OpenAtom Forms Working Committee Created for Open Community of Humanoid Robots
Read more
Pandaily
pandaily.com

OpenAtom Forms Working Committee Created for Open Community of Humanoid Robots

The working committee for the OpenAtom open community of humanoid robots was established at the OpenAtom Developers Conference in Zhengzhou on November 20, 2026, according to Science and Technology Daily.

According to the report, the committee includes nine founding organizations representing robot developers, technology companies, an innovation center, and a research institute.

Among the founders are AgiBot, Leju, EBKernel, PrimeBot, and the Institute of Software of the Chinese Academy of Sciences. Although the Chinese report presents a full list of nine members, the ones listed above are only examples of participants and not an exhaustive list of all founders.

Goals and Objectives of the Committee

The main task of the committee is to organize work on common technical foundations. These areas include operating systems for robots, middleware, embodied intelligence models, algorithms, and data. Its functions are focused on coordination and community building, rather than launching a specific humanoid robot or a ready software platform with a new performance metric.

The report describes common industry challenges: teams having to recreate similar capabilities from scratch, poorly integrated interfaces, and technical resources scattered across different organizations. The proposed community aims to gather common requirements and support the long-term maintenance of reusable components, allowing participants to develop their own applications on top of them.

Joint adaptation and testing are central to this plan. Open source code and common tools can simplify the transfer of technical work between hardware and software platforms, and practical projects will serve as the basis for standard formation. While these goals are stated, they do not prove that the existing robots of each member are already compatible with all others.

The working committee will use the OpenAtom Foundation platform to develop a management system, organize contributions, and coordinate common tasks. Science and Technology Daily notes that transparent discussion rules and cooperation mechanisms are intended to ensure equal participation and give organizations confidence in long-term contribution.

No consolidated manifesto release, unified license covering each participant's technology, or independently measured integration cost reductions were provided upon the announcement of the committee's creation. An organization joining an open-source committee does not automatically make all its proprietary models, datasets, or hardware designs freely available.

The significance of this event lies in creating an institutional structure for common robotic infrastructure. Whether this leads to reusable components will depend on subsequent projects, licensing, maintenance practices, and cross-platform verification, rather than the mere establishment of the committee being considered a completed technical product.

Similar stories

Noitom Robotics releases HiPHI-Hand dataset used to train NVIDIA's EgoHand model
Read more
pandaily.com

Noitom Robotics releases HiPHI-Hand dataset used to train NVIDIA's EgoHand model

Noitom Robotics, a company specializing in data for physical artificial intelligence, has released the HiPHI-Hand dataset on the Hugging Face platform. This dataset consists of hand movement recordings which, according to the company, were used to train NVIDIA's EgoHand model.

The announcement of the release was made at the CoRL 2026 robotics conference held in Austin, Texas, while the actual release on Hugging Face is dated November 9th.

Noitom specified that the presented data is the exact recording that was provided to NVIDIA. It includes 11 hours of first-person synchronized video captured using the SOL headset, accounting for the camera extrinsic relative to the headset's inertial sensor, as well as three-dimensional coordinates of the wrist joints obtained through optical motion capture. To ensure sub-millimeter accuracy, even when hands are obscured from the camera, micro-markers were placed on the fingers.

The Hugging Face page contains five subsets, which collectively comprise 794 clips of stereo video with a resolution of 1360 by 1360 pixels, recorded at 30 frames per second. This data is accompanied by annotations in two skeletal schemas: 21 MediaPipe joints and 25 Vicon joints for each hand, with a total file size of 69.5 GB.

The EgoHand model, developed based on NVIDIA Isaac Capture, is capable of reconstructing the three-dimensional pose and shape of the hand from a single first-person video. NVIDIA distributed this model under the OpenMDW license, which permits its free commercial use. According to Noitom's data on NVIDIA's metrics, the model was trained on 14 million images derived from synthetic data, real video recordings, and multi-camera motion capture, demonstrating state-of-the-art zero-shot accuracy on public egocentric benchmarks and operating in approximately 9 milliseconds per frame.

NVIDIA converted Noitom's recordings into its SOMA-Hand representation as one of several real-world sources, meaning HiPHI-Hand is not a complete training set for EgoHand. It is important to note that the licensing of the dataset differs from the licensing of the model itself. HiPHI-Hand uses the ModalityNet Open Research License v1.0, which is free for non-commercial research, education, and evaluation; commercial use requires a separate license from Noitom, and users must request access via the Hugging Face page. This dataset follows HiPHI, another Noitom dataset on object and full-body interaction, which was released on Hugging Face in August.

At Noitom's booth at the CoRL conference, visitors could try on the SOL headset and observe how their hands were reconstructed in 3D in real time using EgoHand. The company also integrated EgoHand into one of its proprietary egocentric vision data pipelines, where it is used during the hand annotation stage.

The accuracy and speed mentioned in the article are data provided by NVIDIA to Noitom, and there is no independent benchmark that isolates the contribution of HiPHI-Hand to these metrics.

