Axis Robotics completes 10,000 crowdsourced trajectory collections, validating a robot data expansion model without hardware investment

ChainCatcher news: The startup robotics company Axis Robotics announced that in an experiment codenamed “The Little Prince’s Rose,” they mobilized 12,000 users worldwide through their self-developed browser platform to complete 10,000 high-quality trajectory data crowdsourcing within 5 days. The experiment demonstrated that robot intelligence can be built without traditional expensive laboratory environments, achieving large-scale data explosion through mass participation.

Relying on the MetaSim architecture, Axis Robotics has created an end-to-end pipeline from task setup to real machine deployment, decoupling complex physical simulation from lightweight web interaction to ensure the collected data has deep processing capabilities. The platform uses cross-simulation replay technology to replay high-frequency physical data from MuJoCo in Isaac Sim and perform large-scale domain randomization to enhance data diversity.

Currently, the official has successfully reproduced the real machine on a Franka robotic arm based on this data provided by ordinary internet users, verifying the feasibility of this data collection model for industrial applications.