Examples

This folder contains ready-to-use examples for MEMS devices embedding the machine learning core (MLC).

Each folder is dedicated to a specific example and contains a README.md file with general information about that example and a set of sub-folders with the configuration for each of the supported MEMS sensors. Each sensor sub-folder contains:

A README.md file, with information about the specific configuration.
A .json file, containing the device configuration generated with STMicroelectronics tools. This file can be used to load the MLC program and sensor configuration into the sensor. For this purpose, different hardware tools provided by ST can be used (STEVAL-MKI109D, STM32 Nucleo, SensorTile.box PRO, etc.).
A .h file, containing the same information as the .json file, but in the format of a C header file. Examples of .h file integration in C projects are available in the platform independent drivers repository for STMicroelectronics MEMS sensors.

How to use the JSON configuration files to run the examples

Run the example with Professional MEMS tool board and MEMS Studio

To test an MLC example in MEMS Studio, the .json file has to be loaded as described in the following steps:

Download the latest version of MEMS Studio (if not already installed).
Plug the DIL24 adapter of the specific sensor to the Professional MEMS board.
Plug the Professional MEMS board to the PC with a USB cable.
- If necessary, upgrade the firmware of your board using the Firmware Programming functionality embedded in MEMS Studio.
Open MEMS Studio and select the desired device (choose the COM port and the used sensor).
Go to Sensor Evaluation, Load/Save Configuration and load the .json configuration file.
Once the device has been configured, press the ▶ button in the top left corner to start streaming data from the sensor and visualize the MLC results in the MLC Monitor or Data Monitor views.

Run the example with STM32 Nucleo board and MEMS Studio

MEMS Studio can work with Nucleo boards coupled with X-NUCLEO expansion boards (for example, X-NUCLEO-IKS4A1). Optionally, the DIL24 adapter of a sensor can be plugged on top of the X-NUCLEO expansion board. Do not plug a DIL24 adapter if it hosts a sensor with the same I2C address of one of the sensors host by the X-NUCLEO board, as a clash of addresses may cause problems.

To run an MLC example in MEMS Studio, the .json file has to be loaded as described in the following steps:

Download the latest version of MEMS Studio (if not already installed).
Download the latest version of X-CUBE-MEMS1.
Plug the X-NUCLEO expansion board on top of the Nucleo board.
Optionally plug the DIL24 adapter hosting a sensor with MLC in the DIL24 socket of the expansion board.
Plug the Nucleo board to the PC with a USB cable.
The Nucleo board has to be programmed with the DataLogExtended firmware from the X-CUBE-MEMS1 package (if not already programmed).
- Locate the DatalogExtended.bin file in the X-CUBE-MEMS1 package under Projects, STM32xxxxxx-Nucleo, Examples, IKSxxxx, DataLogExtended, Binary.
- Copy the DatalogExtended.bin file to the disk drive associated with the Nucleo board to flash the firmware.
Open MEMS Studio, connect to the Nucleo board, and select the desired sensors.
Go to Sensor Evaluation, Load/Save Configuration and load the .json configuration file.
Once the device has been configured, press the ▶ button in the top left corner to start streaming data from the sensor and visualize the MLC results in the MLC Monitor or Data Monitor views.

Run the example with SensorTile.box PRO / STWIN.box and ST BLE Sensor application

The SensorTile.box PRO / STWIN.box board and the ST BLE Sensor application offer a wireless method to test an MLC example.

Download the latest version of the ST BLE Sensor application (if not already installed).
Download the latest version of FP-SNS-DATALOG2.
Connect the SensorTile.box PRO / STWIN.box board to the device via Bluetooth.
The SensorTile.box PRO has to be programmed with the DATALOG2 firmware from the FP-SNS-DATALOG2 package (if not already programmed).
- Go to Firmware Update in the settings of the ST BLE Sensor application.
- Choose the DATALOG2_Release.bin binary file and click on the Upgrade button.
Go to HighSpeed DataLog2 and find the MLC (LSM6DSV16X) sensor.
Choose the MLC .json file and upload it to the SensorTile.box PRO.
Press the ▶ Start button.
Go to Machine Learning Core and see the live results.

Run the example with SensorTile.box PRO / STWIN.box and ST AIoT Craft

ST AIoT Craft allows user to import a .json configuration file or use one of the already available examples.

