Next to the Code of Honor the following rules apply to the competition.
The Collision Avoidance Challenge is open to everyone, including space experts, data-mining experts and citizen scientists.
No nationality restrictions apply for the submission of solution files.
No, there is no money for the winner. We hope that the chance to improve the safety of operations of satellites in Earth orbit will keep your motivation high.
The winner is determined by the lowest score after the deadline. The final ranking is based on the re-evaluation over the full test set of two of each participant's submissions: the last one, and the submission that ranked best on the public leaderboard. If they are the same, we take also the second-best ranked on the public leaderboard .
In case of a tie, the earlier submission date wins.
The final decision on the winner is subject to whether the participants conformed to the rules, as determined unilaterally by ESA's Advanced Concepts Team.
The winner will be invited to visit the European Space Operation Center (ESOC) in Darmstadt, Germany to discuss the results with experts from ESA's Advanced Concepts Team and ESA's Space Debris Office.
According to the quality of the results, other teams may also be invited to discuss their findings at ESOC/ESTEC.
According to the quality of the results, the winner and possibly other teams will be invited to write a scientific publication with the competition organizers.
By submitting, the authors grant to ESA's Advanced Concepts Team a license to use the submission files for the purpose of verifying and publishing the challenge results.
The use of additional public data or data augmentation for training of machine learning models is allowed.
Before the competition deadline, solutions, notebooks, code snippets may only be shared publicly and should be shared in the discussion page section. Private sharing between teams is not allowed.
The submission limit is two submissions within 24 hours per team.
The leaderboard during the competition uses only a part of the submission data, as an attempt to avoid overfitting on the test data.
The final ranking is based on the re-evaluation over the full test set of two of each participant's submissions: the last one, and the submission that ranked best on the public leaderboard. If they are the same, we take also the second-best ranked on the public leaderboard .