Toborlife AI Launches Teleoperation System For Unitree G1

(MENAFN- Robotics & Automation News) Toborlife AI, developer of the Tobor Harness control system and the official US partner of Unitree Robotics, has announced its full-body teleoperation software for the Unitree G1 Edu humanoid.

The release gives US research labs, universities, and developers the first data collection, training, and support stack for the G1 built and supported domestically.

Training data captured through the system stays on the customer's hardware in the United States and is never transmitted to Unitree or any server outside the customer's control.

As humanoid robotics draws record investment and Chinese-built hardware like the G1 becomes standard in American labs, US teams have lacked a domestic software and support layer to match.

Humanoid robots reaching US labs arrive as capable hardware, with very little between the researcher and the joint controller. Teams spend their first months integrating different open-source libraries just to start using the robots, instead of running experiments.

Toborlife AI builds the layer that closes that gap, and teleoperation is where it starts. An operator wearing a PICO VR headset moves, and the G1 Edu moves with them, in real time, with no scripting in between.

Teleoperation as the backbone, not the product

Full-body teleoperation lets an operator move the Unitree G1 Edu as though they were inside it, guiding arms, legs, and torso naturally.

G1's optional dexterous robotic hands let operators pick up, hold, and manipulate objects with finger-level control, enabling precise grasps and fine motor tasks.

Data capture is built into every session, and the quality of that data is the point. As the operator moves, the system records full-body motion at high fidelity, down to finger-level hand movement, and delivers it in a training-ready format rather than raw logs that need cleaning.

Demonstrations that are laggy or incomplete produce weak models. Toborlife AI's teleoperation produces clean, usable demonstrations the first time, so a lab can record a task, refine it, and move straight to training and validating models.

Toborlife AI backs every deployment with a US-based team of highly trained robotics engineers. Customers get direct technical support, configuration guidance, and hands-on onboarding from people who understand the hardware and software, not a ticket queue in another time zone.

George Ekas, co-founder and CTO at Toborlife AI, says:“Most labs aren't using humanoid robots to do everything. They are focused on a few domain-specific applications. Our end-to-end data collection and training system gives those teams the best foundation to pursue their use case.

“Because we keep optimizing the tools while our customers use them, every skill they build gets better over time without extra work on their end.”

Data sovereignty for customers

Research groups and companies building on humanoid platforms are producing something valuable every time they record a demonstration, and most robotics vendors treat that output as their own.

Toborlife AI does not. Customer data stays with the customer. The company does not monitor it and only accesses it when a customer requests troubleshooting or support.

That policy matters because of what teleoperation makes possible. A research team that owns its data can build on every demonstration it records, turning individual sessions into a growing library of skills and training sets that belong to them.

It also lowers the barrier for newer roboticists: instead of spending months on infrastructure before running a single experiment, they can start collecting real data on real hardware from day one and build a body of work that carries forward past any single project or grant cycle.

David Schulhof, CEO and founder of Toborlife AI, says:“Our mission at Toborlife AI is to bridge the gap between people and robots.

“The Tobor Harness Teleops is a critical milestone in our vision to democratize emerging robotic technology. Ultimately, our goal is to empower anyone, regardless of technical background, to seamlessly work with robots.”

MENAFN23092026005532012229ID1111707270