Guide for researchers
Use Duckietown to teach, learn, and do research.
I am a researcher
Vision
We designed Duckietown to be a modular, inexpensive research platform for studying autonomy in complex systems.
Think of it as an “experimental simulator” which exposes the nuisances fo the real world while preserving control over the environments.
Value proposition
We provide a baseline implementation for you to be able to rapidly and easily test your algorithms on real physical hardware.
- Convenience: You can only change the part that interests you and use the rest of the baselines to have a fully functional system.
- Reproducibility: Your research has a high impact since it uses a standard platform that others can easily replicate. For maximum impact, compete in an AI Driving Olympics (AI-DO).
Example tools for research
- Imitation learning template
- Reinforcement learning template
- Database of Duckiebot driving logs
- Duckiebot driving simulator
- Modularized code (ROS baseline template)
- Low cost, standardized robots and smart city environment
- An international embodied AI competition (AI-DO)
- A community to bounce ideas off
- A simulation-based system for robotic agent benchmarking
- A physical system for reproducible agent benchmarking (Autolab)
- Here is a partial list of published papers using Duckietown.
Researcher next steps
-
- Join the Duckietown Slack
- Learn about the AI Driving Olympics
- Get the hardware
and start testing or training your algorithms with our templates, logs, and simulator.
Want to build a Duckietown Autolab and evaluate your agents locally, at your institution?
For any questions or doubts, do not hesitate to reach out!
Teddy OrtPh. D. Student, Massachusetts Institute of Technology (MIT)
Manfred DiazPh. D. Student, University of Montreal
Matthew WalterProfessor, Toyota Technological Institute at Chicago (TTIC)