Seminars & Events

AeroAstro-CSE PhD Thesis Defense | Rashmi Ravishankar

Program AeroAstro-CSE Advisor Olivier L. de Weck Thesis Title Edge Computing in Space: Design Optimization and Reinforcement Learning Scheduling of Onboard Computing Satellites
Date Monday, August 4, 2025

Rashmi Ravishankar, AeroAstro-CSE PhD Thesis Defense Announcement

Thesis Title: Edge Computing in Space: Design Optimization and Reinforcement Learning Scheduling of Onboard Computing Satellites

Date: Monday, August 4, 2025
Time: 12:30 PM ET
Location: 33-206 / Zoom

Thesis Committee:

  • Olivier L. de Weck, Apollo Program Professor of Astronautics and Engineering Systems, Massachusetts Institute of Technology
  • Kerri Cahoy, Sheila Evans Widnall (1960) Professor of Aeronautics and Astronautics Massachusetts Institute of Technology
  • Luca Carlone, Boeing Career Development Associate Professor, Department of Aeronautics and Astronautics, Massachusetts Institute of Technology
  • Johannes Norheim, Assistant Professor (Chancellor’s Fellow) at University of Strathclyde

Abstract:

As satellite missions heighten in complexity, there is a need for spacecraft equipped with greater intelligence and autonomy. This calls for onboard, real-time decision-making abilities on weight-, power-, and radiation-constrained hardware and processors, one of the most challenging of all “edge computing” problems. Only a small body of literature has considered an environment where space systems make decisions of their own, and none have combined the challenges of edge computing with the multidisciplinary optimization problem of architecting and scheduling a computationally heavy satellite system. This thesis evaluates the potential of edge computing and prescribes an optimal strategy for computer operations. First, an edge computing satellite model, “EdgeSat”, is defined, and its competing subsystems are optimized for a satellite mission where onboard computing and autonomy are needed. Features of the model include data and computing models of space-qualified hardware, thermal models of computing in space, and a dynamic data management protocol. The value of edge computing is defined and quantified, and a pareto optimal utopia point is identified. Next, reinforcement learning is used to prescribe optimal mission profiles in various scenarios using appropriate rewards and visibility windows. The rewards consider the utility of downlinked data, battery preservation, thermal constraints, and downlink windows. Edge computing and design optimization were found to add successive value to the base case, and RL scheduling of the onboard computer adds further value.