Navigation with Dynamic Obstacles
Note: This challenge is not present in AI-DO 1; it will be present in AI-DO 2 (ICRA 2019).
In this task a Duckiebot should drive from its current position to another position in the city with other Duckiebots and obstacles present. Performance is evaluated on speed, how well the Duckiebot adheres to the rules of the road and how much its driving oscillates.
Note: this challenge is already online but it is technically part of AI-DO 2 (ICRA 2019).
Imagine a Mega-Duckietown with 100’000 bots. You are in charge to control them, all alone. You command and the taxis obey. How will you orchestrate the fleet of robotic taxis?
Take a seat in the control center of AMoDeus, choose an existing city and command the fleet. What waiting times can you achieve? Can your fleet sustain the city’s hunger for travel? Is your fleet efficient enough to survive the contest? AMoDeus awaits your algorithms.
Participants are officially taking part in the competition when submitting at least one public entry to one of the tasks described above.
Upon enrollment, participants can submit their code in the form of a docker container to a task of the AI-DO. Detailed instructions and examples of conforming code will be made available via the competition website. The system will schedule to run the code on the cloud on the challenges selected by the user, and, if simulations pass, on the robotariums.
Participants can submit entries as many times as they would like. Access control policies are to be implemented, should certain participants monopolize the computational and physical resources available.
Participants are required to open source their solutions source code. If auxiliary training data are used to train the models, that data must be made available.
Submitted code will be evaluated in simulation and if sufficient on physical robotariums. Scores and logs generated with submitted code will be made available.
Need-based donation programs
The Duckietown Foundation administers a need-based program for obtaining free equipment and AWS credits for preparing for the competition. See here for terms and conditions and the application form.
Prizes and Rewards
Thanks to the 2 sponsors/co-organizers Amazon Web Services and nuTonomy (part of Aptiv Autonomous Mobility) we are able to offer the following prizes:
– 7 trips to a nuTonomy/Aptiv development office (Boston or Singapore, whichever is easier to reach). You will be able to visit the offices, talk with junior and senior engineers, ride in our self-driving cars together with the developers.
– $9500 in AWS credits.
We will divide the prizes among the finalists in the four challenges, plus there are going to be 3 “special prizes”: the nuTonomy prize, the AWS prize, and the Duckietown prize, given to innovative and elegant solutions even if they don’t make it to the top of the leaderboard.
After the competition in the beginning of 2019, we can ask you to contribute a paper to a forthcoming book detailing the lessons learned and successful approaches in the AI Driving Olympics.
Employees and affiliates of nuTonomy and Amazon AWS are ineligible from winning prizes in the competition. Employees and affiliates of nuTonomy and Amazon AWS may submit baseline solutions that will be reported in a special leaderboard.
Students of ETH Zurich, Georgia Tech, NCTU, Tsinghua, UCLA, TTIC, are eligible to participate in the competition as part of coursework, if they do not work in the organization of the competition.
For questions and discussions please visit the AI-DO discussion forums.