Multiple embedded and Cloud AI solutions at ST
summary
ST shows a tooling flow that converts TensorFlow-style neural nets into MCU code with a single-click export via STM32Cube.ai, demonstrated by on-node character recognition running entirely on a microcontroller. The ecosystem stitches sensors, SDKs, evaluation boards, gateways (microprocessors) and AWS cloud services to stream processed telemetry — for example vibration from two motors — so only meaningful events are sent to the cloud, reducing edge-to-cloud bandwidth and engineering time.
AI-written summary · about
related talks
24mRunning AI/Neural networks on microcontrollers made simple with the STM32Cube.AI
18mHow to turn your ideas in AI into products
28mMicrocontrollers (MCUs), Cloud, and a Chance of AI with Tara Walker
55mEdgeAI - AI in the Palm of Your Hand
42mThe Stakes of Green AI for Embedded Systems
19mWhy Machine Learning on Microcontrollers?
21mUnlocking the potential of low-power embedded devices with Edge AI software
59mML and the IoT: Living on the Edge