Multiresolutional Particle Filters With Finite Mixtures for Ground Target Tracking and Identification

MRLets Technologies, Inc. proposes an innovative technology for high performance ground target tracking and identification using powerful multiresolutional sequential Monte Carlo methods (particle filters) with finite mixtures and GMTI/HRR measurements. This innovative technology consists of three major functional components: multiresolutional particle filters for computational efficiency, invariant-based tracking and identification for the performance robustness, and finite mixture models with the expectation maximization (EM) algorithm for effective information exploitation. The coherent combination of these three functional components will provide us with unprecedented power in feature extraction, target modeling, data association, state filtering, target identity description and matching. Particle filters have been proven to be very effective for nonlinear/nonGaussian filtering. However, the most notorious disadvantage of a particle filter is its formidable computational complexity, since thousands particles are usually needed, which is equivalent to running thousands of filters simultaneously. It has also been proven that the only technique of truly solving a computational problem is multiresolutional processing, both in temporal and spatial domains. Therefore, we propose a multiresolutional particle filter to reduce significantly the number of particles and update them in a lower rate, meanwhile maintaining the full strength of a particle filter. In addition, when applying a particle filter to a density function with multimodality due to multiple targets, weak targets could be dropped or merged because of poor localization property of a particle filter. We will improve the localization accuracy by using finite mixture with particles. Information extraction and target modeling for rigid-body targets could be greatly benefited from the invariant principle. In this project, we will extend our invariant-based tracking and identification algorithms to the multiresolutional particle filter implementation. We will also introduce a mixture-EM technique for effective HRR information exploitation. An immediate side product of using multiresolutional tracking is group tracking where targets with similar behaviors are tracked as a group via measurement compression. This proposal is packed with several important innovations. When fully developed, the proposed research should have significant impacts on the target tracking and recognition communities.