The newly revealed BeagleBone AI is a board aimed at developers interested in experimenting with machine-learning and computer vision. Unveiled this week, the computer houses four dedicated chips originally designed to help self-driving cars "see" the world around them. The board's Texas Instruments (TI) Embedded Vision Engine (EVE) chips offer up to 8x the performance per watt when running calculations for computer-vision models compared to running on an Arm Cortex A15-based CPU. This optimized hardware is accessible to developers via the TI Deep Learning OpenCL (Open Computing Language) API. Foundation say the BeagleBone AI board will be able to automate tasks in industrial, commercial and home settings.
Apart from the PocketBeagle, most of the new boards released recently based on the BeagleBone architecture have been built by third parties, exploiting the platform's main advantage over the Pi, the open hardware design. However that has just changed with a new board from the BeagleBoard.org Foundation, the BeagleBone AI, unveiled at Embedded World, which opened today in Nuremberg, Germany. The board shares the familiar BeagleBone form factor, and has compatible headers. But, unlike previous models, it is built around the Texas Instruments AM5729, a dual-core Arm 32-bit Cortex-A15 processor running at 1.5GHz.
Before Raspberry Pi rocked the world of makers, boards from BeagleBoard.org were the computers of choice among developers who were looking to create cool gadgets. One of its boards, BeagleBone, isn't as popular as it used to be, but it still has a loyal following. Seeed Studios has taken a version of the open-source board and given it a much-needed wireless upgrade, adding Wi-Fi and Bluetooth support. The BeagleBone Green Wireless is a significant improvement over predecessors: among other things it now allows makers to add wireless capabilities to smart home devices, wearables, health monitors and other gadgets. The upgrade also brings BeagleBone into the Internet of Things era, in which wirelessly interconnected devices are constantly exchanging data.
The $25 PocketBeagle is based on the Octavo Systems OSD3358 system-in-package (SiP), the same SiP that powers the credit card-sized BeagleBone Black Wireless, but is half the size. The SiP module features a 1GHz Texas Instruments ARM AM3358 processor, 512MB of DDR3, and power management. The processor also has two 200MHz programmable real-time units, which the BeagleBoard foundation says makes them ideal components for building things like drones, 3D printers, robots, and laser cutters. The RPUs are good for handling a lot of little tasks, which require low latency and low overhead, while the ARM processor is for handling high-throughput applications, according to the BeagleBoard foundation. It also features 72 expansion pin headers with power and battery inputs and outputs, high-speed USB, eight analog inputs and 44 digital input/outputs.
A newly released BeagleBone AI board will help developers who are interested in getting more out of machine learning and computer vision. The computer has four dedicated chips designed to help self-driving cars. BeagleBoard has Texas Instruments (TI) Embedded Vision Engine (EVE) chips that offer up to 8x performance per watt while running calculations for computer vision models. The performance is better as compared to running on an Arm Cortex A15-based CPU. The hardware is accessible to developers using TI Deep learning OpenCL (Open Computer Language) API.