NTRS - NASA Technical Reports Server

Back to Results
spike: artificial intelligence scheduling for hubble space telescopeEfficient utilization of spacecraft resources is essential, but the accompanying scheduling problems are often computationally intractable and are difficult to approximate because of the presence of numerous interacting constraints. Artificial intelligence techniques were applied to the scheduling of the NASA/ESA Hubble Space Telescope (HST). This presents a particularly challenging problem since a yearlong observing program can contain some tens of thousands of exposures which are subject to a large number of scientific, operational, spacecraft, and environmental constraints. New techniques were developed for machine reasoning about scheduling constraints and goals, especially in cases where uncertainty is an important scheduling consideration and where resolving conflicts among conflicting preferences is essential. These technique were utilized in a set of workstation based scheduling tools (Spike) for HST. Graphical displays of activities, constraints, and schedules are an important feature of the system. High level scheduling strategies using both rule based and neural network approaches were developed. While the specific constraints implemented are those most relevant to HST, the framework developed is far more general and could easily handle other kinds of scheduling problems. The concept and implementation of the Spike system are described along with some experiments in adapting Spike to other spacecraft scheduling domains.
Document ID
Document Type
Conference Paper
Johnston, Mark
(Space Telescope Science Inst. Baltimore, MD., United States)
Miller, Glenn
(Computer Sciences Corp. Baltimore, MD., United States)
Sponsler, Jeff
(Space Telescope Science Inst. Baltimore, MD., United States)
Vick, Shon
(Space Telescope Science Inst. Baltimore, MD., United States)
Jackson, Robert
(Computer Sciences Corp. Baltimore, MD., United States)
Date Acquired
September 6, 2013
Publication Date
May 1, 1990
Publication Information
Publication: NASA, Marshall Space Flight Center, Fifth Conference on Artificial Intelligence for Space Applications
Subject Category
Distribution Limits
Work of the US Gov. Public Use Permitted.

Available Downloads

NameType 19900017961.pdf STI