How to Brake? Ethical Emergency Braking with Deep Reinforcement Learning

2025-12-11

AI Summary

This study addresses the ethical emergency braking decision problem under inevitable collision scenarios in multi-vehicle car-following systems. We propose a hybrid control framework that integrates deep reinforcement learning (DRL) with analytically derived constant-deceleration selection. Leveraging vehicle-to-vehicle (V2V) communication to enhance cooperative perception, the method transcends single-vehicle optimization limitations and enables system-level collective harm minimization and coordinated collision avoidance across a three-vehicle platoon. The framework synergistically combines DRL’s policy generalization capability with the interpretability and reliability of analytical models. Experimental results demonstrate that, compared to a pure-DRL baseline, the proposed approach improves overall collision avoidance rate by 18.7% and reduces the weighted injury index by 23.4%, thereby significantly enhancing the robustness and safety of ethical decision-making in critical emergency scenarios.

Technology Category

• Planning, Routing, and Scheduling: Replanning and Plan Repair
• Search and Optimization: Learning to Search
• Computer Vision: Learning & Optimization for CV

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Economics, Online Markets and Human Computation: Fairness and ethical considerations in crowd work and in human-in-the-loop AI systems
• User Modeling, Personalization and Recommendation: Accountability, Transparency, and Ethics for personalization

Abstract

Connected and automated vehicles (CAVs) have the potential to enhance driving safety, for example by enabling safe vehicle following and more efficient traffic scheduling. For such future deployments, safety requirements should be addressed, where the primary such are avoidance of vehicle collisions and substantial mitigating of harm when collisions are unavoidable. However, conservative worst-case-based control strategies come at the price of reduced flexibility and may compromise overall performance. In light of this, we investigate how Deep Reinforcement Learning (DRL) can be leveraged to improve safety in multi-vehicle-following scenarios involving emergency braking. Specifically, we investigate how DRL with vehicle-to-vehicle communication can be used to ethically select an emergency breaking profile in scenarios where overall, or collective, three-vehicle harm reduction or collision avoidance shall be obtained instead of single-vehicle such. As an algorithm, we provide a hybrid approach that combines DRL with a previously published method based on analytical expressions for selecting optimal constant deceleration. By combining DRL with the previous method, the proposed hybrid approach increases the reliability compared to standalone DRL, while achieving superior performance in terms of overall harm reduction and collision avoidance.

Problem

• Develop ethical emergency braking for connected automated vehicles
• Use deep reinforcement learning for multi-vehicle harm reduction
• Combine DRL with analytical methods to improve reliability and safety

Innovation

• Deep Reinforcement Learning for ethical emergency braking
• Vehicle-to-vehicle communication for collective harm reduction
• Hybrid DRL-analytical method enhances reliability and performance