Tiny RASynBoard combines Renesas RA6M4 MCU with Syntiant NDP120 ML accelerator, WiFi & BLE module, and some sensors

Avnet RASynBoard is a tiny board that packs a lot of features in a 30x25mm form factor with a Renesas RA6M4 Cortex-M33 microcontroller, a Syntiant NDP120 Neural Decision Processor, a Renesas (Previously Dialog Semi) DA16600 Wi-Fi 4 & Bluetooth 5.1 combo module, and a 6-axis inertial measurement unit (IMU) and digital microphone from TDK.

The RASynBoard is offered as part of an EVK with the Core Board described above plus an IO board with headers, a Pmod connector, a MikroE Shuttle Click header, a microSD card slot for storage, and a built-in debugger, plus two buttons and an RGB LED.

 

RASynBoard specifications:

• Main microcontroller – Renesas RA6M4 Arm Cortex-M33 microcontroller @ 200 MHz with 1 MB flash memory, 256 KB SRAM
• ML accelerator – Syntiant NDP120 Neural Engine based on Syntiant Core 2 Deep Neural Network,  Arm Cortex M0 and HiFi 3 DSP
• Wireless module – Renesas DA16600 2.4 GHz 802.11b/g/n Wi-Fi 4 1×1 and Bluetooth 5.1
• Storage – 16 Mbit SPI NOR Flash
• USB – 1x USB 2.0 Type-C port for power and programming
• Sensors – 6-axis motion sensor (ICM42671), PDM digital microphone (MMICT5838)
• Expansion Connector – 2x 28-pin board-to-board connector for IO board
• Battery Management – LiPo battery management and connector
• Dimensions – 30 x 25 mm

The RA6M4 MCU programming and debugging can be done with Renesas e2 Studio IDE and the on-board the E2OB debugger MCU on the IO board. Pre-trained NDP120 AI/ML models are loaded from local SPI Flash storage for execution on the low-power neural accelerator device.

 

IO board specifications:

• Storage – MicroSD card slot
• Onboard Debugger and USB-Serial interface
  • Renesas E2 OB debugger MCU (USB-C to SWD and VCOM serial interface)
  • 3.3V buck regulator for debugger circuits
• Expansion Interfaces and Storage
  • 2x 28-pin board-to-board connector for RASynBoard Core Board
  • 2x 8-pin MikroE shuttle click header for MikroE Click boards
  • 2x 6-pin Pmod type-6A (I2C) socket
  • 2x 7-pin MCU expansion header
  • 2x 3-pin DMIC expansion headers (two)
  • 3.3V level-translation of expansion
• User Interfaces – 2x Button Switches (Reset and User), 1x User RGB LED
• Dimensions – 50 x 30 mm

The RASynBoard EVK (Core Board + IO Board) comes with a Quick Start Instruction Card and targets a range of applications such as Edge-AI and Machine Learning, battery-powered remote sensor systems, motor predictive maintenance, industrial smart sensors, always-on speech recognition, and sensor fusion processing.

The RASynBoard EVK is listed for $99 on the Avnet website, but it’s unclear when it will become available since we are shown a factory lead time of 52 weeks, which may or may not be accurate. The Core Board is also listed but without price. More details may be found on the product page.

Via Hackster.io

Support CNX Software! Donate via cryptocurrencies, become a Patron on Patreon, or purchase goods on Amazon or Aliexpress