Effective Virtual Reality Teleoperation of an Upper-body Humanoid with Modified Task Jacobians and Relaxed Barrier Functions for Self-Collision Avoidance
This work addresses the challenge of effective and safe teleoperation for humanoid robots in manipulation tasks, though it appears incremental as it builds on existing VR and barrier function methods.
The researchers tackled the problem of teleoperating an upper-body humanoid robot using VR trackers by developing an approach with modified task Jacobians and relaxed barrier functions to ensure self-collision-free motions, and they validated it on hardware by successfully performing tasks like pick-and-place box packing and two-handed box handover.
We present an approach for retartgeting off-the-shelf Virtual Reality (VR) trackers to effectively teleoperate an upper-body humanoid while ensuring self-collision-free motions. Key to the effectiveness was the proper assignment of trackers to joint sets via modified task Jacobians and relaxed barrier functions for self-collision avoidance. The approach was validated on Apptronik's Astro hardware by demonstrating manipulation capabilities on a table-top environment with pick-and-place box packing and a two-handed box pick up and handover task.