makes full use of the space resources of the bus lane and improves the traffic efficiency. In addition, in order to separate CAVs and HVs, this paper also establishes a CAV lane-changing model considering surrounding vehicles, and encourages CAV to change lanes in the direction of the bus lane by establishing the incentive function of right-lane change. In the lane change payoff function, both the leading and following vehicles in the current lane and the target lane of the subjective CAV are considered, the purpose is to reduce the influence of subjective lane-changing CAV on other vehicles. Through the analysis of the results, the proposed strategy in this paper reduces traffic delay by 2.5 seconds compared to the no-control strategy, and by 1.9 seconds compared to the fixed-clear distance strategy when the penetration of CAV rate reaches 0.3 and the number of bus passenger is 5, the efficiency of heterogeneous traffic flow is effectively improved.