Sensor Fusion, the coordination of data from diverse sources to produce a usable perspective, is an important problem which must be addresed today in nearly every system which acquires and responds to data. The feasibility and technical promise of a powerful, innovative, and generic solution to this Sensor Fusion Problem was demonstrated by results in Phase I of this work. Our Dataware Engineering Methodoogy and Ghosting concepts were shown to produce compact, computationally efficient,highlydescriptive, and readily usble "fusions" of sensor data obtained from potentially any set of sensors. Based on these results, it is clear that this Sensor Fusion Technology can be captured, developed, implemented, and delivered in a family of practical Sensor Fusion Algorithms applicable to a wide range of situations and computing environments. We propose to develop and deliver these practical Sensor Fusion Algorithms in this Phase II project. The importance of this work is further evidenced by the fact that we have already received considerable commercial interest in having these Algorithms available, due in part to the difficulty and nearly universal presence of the Sensor Fusion Problem. Patent protection of key Phase I results is already underway, and we expect more patentable innovations to emerge from Phase II.
Keywords: Sensor Fusing Signal Processing Situational Awareness Robotics Image Analysis Algorithms
