Multiregion Mission Planning by Satellite Swarm Using Simulated Annealing and Neighborhood Search

Multiregion planning of the missions of a satellite swarm is regarded as a critical step in cooperative observations involving remote sensing. Given the individual differences between satellites and the constraints of task allocation, an objective function is used in this article to achieve a balance between benefits and costs. To ensure the observation of a large area, the rate of region coverage is regarded as the basic benefit. To encourage the satellite swarm to finish its exhaustive cover as early as possible, the time consumed to complete the observation is regarded as an extra benefit. Furthermore, by punishing satellite storage and maneuvering, the overall observation scheme evolves toward low cost. A region-splitting technique is proposed herein. Each of a satellite's observation opportunities generates two candidate strips at the edge of the region. Once the strips have been confirmed to meet the mission constraints, the original region is reconstructed. Subsequently, the satellite swarm gradually envelops the region through several iterations. A simulated-annealing algorithm is also introduced in this article. To achieve satellite swarm scheduling for multiple regions, optimization variables are established, and a neighborhood search method is used. Simulation results revealed that the proposed technique devised an observation scheme with high coverage, high timeliness, and low cost and also confirmed that the proposed algorithm had an excellent evolution speed of objective function value.