Next-Gen IoT: Arduino and Silicon Labs Make Matter More Accessible with Edge AI and ML
From Partnership to Innovation: Arduino and Silicon Labs Join Forces
In January 2024, Arduino and Silicon Labs announced a partnership that set out to simplify access to the Matter protocol in two phases. In the first phase, we saw the release of an extensive Arduino library for Matter along with a core for Silicon Labs microcontrollers. The second phase saw an addition to Arduino’s Nano family in the form of the Nano Matter board, based on the Silicon Labs MGM240S.
The outcome of this partnership is to bring the user-friendly features of Arduino’s interface to the robust RF chain and technical abilities of Silicon Labs in the field of the Matter protocol which would eventually help drive more innovation in IoT applications.
“This work with Arduino showcases the transformative power of AI in the IoT space,” said Rob Shane, Global VP of Mass Market Sales and Applications at Silicon Labs. “By combining Silicon Labs' cutting-edge AI/ML hardware accelerator equipped MCU with Arduino's intuitive platform, we empower developers to build smart, energy-efficient devices capable of performing real-time computations locally. This partnership is driving the creation of innovative applications that utilize real-time machine learning to boost automation, optimize energy efficiency, and deliver adaptive responses.”
Expanding the Arduino Ecosystem with Matter and Edge AI and ML
Arduino has a proven history of being one of the go-to choices for developers looking to quickly prototype their applications. With more than 100 hardware and software products in multiple categories such as boards, shields, carriers, and kits, as well as an integrated development environment (IDE) that provides support for users of all levels from students to seasoned professionals, Arduino has established itself as the platform of choice for IoT application development. Further, the readily available references in the Arduino user community coupled with the affordable cost of creating versatile end-user applications make it flexible and scalable in a huge variety of settings.
By embracing the Matter protocol, Arduino Nano Matter pairs advanced AI-driven intelligence with seamless interoperability. Real-time machine learning enhances automation, optimizes energy efficiency, and enables adaptive responses to environmental conditions. Compatible across leading ecosystems – including Apple, Google, and Amazon – its integration ensures reliable, efficient, and intelligent control of connected devices, making the future of smart home technology more accessible to all.
Unlocking AI at the Edge with ML Accelerator
In IoT parlance, “edge computing” refers to computations performed locally on the edge devices without storing or transferring data to a centralized, cloud-based server. This method of operation saves power and frees up network bandwidth for other, more critical operations. It also reduces overall development costs while minimizing signal latency. Increasingly, this computation is happening even closer to where the data is generated, such as sensor nodes. This is called the “tiny edge.”
Silicon Labs’ AI/ML SoCs – the EFR32.. xG24, xG26, and xG28 families - make use of this Edge processing to speed up operations and improve performance. The built-in Matrix Vector Processor (MVP) can accelerate ML inferencing by up to 8x while consuming 6x lesser power as compared to the MCU core. A suite of development tools such as Tensor Flow and Silicon Labs Simplicity Studio, together with a broad partner ecosystem for solutions provide developers with a complete and flexible ML ecosystem that enables rapid development and prototyping, thereby reducing time-to-market for the final products.
Experience the Arduino Nano Matter “Magic Wand” Demo in Austin
At the Edge AI Foundation’s event in Austin (February 25-27) Silicon Labs and Arduino will present a demo that exemplifies how their partnership enables the development of new-generation devices for smart home applications. Specifically, visitors will be invited to try an intuitive solution based on movement recognition: a “magic wand” that allows users to simply turn a light on and off by tracing a W (on) or an O (off) in the air.
“Our collaboration with Silicon Labs is all about bringing powerful, accessible technology to developers, and the ‘magic wand’ demo is a perfect example of that vision in action,” said Fabio Violante, CEO of Arduino. “By combining Matter, Edge AI, and TinyML on the Nano Matter board, we’re showing how intuitive and intelligent IoT applications can be – turning a simple hand gesture into a seamless smart home interaction. This is just the beginning of what’s possible when innovation meets accessibility.”
The demo is built on an Arduino Nano Matter, connected via carrier to an Arduino Modulino Movement node, which has an onboard 6-axis accelerometer and gyroscope. The Arduino Nano Matter is equipped with a powerful Silicon Labs MGM240S microcontroller, running Matter over Thread and a TinyML gesture recognition library that leverages TensorFlow Lite.
To try the demo and see it in action, visit the Silicon Labs booth in Austin on February 25-27. Here’s more information about the Edge AI Foundation event in Austin.
