A Conflict Decision Model Based on Game Theory for Intelligent Vehicles at Urban Unsignalized Intersections

This article proposed a novel conflict decision model for intelligent vehicles based on game theory with analyzing the interaction behaviors between vehicles at urban unsignalized intersections. The proposed model can help intelligent vehicles cross intersections safely and more efficiently. Firstly, we developed an inference model for types of interactions among vehicles based on fuzzy logic. Then, the driving data was collected at urban unsignalized intersections by subgrade sensors and a retrofit intelligent vehicle and it was used in verifying the proposed inference model. After that, a conflict decision model considering safety, efficiency and comfort for intelligent vehicles based on game theory, was proposed to select the optimal driving strategies. Finally, a simulation and verification platform was built using Matlab/Simulink & Prescan. And the validity and effectiveness of the model were proved by simulation experiments. The results show the decision model can effectively help vehicles avoid conflicts and save their time spent in crossing intersections by 15 percent.