New Algorithm for Emergency Scheduling in LEO Earth Observation Constellations

A novel algorithm known as Task-Driven Three-Layer Distributed Scheduling (T3L-DS) has been introduced to tackle the dynamic emergency observation scheduling problem (DEOSP) within extensive low-Earth-orbit (LEO) Earth-observation (EO) constellations. This approach, outlined in a recent arXiv preprint (arXiv:2608.14789), focuses on integrating urgent tasks into pre-existing schedules with minimal interference. T3L-DS represents task requirements and sensor coverage on a unified geographic grid, creates temporary clusters based on observational capabilities and inter-satellite connections, and employs onboard dual-plan bidding alongside joint marginal evaluation for intra-cluster coordination. Additionally, it incorporates an inter-cluster coordination mechanism for unmet demands. Extensive computational tests compare T3L-DS to centralized simulated annealing (SA), although specific results are not included in the abstract. This research is significant for the expanding intersection of AI and technology in space operations, especially regarding emergency response and dynamic resource management.