Satellite Navigation and Coordination with Limited Information Sharing

Sydney Dolan, Siddharth Nayak, Hamsa Balakrishnan
Proceedings of The 5th Annual Learning for Dynamics and Control Conference, PMLR 211:1058-1071, 2023.

Abstract

We explore space traffic management as an application of collision-free navigation in multi-agent systems where vehicles have limited observation and communication ranges. We investigate the effectiveness of transferring a collision avoidance multi-agent reinforcement (MARL) model trained on a ground environment to a space one. We demonstrate that the transfer learning model outperforms a model that is trained directly on the space environment. Furthermore, we find that our approach works well even when we consider the perturbations to satellite dynamics caused by the Earth’s oblateness. Finally, we show how our methods can be used to evaluate the benefits of information-sharing between satellite operators in order to improve coordination.

Cite this Paper

BibTeX
@InProceedings{pmlr-v211-dolan23a, title = {Satellite Navigation and Coordination with Limited Information Sharing}, author = {Dolan, Sydney and Nayak, Siddharth and Balakrishnan, Hamsa}, booktitle = {Proceedings of The 5th Annual Learning for Dynamics and Control Conference}, pages = {1058--1071}, year = {2023}, editor = {Matni, Nikolai and Morari, Manfred and Pappas, George J.}, volume = {211}, series = {Proceedings of Machine Learning Research}, month = {15--16 Jun}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v211/dolan23a/dolan23a.pdf}, url = {https://proceedings.mlr.press/v211/dolan23a.html}, abstract = {We explore space traffic management as an application of collision-free navigation in multi-agent systems where vehicles have limited observation and communication ranges. We investigate the effectiveness of transferring a collision avoidance multi-agent reinforcement (MARL) model trained on a ground environment to a space one. We demonstrate that the transfer learning model outperforms a model that is trained directly on the space environment. Furthermore, we find that our approach works well even when we consider the perturbations to satellite dynamics caused by the Earth’s oblateness. Finally, we show how our methods can be used to evaluate the benefits of information-sharing between satellite operators in order to improve coordination.} }
Endnote
%0 Conference Paper %T Satellite Navigation and Coordination with Limited Information Sharing %A Sydney Dolan %A Siddharth Nayak %A Hamsa Balakrishnan %B Proceedings of The 5th Annual Learning for Dynamics and Control Conference %C Proceedings of Machine Learning Research %D 2023 %E Nikolai Matni %E Manfred Morari %E George J. Pappas %F pmlr-v211-dolan23a %I PMLR %P 1058--1071 %U https://proceedings.mlr.press/v211/dolan23a.html %V 211 %X We explore space traffic management as an application of collision-free navigation in multi-agent systems where vehicles have limited observation and communication ranges. We investigate the effectiveness of transferring a collision avoidance multi-agent reinforcement (MARL) model trained on a ground environment to a space one. We demonstrate that the transfer learning model outperforms a model that is trained directly on the space environment. Furthermore, we find that our approach works well even when we consider the perturbations to satellite dynamics caused by the Earth’s oblateness. Finally, we show how our methods can be used to evaluate the benefits of information-sharing between satellite operators in order to improve coordination.
APA
Dolan, S., Nayak, S. & Balakrishnan, H.. (2023). Satellite Navigation and Coordination with Limited Information Sharing. Proceedings of The 5th Annual Learning for Dynamics and Control Conference, in Proceedings of Machine Learning Research 211:1058-1071 Available from https://proceedings.mlr.press/v211/dolan23a.html.

Related Material