Space technology has always been a fascinating topic that encompasses various fields of study. The current major space mission, Artemis II, to fly by the moon, has gripped the world’s attention. However, the amazing advancements in the field come with their own set of challenges now. Space has become congested and resource-constrained, which has now made the need for efficient and sustainable spacecraft operations more critical than ever. Most traditional space missions rely heavily on transmitting large volumes of raw data to Earth for processing, which requires extensive resources and is inefficient. But recent advances in onboard computing and artificial intelligence are reshaping the way such missions can process, filter and interpret data directly in orbit.
Missions such as ESA’s ɸ-Sat series and OPS-SAT show how machine learning algorithms can identify relevant data in orbit, such as anomalous space weather and disruptions from space debris, which significantly reduces unnecessary data transmission. In doing so, these systems can lower power consumption, optimise bandwidth usage, and enable smaller satellites to perform complex tasks.
This topic is a two-part lecture which explores the growing role of intelligent data systems in modern space missions that increase efficiency, enhance mission autonomy and allow spacecrafts to make real-time decisions in dynamic and unpredictable environments where ground intervention can be limited.
The first lecture will focus on the different obstacles that can be tackled with onboard machine learning algorithms and how satellite and ground systems can be modified to support them. We will then discuss the process of implementing such algorithms with brief code demos using an Arduino to simulate an onboard computer. The second part of the lecture will focus on the challenges faced in implementing such intelligent data systems in current space technology, the trade-offs amongst microprocessors used in existing spacecrafts for it, and the possible research required to develop feasible solutions for it.