Ant Group's Robbyant releases LingBot-Touch model for tactile perception of robotic hands
Read more
pandaily.com

Ant Group's Robbyant releases LingBot-Touch model for tactile perception of robotic hands

Robbyant, a subsidiary of Ant Group, has released the LingBot-Touch model to the public. This fundamental tactile model is designed to provide robotic hands with a unified sense of touch.

Announced on November 10, the model was pre-trained on data collected from nearly 5000 hours of real tactile experience. It covers three sensory principles: magnetic, visual-tactile, and piezoresistive. The model code is available under the Apache-2.0 license, and the Tiny and Small checkpoints are hosted on Hugging Face and ModelScope.

The main problem that LingBot-Touch solves is the incompatibility of raw signals generated by different tactile sensors; touch experience gained from one type of sensor is rarely applicable to another. LingBot-Touch addresses this by providing each sensor type with its own input tokenizer, which feeds data into a common Transformer. The model maps contact points into a single fingertip coordinate space and outputs unified tactile tokens that can be used by various action models.

The repository describes a target approach in the JEPA style, which predicts latent representations from the spatio-temporal context of touch. Training data was obtained from both human-worn tactile gloves and teleoperated dexterous hands.

Testing the model on robotics tasks

Robbyant tested the model on two dexterous hands—Sharpa Wave and XHand—across 26 tasks performed by real robots. These tasks included turning pages, twisting bottle caps, and identifying Mahjong tiles. When adding the LingBot-Touch representation to four action models—LingBot-VLA, $\pi$0.5, Diffusion Policy, and T-Rex—the average task success rate increased by approximately 14–26 percentage points compared to versions without tactile feedback.

In one instance, when T-Rex attempted to twist caps on Sharpa Wave, the success rate was 56.25% without tactility, 62.5% using raw sensor signals, and 90% after integrating LingBot-Touch.

All 26 tests were conducted using the Tiny version, which has about 5 million parameters and increases the size of connected policies by only 0.13%–0.16%. The team notes that the larger Small version continues to improve as the volume of training data increases to 5000 hours.

To verify knowledge transfer across different hardware platforms, the team froze the encoder on XHand, whose magnetic sensing module was absent in the pre-training data, and trained only the action model on XHand demonstrations. Knowledge of magnetic touch in this case was acquired exclusively from human glove data.

Robbyant positions this model as a neutral layer between tactile sensors and action models, rather than a component tied to its proprietary LingBot-VLA. This allows hardware manufacturers to choose sensors based on cost, while system teams can retain their own policies.

All presented results belong to Robbyant itself, and the company has not reported independent reproduction of this data. The release on GitHub is an inference release containing the frozen encoder and utilities for extracting tactile features.

X-Humanoid introduces Hi-WBC model for controlling the entire body of Tiangong humanoid robots
Read more
pandaily.com

X-Humanoid introduces Hi-WBC model for controlling the entire body of Tiangong humanoid robots

X-Humanoid, the Center for Humanoid Robotics Innovation in Beijing, has released the Hi-WBC model—a general whole-body control model for humanoid robots via its embodied intelligence platform. The company announced the model on November 2nd and provided detailed information in a Chinese press release on November 7th. The abbreviation Hi-WBC stands for Whole-Body Control for Humanoid Intelligence.

The Hi-WBC model treats movements derived from human motions not as a script to be reproduced exactly, but as a guideline. X-Humanoid asserts that Hi-WBC integrates these guidelines with robot dynamics and environmental constraints. This allows the robot to maintain the intention of movement while adjusting posture, joint motion, and support based on load, external forces, and the terrain it encounters.

The company described three areas of capability. To continuously add new skills without losing old ones, Hi-WBC utilizes an architecture based on expert mixing and policy distillation: specialized expert policies study specific movements or manipulation tasks, and their capabilities are then transferred to the general policy.

For handling heavy loads and physical interaction, a contact-aware learning approach is employed, which assesses external forces from joint currents and motion states. This data is combined with short-term dynamic optimization and reinforcement learning.

When moving over uneven surfaces, a perceptual control method is used, which analyzes terrain geometry to determine suitable footholds and adjust foot placement, leg swing, and body posture on slopes, stairs, and obstacles. According to the statement, the robot can independently decide how and when to bypass an obstacle before resuming walking.

X-Humanoid demonstrated two videos. One shows the model's application to the results of Tiangong robots in a comprehensive five-event robotics sports competition, where the company stated that the Tiangong robots won gold and silver medals, becoming the only bipedal humanoid robots to win gold in scenarios. The other video shows several robots in one continuous shot tidying up a conference room: they pick up a remote control and clutter on the table, move and return chairs, move a whiteboard, pick up a trash bag, and squat to pick up paper from the floor.

These videos are demonstrations by the company, not proof of commercial implementation, and the release does not specify task success rates, payload limits, or customer availability. X-Humanoid reports that Hi-WBC integrates with the decision-making and multi-robot planning system of its platform and will be improved through execution feedback from real robots.

Popular