Evolutionary Game Models to Motivate Cooperation in Mandatory Lane Changing for Bus Exiting

To solve the social dilemma between buses and social vehicles in mandatory lane changing for bus exiting, the evolutionary game models are formulated to analyze the interactions between them. Considering the uncertainties for human-driven and connected autonomous buses around bus bays, the mutation rate is added in the model for infinite populations on the basis of standard replicator dynamics. Furthermore, the lane changing process is also modeled as a network evolutionary game for finite populations due to the characteristics of traffic flow. From the numerical simulations for infinite populations, it can be found that the decision-makings of buses are less sensitive to the mutation rates than those of social vehicles, and the cooperative tendencies of buses are more obvious in the lane changing games. For finite populations, the critical values of the variables which change the evolutionary directions are explored, and more persons’ time can be saved by adjusting the variables of social vehicles. The cooperation of buses is easy to be promoted under the low penetration rates of connected autonomous buses, and that of social vehicles is more available under their high penetration rates. Meanwhile, the consideration of the comfort losses for human-driven buses and that of the energy-saving benefits for social vehicles are both necessary due to their significant impacts on the evolutionary results. The models proposed in this paper can provide decent theoretical and practical support for intelligent decision-makings, design of the vehicles, and traffic management in urban public transportation