Getting lidar data from a QGC mission

[Agm2104]

Hi everyone!

I'm fairly new when it comes to Airsim and I'm currently working on a project where I'm supposed to get the lidar data from a flying multirotor. I'm toying with the idea of launching missions with QGC and then use python and getLidarData() in parallel to retrieve the data along the path. My question is, would this be feasible or would QGC shut any command coming from python?

Currently I'm using:
Windows 10
UE 4.25.4
Python v3.6.12 (Airsim)
Pixhawk v1.10.1
QGC v4.11

Thank you very much beforehand.

[ahmed-elsaharti]

Hi @Agm2104 , AFAIK QGC commands should not interfere with gitLidarData() so your plan sounds doable.
I did a quick test with PX4 SITL and the getLidarData API and I didn't face any issues retrieving the points and plotting them (using matplotlib) as the drone took off and landed:

Here are the settings I used:

{
  "SettingsVersion": 1.2,
  "SimMode": "Multirotor",
  "ClockSpeed": 1,
  "DefaultSensors": {
    "Barometer": {
         "SensorType": 1,
         "Enabled" : true
    },
    "Imu": {
         "SensorType": 2,
         "Enabled" : true
    },
    "Gps": {
         "SensorType": 3,
         "Enabled" : true
    },
    "Magnetometer": {
         "SensorType": 4,
         "Enabled" : true
    },
    "Distance": {
         "SensorType": 5,
         "Enabled" : true
    },
    "Lidar": {
      "SensorType": 6,
      "Enabled" : true,
      "NumberOfChannels": 1,
      "RotationsPerSecond": 10,
      "PointsPerSecond": 10000,
      "X": 0, "Y": 0, "Z": -1,
      "Roll": 0, "Pitch": 0, "Yaw" : 0,
      "VerticalFOVUpper": 0,
      "VerticalFOVLower": 0,
      "DrawDebugPoints": true,
      "DataFrame": "VehicleInertialFrame"
    }
  },
  "Vehicles": {
      "PX4": {
          "VehicleType": "PX4Multirotor",
          "UseSerial": false,
          "UseTcp": true,
          "UdpIp": "127.0.0.1",
          "TcpPort": 4560,
          "ControlIp": "127.0.0.1",
          "ControlPort": 14580
      }
  }
}

and here's the crude python script I wrote for plotting:

from mpl_toolkits import mplot3d

import airsim
import numpy as np
import matplotlib.pyplot as plt
from matplotlib.animation import FuncAnimation

client = airsim.MultirotorClient()
fig = plt.figure()
ax = plt.axes(projection='3d')

def animate(i): 
    ptc = client.getLidarData().point_cloud
    xs = np.empty(shape=[int(len(ptc)/3),])
    ys = np.empty(shape=[int(len(ptc)/3),])
    zs = np.empty(shape=[int(len(ptc)/3),])
    for i in range(0,len(ptc)-1,3):
        xs[int(i/3)]=ptc[i]
        ys[int(i/3)]=ptc[i+1]
        zs[int(i/3)]=-ptc[i+2]
    #plt.cla()
    ax.set_zlim([0, 20])
    ax.scatter(xs,ys,zs,zdir='z',s=10,c='b')


ani = FuncAnimation(plt.gcf(),animate,interval=10)
plt.show()

Hope this helps ^_^

[Agm2104]

Thank you very much for your reply, is really useful!

May I ask how do you manage to install matplotlib properly in the Airsim enviroment? Everytime I try I get a Numpy related error and the code does not compile.

[ahmed-elsaharti]

@Agm2104 , I don't exactly remember how I installed it but probably through pip: pip install matplotlib.
If you can paste the error code or maybe share a screenshot of what happens I'd be happy to try and fix the issue with you ^_^

[Agm2104]

Hey @ahmed-elsaharti!

I checked and the problem seemed to be that I was using conda forge to install matplotlib, which installed a version of Numpy that didn't work. By using pip I have managed to solve the problem and now I can import the library.

Thank you very much for taking the time of helping, I really appreciate it :)

[ahmed-elsaharti]

Glad it worked out 👍
We're all here to help ^_^