Flow Planner combines a UNet with local inpainting conditioning, action-noise data augmentation, and train/inference trajectory splitting to enable flow models to stitch novel trajectories for robotic manipulation.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.RO 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Improving Trajectory Stitching with Flow Models
Flow Planner combines a UNet with local inpainting conditioning, action-noise data augmentation, and train/inference trajectory splitting to enable flow models to stitch novel trajectories for robotic manipulation.