Mission Analysis
1 Mar 2022

GECCO 2022 Space Optimisation Competition (SPoC)

Here you can find the information you need in order to participate in the GECCO 2022 Space Optimisation Competition (SpOC) organised by the Advanced Concepts Team (ACT).

SpOC GECCo Competition Scenario

Since the dawn of space travel, humans have been dreaming about leaving their provincial corner of the Milky Way and exploring the rest of the galaxy. In the year 3275, more than 100 years after the invention of groundbreaking propulsion technology that enables interstellar travel, this dream is finally becoming reality. With the launch of several sophisticated robotic probes, humans have taken the first steps towards establishing a human settlement on one or more planets in another solar system: Trappist-1!

Storyline

Trappist-1 is a small but curious star system. It is suspected to host as many as seven terrestrial planets, four of which are likely in the habitable zone where liquid water might exist. Needless to say, this is of great interest for our 33rd century explorers. However, water alone is not going to be sufficient to sustain a human settlement. In recognition of that, scientists have prepared a series of missions using automated probes to evaluate the prospect of successful settlement and setting up the groundwork for the arrival of humans. Like all large-scale long-term missions, this endeavour is divided into separate individual phases: a system exploration phase, a resource acquisition phase and a resource delivery phase.

1. System Exploration (Trappist Tour)

The pioneering mission is an exploratory tour of the system that aims to evaluate the most suitable planet (or planets) for the settlers. A probe is sent to Trappist-1 with the explicit task of figuring out the optimal settlement spot. As this would effectively be one of humanity's first deep space exploration efforts, there is still a strong focus on performing the task as quickly as possible while using as little fuel as possible.

2. Asteroid Belt Mining (Mine the Belt)

During the planetary tour, an asteroid belt is discovered in the outskirts of the Trappist-1 system, beyond the orbit of the outermost planet. In a lucky twist of fate, the asteroids turn out to contain a rich source of materials considered essential for settling down, including the raw material used for the exotic propulsion system of the probe itself. Thus, the second phase of the mission is geared towards mining as large a quantity of three basic materials as possible in a short amount of time.

3. Resource Delivery (Delivery Scheduling)

While the first small group of settlers is travelling towards their new home, back on Earth scientists and engineers are increasingly voicing their concern that while the materials mined in the preceding phase will be sufficient for a short-term settlement, long-term establishment in the Trappist-1 system will require a lot more materials and fuel than those mined in situ. Therefore, in the third phase of the mission a number of key asteroids are identified as targets for transfer to an orbit much closer to the planet that is being settled. Special unmanned probes that can withstand the effects of titanic acceleration are launched with the task of overtaking the settlers and dispatching as many asteroids to a new orbit within the short time before humans arrive at their destination. This is a time-critical phase, so no effort and resources are spared to complete this in a short amount of time.

Scope of the Competition

Welcome, SpOC competitors! This is your time to put your great minds to work in order to ensure humanity's first successful settlement effort. Your task is to solve as many as three optimisation challenges related to the design of all three phases of the settlement effort. Starting from 1 April 2022, you will have three months to solve these challenges and compete for first place on the overall SpOC leaderboard. Detailed technical descriptions for the three challenges to be solved will be made available on the the European Space Agency's Optimize platform from 1 April 2022. Good luck, and may the best team win!

Competition rules

  • Participants will propose and implement metaheuristic algorithms to solve three complex optimisation problems.
  • We will provide validation code for each of the three problems, implemented in Python, in order for the participants to test and validate their solutions. The code will include problem definitions using the Pygmo user-defined problem format, examples of solutions and visualisation tools.
  • Participants will have up until 30 June 2022 to submit their entries to the competition via the dedicated portal Optimize.
  • The top ten entries for each of the three problems will be awarded points on the overall SpOC leaderboard according to their performance. The best team will be awarded ten points, the second nine points ... up until the tenth-best team which will be awarded one point.
  • The team with the most points on the overall SpOC leaderboard at the end of the competition will be declared the winner and announced on July 1st 2022.
  • The ACT reserves the right to exclude, at any point in time, users from the competition should they not comply with our basic code of honour (e.g., by abusing the evaluation system, use of multiple users accounts, etc..).

Timeline

NOTE: The submission portal will remain open after 30 June 2022. Submissions received after that date will not be taken into consideration for the competition, but they will still appear on the leaderboard.


How to submit

  • The competition will be hosted on the new Optimize platform developed by the Advanced Concepts Team. Teams will need to register online on the platform, and solution entries will need to be submitted via Optimize for validation.
  • Join our Gitter community to stay updated on the competition, interact with us and with the other participants: https://gitter.im/esa/spoc
  • Submission of papers or attendance to GECCO 2022 is not mandatory for the participants.

Prizes

We are in discussions with the organisers of GECCO to have a dedicated session for the best teams to showcase their work during the conference (9-13 July 2022). Stay tuned for more information soon!

CLAIRE (Confederation of Laboratories for Artificial Intelligence Research in Europe) sponsors the conference registration and/or travel costs of 3 students participating in the competition (up to €1000 per student) from any of the over 400 labs and institutions that form the CLAIRE Research Network or the Rising Researcher Network. Students applying for the CLAIRE sponsorship must fill out the dedicated application form.


Contact

All contacts and question regarding the competition should be put through our official gitter channel:

Outcome

Artificial Intelligence Conference paper
Optimization challenges at the European Space Agency
Izzo, D. and Manuel López-Ibáñez, M.
GECCO '22: Proceedings of the Genetic and Evolutionary Computation Conference Companion: 1542--1553
(2022)
Download
BibTex
Hamburger icon
Menu
Advanced Concepts Team