RTX lidar sensor configuration not affecting lidar

[Dante.awaken]

Hello,

Since I saw that PhysX lidars would not work on people, I am turning back to RTX lidars. Im encoutering some issues concerning the latter.

Since I can’t create it on the GUI im using the script that we can find in the documentation isaac_sim_sensors_rtx_based_lidar and here I have created a “My_lidar.json” file where it is exactly a replicate of “Example_Rotary.json”.

Here is the script to create the RTX lidar:

import omni.kit.commands
from pxr import Gf
import omni.replicator.core as rep
lidar_config = "My_lidar"

# 1. Create The Camera
_, sensor = omni.kit.commands.execute(
    "IsaacSensorCreateRtxLidar",
    path="/sensor",
    parent=None,
    config=lidar_config,
    translation=(0, 0, 15.0),
    orientation=Gf.Quatd(1,0,0,0),
)
# 2. Create and Attach a render product to the camera
render_product = rep.create.render_product(sensor.GetPath(), [1, 1])

# 3. Create Annotator to read the data from with annotator.get_data()
annotator = rep.AnnotatorRegistry.get_annotator("RtxSensorCpuIsaacCreateRTXLidarScanBuffer")
annotator.attach(render_product)

# 4. Create a Replicator Writer that "writes" points into the scene for debug viewing
writer = rep.writers.get("RtxLidarDebugDrawPointCloudBuffer")
writer.attach(render_product)

And here is the configuration file I’m using:

{
    "class": "sensor",
    "type": "lidar",
    "name": "My Lidar",
    "driveWorksId": "GENERIC",
    "profile":
    {
      "scanType": "rotary",
      "intensityProcessing": "normalization",
      "rayType": "IDEALIZED",
      "nearRangeM": 5.0,
      "farRangeM": 300.0,
  
      "startAzimuthDeg": 0.0,
      "endAzimuthDeg": 360.0,
  
      "upElevationDeg": 10.0,
      "downElevationDeg": -15.0,
  
      "rangeResolutionM": 0.04,
      "rangeAccuracyM": 0.02,
  
      "avgPowerW": 0.002,
      "minReflectance": 0.1,
      "minReflectanceRange": 300.0,
      "wavelengthNm": 903.0,
      "pulseTimeNs": 6,

      "azimuthErrorMean": 0.0,
      "azimuthErrorStd": 0.015,
      "elevationErrorMean": 0.0,
      "elevationErrorStd": 0.0000,
  

