FastLabel Releases OpenLUTRA, an Open-Source Application for High-Quality Data Collection in Physical AI

FastLabel Inc., a company supporting the development of robot foundation models and domain-specific VLA (Vision-Language-Action) models, announced the open-source release of 'OpenLUTRA,' an application that enables efficient, large-scale, and high-quality data collection required for AI robotics development. OpenLUTRA is an application for recording and quality analysis of data necessary for building AI models for robot control, including VLA models. Through an intuitive UI/UX that is easy for operators at data collection sites to use, and an automatic curation function, it strongly supports the expansion of data collection scale and the improvement of throughput for high-quality data construction. FastLabel released this tool to solve issues occurring at data collection sites, such as the inclusion of data that does not meet requirements, data delays or loss, and low operability for non-engineers. Key features include real-time anomaly monitoring, automatic quality assessment, an intuitive UI/UX, and compatibility with standard robot development environments such as ROS2. The license is Apache License 2.0, allowing for commercial use, and the rights to the data obtained belong to the user. It is currently in beta, with a full release planned within a few months.