Increased Data Processing Capabilities for Remotely Piloted Aircraft (RPA)

ABSTRACT: In this effort Toyon proposes to develop a set of technologies for improving onboard processing capabilities for RPA. This will include small form factor hardware, a library of high-level image processing routines, as well as general development support in order to allow end-users to further enhance the processing unit. Hardware will leverage commercial-off-the-shelf processing modules that conform to industry standard. Examples include COM Express and MXM. Furthermore, this hardware will communicate using standardized, high-speed serialized interfaces, such as PCIe and GigE. Image processing library support will be provided in the form of algorithm implementations that address commonly needed tasks such as registration, background modeling, tracking, and classification. These algorithms will be focused on general purpose graphics processing unit (GPGPU) using OpenCL. However, the platform will be heterogeneous, employing different processing units when appropriate. Toyon will seek to make the platform as extensible and easy to use as possible. Image processing routines will use open-source interfacing in the form of OpenCV. Extending the processing capabilities will be able to take place through not only direct OpenCL programming, but through high level language entry. This may include both OpenCV and MATLAB, with backend OpenCL/BLAS/OpenMP support. BENEFIT: The successful completion of this research will result in the development of hardware and software providing a potentially revolutionary advancement in the ability to deploy advanced image processing to RPA platforms. This capability will be of great interest for future military systems that rely on the automated processing of aerial surveillance video. For example, the developed technology is expected to benefit the Night Stare, Gorgon Stare, Angel Fire, Constant Hawk, TAILWIND, and ARGUS-IS programs as well as numerous smaller platforms. We note that our proposed solution is both modular and scalable. Other potential government applications include any situations where surveillance is required by an imaging sensor on an aerial platform, such as border patrol, crime fighting, traffic analysis, and counter-terrorism.