‘A new way of ‘doing space:’ AIML-supported Space MAITRI mission part of historic launch agreement
Earlier this year, Space Machines Company, an Australian-Indian in-space servicing firm, signed a landmark Launch Service Agreement with NewSpace India Limited (NSIL), the commercial arm of the Indian Space Research Organisation (ISRO).
This collaboration sets the stage for the launch of Space Machines Company’s second Optimus spacecraft weighing 450kg, the largest Australian-designed and built spacecraft so far. The craft is scheduled for launch in 2026 aboard NSIL’s Small Satellite Launch Vehicle (SSLV) in a mission entitled the Space MAITRI मैत्री (Mission for Australia-India’s Technology, Research and Innovation) which is also Sanskrit for ‘friendship.’
“Space MAITRI’s focus is around ‘orbit servicing,’ said Professor Tat-Jun Chin, the SmartSat CRC Professorial Chair of Sentient Satellites for the Australian Institute for Machine Learning (AIML).
‘Instead of creating spacecraft for passive data collection, we’re creating platforms to do useful things such as cleaning space debris, repairing other devices in space and other services,’ said Professor Chin. ‘It’s a new way of ‘doing’ space.’
Professor Chin notes that artificial intelligence could play a key role in creating space tools that could support in-space construction of space craft, among other uses. He and his team took multiple images of spacecraft in order to measure various factors and ensure the mission’s success.
‘AIML used AI-enabled spacecraft perception to determine how to approach spacecraft safely,’ said Professor Chin. ‘Spacecraft perception is when two spacecraft come into proximity with each other; an algorithm allows them to figure out how to approach each other safely. It also allows for the development of autonomous on-orbit servicing systems (OOS).’
‘AIML members took multiple photos of space craft in order to influence spacecraft perception as well as see what we can learn about a spacecraft’s trajectory, intent, functionality, pose (a spacecraft’s relative position and pointing direction), shape and other factors,’ he said. ‘Using AI allows us to analyse the photos very quickly. It can take a few seconds using AI.’
The ability of AI systems to accurately determine the pose of satellites, spacecraft, orbital debris, and asteroids is crucial for autonomous spacecraft, or in missions where direct human control is costly and/or error prone.
Members of the AIML team responsible for taking and analysing the images included AIML Researcher Dr Marcus Martens, and post-doctoral researcher Dr Yasir Latif along with Professor Chin. Professor Chin noted that there was considerable interest in this mission due to its more cost-effective and commercially viable methods.
‘This is Space 2.0,’ he said. ‘We’re learning that much can be achieved even with smaller budgets. We don’t need the big companies or government agencies like we did before. We can use smaller enterprises to create sustainable space operations that are just as effective.’