Active inference as a unified model of collision avoidance behavior in human drivers

2025-06-02

AI Summary

Existing collision-avoidance models are fragmented and scenario-specific, lacking a unified theoretical account of human collision-avoidance behavior. This paper introduces the first computational cognitive model of driving collision avoidance grounded in active inference—specifically, free-energy minimization—systematically integrating Bayesian decision-making, evidence accumulation, and motor control. The model comprehensively characterizes the full behavioral cascade: threat detection, avoidance decision-making, and action execution, across two canonical scenarios—rear-end braking and lateral intrusion. Its key contribution lies in the first systematic application of the active-inference framework to driving collision avoidance, enabling unified, cross-scenario explanation of behavioral phenomena across multiple dimensions: response timing, strategy selection, and motor execution. Validated against high-fidelity driving simulation data, the model accurately reproduces empirically observed patterns—including meta-analytic aggregate effects and scenario-specific response profiles—demonstrating high fidelity to real-world driver behavior.

Technology Category

• Cognitive Modeling & Cognitive Systems: Conceptual Inference and Reasoning
• Planning, Routing, and Scheduling: Model-Based Reasoning
• Humans and AI: Human-Aware Planning and Behavior Prediction

Application Category

• User Modeling, Personalization and Recommendation: Psychology-informed user models and recommender systems
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

Collision avoidance -- involving a rapid threat detection and quick execution of the appropriate evasive maneuver -- is a critical aspect of driving. However, existing models of human collision avoidance behavior are fragmented, focusing on specific scenarios or only describing certain aspects of the avoidance behavior, such as response times. This paper addresses these gaps by proposing a novel computational cognitive model of human collision avoidance behavior based on active inference. Active inference provides a unified approach to modeling human behavior: the minimization of free energy. Building on prior active inference work, our model incorporates established cognitive mechanisms such as evidence accumulation to simulate human responses in two distinct collision avoidance scenarios: front-to-rear lead vehicle braking and lateral incursion by an oncoming vehicle. We demonstrate that our model explains a wide range of previous empirical findings on human collision avoidance behavior. Specifically, the model closely reproduces both aggregate results from meta-analyses previously reported in the literature and detailed, scenario-specific effects observed in a recent driving simulator study, including response timing, maneuver selection, and execution. Our results highlight the potential of active inference as a unified framework for understanding and modeling human behavior in complex real-life driving tasks.

Problem

• Fragmented models of human collision avoidance behavior
• Lack of unified approach for diverse driving scenarios
• Need for cognitive mechanisms in collision avoidance modeling

Innovation

• Active inference models collision avoidance behavior
• Incorporates evidence accumulation for human responses
• Unified framework for complex driving tasks