A Mandatory Lane Changing Integrated Decision Making and Planning Method Using Game Theory for Autonomous Vehicle

The coexistence of human-driven and autonomous vehicles poses challenges for ensuring driving safety and traffic efficiency, particularly in interactions between vehicles in adjacent lanes. To enhance driving safety and traffic efficiency, this paper proposed an integrated decision making and trajectory planning method based on game theory and a path-speed decoupled approach, which optimizing path and speed separately for autonomous vehicles in mandatory lane changing (MLC) scenario. The proposed method employs a path-speed decoupled approach to simultaneously plan two trajectories for lane keeping and lane changing in the Frenet Frame. Subsequently, the decision making module observe the driving styles of surrounding vehicles and determine the optimal strategy using a non-cooperative Stackelberg game between the self vehicle (SV) and the follow vehicle (FV). Based on the optimal trajectories of SV and FV, the SV undergoes trajectory re-planning. To validate the effectiveness of the proposed method, a hardware-in-the-loop co-simulation platform is constructed. The simulation experiments demonstrate the efficacy of the proposed method in MLC scenario. The proposed method leads to a 15% improvement in average speed compared to the rule-based method.