Edge AI: Bringing Intelligence to Embedded Devices
Leonardo Cavagnis, Firmware Engineer @ Arduino
Artificial intelligence is no longer confined to the cloud. Thanks to Edge AI, we can now run AI models directly on embedded devices with limited power and resources. This session will explore the full pipeline of developing a Tiny Machine Learning (TinyML) model, from data collection to deployment, addressing key challenges such as dataset preparation, model training, quantization, and optimization for embedded systems. We’ll explore real-world use cases where AI-powered embedded systems enable smart decision-making in applications like predictive maintenance, anomaly detection, and voice recognition. The talk will include a live hands-on demonstration on how to train and deploy a model using popular tools like Google Colab and TensorFlow, and then run real-time inference on an Arduino board.