23 April 2010 A unification framework for best-of-breed real-time scene generation
Author Affiliations +
Abstract
AMRDEC sought out an improved framework for real-time hardware-in-the-loop (HWIL) scene generation to provide the flexibility needed to adapt to rapidly changing hardware advancements and provide the ability to more seamlessly integrate external third party codes for Best-of-Breed real-time scene generation. As such, AMRDEC has developed Continuum, a new software architecture foundation to allow for the integration of these codes into a HWIL lab facility while enhancing existing AMRDEC HWIL scene generation codes such as the Joint Signature Image Generator (JSIG). This new real-time framework is a minimalistic modular approach based on the National Institute of Standards (NIST) Neutral Messaging Language (NML) that provides the basis for common HWIL scene generation. High speed interconnects and protocols were examined to support distributed scene generation whereby the scene graph, associated phenomenology, and resulting scene can be designed around the data rather than a framework, and the scene elements can be dynamically distributed across multiple high performance computing assets. Because of this open architecture approach, the framework facilitates scaling from a single GPU "traditional" PC scene generation system to a multi-node distributed system requiring load distribution and scene compositing across multiple high performance computing platforms. This takes advantage of the latest advancements in GPU hardware, such as NVIDIA's Tesla and Fermi architectures, providing an increased benefit in both fidelity and performance of the associated scene's phenomenology. Other features of the Continuum easily extend the use of this framework to include visualization, diagnostic, analysis, configuration, and other HWIL and all digital simulation tools.
© (2010) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Joseph W. Morris, Joseph W. Morris, Gary H. Ballard, Gary H. Ballard, Darian E. Trimble, Darian E. Trimble, Dennis H. Bunfield, Dennis H. Bunfield, Anthony J. Mayhall, Anthony J. Mayhall, } "A unification framework for best-of-breed real-time scene generation", Proc. SPIE 7663, Technologies for Synthetic Environments: Hardware-in-the-Loop Testing XV, 76630R (23 April 2010); doi: 10.1117/12.851924; https://doi.org/10.1117/12.851924
PROCEEDINGS
12 PAGES


SHARE
RELATED CONTENT


Back to Top