Dense Points Aggregation for Efficient and Collaborative Earth-Imaging Task Planning

Abstract

The continuous development of high-resolution imagery satellite payloads is featured with lower cost, high-integration and smaller volume, which promotes the wide adoption of Earth-imaging facilities on satellite in order to improve sensing coverage, quality, efficiency and so on. Satellite task planning is crucial in automatically generating observation timelines to fulfill the Earth-imaging tasks by optimizing the usage of satellite resources. As an important category of ground sensing tasks, points of interest(PoIs) are very common for satellite to sense ground changes such as building collapse in earthquake, social hot spots, volcano eruption, etc. As the number of PoIs increases enormously, separately planning for each PoI sensing task is no more realistic and aggregating the dense ground points for collaboratively imaging them through multi-satellite can provide a new solution for constellation applications. In this paper, a novel ground PoIs aggregation method is proposed to decrease task manipulation frequencies of satellites, based on which a collaborative multi-satellite planning is modeled and solved by particle swarm optimization. The efficacy of the aggregation-collaboration manner is evaluated and demonstrated in the experiment.