Onboard Information Fusion for Multisatellite Collaborative Observation: Summary, challenges, and perspectives

Abstract

Onboard information fusion for multisatellites, which is based on spatial computing mode, can improve the satellites’ capability, such as the spatial–temporal coverage, detection accuracy, recognition confidence, position precision, and prediction precision for disaster monitoring, maritime surveillance, and other emergent or continuous persistent observing situations. First, we analyze the necessity of onboard information fusion. Next, the recent onboard processing developments are summarized and the existing problems are discussed. Furthermore, the key technologies and concepts of onboard information fusion are summarized in the fields of feature representation, association, feature-level fusion, spatial computing architecture, and other issues. Finally, the future developments of onboard information fusion are investigated and discussed.