FrodoBots AI develops browser-based tele-operation gaming platforms that allow users globally to control low-cost physical robots in real-world scenarios. These gaming experiences generate valuable tele-operation datasets used to train and advance embodied foundational models for robotics and navigation. The platform incentivizes community participation through tokenomics to expand the distributed network of accessible robotic hardware.
Funding
$6M raised to dateRaised to date based on public sources. This may differ from the amount the company actually raised and is based only on what is publicly available on the internet.
Founders
Product
Problem
Developing robust AI-powered robotic systems requires vast amounts of real-world training data, which is traditionally expensive and time-consuming to collect. Existing methods often lack scalability and struggle to capture the diversity of environments necessary for effective robot learning.
Solution
FrodoBots AI addresses this challenge by gamifying the data collection process, enabling efficient and scalable acquisition of robotics training data. Through engaging games featuring tele-operated robots, FrodoBots incentivizes users to contribute real-world driving data. This crowdsourced approach generates diverse datasets that can be used to train and improve AI navigation and manipulation models for robots. The platform leverages low-cost robots and a low-latency network system to allow gamers from anywhere to participate in robotic gaming titles, effectively turning tele-operations into a fun and rewarding experience.
Target Audience
The primary audience includes AI researchers and robotics developers seeking large, diverse datasets for training embodied AI models, as well as gamers and robot enthusiasts interested in tele-operated robotic gaming.
Features
- Earth Rovers: A global scavenger hunt game where users remotely control sidewalk robots in real-world environments.
- Octo Arms: A 3D puzzle game (coming soon) involving remote control of robotic arms for manipulation tasks.
- Low-latency network system and browser-based controls for accessible remote operation.
- Token-incentivized system to reward users for contributing valuable training data.
- Open-source SDK for researchers to deploy and test AI navigation models on the Earth Rovers platform.
- Datasets of human tele-operated drives collected from various cities, available for training AI models.
- Earth Rovers School: An open-world game where drivers can train and earn rewards by completing missions.
- 4G-enabled robots with front and back cameras, GPS, and IMU for comprehensive data collection.