Event
FLORE: A Force-Reactive Humanoid Mobile Manipulation Framework
1 reports · 1 sources
Sihang Li and other authors propose FLORE, a force-reactive humanoid mobile manipulation learning framework: it uses two force-sensing UMI-style handheld grippers and wearable trackers to collect robot-free demonstrations, and trains a force-reactive diffusion policy that takes vision, proprioception, and measured force/torque as input and outputs actions plus per-hand force commands; a companion hybrid compliant whole-body controller uses force-error integration from real force/torque sensors to achieve single-hand force tracking while maintaining balance. In real-robot tests on three contact-intensive tasks, the force-reactive policy achieved an 80% success rate, whereas a position policy trained on the same set of demonstrations and executed by the same controller achieved 26.7%. In simulation, the controller can track hand force commands of different magnitudes and directions, with little loss in whole-body motion accuracy.
Report timeline
- FLORE: Learning Force-Reactive Humanoid Loco-Manipulation
The research team proposes the FLORE framework, which uses two force-sensing UMI-style handheld grippers and wearable trackers to collect robot-free demonstrations, and trains a force-reactive diffusion policy that takes vision, proprioception, and measured force wrenches as input and outputs actions and per-hand force commands. A companion hybrid compliant whole-body controller uses integral force error from real force/torque sensors to achieve single-hand force tracking while maintaining balance. In real-robot tests on three contact-rich tasks, the force-reactive policy achieved an 80% success rate, compared with 26.7% for a position-only policy under the same controller.
arXiv cs.RO ·