Introduction
RoboX is the first open source data platform for robotics training. It turns smartphones into live, privacy-safe sensors that continuously capture environmental signals such as motion, light, and location context. These streams of anonymized data feed into a global dataset used to train perception models for robotics, navigation, and spatial intelligence.
Instead of deploying expensive fleets of lidar vans or stationary sensors, RoboX distributes the task across people already in motion. Every phone that joins a mission becomes a small but meaningful part of a global sensing grid. The more contributors there are, the more complete and accurate the dataset becomes, improving how autonomous systems understand roads, buildings, and public spaces.
Once the app is installed, it works quietly in the background, collecting anonymized signals while you go about your day, walking, driving, or simply carrying your phone.
Effortless by Design
RoboX is built to work quietly in the background, collecting anonymized signals while you go about your day, walking, driving, or simply carrying your phone.
Join a mission in your region.
Carry your phone. RoboX records data automatically while you move.
Data is anonymized before upload; nothing personal ever leaves the device.
Verified contributions are rewarded with RoboX Points, which later convert to RoboX Tokens (ROBOX).
It’s designed to be invisible, you simply live your routine while your phone contributes to something larger.
Privacy at the Core
RoboX was built to protect people as much as it empowers machines. All personal identifiers, like GPS, device IDs, or visual content, are stripped away before data ever leaves the phone. Everything runs locally and securely.
No faces, voices, or personal visuals are ever recorded.
Data is encrypted and anonymized before being sent.
You can delete your contributions anytime.
This privacy-first model allows RoboX to scale globally without compromising individual safety or trust.
Collective Impact
By joining RoboX, people everywhere can take part in shaping how the next generation of robots learns about the world. Each small movement adds up to a massive, verified dataset: open, transparent, and built for real progress.
Last updated