Conflict modeling at signalized intersection: An evolutionary game theory analysis

The conflict between left-turning and straight-moving vehicles at signalized intersections often stems from the coexistence of these movements within the same signal phase. This arrangement can lead to frequent interactions and potential collisions, causing significant delays and compromising road safety. Therefore, it is necessary to explore the evolution of conflict formation. To our knowledge, there is nearly no research to focus on the topic. To fill this gap, we employ the evolutionary game theory to develop a conflict evolution model for interactions between left-turning and straight-moving vehicles at the micro-level. Then, the excitation coefficient is introduced to expand the model's applicability, which considers the strategy relationship within the same group. Next, we analyze the stability of the equilibrium points and the sensitivity of the improved evolutionary game model, explaining the process of the conflict evolution and providing the basis for the assignment of the parameters in the following. Finally, we calibrate the model parameters and demonstrate the effect of the model combined with the above analysis and the empirical experiment, showing that the improved model can reflect the conflict evolution process effectively by changing the value of the excitation coefficient.