The authors recorded demonstrations using five wearable pose trackers, corresponding to the torso, wrists and feet, plus a chest-mounted camera.
Human demonstrations teach humanoids whole-body manipulation without robot teleoperation
Workhorse converts egocentric video and five human body-link poses into robot actions for contact-rich tasks involving hands, feet and dynamic recovery.
Top University
Songbo Hu · Qiayuan Liao · Yufeng Chi · Kevin Zakka · Yakun Sophia Shao · Pieter Abbeel · +1 more
University of California, Berkeley
Research Digest··3 min read
Hu et al.
Why this paper
From University of California, Berkeley
In one line
Workhorse enables humanoid robots to perform whole-body manipulation tasks by learning from human demonstrations without robot-specific retargeting.
What we could check
- ·No code link found
- ·No weights link found
- ·No dataset link found
- ·No compute details found
- ·No stated limitations found
- ✓Reports numbers on named benchmarks
Observed from the paper text and links we have. Absence here means we did not find it, not that it does not exist.
§