25/03/2024
9 views
0 likes
More than two hundred years into the future, our descendants contemplate creating the largest single structure in human history for the next evolutionary leap: a multi-generational starship capable of bringing people to the first truly Earth-like exoplanet. Yet this interstellar ark – to be self-assembled out of builder units in Earth orbit – will be sufficiently complex as well as vast that even designing it involves formidable mathematical challenges. And this odyssey needs to be preceded by a mammoth astronomy effort to prospect the way ahead, involving a formation of orbital telescopes able to operate together as one, yielding combined images of humankind’s new destination equivalent in resolution to one single giant lens.
The above sci-fi scenario is the throughline of three ‘optimisation problems’ making up ESA’s latest Space Optimisation Competition (SpOC) involving finding the best solutions out of many available options rather than there being a single ‘right’ answer. It will shortly be thrown open to competing research groups around the globe.
This is the third in the annual series of SpOC challenges established by ESA’s Advanced Concepts Team (ACT) in conjunction with the annual Genetic and Evolutionary Computation Conference, GECCO, which takes place this year in Melbourne, Australia, (and online) this July.
“There’s a thriving academic community of people applying evolutionary computation to all kinds of complex optimisation problems,” explains Dario Izzo, coordinating the ACT. “With these challenges we’re seeking to extend a bridge from the space sector, to encourage them to apply what are often very useful algorithms to space-related challenges. Our science fiction setting works to grab the imagination of the community – and we have attracted a lot of contestants to our previous SpOCs – but the core problems are inspired by real scientific issues being looked at in space research.”
Biological evolution has proven to be an extraordinarily powerful problem-solving mechanism during the approximately 3.7-billion-year history of terrestrial life on Earth, succeeding in populating almost all conceivable environmental niches with living things.
Evolutionary computing seeks to borrow this principle to solve complex optimisation problems using software that mutates, mates then reproduces to best fit the solution. The most useful responses are selected to pass their traits onto the next generation in turn. This basic technique has been applied to everything from antenna design to financial trading, power system management to software fault detection.
ACT research fellow Max Bannach, organising this year’s SpOC, explains: “In our contest we have three problems for researchers to apply their methods to. The first problem involves breaking down the construction plan of the megastructure into tractable chunks. Within the competitive context, the Graph Reduction Algorithm for Planetary-scale Hyperoptimization (GRAPH) project models this absurdly complex endeavour via an equally complex mathematical structure.
“The second is to organise a swarm of telescopes in Earth orbit so that they can work together to perform ‘interferometry’ – acquiring combined images with resolution equivalent to one single giant telescope, in the same way that arrays of radio telescopes operate today. Known as the Golomb Ruler Advanced Interferometric Lens, GRAIL, this project is based on a contemporary proposal for optical telescopes, but demands complex yet precise orbital formation flying.
“Finally, when it comes to building this megastructure space ark, the construction work will be done by a huge quantity of smart cubes that will assemble themselves accordingly, but require an algorithm to organise their self-assembly process – the Orbital Assembly of Self-organising Interstellar Spacecraft, or OASIS.”
Any research group is free to enter this year’s SpOC contest. Formally beginning on 1 April, the competition will last for three months. For more information on the SpOC competition, including how to register, click here.