DISCO Satellite Image and ML Pipeline

Supervisors: Julian Priest
Semester: Fall 2022
Tags: Satellite, Image processing, Edge, Constrained Computing, Networks, Machine Learning, Embeded, Radio

ITU is a partner in the Danish Student Cubesat Program DISCO, which will launch a series of small satellites into orbit, starting with DISCO 1 in 2023 and followed by DISCO2 in 2024.

ITU is developing a hi-res multi camera imaging payload for earth observation primarily in the Arctic. We are developing an on satellite machine learning capability using an ML coprocessor, as well as models that can run on the reduced hardware found on board the satellite.There are two related projects to this both of which that working existing systems with the on board computer and the communications network.

The first is working on the image pipeline, how is the camera controlled, scheduled and triggered and how do the images get buffered sent back to earth and stored? The second is working on the ML model pipeline. We would like to be able to upload new models to the satellite as they are improved or new experiments and observation targets are identified. How does this ML iteration/deployment work?

DISCO is a multi-year, multi-university project and there are potential PhD opportunities following up on this work.