Heuristic Onboard Pointing Re-scheduling for an Earth Observing Spacecraft. Chien, S. and Troesch, M. In International Workshop on Planning and Scheduling for Space (IWPSS 2015), Buenos Aires, Argentina, July, 2015.
Heuristic Onboard Pointing Re-scheduling for an Earth Observing Spacecraft [pdf]Paper  abstract   bibtex   
Prior space missions have not routinely used onboard decision-making. The Autonomous Sciencecraft (ASE), flying onboard the Earth Observing One spacecraft, has been flying autonomous agent software for the last decade that enables it to analyze acquired imagery on board and use that analysis to determine future imaging. However ASE takes approximately one hour to analyze and respond. This paper describes a scheduling prototype for the Earth Observing Autonomy (EOA) project to increase the responsiveness of spacecraft flight software for onboard decision-making as well as to increase the capabilities of flight software. Specifically, we target onboard image analysis and response within a single orbital overflight at low Earth orbit (about eight minutes). We focus on the re-scheduling of the future image acquisitions in the context of an existing set of requests along with new requests based on onboard analysis of just acquired imagery. We describe a greedy, constructive, scheduler with O(n2) performance and present preliminary results on its performance.
@inproceedings{chien-troesch-IWPSS-2015,
  author = {Steve Chien and Martina Troesch},
  title = {Heuristic Onboard Pointing Re-scheduling for an Earth Observing Spacecraft},
  booktitle = {International Workshop on Planning and Scheduling for Space (IWPSS 2015)},
  year = {2015},
  month = {July},
  address = {Buenos Aires, Argentina},
  abstract = {Prior space missions have not routinely used onboard decision-making. The Autonomous Sciencecraft (ASE), flying onboard the Earth Observing One spacecraft, has been flying autonomous agent software for the last decade that enables it to analyze acquired imagery on board and use that analysis to determine future imaging. However ASE takes approximately one hour to analyze and respond. This paper describes a scheduling prototype for the Earth Observing Autonomy (EOA) project to increase the responsiveness of spacecraft flight software for onboard decision-making as well as to increase the capabilities of flight software. Specifically, we target onboard image analysis and response within a single orbital overflight at low Earth orbit (about eight minutes). We focus on the re-scheduling of the future image acquisitions in the context of an existing set of requests along with new requests based on onboard analysis of just acquired imagery. We describe a greedy, constructive, scheduler with O(n2) performance and present preliminary results on its performance. },
  url = {https://ai.jpl.nasa.gov/public/papers/chien_iwpss2015_heuristic.pdf},
  clearance = {CL#15-1658},
  project = {EagleEye},
}
Downloads: 0