Robust Capacity-Constrained Scheduling and Data-Based Modeling

Abstract:

More than a half million pieces of space debris are in low-Earth orbit, of which about 5% are considered a threat to operational satellites. Collisions, such as the one that occurred on February 10, 2009 between the decommissioned Russian military communications satellite Kosmos-2251 and a operational U.S. Iridium communications satellite, produce yet more debris, and the accumulation of space debris renders low-Earth orbit increasingly dangerous and un- usable. Space debris are tracked by NORAD, NASA, and the European Space Agency, and orbit-prediction methods, based on state-estimation techniques, play a fundamental role in this activity by combining models and measurements to optimize predictions. Improving the accuracy of space-debris tracking can reduce the frequency of collisions by facilitating evasive maneuvers, thereby protecting valuable space assets and helping to prevent runaway space-debris production. In order to achieve this objective, more accurate nowcasts (current conditions) and forecasts (future con- ditions) of the thermospheric winds and density are needed for improved debris trajectory prediction. This project is focused on the conceptual and technological foundation for a cyber- physical system involving a constellation of Cubesat satellites in orbits that cover the globe. These satellites will collect atmospheric data, which will be communicated through ground stations located worldwide for use in modeling and prediction. One of the challenges of Cubesat technology is that the available space communication in- frastructure is geared toward monolithic satellites with large power sources and large antennas. Collecting data from a constellation of Cubesats with limited on-board power is thus a challenging problem in optimal capacity-constrained scheduling. The data col- lected from the satellites will be used to improve models of the thermosphere through a process called data-based model refinement. These models can then be used within data assimilation algorithms to provide atmospheric data estimates, which, in turn, can be used for orbit trajectory prediction for space debris. This technology has direct appli- cation to space weather prediction, which affects both space- and ground-based assets. The combination of atmospheric science, algorithm development, and systems engi- neering crosses the boundaries of physics, engineering, systems science, and computer science. To realize these objectives, the project team includes experts on space physics, satellite engineering, optimization, model identification, and data assimilation from the University of Michigan and National Center for Atmospheric Research (NCAR).

  • Aerospace
  • CPS Domains
  • Space Systems
  • Defense
  • Modeling
  • Systems Engineering
  • CPS Technologies
  • Foundations
  • National CPS PI Meeting 2013
  • 2013
  • Poster
  • Academia
  • CPS PI Poster Session
Submitted by Dennis Bernstein on