26 October 2005 High-performance computers for unmanned vehicles
Author Affiliations +
The present trend of increasing functionality onboard unmanned vehicles is made possible by rapid advances in high-performance computers (HPCs). An HPC is characterized by very high computational capability (100s of billions of operations per second) contained in lightweight, rugged, low-power packages. HPCs are critical to the processing of sensor data onboard these vehicles. Operations such as radar image formation, target tracking, target recognition, signal intelligence signature collection and analysis, electro-optic image compression, and onboard data exploitation are provided by these machines. The net effect of an HPC is to minimize communication bandwidth requirements and maximize mission flexibility. This paper focuses on new and emerging technologies in the HPC market. Emerging capabilities include new lightweight, low-power computing systems: multi-mission computing (using a common computer to support several sensors); onboard data exploitation; and large image data storage capacities. These new capabilities will enable an entirely new generation of deployed capabilities at reduced cost. New software tools and architectures available to unmanned vehicle developers will enable them to rapidly develop optimum solutions with maximum productivity and return on investment. These new technologies effectively open the trade space for unmanned vehicle designers.
© (2005) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
David Toms, David Toms, Gil J. Ettinger, Gil J. Ettinger, } "High-performance computers for unmanned vehicles", Proc. SPIE 5986, Unmanned/Unattended Sensors and Sensor Networks II, 59860D (26 October 2005); doi: 10.1117/12.630741; https://doi.org/10.1117/12.630741


Tracking for the '90s
Proceedings of SPIE (September 30 1990)
An ATR challenge problem using HRR data
Proceedings of SPIE (April 14 2008)
Survey of radar-based target recognition techniques
Proceedings of SPIE (July 31 1991)
Bayesian probabilistic inference for target recognition
Proceedings of SPIE (June 13 1996)

Back to Top