AI in Satellite Production: A Cautious Leap

Companies are taking baby steps when it comes to using artificial intelligence (AI) in satellite manufacturing. Blue Canyon Technologies, for instance, is keen on understanding how AI can boost production without compromising cybersecurity. "Where does your data go when you're teaching an AI system? " wondered Chris Winslett, Blue Canyon's general manager, at the recent Satellite Innovation conference. "And what about data from other sources? Where does it come from? " Despite these concerns, AI can be a game-changer in engineering design. It can help turn huge amounts of data into useful information, freeing up people's time for decision-making instead of poring over spreadsheets. This is something Blue Canyon, a Raytheon Technologies subsidiary, is looking into. Over at Kongsberg NanoAvionics, engineering operations director Karolis Senvaitis shares similar worries. "Can you trust the AI results? Where do they come from? And if you combine results, are they what you want? " Until these questions are answered, Senvaitis doesn't see AI being directly integrated into manufacturing or testing. However, AI's strength lies in handling and analyzing large datasets. Machina Labs, a Los Angeles startup focusing on robotic manufacturing, has a different approach. Instead of relying on various data sources, they generate their own. "We use sensors and scanning software to check if parts meet requirements, " explained John Borrego, Machina Labs' vice president of production. This data is stored securely in the cloud, opening up opportunities to improve processes and reduce defects in future parts.