Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2603.11655 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866914392393646080 |
|---|---|
| author | Jiang, Hao Wu, Yue Wang, Yue Sukhatme, Gaurav S. Seita, Daniel |
| author_facet | Jiang, Hao Wu, Yue Wang, Yue Sukhatme, Gaurav S. Seita, Daniel |
| contents | Dexterous hands enable concurrent prehensile and nonprehensile manipulation, such as holding one object while interacting with another, a capability essential for everyday tasks yet underexplored in robotics. Learning such long-horizon, contact-rich multi-stage behaviors is challenging because demonstrations are expensive to collect and end-to-end policies require substantial data to generalize across varied object geometries and placements. We present DexMulti, a sample-efficient approach for real-world dexterous multi-task manipulation that decomposes demonstrations into object-centric skills with well-defined temporal boundaries. Rather than learning monolithic policies, our method retrieves demonstrated skills based on current object geometry, aligns them to the observed object state using an uncertainty-aware estimator that tracks centroid and yaw, and executes them via a retrieve-align-execute paradigm. We evaluate on three multi-stage tasks requiring concurrent manipulation (Grasp + Pull, Grasp + Open, and Grasp + Grasp) across two dexterous hands (Allegro and LEAP) in over 1,000 real-world trials. Our approach achieves an average success rate of 66% on training objects with only 3-4 demonstrations per object, outperforming diffusion policy baselines by 2-3x while requiring far fewer demonstrations. Results demonstrate robust generalization to held-out objects and spatial variations up to +/-25 cm. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_11655 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Concurrent Prehensile and Nonprehensile Manipulation: A Practical Approach to Multi-Stage Dexterous Tasks Jiang, Hao Wu, Yue Wang, Yue Sukhatme, Gaurav S. Seita, Daniel Robotics Dexterous hands enable concurrent prehensile and nonprehensile manipulation, such as holding one object while interacting with another, a capability essential for everyday tasks yet underexplored in robotics. Learning such long-horizon, contact-rich multi-stage behaviors is challenging because demonstrations are expensive to collect and end-to-end policies require substantial data to generalize across varied object geometries and placements. We present DexMulti, a sample-efficient approach for real-world dexterous multi-task manipulation that decomposes demonstrations into object-centric skills with well-defined temporal boundaries. Rather than learning monolithic policies, our method retrieves demonstrated skills based on current object geometry, aligns them to the observed object state using an uncertainty-aware estimator that tracks centroid and yaw, and executes them via a retrieve-align-execute paradigm. We evaluate on three multi-stage tasks requiring concurrent manipulation (Grasp + Pull, Grasp + Open, and Grasp + Grasp) across two dexterous hands (Allegro and LEAP) in over 1,000 real-world trials. Our approach achieves an average success rate of 66% on training objects with only 3-4 demonstrations per object, outperforming diffusion policy baselines by 2-3x while requiring far fewer demonstrations. Results demonstrate robust generalization to held-out objects and spatial variations up to +/-25 cm. |
| title | Concurrent Prehensile and Nonprehensile Manipulation: A Practical Approach to Multi-Stage Dexterous Tasks |
| topic | Robotics |
| url | https://arxiv.org/abs/2603.11655 |