Python VS Space Junk
|Countdown link||Open timer|
This talk contains brief discussion of the military and testing of weapons, both in a non-combat context (e.g., firing missiles at uncrewed spacecraft).
A look at some of the systems in place today—data sources, algorithms, libraries, and pipelines—that manage the growing crisis of space debris and protect our critical orbital infrastructure from the ground.
The accumulation of debris in Earth orbit puts critical satellite-based infrastructure at risk. Methods of precise tracking and accurate path prediction for objects in orbit are required to prevent debris-forming collisions, as well as to enable on-orbit satellite repair and debris removal technologies that are in early development.
This tracking and prediction process for objects in Earth orbit is known as Space Traffic Management (STM). It's to satellites and space junk what Air Traffic Control is to aviation, but more. STM is made up of a series of tasks that are each challenging for humans or computers to perform for varied reasons: data collection requires high-fidelity analog capture, data processing requires high-volume stream digitisation and transformation without precision or information loss, data interpretation requires recognition of ambiguous objects, and path extrapolation from current trajectories requires computation of complex and fluctuating physical forces with great precision.
Let's talk about what's known about how STM systems work (or don't work) today, how you can try some of these steps yourself with Python, and how we can make global STM capabilities better with open source.
Mars Buttfield-Addison (@TheMartianLife) is a PhD Candidate in Computer Engineering at The University of Tasmania and CSIRO. Her current work focuses on improving the performance of radio/radar-based satellite and debris tracking with machine learning. On the side, she writes books about ML, runs conferences, and freelances as a creator of STEM educational materials for children.