Kendryte K510 tri-core RISC-V AI processor deliver up to 3 TOPS

Kendryte K510 Block Diagram

Kendryte K510 is a 64-bit tri-core RISC-V processor clocked at up to 800 MHz with AI accelerators that succeed the 400 MHz Kendryte K210 dual-core RISC-V AI processor released a few years ago first in Kendryte KD233 board, and then boards like Maxduino or Grove AI HAT conveniently programmable with Arduino or Micropython. Canaan formally announced the processor yesterday at the 2021 World Artificial Intelligence Conference claiming K510 had three times the performance of K210 making it suitable for UAV high-definition aerial photography, high-definition panoramic video conferences, robotics, STEAM education, driver assistance scenarios, and industrial and professional cameras. The press release did not have much information, but multiple sources provided additional details to CNX Software, so we have Kendryte K510 specifications: Processor – 2x 64-bit RISC-V processor @ 800 MHz, and 1x 64-bit RISC-V core @ 800 MHz with DSP extension AI subsystem with 3 TOPS in total KPU: General […]

TI AM64x 7-core processor is made for PLC’s, motor drives, industrial robots

AM64x GP EVM development platform

Texas Instruments AM64x is a family of 64-bit Arm processors with functional safety designed for Programmable Logic Controllers (PLC), motor drives, remote I/O, and industrial robots. The top-end processor of the family, AM6442, comes with seven cores including two Cortex-A53 application cores, four Cortex-R5F real-time cores, and one Cortex-M4F isolated core. AFAICT, while the documentation is dated January 2021 and TI announced the processor in February in a blog post with a cryptic title, it was only first picked up by Embedded Computing in early May. Besides the processor itself, TI also provides an AM64x starter kit and a full-featured AM64x evaluation kit, and several companies are already preparing development boards and modules as we’ll see further below. TI AM64x processor AM64x key features & specifications: CPU cores Dual-core Arm Cortex-A53 processor @ 1.0 GHz with 256KB L2 shared cache with SECDED ECC, 32KB L1 D-cache, 32KB L1 I-cache Up […]

Aetina DeviceEdge Mini Edge AI computers support NVIDIA Jetson SO-DIMM modules

DeviceEdge Mini M1 Jetson modules

NVIDIA Jetson modules offer a wide range of AI performance starting from the entry-level 492 GFLOPS Jetson Nano module up to the 32 TOPS Jetson AGX Xavier platform. That’s probably Aetina decided to offer its DeviceEdge Mini AI edge solution with either Jetson Xavier NX, Jetson Nano, or Jetson TX2 NX pin-compatible SO-DIMM modules designed for applications in smart transportation, factories, retail, healthcare, AIoT, robotics, etc… The three models M1, M2, and M3 can all take the Jetson modules listed above, and come in a very similar form factor with the following key features: Common features: Storage – 128GB M.2 NVMe SSD (default) Video Output – Micro HDMI 1.4 connector Networking – Gigabit Ethernet RJ45 port, optional WiFi/Bluetooth M.2 module USB – 1x USB Type-C OTG only port Expansion 1x M.2 2242 M-key socket fitted with NVMe SSD 1x M.2 2230 E-key socket for optional Wifi/BT function 1x DB15 male […]

reServer 2-bay multimedia NAS is equipped with an Intel Core Tiger Lake SBC

reServer multimedia NAS

Seeed Studio has just introduced reServer dual-bay multimedia NAS based on Intel Core i3/i5/i7 Tiger Lake 11th generation processor. The networked storage device is based on ODYSSEY X86 V2 single board computer that offers an upgrade to the ODYSSEY-X86J4105 and ODYSSEY-X86J4125 Gemini Lake SBC’s. reServer comes with up to to 64GB dual-channel memory, dual 2.5GbE, support for two 3.5-inch SATA drives, HDMI and DisplayPort video output, and some USB 3.0/2.0 ports. reServer (preliminary) specifications: SoC (one or the other) Intel Core i3-1115G4 dual-core/quad-thread processor with 48 EU Intel UHD Graphics @ 400 – 1250MHz; 28W TDP Unnamed Intel Core i5 with 80 EU Intel Iris Xe Graphics G7 @ 400-1300MHz; 28W TDP Unnamed Intel Core i7 with 96 EU Intel Iris Xe Graphics G7 @ 400-1300MHz; 28W TDP Co-processor – Microchip ATSAMD21G18 32-Bit Arm Cortex-M0+ microcontroller System Memory – Up to 64GB dual-channel DDR4-3200 SDRAM via two SO-DIMM sockets Storage […]

