SES – Gemini – Artificial Intelligence for Dynamic Resource Management

SES – Gemini – Artificial Intelligence for Dynamic Resource Management

Modern high-throughput satellite (HTS) with digital payloads and phased arrays have thousands of configurable variables. This makes it necessary for satellite operators like SES to move from manual resource allocation to a dynamic resource management tool. Therefore, the project’s goal is to develop a holistic and AI-based approach to dynamically allocate resources for satellite constellations. For a given dynamic customer demand, the goal is to optimize the available assets in space to maximize benefits and value for operators and their customers.

Key Personnel

Edward Crawley, Bruce Cameron, Markus Guerster, Juan José Garau Luis, Damon Jones, Rubén Alinque

Publications

  • Static beam placement and frequency plan algorithms for LEO constellations (Nils Pachler, Markus Guerster, Inigo del Portillo, Edward F. Crawley, Bruce G. Cameron), In International Journal of Satellite Communications and Networking, 2020
  • Allocating Power and Bandwidth in Multibeam Satellite Systems using Particle Swarm Optimization (Nils Pachler, Juan Jose Garau Luis, Markus Guerster, Edward F. Crawley, Bruce G. Cameron), In 2020 IEEE Aerospace Conference, 2020
  • Revenue Management for Communication Satellite Operators — Opportunities and Challenges (Markus Guerster, Joel Grotz, Peter Belobaba, Edward F. Crawley, Bruce G. Cameron), In 2020 IEEE Aerospace Conference, 2020
  • Artificial Intelligence Algorithms for Power Allocation in High Throughput Satellites: A Comparison (Juan Jose Garau Luis, Nils Pachler, Markus Guerster, Inigo del Portillo, Edward F. Crawley, Bruce G. Cameron), In 2020 IEEE Aerospace Conference, 2020
  • Deep Reinforcement Learning Architecture for Continuous Power Allocation in High Throughput Satellites (Juan Jose Garau Luis, Markus Guerster, Inigo del Portillo, Edward F. Crawley, Bruce G. Cameron), In Reinforcement Learning for Real Life Workshop at 2019 International Conference on Machine Learning, 201
  • Problem representation of dynamic resource allocation for flexible high throughput satellites (Markus Guerster, Juan Jose Garau Luis, Edward F. Crawley, Bruce G. Cameron), In 2019 IEEE Aerospace Conference, 2019