IngenieriaUP / actino-data-collector

This repository stores the code used to capture telemetry data from a custom drone with Cube (Pixhawk2) and LiDAR sensor data from Velodyne VLP16.

Cube (Pixhawk2) + RPi + LiDAR VLP16

This repository stores the code used to capture telemetry data from a custom drone with Cube (Pixhawk2) and LiDAR sensor data from Velodyne VLP16.

Capture Telemetry Data

• rpi_gps_datange.py: This code store the main object called DataGenerator which is used to generate GPS data as NMEA sentence via the Dronekit Python API and send data to LiDAR sensor via UDP and save it to a file.

RPi + LiDAR communication

• send2velodyne.py: This code is used to schedule the Pulse Per Second and the NMEA Sentence required by the Velodyne VLP16.

Capture LiDAR data

• Velodyne_pcap: External repository used to save a ".pcap" file from Velodyne LiDAR Sensors data.
  Modified to:
  1. Listen to eth0 device by default.
  2. Save data in a mounted usb in the Raspberry Pi (/media/usb/pcap/)

Other uses

• missions.py: Code to set simple missions and test them on a simulated vehicle using the dronekit_sitl library.

Quick Start

1. Open the terminal, clone the repository and cd into it:

$ git clone https://github.com/IngenieriaUP/dronekit-tests.git
$ cd dronekit-tests

2. To communicate between LiDAR and RPi run the following command (usb must be mounted in "/media/usb/"):

$ python3 send2velodyne.py --connect <path_to_pixhawk2> --lidar_port <LiDAR_IP> --filename <filename>

3. To save LiDAR Sensor data as .pcap follow instructions in README.md from the Velodyne_pcap folder (external repository).

Test Dronekit Connection in RPi

1. Conect via terminal to the Raspberry

$ ssh pi@<RPi_IP>

2. To check the Raspberry Pi's IP run:

$ hostname -I

3. cd into the repository

$ cd dronekit-tests

4. (Optional) Install Python 3.6 in the Raspberry Pi and create a virtual environment in the folder

Follow this: https://installvirtual.com/install-python-3-on-raspberry-pi-raspbian/

$ python -m virtualenvironment .env -p python3.6
$ source .env/bin/activate

5. Install Dronekit in the Raspberry Pi *

(.env) $ pip install dronekit

6. Connect the drone to the raspberry pi and check the name of connection (e.g. /dev/ttyACM0).

7. Test the connection using hello_drone.py:

$ python3 hello_drone.py

• Maybe the lxml library won't be found, so its necessary to install it in the Raspberry pi.
  Check this: https://raspberrypi.stackexchange.com/questions/68894/cant-install-lxml

(.env) $ pip install lxml

Run commands when Raspberry Pi boots

1. Edit rc.local file:

$ sudo nano /etc/rc.local

2. Add the command line in rc.local file (e.g. run a python script at boot):

...

sudo python /<path_to_file>/<script_name>.py

exit 0

• For this project the added lines were:

# Capture LiDAR Data from Veloydine VLP16 LiDAR Sensor
# Send NEMA and PPS to Velodyne VLP16 LiDAR Sensor (also save GPS+IMU+timestamp)
sudo /home/pi/dronekit-tests/Velodyne_pcap/Velodyne_pcap VLP16 "LiDAR_data" &
sudo python3 /home/pi/dronekit-tests/send2velodyne.py --connect /dev/ttyACM0 --lidar_port 192.168.1.21 &
wait

3. To save changes and exit press "Ctrl + X" then "Y" and finally "Enter"

4. Reboot to test:

$ sudo reboot