Thanks to everyone who’s following along with this project. For all our subscribers, you still have the chance to grab the early-bird discounts and support open hardware.
Updates to the software continue and we have also released our tentative roadmap to version 1.0 of the firmware that covers the majority of features we plan to support for the Stratus kit. You can read all about it in our repo. Every week more sample applications are being added. This sets a solid foundation for our users to kick start their project using the sample applications we provide and build their awesome projects on top. Please note all our firmware is fully open-source and you are free to use it as you like.
So don’t forget to check it out!
This week we have something special in-house for you!
The concept of Tiny Machine Learning (TinyML) has been around for a while, but only recently with the popularization of more efficient algorithms such as TensorFlow Lite and platforms like Edge Impulse, are we now able to see more applications being easily created and deployed on low-power and constrained embedded devices. Edge Impulse is the leading development platform for machine learning on edge devices and it is free for developers.
The Stratus kit is a versatile cellular IoT platform built around Nordic Semi’s nRF9160 with Cortex-M33. With 1 MB of Flash, 256 KB of RAM, and 500 MB of cellular data making it the perfect board for edge computing and executing machine learning models right on the platform without needing any external MCU or carrier boards.
This week’s tutorial walks through how to build and run your machine learning models using the Stratus kit and Edge Impulse studio. We will show you how to inject the accelerometer sensor data into the Edge Impulse studio from Stratus, test and train your ML model, and finally deploy it onto Stratus.
This is just the starting point for running ML models on the edge using the Conexio Stratus cellular IoT kit. In the upcoming weeks, we will also demonstrate how to perform continuous classification on Stratus and send out the results via a cellular connection to the cloud.
The complete source code for this tutorial, including the accelerometer data injector, ML model, and classification code can be found in our GitHub repo.
That’s all for now with more to come.