Event
DR-LfD: Decomposing Demonstrated Skills and Integrating Them into TAMP Gating
1 reports · 1 sources
Yizhou Chen and other authors propose the DR-LfD framework, which integrates visuomotor policies into a TAMP-gated decision-making system: human demonstrations are decomposed into atomic skills according to contact relationships, then reproduced and recombined using visuomotor policies or object-centric primitives. The authors claim that this shifts the demonstration data requirement from growing exponentially with the skill sequence to growing with the number of skill types. On two tasks in DexMimicGen, DR-LfD with 100 demonstrations outperforms a diffusion imitation policy using 1000 demonstrations; real-world and simulation benchmarks show that it can compose skills under unseen settings and physical constraints. The paper has been accepted by T-RO.
Report timeline
- DR-LfD: Decomposing and Reorganizing Demonstrations with Atomic Skills and Visuomotor Policies for Bimanual Manipulation Planning
The authors propose DR-LfD, a framework that integrates visuomotor policies into a TAMP-gated decision-making system, decomposes human demonstrations into atomic skills based on contact relationships, and then reproduces and reorganizes them using visuomotor policies or object-centric primitives. This method shifts the demonstration data requirement from growing exponentially with the skill sequence to growing with the number of skill types, and on two DexMimicGen tasks it outperforms a diffusion imitation policy trained on 1,000 demonstrations while using only 100 demonstrations. The paper has been accepted by T-RO.
arXiv cs.RO ·
Related topics
Interest trend
- Current
- 3
- Peak
- 4Nov 4 01:00
- Past 24 hours
- –
The trend compares participants observed throughout the same period, so its range may be smaller than the current interest count. Move or tap to inspect each hour; use the arrow keys on a keyboard.