MyCobot robotic arm is offered with Raspberry Pi 4 or M5Stack ESP32 modules

myCobot-Pi and myCobot-M5

Elephant Robotics’ MyCobot robotic arm was introduced last year with M5Stack Atom & Basic ESP32 modules with 6-degrees of freedom, a 250 grams max payload, available now for $599. The company has launched a new $699  version – MyCobot-Pi – with most of the same features, except for an upgrade to Raspberry Pi 4B which offers more performance and flexibility. Both versions of MyCobot arm robotic arm support accessories such as a gripper and a suction pump attachable through a LEGO connector, but while the original version – MyCobot-M5 – relies on Arduino programming, the new MyCobot-Pi runs Debian/Ubuntu and ROS operating systems with image recognition algorithms that can work with a variety of cameras. Since we’ve missed last year’s announcement, we’ll cover both models in this article. MyCobot Robotic Arm Despite the mechanical parts being the same, the specifications for MyCobot-M5 and MyCobot-Pi differ quite a lot: Controller MyCobot-M5 […]

Edge AI and robotics embedded computers feature NVIDIA Jetson TX2 NX module

NVIDIA Jetson TX2 NX computers

NVIDIA unveiled a cost-down version of the Jetson TX2 module with the Jetson TX2 NX SO-DIMM system-on-module last month, and AAEON is one of the first companies to have updated some of their products with the new module. AAEON’s BOXER-823x-series of fanless embedded computers feature the NVIDIA Jetson TX2 NX module, and multiple Ethernet ports, including some PoE capable for the development of edge AI and robotics applications. BOXER-8230AI AI Edge Fanless Embedded Box PC This model offers a mid-range alternative between the Jetson Xavier AGX NX-based BOXER-8250AI embedded box PC and the Jetson Nano-powered BOXER-8220AI mini PC both of which offer the same ports. BOXER-8230AI specifications: SoM (CPU/Memory/Storage) – NVIDIA Jetson TX2 NX with dual-core NVIDIA Denver 2 64-Bit CPU and quad-core Arm Cortex-A57 MPCore processor @ up to 2.0 GHz, 256-core NVIDIA Pascal GPU @ up to 1.3 GHz, 4GB LPDDR4, 16GB eMMC flash Storage – MicroSD card […]

GMSL2 camera kit supports 15-meter long cables, up to six cameras with NVIDIA Jetson Xavier AGX

6x gmsl2 cameras nvidia agx xavier devkit

We’ve recently seen it’s possible to use a long cable with the Raspberry Pi camera thanks to THine camera extension kit that works with 20-meter LAN cables using V-by-One HS technology. e-con Systems has now launched a similar solution for NVIDIA Jetson Xavier AGX, one of the most powerful Arm devkits available in 2021, with NileCAM21 Full HD GMSL2 HDR Camera that supports up to 15-meter long cable as well as LFM (LED Flickering Mitigation) technology. NileCAM21 camera features and specifications: Based on OnSemi AR0233 Full HD camera module with S-mount interchangeable lens holder Gigabit Multimedia Serial Link 2 (GMSL2) interface with FAKRA connector Shielded coaxial cable for transmission of power and data over long distances (up to 15m) High Dynamic Range (HDR) with LED Flickering Mitigation (LFM) Supported resolutions and max frame rates for uncompressed UYVY streaming VGA (640 x 480) up to 60 fps qHD (960 x 540) […]

This mini-computer is set to outperform NVIDIA Xavier by eight times

Fast Sense X Robotics AI Platform

When it comes to high-end AI-edge computer boards, we have seen Hackboard 2 with support for Windows 10 operating system and optional 4G or 5G cellular connectivity. Fast Sense Studios is a Russian-based startup designing powerful on-board computers to bring edge AI capabilities to mobile robotics. At Embedded World 2021, the company has unveiled the Fast Sense X Robotics AI Platform that integrates the power of six computers into a single module. The company claims the newly launched edge AI computer outperforms the NVIDIA Xavier with over 8x higher performance in mobile robots thanks to the advanced DNN inference hardware. Apart from the Intel UHD GPU, the computer also integrates up to 3 Coral Edge TPU devices and up to 2 Myriad X devices allowing it to give optimum results even when on heavy workloads. “We believe in the future where reliable physical agents become relevant for routine tasks that […]