Thoughts on Research Imperatives in Data Fusion
James Llinas · AIAA Scitech 2021 Forum · 2021
View Video Presentation: https://doi.org/10.2514/6.2021-0914.vid Concepts and principles of Data Fusion gained some specificity as far back as in the mid-1980’s, and the field has been maturing now for some decades, with societies, journals and expansive research thrusts in academia, industry and government. Nevertheless, there are a number of major deficiencies in capability that remain to be addressed and have prevented the field from realizing some of its major potential contributions to a wide range of application domains. Among such shortfalls are: cross-level synergies (meaning fusion model Levels; see the references cited below), robust methods for association of disparate/non-commensurate data, advances in capability to associate and fuse hard and soft data, realization of adaptable, multi-algorithm systems, and well-defined, efficient and effective entry points for human involvement and exploitation, among others. This paper reviews the state of the art in these named areas and is a statement about the need for framing research imperatives for potentially major advances in the field.