The Military & Aerospace Electronics take:
10 Jan. 2019 -- The embedded computing board from AlphaICs features 60 total operations processing (TOPs) per second, and operates at 40 Watts. Intended applications include drones, robots, and driverless cars. It is powered by AlphaICs's proprietary Real AI Processor (RAP) inference chip.
The RAP inference chip offers performance of 30 TOPs and consumes 30 Watts, with a software tool chain and set of libraries. RAP supports neural networks like convolutional neural networks recurrent neural networks, long short-term memory neural networks, and reinforcement learning neural networks.
The alphaEdge embedded computing board will be available in variants ranging from 8 TOPs at 8 Watts to 60 TOPs at 40 Watts. The board comes with a software tool chain that supports AI frameworks like TensorFlow, Caffe2, ONNX, and PyTorch.
John Keller, chief editor
Military & Aerospace Electronics
Ready to make a purchase? Search the Military & Aerospace Electronics Buyer's Guide for companies, new products, press releases, and videos