National-level Platform Launches First Embodied Intelligence Open Source Dataset Community, Globally Releases OpenLET Full-Body Motion Control Dataset

At the Open Atom 'Park Tour' open source dataset special session, the first embodied intelligence open source dataset community initiated by a national-level platform was officially launched. Founded by the Open Atom Open Source Foundation and co-built with core industry-university-research units, this community aims to establish an industry-level data collaboration mechanism, formulate open source governance frameworks and quality evaluation standards for AI datasets, and promote standardized circulation and sharing of data.

The OpenLET 'Tactile Dexterous Operation + Full-Body Motion' dataset was globally released at the event. It is the world's first multi-modal embodied intelligence dataset integrating tactile dexterous operation and full-body high-dynamic motion, and also the first domestic open source of multi-modal real-machine data for full-size humanoid robot full-body motion control. The dataset includes dexterous operation data (tactile array, 6D force, RGB-D vision) and full-body motion control data (41 joint control signals covering high-difficulty linked tasks), filling industry gaps.