Published in Proceedings of BRL-CAD Symposium '95 Aberdeen Proving Ground, MD, 5-9 June 1995.

Modern missiles utilize multiple sensors to locate their targets. This paper describes a project to produce a physically accurate digital simulation of all sensor inputs for the missile's automatic target recognizer in real time, with image quality equal to that of real military sensors. The approach taken is to combine highly detailed terrain, tree, vegetation, and man-made structure models, highly detailed target geometry, and highly detailed target signature modeling, using ray-tracing to interrogate the scene. The simulations employed will be as true to the underlying physics of the situation as is possible.

In order to approach real-time performance, the initial system is being implemented using an 8-node, 96-processor Silicon Graphics Incorporated (SGI) Power Challenge Array, with a peak system performance of over 28 billion floating point operations per second (28 GFLOPS). The goal is to produce five television-resolution frames per second simultaneously in three spectral bands: optical, thermal, and radar.

The primary goal is to produce real-time imagery for the development and testing of missile automatic target recognition (ATR) systems. This work also addresses one of the pressing needs of the Defense Interactive Simulation (DIS) community by providing the ability to add a physically accurate high-resolution multispectral signature generation node to a distributed simulation when new sensor technology needs to be explored.

HTML Version, Postscript Version with diagrams. Form1.

Published in Proceedings of BRL-CAD Symposium '95 Aberdeen Proving Ground, MD, 5-9 June 1995.

Ray-tracing is essential when it is necessary to accurately compute reflection and refraction effects, shadows with penumbrae, and energy transport through participating media. In this new effort, a combination of hardware and software techniques is employed to produce ray-traced renderings of models of enormous complexity (>500,000,000 polygon equivalents) in near-real-time, with the goal being to achieve performance of five television-resolution frames per second. For this class of problem, the ray-tracing technique stands to produce images at a substantially faster rate than any known scanline-rendering system, while simultaneously bestowing all the advantages of ray-traced rendering.

This effort uses a commercial off-the-shelf (COTS) parallel processor array, high-speed HIPPI and ATM networks, and real-time image display hardware. The initial system is being implemented using an 8-node, 96-processor Silicon Graphics Incorporated (SGI) Power Challenge Array. By adding existing BRL-CAD network-distributed ray-tracing software, the result is a system capable of producing real-time ray-traced optical image generation of low-complexity geometry.

The primary goal is to produce real-time imagery for the development and testing of missile automatic target recognition (ATR) systems. This work also addresses one of the pressing needs of the Defense Interactive Simulation (DIS) community by providing the ability to add a physically accurate high-resolution multispectral signature generation node to a distributed simulation when new sensor technology needs to be explored.

HTML Version, Postscript Version with diagrams. Form1.