The latest member of Nordic Semiconductor’s Thingy series, the Thingy 53, combines a flagship dual-core SoC, power management IC, PA/LNA range extender, and multiple sensors, along with embedded machine learning firmware, accelerating the realization of advanced wireless proofs-of-concept.
Nordic Semiconductor has released the multi-sensor prototyping platform ‘Nordic Thingy:53’ (“Thingy:53”), which features multi-protocol short-range wireless connectivity and supports embedded machine learning (ML). It is an ideal platform for building advanced wireless proofs-of-concept with ML capabilities in the shortest development time.
The Thingy:53 platform is based on Nordic’s nRF5340 advanced multi-protocol system-on-chip (SoC), which features a dual-core Arm® Cortex® M-33 and integrates the company’s nPM1100 power management IC (PMIC), as well as an nRF21540 front-end module (FEM) with a power amplifier/low-noise amplifier (PA/LNA) range extender. This prototyping platform is equipped with a rechargeable 1350 mAh lithium polymer battery and multiple motion and environmental sensors, supporting Bluetooth® Low Energy (Bluetooth LE), Thread, Matter, Zigbee, IEEE 802.15.4, NFC, and Bluetooth mesh protocols, and is pre-installed with embedded ML firmware.
The core component of the Thingy:53, the nRF5340 SoC, features a dedicated application processor and a network processor. The Arm Cortex-M33 application processor has a clock frequency of 128 MHz, ensuring that the prototyping platform can handle advanced algorithms and complex computational tasks related to embedded machine learning, with 1 MB of flash memory and 512 KB of RAM providing ample memory for the most complex applications. The network processor is designed for energy-efficient operation, with its own computing resources, ensuring robust wireless connectivity without impacting the execution of application software.
The Thingy:53 is the result of close collaboration between Nordic and the U.S. “tinyML” specialist Edge Impulse. This platform is equipped with Edge Impulse firmware and the nRF Edge Impulse mobile application. The firmware collects training and testing data through the sensors on the Thingy:53 and wirelessly forwards the data to the mobile application via Bluetooth. Subsequently, the mobile application forwards the data to Edge Impulse Studio (a cloud-based ML development platform) for building and testing embedded ML models. Users can deploy ML models to the Thingy:53 wirelessly using the same mobile application, which also directly displays inference results. This firmware and application enable developers to quickly collect data and test ML models without needing to connect any cables to the Thingy:53.
The Thingy:53 platform comes with the nRF Programmer application, allowing developers to select pre-made firmware and wirelessly update the Thingy:53 firmware directly through iOS or Android devices, significantly simplifying the prototyping process. This feature means users can deploy new firmware anytime, anywhere, without needing a desktop computer. The Thingy:53 is also compatible with Nordic’s mobile applications, nRF Mesh, nRF Connect, nRF Toolbox, and nRF Blinky.
The Thingy:53 includes a six-axis inertial measurement unit and an additional low-power accelerometer, with environmental sensors for temperature, humidity, air quality, and pressure. The product also integrates color and light sensors, a buzzer, and a pulse density modulation (PDM) microphone. The embedded ML capabilities allow developers to use the Thingy:53’s sensors for applications such as voice recognition or motion pattern detection. When triggered by motion or sound events, users can wake the nRF5340 SoC from sleep mode using the accelerometer and PDM microphone. Since the Thingy:53 can remain in sleep mode without external stimuli (extending battery life), it is very useful for creating low-power embedded ML applications.
The Thingy:53 features a USB-C charging and data port, as well as a four-pin JST connector compatible with Qwiic, Stemma, and Grove. Each Thingy:53 comes with a debugging and current measurement board, allowing access to relevant pins for application code troubleshooting and monitoring power consumption with other debugging tools.
Pär Håkansson, Product Manager for Thingy:53 at Nordic, stated: “Nordic’s ‘Thingy’ series has already been recognized by the developer community, and this platform is the fastest way to build and develop innovative short-range wireless and cellular IoT applications. Now, Thingy:53 is the prototyping platform using Nordic’s most powerful multi-protocol SoC, expanding our product range.”
“The new Thingy platform not only features the nRF5340 but also integrates Nordic’s PMIC and range extender for excellent battery performance and link budget, using multiple advanced sensors. By utilizing the pre-programmed Edge Impulse firmware and nRF Edge Impulse mobile application, even developers without machine learning experience can easily grasp this technology for the first time. Machine learning is key to advanced IoT edge processing in the future.”
Nordic will host a webinar on July 6 to demonstrate the main features of Thingy:53. To register and learn more, please contact Nordic agent Ruozhuo Product Sales Manager Hong Lan: +86 18575539388