Microsoft explores machine learning in space with AIML

machine learning

Nicholas Moretti and Professor Tat-Jun Chin. Image credit: University of Adelaide.

Microsoft has signed a MoU with the University of Adelaide’s Australian Institute for Machine Learning (AIML) to explore how advanced cloud computing, AI, computer vision and machine learning can be used in space. 

“Project AI Off Earth” will model, emulate and simulate complex space operations and systems, and build algorithms for on-board satellite data processing. It will develop solutions to remotely operate and optimise satellites, constellations and swarms, and address space domain awareness and debris monitoring.  

AIML will bring experience in applying AI, computer vision and machine learning to real world problems. Microsoft’s experience in advanced cloud computing and cognitive systems is informing Azure Space. Azure Space is being built as a set of cloud offerings that allow organisations to leverage geospatial data, access bandwidth anywhere, digitally engineer space systems and engage in remote edge computing. This can all be applied in space.  

“The relationship with Microsoft will give us access to cloud-based platforms that will allow us to focus on the investigation on the performance of algorithms, used to analyse large amounts of earth-observation data from satellites without needing to be concerned about gaining access to space at the onset,” University of Adelaide professor Tat-Jun Chin said. 

“Our work on these algorithms has the potential to contribute to many applications, including agricultural land management, water management, mining practices and understanding of economic activity.” 

Microsoft Australia Azure Space engineer Nicholas Moretti, is now based at Adelaide’s innovation precinct, Lot Fourteen. 

“I first got exposed to the space industry while I was studying for my undergraduate degree at the University of Adelaide and crossed paths with Professor Chin,” Moretti said. 

“Although focused on in-space technologies, Project AI Off Earth will explore how space related technologies and data, and cognitive systems can be used to support automation of multiple different industries, help establish smart cities, as well as address sustainability and important environment challenges.” 

Microsoft and AIML are already collaborating using Microsoft Azure Orbital Emulator, a cloud-based native space emulation environment that enables massive satellite constellation simulations. Using this, AIML and Project AI Off Earth can develop, evaluate and train algorithms, machine learning models and AI intended for space without the need to launch a satellite. 

“Adelaide has established itself as the very heart of Australia’s space industry,” SA minister for Trade and Investment Stephen Patterson said. 

“This agreement between AIML and Microsoft, which is building a space team, is a signal of what’s to come. Australia has the opportunity to be a leading player in the global space industry and this sort of international collaboration – centred on Adelaide but with a truly global focus – will strengthen the local industry, help build skills in this important area and reinforce Adelaide’s reputation as the epicentre of space activity in this part of the world.” 

Moretti and Chin discuss the new partnership between Microsoft and the University of Adelaide’s AIML in the video below.

Leave a Reply