Group 1 - The core issue of data quality is a bottleneck for the application of embodied intelligence in humanoid robots, with the industry moving towards standardization and quality evaluation [1] - Zhiyuan Robotics has obtained the first CR certification for humanoid robot datasets, indicating a significant step in ensuring product quality and safety in the robotics market [1][2] - The CR certification evaluates four core elements of humanoid robot dataset construction: classification and coding standards, data labeling standards, unified dataset format requirements, and comprehensive quality evaluation criteria [2] Group 2 - The AgiBot World dataset released by Zhiyuan Robotics has achieved breakthroughs in both scale and quality, covering a wide range of tasks from basic operations to complex long-range tasks [2][3] - The dataset is designed based on a self-built data collection factory and application laboratory, replicating five core scenarios: home, dining, industrial, supermarket, and office, with hundreds of real sub-scenes and over 3,000 real items [3] - The successful certification of the AgiBot World dataset positions it as a model for standardization in humanoid robot datasets, with future plans to promote a "standardized, open, and efficient" data co-construction and sharing mechanism in the field of embodied intelligence [3]
编号“001” 智元获业内首张人形机器人数据集CR认证