Machine learning has incredible applications for the science community when it comes to processing data once it’s been downlinked and perhaps, even processing data on a spacecraft. Applying machine learning algorithms to these applications allows scientific analysis to become much more efficient and therefore have a greater impact not just in planetary science but also in industries which impact our lives every day.

In this episode, I interview Dr. Hannah Kerner on what goes into machine learning for planetary science and what challenges and limitations arise from applying algorithms both on the ground and onboard spacecraft. Dr. Kerner earned her PhD in Systems Exploration Design from ASU, with a focus on novelty detection methods within machine learning. Her dissertation presented how these algorithms could be harnessed onboard the Mars Curiosity Rover to extract meaningful information from data sets which could then inform scientific observations more efficiently. She is now an assistant professor at the University of Maryland (UMD) where she is working with NASA harvest to apply machine learning solutions in remote sensing.