FP-AI-PDMWBSOC2

Active
Design Win

STM32Cube function pack for STEVAL-PROTEUS1 for AI anomaly detection and classification based on AzureRTOS

Get Software Download databrief

Product overview

Description

FP-AI-PDMWBSOC2 is an STM32Cube function pack for the STEVAL-PROTEUS1, able to get motion sensor data, process them to make anomaly detection and faults classification, and send the results to the STBLESensClassic mobile app, a PC terminal console and on Azure IoT central dashboard.

The function pack helps to explore the implementation and development of a predictive maintenance application that embeds the NanoEdge AI library middleware, capable to provide an AI-based solution (the NanoEdge AI libraries are generated using NanoEdgeAIStudio).

Using the datalogging binary file, the raw data can be extracted from the sensors inside the STEVAL-PROTEUS1 board and provided to the NanoEdgeAIStudio software to extract machine learning libraries for anomaly detection and classification.

These NanoEdge AI libraries, customized and built on the classified set of data, can be easily updated inside the framework architecture proposed, facilitating the early detection of warning status and potential failure within the equipment.

FP-AI-PDMWBSOC2 implements two different HMI communication tools at user level: a wired interactive CLI (USB CDC) to configure the node and manage the learning, detecting and classifying phases, and the STBLESensClassic app with dedicated pages to provide the same functionalities. The STBLESensClassic mobile app works also as a bridge to display data on Azure IoT Central dashboard.

To start/stop the learning, detecting and classifying phases, an additional control, indicated by LEDs, can be performed by just pressing the user button.

The NanoEdge libraries functionalities are based on a set of application-level modules (Sensor Manager, Digital Processing Unit, EM Data, PnPLCompManager), useful to reuse and easily extendable to build other customized applications.

  • All features

    • Firmware to develop a WPAN sensor node for predictive maintenance applications, featuring motion sensors and performing anomaly detection or classification controlled via Bluetooth® Low Energy connectivity
    • STM32 wireless personal area network middleware developed within the STM32WB framework used to support Bluetooth® Low Energy 5
    • Compatible with NanoEdgeAIStudio solution to enable AI-based applications
    • On-board battery status monitor
    • Compatible with STBLESensClassic app (Android and iOS) to enable AI library control and monitoring, settings by PnP-Like protocol messages, firmware update via fast FUOTA, and bridging data to the Azure IoT central PnP cloud dashboard
    • Based on accelerometer data up to 6 kHz bandwitdh
    • NanoEdge AI library generated to run in the STM32WB module. Anomaly detection model storable in the external NOR flash memory
    • Firmware modular example based on the embedded light object-oriented framework (eLooM) to enable code reusability at application level
    • Utilities: high speed datalog in binary format, python scripts and batch files to use it
    • AzureRTOS: ThreadX, USBX
    • Free, user-friendly license terms

Get Software