Control-Ready Uncertainty for Trajectory Diffusion

Key points

• Diffusion models can represent complex, multimodal trajectory distributions, but extracting uncertainty from them typically requires costly Monte Carlo sampling.1
• SCOPE learns a structured precision matrix around each nominal trajectory by distilling score-curvature information and producing calibrated Gaussian tubes with low overhead and without repeated Monte Carlo sampling.1
• These tubes provide per-timestep covariance estimates that can be used both as predicted occupancy for moving agents and as adaptive exploration guides for robot control.1
• We evaluate SCOPE with mode-conditioned multimodal diffusion backbones in pedestrian forecasting, crowd navigation, Maze2D control, and real-world Franka Panda manipulation.1

Sources (1)

• [1]Control-Ready Uncertainty for Trajectory Diffusion
  arXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 8, 05:57 PM

  We introduce Score-Curvature for Online Precision Estimation (SCOPE), a lightweight module that augments diffusion trajectory models with control-ready uncertainty.

  Diffusion models can represent complex, multimodal trajectory distributions, but extracting uncertainty from them typically requires costly Monte Carlo sampling.

Extractive summary: sentences quoted from the sources.