Physical AI News (Japan)

Physical AI, robotics and compute — reported from Japan.

日本語
· Partnerships & M&A· 日本語で読む

HELTEC to Become Japan Distributor for Melt Interface Technologies Data Gloves

The products include data gloves such as the ContactGlove3 Pro and data collection devices. Sales are scheduled to begin in November 2026, with shipments planned to start in January 2027.

Key facts

  • Sales are scheduled to begin in November 2026, with shipments planned to start in January 2027.
  • The ContactGlove3 Pro directly measures the position and orientation of all five fingertips in six degrees of freedom.
  • Under the company’s recommended conditions, median fingertip position error is 0.5 mm and median angular error is 0.4 degrees.

What happened

The products include the ContactGlove3 Pro, announced in June 2026. Sales are scheduled to begin in November 2026, with shipments planned to start in January 2027.

Why it matters

The design records not only fingertip position and orientation but also contact and pinching movements, making it relevant to robot development that seeks to include fine hand operations in training data. HELTEC’s agreement to serve as the domestic distributor paves the way for sales in Japan, but the announced specifications alone do not establish the product’s effectiveness in training or teleoperation.

Analysis

The distribution agreement covers not only the data gloves themselves but also data collection devices. The focus is not on robot hardware, but on a domestic sales network for input devices that measure human finger movements and turn them into data for robot development. The products are said to record contact and pinching movements as well as fingertip position and orientation, distinguishing them by their design for digitizing interactions with objects, not just tracking hand shape. However, the stated error figures are measurements taken under conditions recommended by the manufacturer; they do not indicate the accuracy of operations after training a robot or reproducibility in different environments. The data collection process is envisioned as moving from measuring human finger movements to importing them into development environments through an SDK, combining them with full-body motion capture, and using them for teleoperation or imitation learning. The announcement describes supported environments and potential uses, but does not specify data formats, sampling rates or how the data is converted into motion commands for robots. The ability to collect data is a separate step from being able to use it as is across multiple robots and learning systems, making integration specifications a key implementation issue. Humanoid.guide reported on Booster T2 as a humanoid robot for developers, but that report concerns a robot, while this announcement covers a device for collecting hand data and its distribution. The two cannot be treated as the same product trend, but viewing robot development as involving both hardware and inputs for training data makes the glove’s role limited but specific. With planned dates announced for the domestic distributor’s sales and shipments, the next points to verify are connection procedures in actual development environments, reproducibility under measurement conditions, synchronization of hand and full-body data, and results from use in teleoperation or imitation learning. These details were not provided in the announcement, and the product specifications alone are not enough to assess training effectiveness or reductions in work after adoption.

Details

The ContactGlove3 Pro uses electromagnetic fields to directly measure the position and orientation of all five fingertips in six degrees of freedom. According to the company, under its recommended conditions, the median fingertip position error is 0.5 mm and the median angular error is 0.4 degrees. Measurement coils at each fingertip can also record finger-to-finger contact and pinching movements, and the product is said to include a function that reduces interference from metal. It runs on Linux, macOS and Windows, and has a maximum continuous battery life of 10 hours. The company says combining it with full-body motion capture equipment expands options for collecting training data for the entire body, including the fingers, and for teleoperation.

Implications for Japan

The company announced an agreement to appoint a domestic distributor and plans to begin sales in Japan. The materials provide no information on effects on specific Japanese industries, so no particular advantages or competitive implications are stated.

Sources

Translated from the Japanese original: /news/heltec-melt-interface-technologies-sei-deetaguroobu-no-kokun-45c2c7. The Japanese version is authoritative if they differ.