DualShield: Safe Model Predictive Diffusion via Reachability Analysis for Interactive Autonomous Driving

2026-01-22

AI Summary

This work addresses the safety risks in autonomous driving motion planning posed by diffusion models that often neglect vehicle dynamics under interactive uncertainty. The paper introduces, for the first time, the Hamilton-Jacobi (HJ) reachability value function into diffusion-based planning with a dual role: it proactively guides the denoising process to ensure dynamic feasibility and simultaneously constructs a Control Barrier Value Function (CBVF) for real-time safety correction. Integrated with model predictive control, the proposed method significantly enhances both safety and task completion rates in high-risk scenarios such as unprotected U-turns, outperforming several state-of-the-art planning approaches.

Technology Category

• Planning, Routing, and Scheduling: Replanning and Plan Repair
• Humans and AI: Human-Aware Planning and Behavior Prediction
• Computer Vision: Diffusion Models for Vision

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Security and Privacy: Large-scale security measurements
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

Diffusion models have emerged as a powerful approach for multimodal motion planning in autonomous driving. However, their practical deployment is typically hindered by the inherent difficulty in enforcing vehicle dynamics and a critical reliance on accurate predictions of other agents, making them prone to safety issues under uncertain interactions. To address these limitations, we introduce DualShield, a planning and control framework that leverages Hamilton-Jacobi (HJ) reachability value functions in a dual capacity. First, the value functions act as proactive guidance, steering the diffusion denoising process towards safe and dynamically feasible regions. Second, they form a reactive safety shield using control barrier-value functions (CBVFs) to modify the executed actions and ensure safety. This dual mechanism preserves the rich exploration capabilities of diffusion models while providing principled safety assurance under uncertain and even adversarial interactions. Simulations in challenging unprotected U-turn scenarios demonstrate that DualShield significantly improves both safety and task efficiency compared to leading methods from different planning paradigms under uncertainty.

Problem

• diffusion models
• autonomous driving
• safety
• uncertain interactions
• motion planning

Innovation

• Diffusion Models
• Hamilton-Jacobi Reachability
• Control Barrier-Value Functions
• Safe Motion Planning
• Autonomous Driving