Game-Theoretic Approach to Autonomous and Human-Driven Vehicle Interactions

This article presents a game-theoretic approach to modeling interactions between autonomous vehicles (AVs) and human-driven vehicles (HDVs) as strategic games. Each driver, including the AV, aims to optimize specific objectives such as safety, efficiency, and compliance with traffic laws. We define the interaction dynamics as a non-cooperative game and analyze Nash equilibria in different traffic scenarios. Simulation results using MATLAB illustrate the effectiveness of our model in predicting real-world driving behaviors. We also generate a comparison between game-theoretic approach and rule-based approach in terms of performance metrics (accuracy, precision, recall and F1-score).