Team
- Giacomo Acciarini (University of Strathclyde)
- Francesco Pinto (University of Oxford)
- Sascha Metz (Technische Universitaet Darmstadt)
- Sarah Boufelja (IBM)
- Sylvester Kaczmarek (Imperial College London)
- Klaus Merz (European Space Agency)
- José A. Martinez-Heras (European Space Agency)
- Francesca Letizia (European Space Agency)
- Christopher Bridges (Surrey Space Centre)
- Atılım Güneş Baydin (University of Oxford)
Abstract
After decades of space travel, low Earth orbit is a junkyard of discarded rocket bodies, dead satellites, and millions of pieces of debris from collisions and explosions. Objects in high enough altitudes do not re-enter and burn up in the atmosphere, but stay in orbit around Earth for a long time. With a speed of 28,000 km/h, collisions in these orbits can generate fragments and potentially trigger a cascade of more collisions known as the Kessler syndrome. This could pose a planetary challenge, because the phenomenon could escalate to the point of hindering future space operations and damaging satellite infrastructure critical for space and Earth science applications. As commercial entities place mega-constellations of satellites in orbit, the burden on operators conducting collision avoidance manoeuvres will increase. For this reason, development of automated tools that predict potential collision events (conjunctions) is critical. We introduce a Bayesian deep learning approach to this problem, and develop recurrent neural network architectures (LSTMs) that work with time series of conjunction data messages (CDMs), a standard data format used by the space community. We show that our method can be used to model all CDM features simultaneously, including the time of arrival of future CDMs, providing predictions of conjunction event evolution with associated uncertainties.
Video
Publications
- Acciarini, Giacomo, Francesco Pinto, Sascha Metz, Sarah Boufelja, Sylvester Kaczmarek, Klaus Merz, José A. Martinez-Heras, Francesca Letizia, Christopher Bridges, and Atılım Güneş Baydin. 2021. “Kessler: a Machine Learning Library for Space Collision Avoidance.” In 8th European Conference on Space Debris.
@inproceedings{acciarini-2020-automated, title = {Kessler: a Machine Learning Library for Space Collision Avoidance}, author = {Acciarini, Giacomo and Pinto, Francesco and Metz, Sascha and Boufelja, Sarah and Kaczmarek, Sylvester and Merz, Klaus and Martinez-Heras, José A. and Letizia, Francesca and Bridges, Christopher and Baydin, Atılım Güneş}, booktitle = {8th European Conference on Space Debris}, year = {2021} }
- Pinto, Francesco, Giacomo Acciarini, Sascha Metz, Sarah Boufelja, Sylvester Kaczmarek, Klaus Merz, José A. Martinez-Heras, Francesca Letizia, Christopher Bridges, and Atılım Güneş Baydin. 2020. “Towards Automated Satellite Conjunction Management with Bayesian Deep Learning.” In AI for Earth Sciences Workshop at NeurIPS 2020, Vancouver, Canada.
After decades of space travel, low Earth orbit is a junkyard of discarded rocket bod-ies, dead satellites, and millions of pieces of debris from collisions and explosions.Objects in high enough altitudes do not re-enter and burn up in the atmosphere, butstay in orbit around Earth for a long time. With a speed of 28,000 km/h, collisionsin these orbits can generate fragments and potentially trigger a cascade of morecollisions known as the Kessler syndrome. This could pose a planetary challenge,because the phenomenon could escalate to the point of hindering future spaceoperations and damaging satellite infrastructure critical for space and Earth scienceapplications. As commercial entities place mega-constellations of satellites in orbit,the burden on operators conducting collision avoidance manoeuvres will increase.For this reason, development of automated tools that predict potential collisionevents (conjunctions) is critical. We introduce a Bayesian deep learning approachto this problem, and develop recurrent neural network architectures (LSTMs) thatwork with time series of conjunction data messages (CDMs), a standard data formatused by the space community. We show that our method can be used to modelall CDM features simultaneously, including the time of arrival of future CDMs,providing predictions of conjunction event evolution with associated uncertainties.
@inproceedings{pinto-2020-automated, title = {Towards Automated Satellite Conjunction Management with Bayesian Deep Learning}, author = {Pinto, Francesco and Acciarini, Giacomo and Metz, Sascha and Boufelja, Sarah and Kaczmarek, Sylvester and Merz, Klaus and Martinez-Heras, José A. and Letizia, Francesca and Bridges, Christopher and Baydin, Atılım Güneş}, booktitle = {AI for Earth Sciences Workshop at NeurIPS 2020, Vancouver, Canada}, year = {2020} } - Acciarini, Giacomo, Francesco Pinto, Sascha Metz, Sarah Boufelja, Sylvester Kaczmarek, Klaus Merz, José A. Martinez-Heras, Francesca Letizia, Christopher Bridges, and Atılım Güneş Baydin. 2020. “Spacecraft Collision Risk Assessment with Probabilistic Programming.” In Third Workshop on Machine Learning and the Physical Sciences (NeurIPS 2020), Vancouver, Canada.
Over 34,000 objects bigger than 10 cm in length are known to orbit Earth. Amongthem, only a small percentage are active satellites, while the rest of the populationis made of dead satellites, rocket bodies, and debris that pose a collision threatto operational spacecraft. Furthermore, the predicted growth of the space sectorand the planned launch of megaconstellations will add even more complexity,therefore causing the collision risk and the burden on space operators to increase.Managing this complex framework with internationally agreed methods is pivotaland urgent. In this context, we build a novel physics-based probabilistic generativemodel for synthetically generating conjunction data messages, calibrated usingreal data. By conditioning on observations, we use the model to obtain posteriordistributions via Bayesian inference. We show that the probabilistic programmingapproach to conjunction assessment can help in making predictions and in findingthe parameters that explain the observed data in conjunction data messages, thusshedding more light on key variables and orbital characteristics that more likelylead to conjunction events. Moreover, our technique enables the generation ofphysically accurate synthetic datasets of collisions, answering a fundamental needof the space and machine learning communities working in this area.
@inproceedings{acciarini-2020-spacecraft, title = {Spacecraft Collision Risk Assessment with Probabilistic Programming}, author = {Acciarini, Giacomo and Pinto, Francesco and Metz, Sascha and Boufelja, Sarah and Kaczmarek, Sylvester and Merz, Klaus and Martinez-Heras, José A. and Letizia, Francesca and Bridges, Christopher and Baydin, Atılım Güneş}, booktitle = {Third Workshop on Machine Learning and the Physical Sciences (NeurIPS 2020), Vancouver, Canada}, year = {2020} }
Acknowledgments
This work has been enabled by Frontier Development Lab (FDL) Europe, a public–private partnership between the European Space Agency (ESA), Trillium Technologies and the University of Oxford, and supported by Google Cloud. We would like to thank Dario Izzo and Moriba Jah for sharing their technical expertise and James Parr, Jodie Hughes, Leo Silverberg, Alessandro Donati for their support. AGB is supported by EPSRC/MURI grant EP/N019474/1 and by Lawrence Berkeley National Lab.