Optimisation problems and resolution methods in satellite scheduling and space-craft operation: a survey

Abstract

Nowadays, satellite technologies are developing very fast, including the production of small, low cost satellites. This is propelling an important increase in the number of satellite mission planning and operations and the need for intelligent scheduling systems to automatically and optimally handle such mission planning, i.e., the assignment of tasks to space missions through ground station services. Central to satellite mission planning is the satellite scheduling problem, whose resolution is the basis for an optimised allocation of user requests for efficient communication between operations teams at ground and spacecraft systems. The aim of this paper is to survey the state of the art in the satellite scheduling problem, analyse its most significant mathematical formulations, seen as a family of time window scheduling problems, as well as examining its multi-objective nature. We particularly stress the computational complexity of the problem, its highly constrained features and the conflicting objectives due to windows accessibility and visibility clashes caused by requirements of different missions to be scheduled. In view of the computationally hardness to solve to optimality, the resolution of the problem is addressed through heuristics and meta-heuristics methods, namely, local search and population-based methods. While, for validating and evaluating the performance of resolution methods, a simulation toolkit in the literature, called STK, is used. Finally, we consider some optimisation problems arising in spacecraft design, operation and satellite deployment systems.

Multi-objective approaches to ground station scheduling for optimization of communication with satellites

“…There are 16 benchmark instances for each size that can be found in Xhafa and Ip (2019). Due to the similarities of these instances within the same dimensionality of the scheduling problem, and the computational expense, we decided to test only five instances of each size.…”

“…The formulation of the problem and its objectives used in this paper are taken from Xhafa et al (2013), Xhafa and Ip (2019), along with the benchmark problems. The benchmark problems are of 3 different sizes.…”

“…Schedules that are to be optimized can be encoded as a chromosome in multiple ways. Paper (Xhafa and Ip 2019) splits the schedule encoding into two separate chromosomes. The first chromosome encodes only the windows of communication for a certain satellite without defining a concrete ground station.…”

“…These include day-today activities of an operational satellite, like configuration of different sensors on the satellite to collect data, maintenance of the satellite status, and transmission of data/ commands between satellites and ground stations. During the allocated communication window the task to be performed should be completed (Xhafa and Ip 2019). The goal of scheduled data downloads from satellites to ground stations is to maximize the total amount of downloaded data while satisfying all visibility, energy, and data requirements; this is also known as Multiple-Satellite, Multiple Ground Station Scheduling Problem (MMSP) (Castaing 2014).…”

On Formulating the Ground Scheduling Problem as a Multi-objective Bilevel Problem

“…The problem is most of the time formulated as a multi-objective problem, consisting of several and conflicting objectives [11], [12]. Moreover, the optimisation of each of the objectives can be modeled in a hierarchical or simultaneous fashion [13]. In the simultaneous optimisation, the objectives are optimised at the same time, obtaining a Pareto front of the solutions, ignoring their hierarchy.…”