AUTONOMOUS AND ADAPTIVE SUBSEA SENSOR FOR OPTICAL MEASUREMENT OF VERY-NEAR SEAFLOOR FLUID FLOWS

ACCEPTABLE DEVELOPMENT OF OCEAN RESOURCES REQUIRES ACCURATE CHARACTERIZATION OF THAT ENVIRONMENT. HOWEVER, MANY VERY-NEAR SEAFLOOR PROCESSES EITHER CANNOT BE MEASURED REMOTELY, OR REQUIRE SEA TRUTH DATA. RESEARCHERS ARE DEVELOPING AN AUTONOMOUS AND ADAPTIVE SUBSEA SENSOR THAT MEASURES VERY-NEAR SEAFLOOR FLUID FLOWS, SUSPENDED PARTICULATES, AND MOMENTUM EXCHANGES IN THE COASTAL ENVIRONMENT. THEY ARE USING COMPUTER-AIDED SPECKLE INTERFEROMETRY TO DEDUCE A TWO-DIMENSIONAL FLOW FIELD FROM PARTICULATE SPECKLE PATTERNS. PARTICULATES ARE SAMPLED THROUGH LONG-DISTANCE MICROSCOPY. THE SENSOR HAS A TEMPORAL RESOLUTION GREATER THAN 33 MS OVER SUB-MILLIMETER TO METER SCALES. THE SYSTEM INCLUDES A LASER, MICROPROCESSOR, VIDEO CAMERAS, OPTICS, AND A RECORDER. POWER REQUIREMENTS ARE LOW-THE SENSOR OPERATES AUTONOMOUSLY FOR MONTHS. POWER IS CONSERVED WITH NEURAL NETWORKS-THE SENSOR MONITORS ENVIRONMENTAL CHANGES AND SAMPLES ACCORDINGLY. INVESTIGATORS VERIFY FEASIBILITY THROUGH HARDWARE TESTS. SPECIFICALLY, A LABORATORY INTERFEROMETER THAT DEDUCES 2-D FLOW FIELDS FROM NATURAL PARTICULATE SIDE-SCATTER WILL BE CONSTRUCTED. BOTH FLOW AND PARTICULATE DATA WILL BE USED TO TRAIN NEURAL NETWORKS TO RECOGNIZE ENVIRONMENTAL CHANGES. A DETAILED SENSOR DESIGN THAT INCLUDES POWER REQUIREMENTS, OPERATIONAL ESTIMATES, AND BIOFOULING AND CORROSION COUNTERMEASURES WILL BE PRODUCED.