{"id":21922,"date":"2020-05-28T17:47:42","date_gmt":"2020-05-28T21:47:42","guid":{"rendered":"https:\/\/hobbyspace.com\/Blog\/?p=21922"},"modified":"2020-05-28T17:47:42","modified_gmt":"2020-05-28T21:47:42","slug":"esa-sponsors-ai-competition-to-spot-geo-objects-in-low-cost-telescope-images","status":"publish","type":"post","link":"https:\/\/hobbyspace.com\/Blog\/?p=21922","title":{"rendered":"ESA sponsors AI competition to spot GEO objects in low-cost telescope images"},"content":{"rendered":"

ESA sponsoring a new citizen science challenge at Kelvins- “ESA’s Advanced Concepts Competitions”<\/a>. In the spotGEO Challenge<\/a>, held in partnership with the\u00a0 University of Adelaide<\/a>,<\/p>\n

… you are asked to develop an algorithm to detect Geostationary orbiting objects from simple png images (or frames) acquired by an unknown, low-cost ground-based telescope. Can you learn on how to cope with cloud cover, atmospheric\/weather effects, light pollution, sensor noise\/defects, star occlusions and more?<\/em><\/p>\n

Check out the problem description<\/a>, the dataset<\/a> and the starter-kit<\/a> (the dataset and the starter kit will only be available at the competition start date).<\/em><\/p>\n

The start date is June 8, 2020. Here’s the official announcement from ESA:<\/p>\n

Telescope-peering AI challenged to spot mystery space objects<\/strong><\/a><\/p>\n

\"\"<\/a>
“Artificial objects in or near geostationary orbit appear stationary or nearly stationary compared to the background stars, which appear to move due to Earth’s rotation, as shown in this sequence of observed images.” Credits: ESA\/Univ. Adelaide<\/a><\/figcaption><\/figure>\n

ESA\u2019s latest public competition challenges \u2018citizen scientists\u2019 to combine AI with observations from low-cost telescopes to pick out mystery objects in and around geostationary orbit, thousands of kilometres above Earth.<\/p>\n

Geostationary orbit is also known as the \u2018Clarke belt\u2019 \u2013 science fiction writer Sir Arthur C Clarke forecast it back in 1945. The further up that satellites orbit, the slower they need to travel to overcome Earth\u2019s gravity. Orbiting at approximately 36 000 km altitude directly above the equator, satellite velocity precisely matches Earth\u2019s rotation, enabling them hover above the same spots in the sky.<\/p>\n

The result has been called the most valuable real estate in our solar system: a 265 000 km ring of telecommunications, meteorology and other satellites around our planet, carefully regulated by the International Telecommunication Union.<\/p>\n

Despite its economic value however, geostationary orbit \u2013 as well as adjacent \u2018geosynchronous\u2019 orbits \u2013 must contend with the same problems of space debris also seen in lower orbits. ESA and other space agencies perform regular monitoring to identify and track potentially-hazardous debris items. This is usually done using either high-power radar or high-performance astronomical telescopes.<\/p>\n

\u201cGeostationary orbit is generally well managed and documented, partly because of its immense practical and commercial value,\u201d notes Tat-Jun Chin<\/a> of the University of Adelaide<\/a>, partnering with ESA on the competition. \u201cHowever, precisely because of that value we should put more efforts into further understanding and protecting it.\u201d<\/em><\/p>\n

Dario Izzo of ESA\u2019s Advanced Concepts Team (ACT<\/a>) adds:<\/p>\n

\u201cSo, for our new \u2018spotGEO<\/a>\u2019 competition, we want to see how well low-cost telescopes combined with tailored AI algorithms can identify \u2018resident space objects\u2019 at these altitudes.\u201d<\/em><\/p>\n

Competition entrants will receive a dataset made up of sets of five sequential images of unspecified segments of the geostationary belt, then challenged to pick out artificial objects against the surrounding stars.<\/p>\n

\"\"<\/a>
This small telescope at the Univ. of Adelaide was used to “gather imagery of geostationary artificial objects for ESA’s GEOspot competition”. Credits: Univ. Adelaide\/ESA<\/a><\/figcaption><\/figure>\n

In theory this is made easier because such objects will remain static (or nearly static) compared to the background starfield, which appears to move because of Earth\u2019s rotation. In practice, with atmospheric distortion and an approximately 40-second exposure time for each single image the objects will be smeared out and dimmed. Clouds, light pollution and sensor noise also add to the challenge.<\/p>\n

\u201cThe sheer distance between the observer and the target objects makes this a difficult problem,\u201d adds Dario.<\/em><\/p>\n

\u201cEach pixel observed at this altitude corresponds to an arc length of about 800 m \u2013 so the objects of interest are much smaller than a single pixel. But success should help us keep better watch on this essential region of space around our planet.\u201d<\/em><\/p>\n

Tat-Jun Chin and his team made contact with the ACT after winning the Pose Estimation Challenge<\/a>, their previous space-themed AI competition, on estimating the orientation of distant satellites from a dataset of still images.<\/p>\n

\u201cDeep learning algorithms can be trained through such datasets to detect visual features of interest,\u201d he notes. \u201cResearchers in AI \u2013 particularly computer vision and machine learning \u2013 understand that having common datasets is vital towards making progress. These allow different methods to be compared objectively, so that the community can learn the best practices then apply them for their respective problems.<\/em><\/p>\n

\u201cGenerally speaking, sharing datasets in space research is not so common, but the excellent Kelvins competitions are changing this, and after getting to know the ACT we decided to contribute our own.\u201d<\/em><\/p>\n

The University of Adelaide team coincidentally acquired these images during an observing campaign during the last Australian summer, so that forest fire smoke adds to the observing difficulty.<\/p>\n

This is the latest competition hosted at the ACT\u2019s Kelvins website, named after the temperature unit of measurement \u2013 with the idea that competitors should aim to reach the lowest possible error, as close as possible to absolute zero. The spotGEO dataset will be available there<\/a> from 8 June, at the start of the three-month competition.<\/p>\n

===<\/em> Amazon Ad <\/em><\/strong>===<\/em><\/p>\n

Xtronaut: The Game of Solar System Exploration<\/a>\"\"<\/strong><\/p>\n