Elma Electronics' new JetSYS-5320 addresses the growing data processing needs of very rugged and mobile embedded computing applications by easily handling data-intensive computing tasks. It provides complex GPGPU edge inferential calculations for DL and ML operations in AI applications.
The system is designed for high-performance intelligent video analytics, VR, AR and AI edges, as well as robotics and unmanned or autonomous vehicles, fusing NVIDIa-based computing power and unmatched ruggedness on a compact, low-power platform.
New devices enhance visual intelligence in transportation and defense applications, including high-resolution sensor systems, automatic object recognition, motion-tracking safety systems, threat location detection and prediction, as well as machine condition-based monitoring and predictive maintenance, semi-autonomous driving and driving staff consultation system.
"Ai-based computing is driving the embedded computing industry to achieve multiple levels of intelligence and security advances," said Ram Rajan, vice President of engineering at Elma Electronics. We saw huge growth in target detection and tracking, video surveillance, target recognition and condition-based monitoring, so we built an AI platform that can handle the processing requirements of such applications used in challenging environments."
At the heart of the device is the NVIDIA Jetson TX2i SoM, which allows for real-time AI inference and DL/ML capabilities. Equipped with a dual-core Denver 2-bit CPU, quad-core ARM A57 Complex and 256 CUDA cores, NVIDIA's Pascal architecture delivers 1.3TFLOPs of performance.
The rugged system also offers HD-SDI, Gigabit Ethernet with PoE, USB 3.0 port for video capture, and miniPCIe expansion slots. It operates in the -40C to 71C temperature range, provides IP67 protection and meets mil-STD-810G harsh environment requirements.
