An AI scheduling environment for the Hubble Space Telescope
Jeffrey L. Sponsler, Mark Johnston, Glenn E. Miller, Anthony P. Krueger, Michael Lucks, Mark Giuliano · 1991
This report describes the SPIKE (Science Planning Intelligent Knowledge Environment) system that is currently used to create long term schedules for the Hubble Space Telescope. SPIKE employs the suitability function ("maps goodness over time") as its fundamental knowledge representation for scheduling constraints. To support the search for optimal schedules, a constraint satisfaction system is used. An initial long range schedule is generated using a greedy algorithm and a heuristic algorithm is applied in order to repair the schedule (specifically portions where constraint violations have occurred). Work in progress includes the scheduling of parallel exposures and the design of a machine independent windowing environment. 1.INTRODUCTION This report describes a system called SPIKE. SPIKE is used by astronomers at the STScI to verify pre-defined long-term schedules and to create new schedules for the Hubble Space Telescope. The system employs techniques of artificial intelligence incl...