HRT1: One-Shot Human-to-Robot Trajectory Transfer for Mobile Manipulation
Sai Haneesh Allu, Jishnu Jaykumar P, Ninad Khargonkar, Tyler Summers, Jian Yao, Yu Xiang · 2025
Read HRT1: One-Shot Human-to-Robot Trajectory Transfer for Mobile Manipulation by Sai Haneesh Allu, Jishnu Jaykumar P, Ninad Khargonkar, Tyler Summers, Jian Yao, Yu Xiang (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2510-21026.txt.
Description
Sai Haneesh Allu and five co-authors present HRT1, a robot learning system. HRT1 lets a robot copy a task from one human video. The task type is mobile manipulation, or moving while handling objects. A person shows the task once through an augmented reality headset. Augmented reality overlays digital views on the real world.
Four modules make the transfer work in sequence. The first records human demo videos from a robot like view. The second finds objects and pulls out three dimensional hand paths. The third turns a hand path into a robot gripper path. The last module plans joint moves that follow that path. With these steps, the robot can repeat the task elsewhere. It works even when objects sit in new places. Tests on a mobile manipulator covered several handling tasks. Results support the claim that one showing can be enough.
Catalog record
| CREATOR | Sai Haneesh Allu, Jishnu Jaykumar P, Ninad Khargonkar, Tyler Summers, Jian Yao, Yu Xiang |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 61 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2510-21026