Cooperative Decision Making of Lane-change for Automated Vehicles Considering Human-like Driving Characteristics

To deal with the cooperative lane-change decision-making issue of autonomous vehicles (AVs), a cooperative coalitional game approach is utilized considering human-like driving characteristics. Firstly, three different driving characteristics are defined for AVs, i.e., aggressive, moderative and conservative. Then, the single-track vehicle model is applied to the system modeling for lane-change decision making. Moreover, the cost function of decision making is constructed considering three vital performance indexes, i.e., safety, comfort and efficiency. Furthermore, the cooperative lane-change decision-making issue is transformed into an optimization problem with multi-constraints via the cooperative coalitional game approach. Finally, testing cases are carried out to verify the feasibility and effectiveness of the proposed approach. Testing results indicate that the designed algorithm is able to make safe and correct lane-change decisions for AVs. Additionally, it can adapt to different driving characteristics of AVs.