Deep Fictitious Play-Based Potential Differential Games for Learning Human-Like Interaction at Unsignalized Intersections

2025-06-14

AI Summary

This work addresses the challenge of modeling human-like vehicle interactions at unsignalized intersections. Methodologically, it introduces an interpretable and convergent potential differential game framework, pioneering the integration of Deep Fictitious Play (DFP) into interactive driving policy learning. A learnable weighting mechanism explicitly models driver-specific behavioral styles—such as aggressiveness and preference biases—while supervised training and game alignment optimization are performed on the INTERACTION dataset. Theoretically, the framework guarantees Nash equilibrium convergence; empirically, it successfully reproduces diverse human driving behaviors. Ablation studies confirm the necessity of the potential game structure, the DFP solver, and the style-weighting module. By preserving game-theoretic rationality, the approach enhances both policy interpretability and human behavioral consistency, establishing a novel paradigm for cooperative autonomous driving.

Technology Category

• Humans and AI: Human-Aware Planning and Behavior Prediction
• Game Theory and Economic Paradigms: Adversarial Learning
• Multiagent Systems: Mechanism Design

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Economics, Online Markets and Human Computation: Social networks and social learning
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

Modeling vehicle interactions at unsignalized intersections is a challenging task due to the complexity of the underlying game-theoretic processes. Although prior studies have attempted to capture interactive driving behaviors, most approaches relied solely on game-theoretic formulations and did not leverage naturalistic driving datasets. In this study, we learn human-like interactive driving policies at unsignalized intersections using Deep Fictitious Play. Specifically, we first model vehicle interactions as a Differential Game, which is then reformulated as a Potential Differential Game. The weights in the cost function are learned from the dataset and capture diverse driving styles. We also demonstrate that our framework provides a theoretical guarantee of convergence to a Nash equilibrium. To the best of our knowledge, this is the first study to train interactive driving policies using Deep Fictitious Play. We validate the effectiveness of our Deep Fictitious Play-Based Potential Differential Game (DFP-PDG) framework using the INTERACTION dataset. The results demonstrate that the proposed framework achieves satisfactory performance in learning human-like driving policies. The learned individual weights effectively capture variations in driver aggressiveness and preferences. Furthermore, the ablation study highlights the importance of each component within our model.

Problem

• Learning human-like driving at unsignalized intersections
• Modeling vehicle interactions using game theory
• Capturing diverse driving styles from data

Innovation

• Deep Fictitious Play for human-like driving
• Potential Differential Game modeling interactions
• Cost function weights learned from dataset