Enhanced Long-Range Maritime Vessel Classification

RDRTec has made significant advancements in the automated classification of ships at long ranges using real time feature extraction from Inverse Synthetic Aperture Radar (ISAR) imagery. RDRTec’s Maritime Classification Aid (MCA) can classify a ship to its fine naval class level using ISAR data collected at long ranges in all environments. While physical dimensions of major structural elements of the ship provide the primary features that feed the classification expert system, micro-Doppler based signatures associated with rotating antennas have been shown to provide important additional information to support separation between ship classes with similar physical features. The objective of this research project is to expand the exploitation of micro-Doppler signatures further by identifying ship structural vibrations due to the movement of the ship through water and the ships power plant.  In addition to improving classification of ship types that are difficult to separate from similar ship types (confusers), we will explore the use of vibration induced micro-Doppler artifacts for battle damage assessment and fingerprinting of a specific hull. RDRTec proposes to attack this topic in collaboration with a diverse, collaborative set of partners in order to examine the problem from multiple points of view, leverage a broad range of expertise, and increase the likelihood of transitioning the results to the US war fighter. This team is comprised of: RDRTec (small business research prime), Duke University (research institute for this topic), and the University of Michigan (additional subcontractor). Dr. Nickolas Vlahopoulos from the University of Michigan will apply the Finite Element Analysis (FEA) methods which he developed for vibro-acoustic analysis of complex Naval systems to generate simulated vibration data sets for ship cases to use in assessing the feasibility of the extraction of micro-Doppler signatures.  RDRTec will generate simulated ISAR returns based on the vibration datasets.  The resulting radar measurements will be processed at RDRTec using a modified version of algorithms currently used to extract rotator micro-Doppler artifacts and in parallel processed by Dr. Jeffrey Krolik from Duke University using two stage neural network (NN) techniques.  This will allow us to examine the feasibility of both physics and NN based approaches for exploiting vibration induced micro-Doppler.  Data sets will be generated for ships of the same size and shape but with different vibration characteristics to assess the ability to distinguish ships which are identical in all other regards.