REAL-TIME SENSOR DATA FUSION

TRACKING SADARM SUBMUNITIONS FOR TEST EVALUATION IS A DIFFICULT TASK. THE RATIO OF THE TOTAL NUMBER OF AIRBORNE EXTRANEOUS TARGETS OF INTEREST IS GREATER THAN 4. IN FACT, THE SAME RATIO, BUT FOR TARGETS WITHIN THE FIELD OF VIEW OF INDIVIDUAL SENSORS DURING THE "EARLY PHASE" OF THE SUBMUNITION'S FLIGHT, CAN BE A LOT HIGHER. DURING THIS TIME, WHICH SPANS EJECTION OG THE SUBMUNITION FROM THE MLRS ROCKET TO THE ONSET OF STABILIZED VERTICAL DESCENT, THE SUBMUNITION UNDERGOES 2 DISTINCTLY VIOLENT MANEUVERS WHEREAS THE CLUTTER TARGETS MOVE ALONG CONTINUOUS AEROBALLISTIC TRAJECTORIES. WE PROPOSE TO INTEGRATE A VARIETY OF TEST RANGE SENSORS INTO A UNIFIED TRACKING NETWROK BASED UPON NEW MTI ALGORITHMS FOR DISTRIBUTED FILTERING AND MANEUVER DETECTION. THE MANEUVER PARAMETERS WILL HELP IN THE DISCRIMINATION OF TARGETS OF EVALUATION FROM CLUTTER, AND THUS REDUCE THE NUMBER OF HYPOTHESES TO CONSIDER FOR DATA ASSOCIATION DURING THE EARLY PHASE.