Citation

Niu X, Tang H, Wu L. Int. J. Disaster Risk Reduct. 2018; 28: 813-825.

Copyright

(Copyright © 2018, Elsevier Publishing)

DOI

10.1016/j.ijdrr.2018.02.013

PMID

unavailable

Abstract

Earth satellite observations are very useful during the response phase of disaster management, since satellites could provide accurate, frequent and almost instantaneous data for large areas anywhere in the world. To rapidly respond to natural disasters, a key problem is how to efficiently schedule multiple earth observation satellites to acquire image data of a large stricken area by coordinating multiple different even conflicting needs of disaster relief, such as the extent of coverage over the stricken area, timeliness, and the spatial resolution. In this paper, considering two typical application scenarios during the response phase, we propose a multi-objective optimization method to solve the problem of satellite scheduling of a large area target. First, we design a decomposition method to partition a areal task into a series of observation strips. Next, the multiple satellite tasking problem is expressed as a multi-objective integer-programming model including optimizing objectives of the coverage rate, the imaging completion time, the average spatial resolution and the average slewing angle. Finally, the multi-objective genetic algorithm NSGA-II is designed to obtain optimal solutions of satellite scheduling. A real disaster scenario, i.e., 2008 Wenchuan earthquake, is revisited in terms of satellite image acquisition in the context of emergency response. To prove the advantage of NSGA-II, a comparison with state-of-the-art approaches is performed. Furthermore, we discuss the applicability of the proposed method under two kinds of situations: (1) roughly grasping the damage of affected area as soon as possible and (2) accurately assessing the damage of buildings in the worst-hit area.

Language: en

Keywords

Area target; Decomposition; Disaster emergency response; Multi-objective genetic algorithm; Multi-satellite scheduling