PC-Diffuser: Path-Consistent Capsule CBF Safety Filtering for Diffusion-Based Trajectory Planner

2026-03-11

AI Summary

Existing diffusion-based trajectory planners lack certifiable safety in out-of-distribution scenarios, often leading to catastrophic failures. This work proposes a path-consistent safety filtering mechanism that integrates capsule-distance control barrier functions with a kinematic bicycle model into the diffusion denoising process, enabling context-aware safety corrections at every iterative step. By embedding safety directly into the trajectory generation process rather than treating it as a post-hoc remedy, the method ensures certifiable safety while minimally distorting the original trajectory geometry. Experimental results demonstrate that the proposed framework achieves a unified balance between high safety assurance and high-fidelity trajectory generation in complex traffic scenarios, effectively avoiding collisions without compromising planning performance.

Technology Category

• Planning, Routing, and Scheduling: Replanning and Plan Repair
• Computer Vision: Diffusion Models for Vision
• Natural Language Processing: Safety and Robustness

Application Category

• Security and Privacy: Data transparency and provenance
• Search and Retrieval-Augmented AI: Agentic search
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Autonomous driving in complex traffic requires planners that generalize beyond hand-crafted rules, motivating data-driven approaches that learn behavior from expert demonstrations. Diffusion-based trajectory planners have recently shown strong closed-loop performance by iteratively denoising a full-horizon plan, but they remain difficult to certify and can fail catastrophically in rare or out-of-distribution scenarios. To address this challenge, we present PC-Diffuser, a safety augmentation framework that embeds a certifiable, path-consistent barrier-function structure directly into the denoising loop of diffusion planning. The key idea is to make safety an intrinsic part of trajectory generation rather than a post-hoc fix: we enforce forward invariance along the rollout while preserving the diffusion model's intended path geometry. Specifically, PC-Diffuser (i) evaluates collision risk using a capsule-distance barrier function that better reflects vehicle geometry and reduces unnecessary conservativeness, (ii) converts denoised waypoints into dynamically feasible motion under a kinematic bicycle model, and (iii) applies a path-consistent safety filter that eliminates residual constraint violations without geometric distortion, so the corrected plan remains close to the learned distribution. By injecting these safety-consistent corrections at every denoising step and feeding the refined trajectory back into the diffusion process, PC-Diffuser enables iterative, context-aware safeguarding instead of post-hoc repair...

Problem

• diffusion-based planning
• safety certification
• autonomous driving
• trajectory safety
• out-of-distribution failure

Innovation

• diffusion-based planning
• control barrier function
• path consistency
• capsule distance
• safety filtering