Target Characterization using McMODAF (Markov-chain Augmented Multiple-Modality Data Fusion) Technique

ABSTRACT: The proposed technique employs a variety of sensors that combine data to form a comprehensive picture of a situation of interest. The goal is to increase the effectiveness of military surveillance by giving a more complete, integrated picture of situations to enable a quicker response while eliminating errors from potential failures of individual sensors. In Phase I of this program, AS&T developed and performed a proof-of-the-concept demonstration of decision-support operations based on the fusion of heterogeneous data. This was done by utilizing sensors with differing modalities, some typical for imaging (visible, MWIR, LWIR) and some associated with non-imaging detection. The collected data used was to demonstrate optimal methods of identification, discrimination and tracking of the target necessary to determine various aspects of its state and features for augmented space surveillance. During Phase II, we will extend the proposed technology and enhance the operational level of the software for data fusion. By designing and assembling an FPGA-based multi-sensor data-fusion platform, AS&T will be able to verify and optimize performance of the predictive system. The program will culminate with technology laboratory tests and field demonstrations in an operational environment. BENEFIT: The US military relies upon a diverse array of sensing resources to achieve enhanced space situation awareness. As a result, given the vast resources providing a war-fighter with information, the military incorporates stochastic algorithms into sensor fusion systems to process the sensors' information. Military applications include, but are not limited to remote sensing, active tracking of fast-moving targets including space objects (satellites, space debris), and cross-communication systems. Commercial sectors that are effectually use the multi-parametric sensor fusion technology include robotics, geospatial information systems, business intelligence, and a broad variety of medical applications, including diagnostics, recovery prognosis and optimal recovery estimates, etc.