Skip to main content
Virtuix and UCF Demonstrate Humanoid Robot Controlled in Real Time Via Omni One Treadmill
Teleoperation·GlobeNewsWire / Virtuix·1 min read

Virtuix and UCF Demonstrate Humanoid Robot Controlled in Real Time Via Omni One Treadmill

Virtuix and the University of Central Florida's Institute for Simulation & Training demonstrated a humanoid robot steered in real time using Virtuix's Omni One omnidirectional treadmill. The operator walks naturally in any direction on the treadmill while the robot mirrors the movement, addressing a key limitation in humanoid teleoperation — unconstrained locomotion without physical boundaries. The operator sees through the robot's eyes via UCF's 'vDen,' a surround-projection CAVE system that doesn't require a headset. Virtuix also announced a partnership with 1HMX on the Nexus NX1, a full-body teleoperation and physical AI training system that integrates the Omni One treadmill with upper-body tracking and HaptX haptic gloves for force feedback. The company sees dual use: immediate teleoperation of humanoid robots plus capturing full-body human movement data to train autonomous AI systems.

VirtuixUCFTeleoperationVROmni One1HMXNX1Training Data
Read original source

Related News

View All News →
The Human Work Behind Humanoid Robots Is Being HiddenIndustry

The Human Work Behind Humanoid Robots Is Being Hidden

MIT Technology Review reports that robotics companies are obscuring the extent of human labor involved in training and operating humanoid robots. Workers in exoskeletons and VR headsets spend days repeating mundane tasks — opening microwaves, moving boxes — to generate training data for robot AI. Teleoperation is widespread too: 1X's $20,000 Neo home robot ships this year, but when it gets stuck, a remote operator in Palo Alto takes over through its cameras. Figure AI partnered with Brookfield to capture movement data across 100,000 residential units. The piece argues that if physical AI is coming for our workplaces and homes, the gap between perceived autonomy and actual human involvement needs honest scrutiny — otherwise we risk mistaking concealed labor for machine intelligence.

·MIT Technology Review
MIT Technology Review: The Human Work Behind Humanoid Robots Is Being HiddenIndustry

MIT Technology Review: The Human Work Behind Humanoid Robots Is Being Hidden

MIT Technology Review published an investigation into the invisible human labor powering today's humanoid robots. The piece reveals how workers in Shanghai spend weeks wearing VR headsets and exoskeletons to repeatedly open microwaves hundreds of times a day — generating training data for the robots beside them. Figure AI partnered with Brookfield, which manages 100,000 residential units, to capture 'massive amounts' of household movement data. Meanwhile, 1X's $20,000 Neo humanoid shipping to homes this year relies on remote teleoperators in Palo Alto who pilot the robot through its cameras when it gets stuck — effectively a form of wage arbitrage that re-creates gig work dynamics for physical tasks. The article argues that keeping these human workforces invisible causes the public to consistently overestimate what robots can actually do, drawing a parallel to Tesla's 'Autopilot' branding that a Miami jury found contributed to a fatal crash resulting in a $240 million verdict.

·MIT Technology Review
No image available
Research

NTT DOCOMO and Keio University Achieve First Haptic Robot Teleoperation Over Commercial 5G

NTT DOCOMO and Keio University demonstrated high-precision remote robot control with real-time force feedback over a commercial 5G standalone network — a first for practical teleoperation. The system pairs DOCOMO's Configured Grant low-latency network slicing with Keio's Real Haptics technology, which synchronizes position, force, and speed between a human operator and a remote robot in real time. In tests, a hand-type robot grasped and transported objects while force-feedback reproduction improved by 40% and motion smoothness by 59% compared to standard 5G. Configured Grant pre-allocates radio resources so the robot can transmit instantly without waiting for network permission, eliminating the scheduling delay that makes remote manipulation jerky and imprecise. The breakthrough matters for home and humanoid robots because teleoperation remains a key fallback when autonomous AI gets stuck — as seen with 1X's NEO, where remote operators take over through the robot's cameras. Lower-latency, higher-fidelity remote control over existing 5G infrastructure could make teleop-assisted home robots far more practical.

·Interesting Engineering / NTT DOCOMO