This repo is the open-source implementation for the Predictive Maintenance of Xilinx made with MakarenaLabs. For more information about the other projects involved as dependencies please visit:
The project currently works with the Motor Control Kit provided by Trenz Electronic and supports the Ultra96-V2 board. To connect to the power stage board of the motor is needed the Adapter provided by Trenz Electronic.
To build the project you need to:
- Flash in a SD-Card the Pynq image V2.6 for Ultra96-V2 from the ones available and follow the steps to setup the board.
- After having created a working PYNQ image, connect to the board and clone the repo.
- cd into the folder cloned and tap
chmod 755 init.sh && ./init.sh. It will ask you the password, which isxilinx(same name as the account). - Once the installation has completed you can try the Jupyter Notebooks provided as an example under
pynq-dpuinstalled in the jupyter folder.
To run the flask server you need to tap sudo python3 main.py in the folder pynq-foc-dpu-python-code.
Currently it is supported just the FOC interfacing.
The server will run in local (w.r.t. the Ultra96-V2) on the port 5000. For example, if your board has 192.168.2.5 IP address, the server is accessible through this address:
192.168.2.5:5000
Note that the python process is not detached, so you need to keep your SSH terminal active.
The documentation of Flask API server is here: https://github.com/MakarenaLabs/Predictive-Maintenance-PYNQ/tree/main/pynq-foc-dpu-python-code
The script init.sh will download and install the project DPU-PYNQ needed to interface the DPU.