1 A Logic-Constrained Knapsack Formulation and a Tabu Algorithm for the Daily Photograph S.pdf
文本预览下载声明
1A Logic-Constrained Knapsack Formulation and a Tabu Algorithm
for the Daily Photograph Scheduling of an Earth Observation Satellite
Michel Vasquez and Jin-Kao Hao
*
(Submitted to Journal of Computational Optimization and Applications )
Abstract: The daily photograph scheduling problem of earth observation satellites like Spot 5
consists in scheduling a subset of mono or stereo photographs from a set of candidates to
different cameras. The scheduling must optimize a cost function while satisfying a large number
of constraints. In this paper, we first present a formulation of the problem as a generalized
version of the well-known knapsack model, which includes large numbers of logical
constraints that bound the sums of pairs and triples of variables. We then develop a tabu search
algorithm which integrates some important features. Extensive experiments on a set of large and
realistic benchmark instances show the high effectiveness of this approach.
Key words: tabu search, heuristics, satellite photograph scheduling, multidimensional knapsack,
constrained combinatorial optimization
1. Introduction
The daily photograph scheduling problem (DPSP) is one of the key applications for an earth
observation satellite like Spot 5. The main purpose of the DPSP is to schedule a subset of photographs
from a set of candidate photographs which will be effectively achieved by the satellite. The resulting
subset of photographs must satisfy a large number of imperative constraints of different types and at
the same time maximize a given cost function.
The cost function reflects several criteria like client importance, demand urgency, meteorological
forecasts and so on. The constraints include both physical constraints such as the recording capacity
on board and logic constraints such as non overlapping trials and minimal transition time between two
successive trials on the same camera.
This problem is also important and interesting from a complexity point of view. Indeed it can be
mode
显示全部