Guide morphing rovers across an alien world in an evolutionary computer race

In any case, that’s the scenario of ESA’s new Space Optimization Competition (SpOC), designed to test the abilities of participants around the world.

“Last year’s inaugural challenge involving the settlement of Trappist exoplanets proved to be a great success,” commented Dario Izzo of ESA’s Advanced Concepts Team, ACT. “The setting worked well, making the competition attractive and interesting for evolutionary computing faculty and students. This time we set out to come up with an even crazier sci-fi scenario, while at the same time incorporating tangible challenges for the kind of work we do here at ESA.”

Biological evolution has proven to be an extremely powerful problem-solving mechanism during the approximately 3.7 billion-year history of terrestrial life on Earth, succeeding in populating almost every possible ecological niche with living things.

Evolutionary computers seek to borrow this principle to solve complex “optimization problems”—meaning how to find the best answer out of the many available options—using software that mutates, binds, then reproduces itself to best fit good on the decision. The most useful responses are selected to pass their traits on to the next generation in turn. This basic technique has been applied to everything from antenna design to financial trading, power system management to software error detection.

Any evolutionary computing research group is free to enter this year’s SpOC competition. The competition officially starts on April 1 and will run for three months. The winners will be announced at this July’s Genetic and Evolutionary Computation Conference, GECCO, in Lisbon, Portugal.

Organized by ESA’s ACT and hosted by GECCO, SpOC 2023 involves three specific optimization challenges, explains ACT researcher Emmanuel Blasquez: “The scenario takes place in a future where humans have spread across many different solar systems. They seek to meet on a single planet designated for exploration using a network of wormholes that teleport them to various nodes through time and space, aboard motherships that aim to arrive at roughly the same time . So the first challenge is to get them there in a synchronized and efficient way.

“The second challenge involves creating a quantum communication satellite network to connect both the entire surface of the planet New Mars itself and the various human faction motherships located around it, which will require complex planning for optimal efficiency.

“Finally, systematic exploration of the planet will be carried out by AI rovers that are able to plan their onward paths and change their morphology to cope with certain planetary terrains – for example, grow legs to cross chasms . The idea is to plan the optimal survey solution, bearing in mind that there will be trade-offs involved that will limit shape change too much. And for added realism and motivation, this challenge will be based on actual terrain maps of the planet Mars.”

Designing SpOC involved taking lessons from last year’s event, adds Dario: “The overall feedback was very positive, but some people felt that the aerospace angle was too strong, that the mission analysis and trajectory design aspects favored the aerospace teams. domains. So this year we aimed to come up with problems that were more general in nature but still related to the kind of real-world problems that ACT and ESA are grappling with.”

For more information on the SpOC competition, including how to register, click here.