Cooperative Decision Making of Automated Vehicles at the Multi-lane Merging Zone: A Coalitional Game Approach

To address the decision-making issue of automated vehicles (AVs) at the multi-lane merging zone, a cooperative decision-making framework is designed with the coalitional game approach via the vehicle to vehicle (V2V) technique. Firstly, the motion prediction module is established based on the simplified single-track vehicle model to advance the accuracy and reliability of the decision-making algorithm. Then, the cost function and constraints of decision making are designed considering the performance indexes, i.e., safety, comfort and efficiency. Moreover, from the consideration of personalized and human-like AVs, different driving characteristics are reflected in the modeling process. Furthermore, four typical coalition types are defined, and the coalitional game approach is combined with model predictive control (MPC) to deal with the decision-making issue of AVs at the multi-lane merging zone. Finally, two testing cases considering different driving characteristics are carried out to evaluate the performance of the cooperative decision-making framework. The results show that the proposed approach is able to make reasonable decisions for AVs at the multi-lane merging zone. In addition, the coalitional game strategy is able to adapt to different driving characteristics of AVs.