Embedded

Equipped with AMD’s first desktop processor, the Ryzen Embedded 5000, the AIMB-522 is an industrial Micro-ATX motherboard for AI image processing in automation and surveillance applications.

According to Advantech, the Rysen Embedded 5000’s 16 Zen 3 cores are complemented with “a suite of technologies that empower high computing performance efficiency”. The board has PCIe Gen 4 expandability, four Gigabit Ethernet ports, and eight USB 3.2. The AIMB-522 is suitable for factory automation, smart logistics, and intelligent surveillance applications.

The AMD Desktop Processor Ryzen Embedded 5000 processor provides instructions per clock, 16 cores and L3 cache for latency-sensitive applications in smart manufacturing, automated visual inspection and intelligent surveillance. These performance gains are delivered with no increase in power consumption or TDP (thermal dynamic performance) said Advantech. The 7nm process enables the processor to deliver a 24 per cent improvement in energy efficiency and 2.8-fold lead compared to previous generation processors and competing architectures respectively, said Advantech.

Computer vision applications require the seamless integration of high speed digital cameras. The AIMB-522 features four Gigabit Ethernet ports and eight USB 3.2 10Gbits per second ports. The interfaces provide connectivity for high data throughput cameras with transfer bandwidths up to 350Mbyte per second and / or 60 frames per second. As a result, the AIMB-522 is capable of integrating more than 10 video cameras without additional peripherals cards. For extra connectivity, the onboard PCI-Express four slots offer the flexibility needed to support more camera devices or industrial controllers with add-on cards.

The Advantech AIMB-522 supports one PCI-Express x16 Gen4 bus standard for graphics demanding applications. Its two PCI-Express x4 slots offer the expandability needed for robotic controller card integration, while the onboard M.2 M-Key socket supports high speed SSD for real time operating system (OS) operations. 

The Advantech AIMB-522 industrial motherboard is available now. 

http://www.advantech.eu

BeagleBoard.org’s first 64bit open hardware single board computer (SBC) the BeagleBone AI-64 SBC is available from stock from Farnell (in EMEA, Newark in North America and element14 in APAC).

It is designed for building embedded applications and brings a complete artificial intelligence (AI) and machine learning system to developers, said Farnell.

“We believe this board will capture the imaginations of designers and empower them to build complete and powerful AI systems,” said Christine Long, CEO of the BeagleBoard.org Foundation. She continued that it is offered at “an extremely competitive price point, we are excited about the new applications that BeagleBone AI-64 will enable for new and experienced users.”

The BeagleBone AI-64 SBC provides all the on-board peripherals required to start learning and building applications. The BeagleBone cape headers provide expansion possibilities with hundreds of open source hardware examples and dozens of off-the-shelf embedded expansion options. The user needs only to add a simple web browser, power source and network connection, confirmed the company. 

The SBC is described as a milestone for BeagleBoard.org by Jason Kridner, president of the BeagleBoard.org Foundation board. “With multiple SuperSpeed USB ports, familiar BeagleBone cape expansion headers and desktop-capable performance, the general purpose embedded applications for this board are endless, with eight TOPS [terra operations per second] neural network performance accessible through familiar Python libraries to boot.”

BeagleBone AI-64 builds on more than a decade of success in open hardware single board Linux computers. It combines the power of the Texas Instruments TDA4VM SoC with dual Arm Cortex-A72 cores, a programmable C7x DSP core and deep learning, vision and multimedia accelerators to offer developers access to fast analytics, data storage options and high speed interfaces. In addition to “a vast amount of computing power”, the SBC includes connectors to create applications such as autonomous robots and drones, smart buildings and factories, home security, retail automation, media servers, machine learning, machine vision and computer vision.

The BeagleBone AI-64 board feature set includes the BeagleBone cape header compatibility with existing add-on boards and ikroBUS Shuttle header, providing access to hundreds of existing Click sensors and actuators. The memory capability is 4Gbyte LPDDR4, 6Gbyte eMMC flash with high speed interface and a MicroSD card. An M.2 E-key PCIe connector interfaces with Wi-Fi / Bluetooth adapters and there is a USB 3.0 Type-C SuperSpeed interface for power input and data and two USB 3.0 Type-A SuperSpeed interfaces and Gigabit Ethernet.

For camera and display options there is a Mini DisplayPort interface, a two or four-lane CSI connector for camera options, and a four-Lane DSI connector for popular display types.

User interfaces include boot, reset and power buttons, a power indication LED and five user LEDs. In addition, there is a 5V DC input power, UART and JTAG 10pin Tag-Connect for debug also included.

In addition to the TDA4VM, there are vision processing accelerators with an image signal processor (ISP) and multiple vision assist accelerators including DMPAC, dual 64-bit Arm Cortex-A72 microprocessors, six Arm Cortex-R5F MCUs at up to 1.0GHz, memory subsystem with up to 8Mbyte of on-chip L3 RAM with ECC and coherency and 12 multi-channel audio serial port modules. 

http://www.element14.com