As the volume of data generated by space missions continues to grow exponentially, the need for intelligent automation becomes increasingly urgent. In a recent episode of Federal News Network’s Space Hour Podcast, Victoria Da Poian, Lead Data Scientist at Tyto Athene, joined host Eric White to explore how artificial intelligence (AI) and machine learning (ML) are reshaping the future of space exploration. 

Victoria, who works closely with NASA as part of Tyto Athene’s mission enablement portfolio, discussed how AI and ML are being used to make spacecraft more autonomous, data collection more strategic, and science operations more adaptive. 

“We are trying to enhance and enable different space missions through flight and ground software tools, through data processing, and also through… the development of AI and some machine learning tools to make our instruments a little bit smarter for the missions we send them to,” Victoria explained. “Because we’re going further away in our solar system.” 

Prioritizing Data in Deep Space 

In traditional mission architecture, data is gathered, transmitted to Earth, and analyzed post-collection. However, limited bandwidth and the sheer scale of information present new challenges. As Victoria noted, “One of the main things that we want [long term] is… onboard machine learning algorithms to help the mission, to help the spacecraft, to handle the data that is being collected, to see what is the most relevant, to see what is the most interesting information, and prioritize the data we send back.” 

With AI-driven prioritization, spacecraft can intelligently assess what data matters most, which is critical in environments where transmission windows are narrow and scientific opportunity is fleeting. 

Integrating Operational and Scientific Intelligence 

The conversation highlighted two core use cases: optimizing operational functions (like fuel and power usage) and supporting scientific discovery in real time. Tyto Athene’s data scientists work in direct collaboration with NASA scientists to tailor ML algorithms to specific scientific goals while minimizing human bias. 

Looking Ahead: Collaboration and Trust 

As AI capabilities mature, Victoria emphasized that success in space won’t just come from smarter algorithms; it will require trust between human operators and machine intelligence. 

“One of the main challenges… is developing the trust in these AI-driven strategies,” she said. “Machine learning is often seen as black boxes. The beauty of our team within Tyto… is to closely work with [NASA scientists] from day one to develop this trust in the solutions we’re offering.” 

Listen to the full episode on The Federal News Network