BridgeGuard: Explicit Safety Drift for Diffusion-based Autonomous Driving
We propose BridgeGuard, a safety-constrained diffusion planning method that progressively strengthens a constraint term during denoising to drive intermediate trajectories toward a scene-dependent safety domain.
ProofPaper ↗
Key points
- Diffusion-based driving planners capture diverse behaviors but can generate unsafe trajectories under distribution shift.
- A learned module, DistanceFieldNet, predicts a time-dependent distance field from bird's-eye-view features.
- Value and spatial-gradient supervision at queries sampled beyond expert trajectories teaches this field about both safe and unsafe regions.
- On Bench2Drive, BridgeGuard improves driving score/success rate from 87.99/74.99% to 90.88/76.36% for BridgeDrive and from 80.79/58.18% to 90.46/74.09% for $DiffusionDrive^{geo}$, demonstrating cross-model generalization.
Sources (1)
- [1]BridgeGuard: Explicit Safety Drift for Diffusion-based Autonomous DrivingarXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 8, 08:26 AM
We propose BridgeGuard, a safety-constrained diffusion planning method that progressively strengthens a constraint term during denoising to drive intermediate trajectories toward a scene-dependent safety domain.
Diffusion-based driving planners capture diverse behaviors but can generate unsafe trajectories under distribution shift.
Extractive summary: sentences quoted from the sources.