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

TL;DR

DualShield integrates model-based diffusion planning with Hamilton–Jacobi reachability to simultaneously guide trajectory generation toward safe, dynamically feasible regions and enforce real-time safety via a control barrier-value function. By reusing pre-computed HJ value functions in both proactive denoising guidance and a reactive CBVF-QP safety shield, the framework achieves a principled balance between multimodal exploration and formal safety under uncertain interactions. In challenging interactive driving simulations, DualShield delivers perfect task success with zero collisions, outperforming baselines in safety and efficiency, though with higher computational cost that could be mitigated via GPU parallelization and batched value queries. The work highlights a promising path toward trustworthy autonomous systems, with future directions including online HJ value approximation with reinforcement learning to remove offline computation requirements.

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.

Figures (5)

• Figure 1: Overview of the DualShield framework. The model-based diffusion generates candidate trajectories in a receding horizon scheme, leveraging a pre-computed HJ value function in a dual capacity: it proactively guides the denoising process away from high-risk regions, and serves as the final shield by forming a reactive CBVF-QP to filter the executed controls.
• Figure 2: Visualization of the safety-guided denoising process at different steps. The planner begins with a cloud of noisy trajectories ($i=99$). As the process continues, the safety-guided score steers the distribution until it converges to a high-quality, safe trajectory distribution ($i=1$).
• Figure 3: Comparison of closed-loop trajectories in a representative scenario with adversarial HV1 and HV2. (a) MBD: Collides with the HV1 due to reliance on soft penalties. (b) NMPC: Fails by getting trapped in a local minimum. (c) DualGuard-MPPI: Ensures safety but results in an inefficient, hesitant trajectory. (d) DualShield: Achieves a safe and efficient trajectory by proactively navigating the interaction (see Fig. \ref{['fig:six_frame_subplot_with_brt']} for the snapshots of dynamic interaction).
• Figure 4: Snapshots of DualShield executing a tactical U-turn. It first yields to the adversarial HV1, then accelerates to merge safely behind the adversarial HV2, ultimately achieving a stable lane-keeping state in the target lane. This safe maneuver is achieved despite relying on a simple constant velocity prediction for both HVs, as the underlying reachability analysis accounts for the full range of their potential actions. The contours depict the HJ value function relative to the most threatening HV, with the black line marking the BRT boundary.
• Figure 5: Flexible multimodal planning of DualShield in varying multi-agent interactions. (a) When both HVs are cooperative, DualShield plans an assertive merge. The planner then demonstrates robust adaptation to mixed-intent scenarios: (b) It safely navigates a contested space created by a yielding HV1 and an adversarial HV2. (c) it adeptly handles the mirrored case with an adversarial HV1 and a yielding HV2. (d) Finally, when both HVs act adversarially, the planner correctly identifies the high risk and switches to a safe, defensive yielding maneuver, robustly selecting the appropriate behavioral mode. (The opacity indicates temporal progression, with current states at full opacity and past positions fading progressively.)