build details

Show: section status errors & todos local changes recent changes last change in-page changes feedback controls

The AI Driving Olympics

Modified 2021-10-31 by tanij

The AI Driving Olympics (AI-DO) is a set of competitions with the objective of evaluating the state of the art in machine learning and artificial intelligence for mobile robotics.

For a detailed description of the scientific objectives and outcomes please see our paper about the AI-DO at NeurIPS.

The video is at

The AI Driving Olympics at ICRA 2020


Modified 2021-10-29 by liampaull

  • AI-DO 1 was in conjunction with NeurIPS 2018.

  • AI-DO 2 was in conjunction with ICRA 2019.

  • AI-DO 3 was in conjunction with NeurIPS 2019.

  • AI-DO 4 was supposed to be in conjunction with ICRA 2020, but was canceled due to COVID-19.

  • AI-DO 5 was in conjunction with NeurIPS 2020.

  • AI-DO 6 is in conjunction with NeurIPS 2021.

Where it all started: AI-DO 1 at NeurIPS 2018 in Montreal.


Modified 2021-10-31 by tanij

There are currently three leagues in the AI Driving Olympics.

The Urban League is based on the Duckietown platform, and includes a series of tasks of increasing complexity. For each task, we provide tools for competitors to use in the form of simulators, logs, code templates, baseline implementations and low-cost access to robotic hardware. We evaluate submissions in simulation online, on standardized hardware environments, and finally at the competition event.

Participants will not need to be physically present at any stage of the competition — they will just need to send their source code.
There will be qualifying rounds in simulation, similar to recent DARPA Robotics Challenges, and, for evaluation, we make available the use of “Duckietown Autolabs (unknown ref opmanual_autolab/book)

warning next (1 of 18) index
I will ignore this because it is an external link. 

 > I do not know what is indicated by the link '#opmanual_autolab/book'.

Location not known more precisely.

Created by function n/a in module n/a.
” which are facilities that allow remote experimentation in a reproducible setting.

See the leaderboards and many other things at the challenges site.

The Advanced Perception League is organized by Motional (ex nuTonomy, Aptiv Mobility).

All information about the Advanced Perception League is at

The Racing League is organized by the AWS Deepracer team. All information about the racing league is available on

What’s new in the Urban League in AI-DO 6

Modified 2021-10-31 by tanij

There have been cool new improvements for the 6th edition of the AI-DO Urban League:

How to use this documentation

Modified 2021-10-31 by tanij

If you would like to compete in the AI-DO Urban League, you will want to:

At this point you are all set up: your environment is operational, and you can make a submission. But you should want to make your submission perform better than the provided baselines.

To do this the following tools might prove useful:

  • The AIDO API so that your workflow is efficient using the available tools.
  • The reference algorithms where we have implemented some different approaches to approach the challenges.

How to get help

Modified 2021-10-31 by tanij

If you are stuck try one of the following things:

  • Look through the contents of this documentation using the links on the left. Note that the “Parts” have many “Chapters” that you can see when you click on the Part title,
  • Join our slack community,

  • Look on the Duckietown Stack Overflow to see if someone already answered your question (you can ask to be invited in the Slack channel #help-accounts)

  • If you are sure you actually found a bug, file a Github issue in the appropriate repo.

How to cite

Modified 2021-10-29 by liampaull

If you use the AI-DO platform in your work and want to cite it please use:

  title={The AI Driving Olympics at NeurIPS 2018},
  author={Julian Zilly and Jacopo Tani and Breandan Considine and Bhairav Mehta and Andrea F. Daniele and Manfred Diaz and Gianmarco Bernasconi and Claudio Ruch and Jan Hakenberg and Florian Golemo and A. Kirsten Bowser and Matthew R. Walter and Ruslan Hristov and Sunil Mallya and Emilio Frazzoli and Andrea Censi and Liam Paull},
  journal={arXiv preprint arXiv:1903.02503},

If you use the Duckietown platform in your work and want to cite it please use:

    author={Paull, Liam and Tani, Jacopo and Ahn, Heejin and Alonso-Mora, Javier and Carlone, Luca and Cap, Michal and Chen, Yu Fan and Choi, Changhyun and Dusek, Jeff and Fang, Yajun and Hoehener, Daniel and Liu, Shih-Yuan and Novitzky, Michael and Okuyama, Igor Franzoni and Pazis, Jason and Rosman, Guy and Varricchio, Valerio and Wang, Hsueh-Cheng and Yershov, Dmitry and Zhao, Hang and Benjamin, Michael and Carr, Christopher and Zuber, Maria and Karaman, Sertac and Frazzoli, Emilio and Del Vecchio, Domitilla and Rus, Daniela and How, Jonathan and Leonard, John and Censi, Andrea},
    booktitle={2017 IEEE International Conference on Robotics and Automation (ICRA)}, title={Duckietown: An open, inexpensive and flexible platform for autonomy education and research},