Edge Impulse logo
Blog Post

RASynBoard Brings Low-Power, Edge AI Processing to Embedded Sensor Applications

Artificial Intelligence, Machine LearningAvnet's RASynBoard combines a  Renesas RA6M4 MCU, a Sentient NDP120, TDK sensors, and Edge Impulse support for ultra-low-power edge AI application.

Avnet

March 7, 2023

    Avnet offers compact sensor module and development kit for always-on sensing

    The RASynBoard development kit from Avnet can help hardware and software design engineers and embedded AI developers quickly adopt edge AI capabilities into their products, making them more efficient, accurate and reliable. It was created in collaboration with Renesas, Syntiant, TDK, and Edge Impulse, and will be demonstrated in the Avnet Embedded booth (1-510) at the Embedded World Exhibition and Conference, March 14-16, in Nuremberg, Germany. 

    RASynBoard targets industrial and consumer applications needing battery powered, always-on, smart sensing capabilities, such as sound recognition, command recognition, vibration sensing and environmental sensing. Its small (30mm x 25mm) size makes it ideal as an add-on sensor module for existing products, or its design can be customized and integrated in a larger system for enhanced sensing needs.

    The RASynBoard module, or Core Board, comprises two domains: an always-on AI processor with on-board sensors and an MCU-based controller with a low-power Wi-Fi/BT wireless module. The AI processing system has the ability to wake up the MCU controller when it detects an AI triggered “event,” allowing the MCU system to take action or send notification through the wireless module and then return to the low-power AI sensing mode. Developers can customize both the sensors connected to the AI processor as well as the AI algorithm running on the AI processor.

    The Syntiant NDP120 Neural Decision Processor is the AI processor used on RASynBoard and the Renesas RA6M4 MCU is the MCU controller. Also included on the Core Board is the Renesas DA16600 low-power Wi-Fi/BT combo module, and two TDK sensors: the ICM-42670-P 6-axis IMU and the MMICT5838 PDM digital microphone.

    RASynBoard can work standalone with nothing more than a battery. For development and prototyping it can be plugged into the IO Board, which is included in the RASynBoard Development Kit. The IO Board provides on-board debug and programming capabilities for the RA6M4 MCU, along with expansion capabilities for both the MCU and NDP120. Both Pmod™ and Click™ interfaces are available for adding different sensors or control functions, offering maximum flexibility for prototyping. The IO Board also includes a micro-SD card interface to the RA6M4 MCU, for additional data storage if needed.

    “Avnet’s strong relationship with industry leading suppliers such as Renesas, TDK and our new supplier Syntiant allows us to combine the unique offerings and capabilities of each partner into a complete system-level, production-ready solution,” said Jim Beneke, vice president Advanced Applications Group, Avnet. “Our design customization and manufacturing capabilities also allow us to offer production-ready solutions for customers needing a full-turnkey option.”

    "The RASynBoard provides a powerful, low-power solution that allows our customers to seamlessly implement IoT design using BLE or Wi-Fi,” said Mohammed Dogar, Vice President of Renesas MCU Business Development Division. “This opens up new possibilities for AI-enabled systems across a broad range of use cases. We have been working closely with Avnet to bring this exciting new technology to our customers and we look forward to continuing our collaboration in the future.”

    “Our hardware technology brings advanced multimodal neural processing to Avnet's new RASynBoard with very low power consumption,” said Mallik Moturi, chief business officer at Syntiant. “The NDP120’s ability to provide highly accurate, always-on sensor processing with relatively no impact to battery life was a key edge AI requirement in the development of the module, which also makes the device ideal for supporting a wide range of field applications in smart buildings, factories and cities.” 

    “As a market leader in MEMS motion sensors and microphones, TDK InvenSense offers high performance 6-axis IMU’s and digital microphones with the industry’s lowest power consumption to enable aways-on sensing,” said Song Li, director of product marketing at TDK InvenSense. “The RASynBoard module adds AI capability to the sensors and provides customers with easy access to smart sensing solutions and the AI algorithms behind it."

    “Edge AI has become a valuable enterprise tool due to advancements in compact devices that very efficiently offer high levels of compute,” says Raul Vergara, Edge Impulse executive vice president of growth. “Edge impulse is excited to collaborate with Avnet on the launch of the RASynBoard, an ideal solution for ultra-low-power machine learning applications thanks to its Renesas MCU and sub-mW Syntiant NDP120 processor, and flexible array of sensors. Customers can quickly develop advanced models in Edge Impulse Studio and deploy them to the board for always-on inferencing in almost any location or environment.”

     The RASynBoard kit features:

    • Always-on, edge AI sensing capability to detect events based on real-time sensor data
    • Dedicated AI processor offering the flexibility to run single neural network or multiple neural networks in parallel, providing true sensor fusion AI processing capability
    • A small sensor board that can work standalone or plugged into the included IO Board for easier sensor prototyping and software development
    • Included Wi-Fi/BT module providing wireless connectivity for enhanced cloud capabilities with Avnet’s IoT Connect and AWS and Azure support
    • Edge Impulse Studio support for AI model development, training, optimization, and deployment
    • e2 studio IDE from Renesas enabling application development for the RA6M4 MCU

    The kit will be available initially in the U.S., Canada, Europe and Japan. It can be pre-ordered using part number: AES-RASYNB-120-SK-G at Avnet.me/RASynBoard for $99 (USD), with delivery in late Q2 2023.

    Subscribe

    Are you interested in bringing machine learning intelligence to your devices? We're happy to help.

    Subscribe to our newsletter