Dynamic Data Driven Multistatic EO/RF Sensor Fusion
Abstract
To deal with the large amount of data generated in multi-level, cross-domain representation ofmulti-modal signals for efficient transmission, storage, manipulation, multi-modal data mining,machine learning and identification of signals, new sensory methods of information analysis anddetection are needed. Electro-optical (EO) and radio frequency (RF) sensor fusion couples thepoint detection range of the radar feature return with the EO sensors to reduce the large amount ofEO and RF sensory data for fast computation and robust detection of objects. This project willaddress the problem of information fusion when using multiple, spatially distributed EO and RFsensors for automatic target recognition and tracking (ATRT). With sensor, target andenvironment variations over space and time, the capability of adaptive reconfiguration is criticalto achieving optimal performance in real-time. The existing work in applying EO/RF multi-modalsensors for target recognition and tracking are either inadaptive, or adaptive only in the sense ofsteering measurement components, such as waveform adaptivity of RF sensors or zoom andpointing angle of EO sensors. There is a lack of reconfigurable design in different levels of theEO/RF fusion system. In this project, we will adopt dynamic data driven application system(DDDAS) concept to coordinate the tasks of data fusion, sensor management and performanceprediction dynamically. The proposed research will deal with challenges in major aspects ofDDDAS, including applications modeling, mathematical and statistical algorithms, measurements,and software infrastructure.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Sep 19, 2018
- Source ID
- FA95501810287
Entities
People
- Jia Li
Organizations
- Air Force Office of Scientific Research
- Oakland University
- United States Air Force