This is a monorepo for the collaborative project (between UFSC and Coontrol) to calculate the volume of truck loads using Pepperl-Fuchs LiDAR sensors in a proper arrangement.
The project's root folder has a Python library sensor_library.py
which encapsulates most of the useful HTTP requests we send to the sensors to either configure them or start scanning and outputting data.
The receiver.py
is the actual main file that when run, triggers the required requests to the sensors, and handles data reception using an UDP server. The default configuration for the server address is the IP 192.168.1.50
with PORT 54321
.
For everything to work, the computer must be plugged into the switch (which is also plugged into all sensors), and you must also configure the local network connection to set the appropriate IP address. If using 192.168.1.50
, you need to set that into the local connection to the switch and sensors like in the image, so the network adapter in each sensor knows what or where is this IP address to send the UDP packets to.
While most functionalities use the default Python packages (requests
, json
, etc.), the interface requires PyQt5, which can be installed in your environment with pip install PyQt5
.
Pepperl-Fuchs R2000 HD Series LiDAR Ethernet Communication Documentation
Pepperl-Fuchs R2000 HD Series LiDAR Technical Specifications Page
Pepperl-Fuchs R2000 HD Series LiDAR Engineering Documentation