Explore the journey of building a TinyML-powered artificial nose in this 25-minute conference talk from the EMEA 2021 tiny Talks series. Discover how a simple Cortex-M4 microcontroller and a gas sensor evolved into a complete open-source and open-hardware DIY project with wide-ranging applications. Learn about the challenges faced in understanding neural networks, the development process, and the potential of TinyML in IoT and connected environments. Gain insights into hardware selection, sensor integration, model tradeoffs, and the concept of digital twins. Perfect for embedded developers interested in TinyML applications and their real-world implementations.
Lessons Learned from Building a TinyML-Powered Artificial Nose