AI for Embedded Defense is Here

July 18, 2019

An armed Humvee is moving at night thru a smoke-filled, urban battlefield. Using multiple displays, the crew has a 360˚ view that is daylight clear, delivered by an imaging system fusing data from optical, infrared, lidar, and radar sensors.

The imaging data, combined with RF input, is also feeding into an onboard AI capability that identifies objects and categorizes them into threat levels. A moving object is instantly focused on
by sensors and a window on the crew’s display zooms in; both the crew and the AI deep learning algorithms identify the moving object as a stray dog – no threat. The Humvee never pauses.

Previous Article
Autotalks and Datang Fulfill Chinese C-V2X Direct Communication Interoperability Testing

Autotalks and Datang have completed assessing chipset-level interoperability of their corresponding solutio...

Next Article
Programmable RISC-V-based AI Reference Camera Supports MicroPython, OpenMV
Programmable RISC-V-based AI Reference Camera Supports MicroPython, OpenMV

The 28 nm Kendryte K210 SoC at the heart of the camera leverages dual 64-bit RISC-V CPU cores running at 40...