Efficient Operation of Deep Learning for In-Orbit Inference

Jon Brookshire, Latent AI’s Director of Integrated Systems recently gave a virtual presentation in the "AI/ML for Edge Processing in Aerospace and Defense" mini symposium during the 2nd IACM Community MMLDE-CSET (Mechanistic Machine Learning and Digital Engineering for Computational Science, Engineering and Technology).

Deep learning models are already being used on geo-spatial imagery on both ground and air domains, and their application in the space domain is imminent. Jon discusses how to enable #deeplearning processing on satellites with an edge #machinelearning operations (#MLOps) pipeline that addresses challenges in the #space domain.

As more powerful processors are made available in low-size, weight, and power (#SWaP), ruggedized, and radiation-tolerant packages, it becomes feasible to deploy large #AI/#ML models to edge nodes for processing data at the sensor. AI/ML processing on edge platforms from #drones to #satellites allows for advanced data analysis even when limited communications precludes ground in-the-loop processing.