A Model to Manage the Lane-Changing Conflict for Automated Vehicles Based on Game Theory

Abstract

In this study, we propose a lane-changing conflict management model based on game theory for automated vehicles. When a vehicle plans to change to the adjacent lane, and if there is a closely following vehicle on that lane, the following vehicle must sacrifice its speed to make space for the lane-changing vehicle, which means there are conflicts of interest between two vehicles. So far, there is no clear answer if the following vehicle should make space for the lane-changing vehicle. These individualistic lane-changing models may lead to suboptimal traffic flow or even traffic safety issues. To solve this problem, this study designed a model based on game theory to solve lane-changing conflicts between the lane-changing vehicle and the following vehicle in the target lane. When the two vehicles enter a lane-changing conflict, the payoffs of the two vehicles under various combinations of strategies were evaluated, and the final strategy and the acceleration for each vehicle were obtained based on the principle of benefit equilibrium. The simulation is conducted to analyze the game strategy of the lane-changing vehicle (LV) and the close rear vehicle (RV) in the process of lane-changing from different initial positions. The results show that, under the hypothesis scenario in the simulation, the strategy {changing a lane, avoiding } will be chosen when the RV is initially located in the range of [0, 40 m], while {not changing a lane, not avoiding} is more appropriate when the initial position of the RV is in the range of [41 m, 90 m].