SOURCE-LINKED INTELLIGENCE

ObstaDiff: Generalizable Diffusion Policy Learning via Obstacle-aware Representations

arXiv · AI, language, vision and robotics · article · Sep 10, 2026 · UTC

Imitation learning has achieved impressive results in robotic manipulation, yet most existing approaches assume clean backgrounds and lack explicit mechanisms for obstacle-aware motion generation. Extending such policies to cluttered, real-world scenes with unstructured obstacles remains a key generalization challenge. We present ObstaDiff, a decomposed diffusion-policy framework with a lightweight obstacle-aware visual encoder. ObstaDiff extracts a structured target-obstacle-background representation, enabling the downstream alignment policy to generate end-effector trajectories toward a targ

Read original source ↗ Open in workspace

recordType

paper

region

Global