As part of Earth observation missions, satellites are used to acquire areas on the Earth's surface. Large area coverage missions are large-scale acquisition projects. They require the use of a set of satellites that make several passes over a long acquisition period. Cutting an area of interest into sub-areas called meshes that can be acquired by satellites during their passes is a difficult optimization problem. In this paper, we present the Glimpse system based on adaptive multi-agent systems for the dynamic cutting and positioning of meshes on a grid of elementary sub-areas. The experiments carried out and the comparison with an approach that is part of the operational state of the art show that a decentralised and dynamic approach to solving this problem is adequate.