ST Microelectronics, an electronics and semiconductor manufacturer, has announced the availability of Version 3 of its NanoEdge AI (artificial intelligence) Studio, the first major upgrade of the software tool for machine-learning applications.

The new version of NanoEdge AI Studio comes as the shift of AI capabilities from the cloud to the edge offers manufacturers the potential to improve industrial processes, optimize maintenance costs and deliver innovative functions in equipment that can sense, process data, and act to improve latency and information security. Applications include connected devices, household appliances and industrial automation.

Source: ST MicroelectronicsSource: ST Microelectronics

NanoEdge AI Studio simplifies the creation of machine learning, anomaly learning, detection and classification on any STM32 microcontroller. This new release also includes prediction capabilities such as regression and outliers libraries. The tool makes it easier for users to integrate machine-learning capabilities quickly, easily and cost-effectively into their equipment without data-science expertise.

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

Other features of the NanoEdge AI Studio V3:

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

Additional information is available on the ST website.

To contact the author of this article, email