Controlling stepper motors

[bill25]

Hi all,

Just curious if anyone here is experimenting with stepper motors controlled by Edge AI systems (like Raspberry Pi + AI accelerators, Jetson Nano, etc.). I'm exploring the idea of using local AI models for adaptive motor control — for example, adjusting speed or torque in real time based on sensor input, without relying on a cloud connection.

Some use cases I’m thinking of:

Smart 3D printers that self-optimize print parameters
Industrial robots with more responsive motor control
Predictive maintenance based on motion patterns
Has anyone tried this? Any tips on frameworks or libraries that work well for real-time AI + motor control?

Would love to hear your thoughts or see your projects!

[Julian E.]

Hello @bill25,

I am not aware of any example using neural network to do that. You may have to look for litterature about that.

If you find a model, you can then use our st edge ai tool to convert tlite or onnx model to C code to be able to run it on micro controller. Depending on the size of your model, you could be interested in the stm32n6 with neural acceleration.

Doc: https://stedgeai-dc.st.com/assets/embedded-docs/command_line_interface.html 

For predictive maintenance using motion patterns, you could be interested in another of our tool, NanoEdge AI Studio. This is an autoML software, meaning that you need to provide data and the tool will look for a model and output a C library of a machine learning model. You can look at this tutorial to see what I am talking about: AI:How to Build an Anomaly Detection Project for Predictive Maintenance with NanoEdge AI Studio - stm32mcu

To sum up, in ST, you can either come with your model and convert it with the st edge ai core (and different tools using it, the st dev cloud, cubeMX X cube AI) or if you don't have a model, you can look at NanoEdge.

NanoEdge is only using small machine learning model with sensing data though.

Have a good day,

Julian

[SimonePradolini]

Hello @bill25 

If you’re interested in ST solutions, I can suggest you FP-IND-DATALOGMC for STEVAL-STWINBX1 and EVLSPIN32G4-ACT.

The function pack contains a firmware example called DATALOGMC_AI that implements a motor fault classification based on a machine learning solution developed through NanoEdgeAIStudio. The machine learning model allows an accurate classification of motor behavior into two states: good and faulty.

The package also includes a portable mechanical setup that can be replicated with a 3D printer.

Have a look into the package and the documentation and feel free to ask for further assistance if needed.

Best regards,

Simone