- Detects 10+ signs
- Follows a road
To get the robot to follow the road with sign detection a few programs must be installed. To have it function one should use Linux. This setup guid shows how to install for Debian based OS's and Arch based OS's.
- Computer
- Arduino Nicla Vision
- Pololu Zumo 32U4
- Nicla Vision to Zumo printplate
- Router
- Python
- npm
- openMV
- Arduino IDE or similar
Since the Nicla and the computer communicate via WiFi a kind of router is necessary. The router can be anything. A hotspot on a mobile phone or laptop is already good enough as long as the router doesn't introduce a significant delay.
Make sure the routers WiFi name is Nicla and the password should be Weerstation.
On the nicla flash the code from Integration/WiFiToZumo.py using the openMV application on the Nicla Vision. The code will get lost on power loss so make sure either the Zumo or the Nicla has power.
In the IDE make sure you have installed the 32U4 library and its dependencies. Then connect to the Zumo and flash it. The board is an Arduino Leonardo board.
first install openCV for python
pip3 install openCV-python
If for either Debian based or Arch based there is the error:
error: externally-managed-environment
Anywhere in the process. Make sure you put --break-system-packages at the end of the pip3 command.
pip3 install edge_impulse_linux
sudo apt install -y curl
curl -sL https://deb.nodesource.com/setup_14.x | sudo bash -
sudo apt install -y gcc g++ make build-essential nodejs sox gstreamer1.0-tools gstreamer1.0-plugins-good gstreamer1.0-plugins-base gstreamer1.0-plugins-base-apps
npm config set user root && sudo npm install edge-impulse-linux -g --unsafe-perm
edge-impulse-linux
Access to the edge impulse is only necessary if you want to download the model from edge impulse. The model is also available on the github.
Building model(optional):
edge-impulse-linux-runner --download modelfile.eim
Building the model file is only necessary if not using the one supplied on github.
running model
python3 openCV.py
To run the program use openCV.py in the folder opencv. Make sure that the model is in the same folder and named modelfileV3.eim.
pip3 install edge_impulse_linux
Paru -S curl
curl -sL https://deb.nodesource.com/setup_14.x | sudo bash -
npm install edge-impulse-linux -g --unsafe-perm
edge-impulse-linux
Access to the edge impulse is only necessary if you want to download the model from edge impulse. The model is also available on the github.
Building model(optional):
edge-impulse-linux-runner --download modelfile.eim
Building the model file is only necessary if not using the one supplied on github.
running model
python3 openCV.py
To run the program use openCV.py in the folder opencv. Make sure that the model is in the same folder and named modelfileV3.eim.