Δ-MILP: Deep Space Network Scheduling via Mixed-Integer Linear Programming

被引：2

作者：

Claudet, Thomas ^{[1
]}

Alimo, Ryan ^{[1
]}

Goh, Edwin ^{[1
]}

Johnston, Mark D. ^{[1
]}

Madani, Ramtin ^{[2
]}

Wilson, Brian ^{[1
]}

机构：

[1] CALTECH, Jet Prop Lab, Pasadena, CA 91109 USA

[2] Univ Texas Arlington, Dept Elect Engn, Arlington, TX 76015 USA

来源：

IEEE ACCESS | 2022年 / 10卷

基金：

美国国家航空航天局;

关键词：

Optimization; optimization methods; scheduling;

D O I：

10.1109/ACCESS.2022.3164213

中图分类号：

TP [自动化技术、计算机技术];

学科分类号：

0812 ;

摘要：

This paper introduces Delta-MILP, a powerful variant of the mixed-integer linear programming (MILP) optimization framework to solve NASA's Deep Space Network (DSN) scheduling problem. This work is an extension of our original MILP framework (DOI:10.1109/ACCESS.2021.3064928), and inherits many of its constructions and strengths, including the base MILP formulation for DSN scheduling. To provide more feasible schedules with respect to the DSN requirements, Delta-MILP incorporates new sets of constraints including 1) splitting larger tracks into shorter segments and 2) preventing overlapping between tracks on different antennas. Additionally, Delta-MILP leverages a heuristic to balance mission satisfaction and allows to prioritize certain missions in special scenarios including emergencies and landings. Numerical validations demonstrate that Delta-MILP now satisfies 100% of the requested constraints and provides fair schedules amongst missions with respect to the state-of-the-art for the most oversubscribed weeks of the years 2016 and 2018.

引用

页码：41330 / 41340

页数：11

共 32 条

[1] Delaunay-based derivative-free optimization via global surrogates. Part III: nonconvex constraints [J].