Open the ST AIoT Craft webpage.
Plug the SensorTile.box PRO / STWIN.box to the PC with a USB cable.
The SensorTile.box PRO / STWIN.box has to be programmed with the DATALOG2 firmware.
- While pairing your device click on Update the firmware link.
- Choose whether you want to flash from PC or mobile app, and follow the steps.
- Finally click on the Flash firmware button.
Choose an example from the offer or import your own configuration.
- Example from the offer
  - Choose an example from Project Examples and click on the Try out button.
  - Select environment.
  - Connect the device.
  - Press the Start button to see the live results.
- Importing your own configuration
  - Go to My datasets and create a new dataset. Choose name and insert the classes corresponding with the .json file configuration.
  - Open the dataset.
  - Go to Log data from board.
  - Connect the device.
  - In Sensor Settings find the MLC tab and click on Upload JSON.
  - Select the desired .json file and upload it.
  - Select mode Monitoring.
  - Finally click on the Start button and see the live results.

Examples per device

ASM330LHB
- tow_detection
- vehicle_stationary_detection
ASM330LHBG1
- tow_detection
- vehicle_stationary_detection
ASM330LHHX
- tow_detection
- vehicle_stationary_detection
ASM330LHHXG1
- tow_detection
- vehicle_stationary_detection
IIS2DULPX
- 6d_position_recognition
- vibration_monitoring
IIS2ICLX
- tilt_angle
- tilt_angle_dual_plane
ISM330BX
- 6d_position_recognition
- vibration_monitoring
ISM330DHCX
- 6d_position_recognition
- motion_intensity
- vehicle_stationary_detection
- vibration_monitoring
ISM6HG256X
- 6d_position_recognition
- asset_tracking
- door_opening_closing_still
- motion_intensity
- vibration_monitoring
LIS2DUX12
- 6d_position_recognition
- activity_recognition_for_mobile
- activity_recognition_for_wrist
- asset_tracking
- motion_intensity
- vibration_monitoring
LIS2DUXS12
- 6d_position_recognition
- activity_recognition_for_mobile
- activity_recognition_for_wrist
- asset_tracking
- motion_intensity
- vibration_monitoring
LSM6DSO32X
- 6d_position_recognition
- activity_recognition_for_wrist
- gym_activity_recognition
- motion_intensity
- vibration_monitoring
LSM6DSOX
- 6d_position_recognition
- activity_recognition_for_mobile
- activity_recognition_for_wrist
- asset_tracking
- door_opening_closing_still
- gym_activity_recognition
- head_gestures
- motion_intensity
- tennis_activity
- vehicle_stationary_detection
- vibration_monitoring
- yoga_pose_recognition
LSM6DSRX
- 6d_position_recognition
- activity_recognition_for_mobile
- activity_recognition_for_wrist
- asset_tracking
- head_gestures
- motion_intensity
- vehicle_stationary_detection
- vibration_monitoring
LSM6DSV16BX
- 6d_position_recognition
- activity_recognition_for_wrist
- head_gestures
- motion_intensity
- vibration_monitoring
LSM6DSV16X
- 6d_position_recognition
- activity_recognition_for_mobile
- activity_recognition_for_wrist
- asset_tracking
- door_opening_closing_still
- gym_activity_recognition
- head_gestures
- motion_intensity
- vibration_monitoring
- yoga_pose_recognition
LSM6DSV320X
- 6d_position_recognition
- activity_recognition_for_mobile
- activity_recognition_for_wrist
- asset_tracking
- door_opening_closing_still
- gym_activity_recognition
- head_gestures
- motion_intensity
- tennis_activity
- vibration_monitoring
- yoga_pose_recognition
LSM6DSV32X
- 6d_position_recognition
- activity_recognition_for_mobile
- activity_recognition_for_wrist
- asset_tracking
- gym_activity_recognition
- head_gestures
- motion_intensity
- tennis_activity
- vibration_monitoring
LSM6DSV80X
- 6d_position_recognition
- activity_recognition_for_mobile
- activity_recognition_for_wrist
- asset_tracking
- door_opening_closing_still
- gym_activity_recognition
- head_gestures
- motion_intensity
- tennis_activity
- vibration_monitoring
- yoga_pose_recognition
ST1VAFE3BX
- 6d_position_recognition
- activity_recognition_for_wrist
- motion_intensity
ST1VAFE6AX
- 6d_position_recognition
- activity_recognition_for_wrist
- motion_intensity

More information: http://www.st.com

Copyright © 2025 STMicroelectronics