Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2503.15371 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866911735219224576 |
|---|---|
| author | de Farias, Cristiana Figueredo, Luis Laha, Riddhiman Adjigble, Maxime Tamadazte, Brahim Stolkin, Rustam Haddadin, Sami Marturi, Naresh |
| author_facet | de Farias, Cristiana Figueredo, Luis Laha, Riddhiman Adjigble, Maxime Tamadazte, Brahim Stolkin, Rustam Haddadin, Sami Marturi, Naresh |
| contents | Robotic manipulation of unfamiliar objects in new environments is challenging due to limited generalisation capabilities. We propose a new skill transfer framework, GIFT (Geometry-Induced Functional Transfer), which enables a robot to transfer complex object manipulation skills and constraints from a single human demonstration. Our approach addresses the challenge of skill acquisition and task execution by deriving geometric representations from demonstrations focusing on object-centric interactions. By leveraging the Functional Maps (FMC) framework, we efficiently map interaction functions between objects and their environments, allowing the robot to replicate task operations across objects of similar topologies or categories, even when they have significantly different shapes. Additionally, our method incorporates screw interpolation (ScLERP) for generating smooth, geometrically-aware robot paths to ensure the transferred skills adhere to the demonstrated task constraints. We validate the effectiveness and adaptability of our approach through extensive experiments, demonstrating successful skill transfer and task execution in diverse real-world environments without requiring additional training. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2503_15371 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | GIFT: Geometry-Induced Functional Transfer for Category-level Object Manipulation de Farias, Cristiana Figueredo, Luis Laha, Riddhiman Adjigble, Maxime Tamadazte, Brahim Stolkin, Rustam Haddadin, Sami Marturi, Naresh Robotics Machine Learning Robotic manipulation of unfamiliar objects in new environments is challenging due to limited generalisation capabilities. We propose a new skill transfer framework, GIFT (Geometry-Induced Functional Transfer), which enables a robot to transfer complex object manipulation skills and constraints from a single human demonstration. Our approach addresses the challenge of skill acquisition and task execution by deriving geometric representations from demonstrations focusing on object-centric interactions. By leveraging the Functional Maps (FMC) framework, we efficiently map interaction functions between objects and their environments, allowing the robot to replicate task operations across objects of similar topologies or categories, even when they have significantly different shapes. Additionally, our method incorporates screw interpolation (ScLERP) for generating smooth, geometrically-aware robot paths to ensure the transferred skills adhere to the demonstrated task constraints. We validate the effectiveness and adaptability of our approach through extensive experiments, demonstrating successful skill transfer and task execution in diverse real-world environments without requiring additional training. |
| title | GIFT: Geometry-Induced Functional Transfer for Category-level Object Manipulation |
| topic | Robotics Machine Learning |
| url | https://arxiv.org/abs/2503.15371 |