Global constraints for scheduling data transfer in space missions

International audienceIn the context of space missions, where onboard memory resources are limited, efficient data transfer is crucial to avoid buffer overflow. In this paper, we are interested in scheduling data transfer for ESA's Rosetta mission. This problem has been addressed with heuristic methods that show good results, and even optimal results in few cases. In line with a previous work, we propose an exact method to solve this problem. We use constraint programming with two new global constraints. The first constraint links the decisions on priority to the memory usage by reasoning about bandwidth allocation. The second constraint breaks some symmetries of the model. The first results show that the exact method is relevant to solve this data transfer problem. This method aims at being generic, which means that it could be applied for other space missions, or any problem that involves a dense ranking