1. Automating the Swift scheduling pipeline
- Author
-
Alex Deich, Sam LaPorte, and Jeff Gropp
- Subjects
Swift ,Iterative and incremental development ,Spacecraft ,Computer science ,business.industry ,Real-time computing ,Timeline ,business ,computer ,Automation ,Scheduling (computing) ,computer.programming_language - Abstract
Creating an observing timeline for the Swift spacecraft is a non-trivial problem. Currently, it takes a human an average of 5 hours to produce a plan which maximizes science returns while balancing numerous spacecraft constraints which preserve spacecraft health. We present the Swift scheduling automation pipeline, which aims to totally automate the scheduling process. We use an iterative process to search precomputed databases to construct robust observation plans. Our system can run several orders of magnitude faster than a human, is computationally cheap, and produces much more optimized plans while also enabling fast optimal responses to unpredictable phenomena including GRB, GW and neutrino triggers.
- Published
- 2018
- Full Text
- View/download PDF