Dual-Population Artificial Bee Colony Algorithm for Joint Observation Satellite Mission Planning Problem
Xuemei Jiang, Yanjie Song, Lining Xing
Abstract
Earth observation satellites provide services for users by taking images. The rapid increase in the number of satellites and missions makes mission planning more difficult. In order to solve this problem, this paper adopts a multi-population evolutionary algorithm. First, a mixed-integer programming model of the joint observation satellites mission planning problem (JOSMPP) is constructed. After that, a dual-population artificial bee colony algorithm (DPABC) and a heuristic task scheduling algorithm are proposed. Two learning-based nectar generation methods are used to guide the direction of population optimization. The search population is used to lead the nectar search at the stage of employed bees and supplement population is used to improve search performance. According to the performance of these two populations, candidate populations are generated to realize the search of onlooker bees. After each generation of population search, the composition of the two populations is adjusted according to the performance. Finally, the experimental results show that the DPABC algorithm has more advantages than multiple state-of-the-art algorithms to solve the joint mission planning problem of earth observation satellites.