STMicroelectronics Streamlines Machine Learning Software Development for Connected Devices and Industrial Equipment with Upgrades to NanoEdge AI Studio


STMicroelectronics, today announced the availability of version 3 of NanoEdge AI Studio, the first major software tool upgrade for machine learning applications that ST acquired with Cartesiam earlier this year.

The new version of NanoEdge AI Studio comes as the shift in AI capabilities from the cloud to the edge offers manufacturers phenomenal potential to fundamentally improve industrial processes, optimize maintenance costs and offer innovative functions in equipment capable of detecting, processing data and acting locally to improve latency and information security. Applications include connected devices, household appliances and industrial automation.

NanoEdge AI Studio simplifies the creation of machine learning, anomaly learning, detection and classification on any STM32 microcontroller. This new version also includes prediction features such as regression and outliers libraries. The tool makes it easier for users to integrate these cutting-edge machine learning capabilities quickly, easily, and cost-effectively into their equipment. No data science expertise is required.

By adding native support for all STM32 development boards, ST has also eliminated the need to write code for its industrial-grade sensors with new high-speed data acquisition and management capabilities. NanoEdge AI Studio software improves security by using local data storage and processing, instead of transferring and processing data in the cloud.

What our clients say:
Steve Peguet, Scientific Director, Innovation Department of Alten Group, an international technology consulting and engineering company, said: “We had the opportunity to use NanoEdge AI Studio with one of our major aerospace customers. For mechanical drilling when manufacturing expensive parts, where a worn drill bit or even the smallest anomaly can have serious consequences, Alten used NanoEdge AI Studio to integrate machine learning algorithms into the drilling equipment. The solution tested on a production line was so effective that Alten launched a practice around this technology to support its customers and industrialize these initial results to deploy a disruptive solution for the prescriptive maintenance of drilling tools in their factories.

David Dorval, CEO and Founder of Stimio, a company specializing in the development of industrial IoT solutions for rail and other industries (IIoT), said: “Our major rail customers ask us to provide them with standalone, low-power wireless predictive maintenance solutions to increase uptime, optimize costs and avoid costly downtime. The contribution of low power AI Edge is at the heart of our strategy and after comparing several Edge AI software solutions, we chose NanoEdge AI Studio from STMicroelectronics to enrich our Oxygen Edge offering with powerful low power AI algorithms.. “

Deepak Arora, President and CEO of Wearable Technologies Inc, said: “To protect our loved ones so that they can have healthy and fulfilling lives, NanoEdge AI enables us to reduce the development time of machine learning for our next generation personal security devices. AI running at the edge of our devices will allow us to make informed decisions quickly with greater accuracy and reduced false positives.

Main features of NanoEdge AI Studio V3

  • Completely redesigned user interface to make it easier for non-specialists to develop the state of the art machine learning libraries.
  • New high speed data acquisition and management on the STWIN development board making all industrial grade sensors easily manageable without having to write a single line of code.
  • Improved support for anomaly detection, particularly useful for predictive maintenance to anticipate wear phenomena or to better cope with equipment obsolescence.
  • Learn normalcy directly on STM32 MCUs using a small data set or use new algorithms to practice without ever seeing anomalous patterns before.
  • Added regression algorithms to extrapolate data and predict future data models for energy management or forecasting remaining equipment life.
  • Native support for all STM32 development boards, no configuration required.

For more information visit www.st.com


Gordon K. Morehouse