Uncertainty-Aware Safety-Critical Decision and Control for Autonomous Vehicles at Unsignalized Intersections

TL;DR

This paper introduces an uncertainty-aware framework for autonomous vehicle decision-making at unsignalized intersections, combining risk-aware reinforcement learning with safety filters to enhance safety without sacrificing efficiency.

Contribution

It proposes a novel USDC framework that integrates ensemble distributional RL with high-order safety filters, enabling dynamic safety-efficiency trade-offs in urban driving scenarios.

Findings

USDC improves safety in intersection navigation

The framework maintains traffic efficiency

Uncertainty estimation enhances decision reliability

Abstract

Reinforcement learning (RL) has demonstrated potential in autonomous driving (AD) decision tasks. However, applying RL to urban AD, particularly in intersection scenarios, still faces significant challenges. The lack of safety constraints makes RL vulnerable to risks. Additionally, cognitive limitations and environmental randomness can lead to unreliable decisions in safety-critical scenarios. Therefore, it is essential to quantify confidence in RL decisions to improve safety. This paper proposes an Uncertainty-aware Safety-Critical Decision and Control (USDC) framework, which generates a risk-averse policy by constructing a risk-aware ensemble distributional RL, while estimating uncertainty to quantify the policy's reliability. Subsequently, a high-order control barrier function (HOCBF) is employed as a safety filter to minimize intervention policy while dynamically enhancing constraints based on uncertainty. The ensemble critics evaluate both HOCBF and RL policies, embedding uncertainty to achieve dynamic switching between safe and flexible strategies, thereby balancing safety and efficiency. Simulation tests on unsignalized intersections in multiple tasks indicate that USDC can improve safety while maintaining traffic efficiency compared to baselines.