Multi-Agent Planning and Execution Approach for Controlling Multiple Satellites

Reducing the costs of space operations is an increasing demand which may be achieved by automating space operation planning and execution. Towards this aim, this work proposes a Multi-Agent Space Operation Automation architecture. Multi-Agent Space Operation Automation architecture manages ground resource allocation for multi-satellite tracking, plans the satellite control operations, and automates plan execution. For managing the sharing of ground resources, it identifies satellites with time-conflicting visibility periods and reduces the least priority satellite tracking. For the planning process, Multi-Agent Space Operation Automation architecture automatically generates planning problem descriptions and analyzes whether a satellite tracking period is sufficient to achieve all the tracking goals. If there is insufficient time, it allows the elimination of lower priority goals. The satellite control planning problem was specified in PDDL 2.2 and the satellite control plan generated according to the temporal planning paradigm.