DISTRIBUTED KALMAN FILTER ARCHITECTURES

THIS INVESTIGATION WILL DEVELOP DISTRIBUTED KALMAN FILTER ARCHITECTURES APPLICABLE TO MULTI-SENSOR INTEGRATED NAVIGATION SYSTEMS TYPICAL OF THOSE ENVISIONED FOR ADVANCED TACTICAL AIRCRAFT. WHILE THESE MULTI-SENSOR NAVIGATION SYSTEMS POTENTIALLY PROVIDE HIGH LEVELS OF ACCURACY, A NUMBER OF DIFFICULTIES CAN ARISE WHEN CLASSICAL KALMAN FILTERING TECHNIQUES ARE APPLIED. THESE DIFFICULTIES INCLUDE POOR ACCURACY, LOSS OF STABILITY, AND/OR SEVERE COMPUTATIONAL BURDENS IN VARIOUS DEGREES. THE FILTER ARCHITECTURES AND DESIGN TECHNIQUES DEVELOPED HERE WILL CIRCUMVENT OR AMELIORATE THESE DIFFICULTIES. SPECIFIC ARCHITECTURES TO BE DEVELOPED INCLUDE MULTIRATE FILTERS, CASCADED FILTERS, AND DISTRIBUTED (DECENTRALIZED) FILTERS.