无需训练，让扩散模型生成更安全的导航路径。
Fisher-Preserving Guidance: Training-Free Manifold Constraints for Safe Diffusion Control
- 通过低秩雅可比分解实现无训练的曼达托约束更新
- 在多个基准上优于现有扩散策略，提升路径可靠性
- 适合需要实时安全控制的机器人导航场景
扩散模型在视觉导航的路径点预测中表现良好，但标准采样和测试时引导可能导致轨迹不可靠或效率低下，当动作偏离训练流形时尤为明显。本文提出 Fisher 保持引导与外积张量投影方法，一种无需训练的推理技术，在优化任务目标的同时避免大范围 Fisher 漂移。该方法通过低秩雅可比分解计算 Fisher 保持更新，每步仅需一次反向传播，支持实时应用。我们进一步引入截断 Fisher 去噪敏感度作为不确定性信号，并用于鲁棒的多样本动作融合。在玩具和真实导航基准上的实验，包括基于 TSDF 指导的 Maze2D、使用官方扩散策略权重的 PushT，以及仿真和真实机器人上的视觉导航任务，均显示本方法在不增加训练成本的前提下，持续优于强基线扩散策略。
原文摘要 · Abstract (English)
Diffusion models are effective for waypoint prediction in visual navigation, but standard sampling and test time guidance can produce unreliable or inefficient trajectories when updates drift off the training manifold. We propose Fisher Preserving Guidance with Outer Product Span Projection, a training-free inference method that avoids large Fisher drift associated with off-distribution actions while optimizing a task objective. Our method computes the Fisher-preserving update via a low-rank Jacobian factorization, requiring only a single backward pass per step and enabling real-time use. We further introduce Truncated Fisher Denoising Sensitivity as an uncertainty signal and use it for robust multi-sample action blending. Experiments on toy and realistic navigation benchmarks, including Maze2D with TSDF-based guidance, PushT with official Diffusion Policy weights, and visual navigation in simulation and on real robots, demonstrate consistent improvements in performance over strong diffusion-policy baselines without additional training.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。