Feel the Force: Contact-Driven Learning from Humans
Ademi Adeniji, Zhuoran Chen, Vincent Liu, Venkatesh Pattabiraman, Raunaq Bhirangi, Siddhant Haldar, Pieter Abbeel, Lerrel Pinto · 2025
Read Feel the Force: Contact-Driven Learning from Humans by Ademi Adeniji, Zhuoran Chen, Vincent Liu, Venkatesh Pattabiraman, Raunaq Bhirangi, Siddhant Haldar, Pieter Abbeel, Lerrel Pinto (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-2506-01944.txt.
Description
Ademi Adeniji and seven co-authors teach robots force-sensitive touch by recording how human hands press and grip. Their paper from twenty twenty-five tackles fine force control in manipulation. Robot-gathered data and simulation both fail to cover real contact variety. Human demos scale well but video alone hides the forces. Their FeelTheForce system closes that gap with a tactile glove.
The glove gauges contact force while a vision model tracks hand pose. A closed-loop policy learns to predict the forces a task needs as it unfolds. Shared sight and action codes port the policy to a Franka arm with tactile grippers. A PD controller then tracks the predicted forces by gripper squeeze. Across five force-sensitive tasks, success rates run high. Code and videos are public.
Catalog record
| CREATOR | Ademi Adeniji, Zhuoran Chen, Vincent Liu, Venkatesh Pattabiraman, Raunaq Bhirangi, Siddhant Haldar, Pieter Abbeel, Lerrel Pinto |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 31 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2506-01944