Georgia Institute of Technology
Feasibility of a dedicated satellite constellation for orbital edge computing services
Abstract
dc:description.abstractIn-orbit computing addresses the limitations of ground satellite link (GSL) capacity by shifting part of the data processing workload to the satellite itself. Satellite constellations equipped with machine learning-based computing payloads can process large volumes of raw data and downlink only the processed results to the ground. However, scaling up computing resources to meet the increasing complexity of in-orbit applications remains a challenge. In this work, we propose to investigate the feasibility of a dedicated small satellite constellation that provides orbital edge computing (OEC) services to spacecraft in low Earth orbit (LEO). We demonstrated that a 27-satellite OEC constellation can handle all data from 95 Planet Dove satellites while supporting compute-intensive applications such as vessel tracking. By offloading all of the computing tasks from individual client satellites and storing all data they collect, OEC allows mission designers to allocate more power and volume budgets to other subsystems. Additionally, this approach allows for more flexible and agile development of data processing applications.
Degree
thesis:*- Level thesis:degree_level
- Masters
- Department dc:contributor.department
- Aerospace Engineering
- Grantor dc:publisher
- Georgia Institute of Technology
- Year dc:date.issued
- 2025
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Duan, Yunpeng
- Advisor dc:contributor.advisor
-
- Gunter, Brian C.
- Committee members dc:contributor.committeemember
-
- Ho, Koki
- Bhardwaj, Ketan
Subjects
dc:subject × 2Identifiers
dc:identifier.*- Handle dc:identifier.uri
- https://hdl.handle.net/1853/81517