      "maxReturns": 2,
  
      "scanRateBaseHz": 10.0,
  
      "reportRateBaseHz": 36000,
  
      "numberOfEmitters": 128,
      
      
      "rangeOffset": 5.0,
  
      "emitters":
      {
        "azimuthDeg": [-3.0,-3.0,-3.0,-3.0,-3.0,-3.0,-3.0,-3.0,-3.0,-3.0,-3.0,-3.0,-3.0,-3.0,-3.0,-3.0,-3.0,-3.0,-3.0,-3.0,-3.0,-3.0,-3.0,-3.0,-3.0,-3.0,-3.0,-3.0,-3.0,-3.0,-3.0,-3.0,-1.0,-1.0,-1.0,-1.0,-1.0,-1.0,-1.0,-1.0,-1.0,-1.0,-1.0,-1.0,-1.0,-1.0,-1.0,-1.0,-1.0,-1.0,-1.0,-1.0,-1.0,-1.0,-1.0,-1.0,-1.0,-1.0,-1.0,-1.0,-1.0,-1.0,-1.0,-1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,3.0,3.0,3.0,3.0,3.0,3.0,3.0,3.0,3.0,3.0,3.0,3.0,3.0,3.0,3.0,3.0,3.0,3.0,3.0,3.0,3.0,3.0,3.0,3.0,3.0,3.0,3.0,3.0,3.0,3.0,3.0,3.0],
        "elevationDeg": [-15.0,-14.19,-13.39,-12.58,-11.77,-10.97,-10.16,-9.35,-8.55,-7.74,-6.94,-6.13,-5.32,-4.52,-3.71,-2.9,-2.1,-1.29,-0.48,0.32,1.13,1.94,2.74,3.55,4.35,5.16,5.97,6.77,7.58,8.39,9.19,10.0,-15.0,-14.19,-13.39,-12.58,-11.77,-10.97,-10.16,-9.35,-8.55,-7.74,-6.94,-6.13,-5.32,-4.52,-3.71,-2.9,-2.1,-1.29,-0.48,0.32,1.13,1.94,2.74,3.55,4.35,5.16,5.97,6.77,7.58,8.39,9.19,10.0,-15.0,-14.19,-13.39,-12.58,-11.77,-10.97,-10.16,-9.35,-8.55,-7.74,-6.94,-6.13,-5.32,-4.52,-3.71,-2.9,-2.1,-1.29,-0.48,0.32,1.13,1.94,2.74,3.55,4.35,5.16,5.97,6.77,7.58,8.39,9.19,10.0,-15.0,-14.19,-13.39,-12.58,-11.77,-10.97,-10.16,-9.35,-8.55,-7.74,-6.94,-6.13,-5.32,-4.52,-3.71,-2.9,-2.1,-1.29,-0.48,0.32,1.13,1.94,2.74,3.55,4.35,5.16,5.97,6.77,7.58,8.39,9.19,10.0],
        "fireTimeNs": [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 3500, 3500, 3500, 3500, 3500, 3500, 3500, 3500, 3500, 3500, 3500, 3500, 3500, 3500, 3500, 3500, 7000, 7000, 7000, 7000, 7000, 7000, 7000, 7000, 7000, 7000, 7000, 7000, 7000, 7000, 7000, 7000, 10500, 10500, 10500, 10500, 10500, 10500, 10500, 10500, 10500, 10500, 10500, 10500, 10500, 10500, 10500, 10500, 14000, 14000, 14000, 14000, 14000, 14000, 14000, 14000, 14000, 14000, 14000, 14000, 14000, 14000, 14000, 14000, 17500, 17500, 17500, 17500, 17500, 17500, 17500, 17500, 17500, 17500, 17500, 17500, 17500, 17500, 17500, 17500, 21000, 21000, 21000, 21000, 21000, 21000, 21000, 21000, 21000, 21000, 21000, 21000, 21000, 21000, 21000, 21000, 24500, 24500, 24500, 24500, 24500, 24500, 24500, 24500, 24500, 24500, 24500, 24500, 24500, 24500, 24500, 24500]

      },
  
      "intensityMappingType": "LINEAR"
    }
  }

1. The problem I am encountering is that changing the “nearRangeM” and “farRangeM” on this config file doesnt affect at all the Lidar lines produced. I dont know exactly what I’m doing wrong.

2. Is there a possible workaround to not use the script but instead use the GUI and have it work properly ? (as the PhysX one which is more intuitive with the appropriate parameter just along with it).

3. As I understood, if I want to publish the point cloud in ROS1 I am obliged to modify the SGDpipeline (once created) / add in the script the ROS1 Helper to publish my point cloud ? Or is there a simpler or replacable way to do it ? [For some reasons I need to have python 3.11 and I can’t run standalone python code which requires specifically python 3.10] so the only accessible way for me is the script editor.

4. When the lidar / SGDpipeline is created, is it possible to delete it to recreate another one using the same script ?

Thank you for your answer !

[Sirens_uk]

I cant answer all your questions, But I would say you might be better to stick with the existing lidars provided in the RTX Lidar config set. You cant just swap between the two as you mentioned.

1. Creating your own lidar might not be a great idea as your data could be off, it wont be the same as the lidar you are planning to use further down the line ie in deployment.

2. Use the RTX Lidar but go into

window > visual scripting > Action graph > edit action graph > /Render/Postprocess/SDGPipeline

Adjust the point size in

If you set it to about 0.2 and change it to red its much easier to work with i find.

I dont think there is a way to use the GUI atm for this as it wont create the SDG pipline. I had a Dev at GTC try that and they couldnt get it to work.

3. “I need to have python 3.11 and I can’t run standalone python code which requires specifically python 3.10] so the only accessible way for me is the script editor.”

In VScode change your interpreter to 3.10 with the Isaac Sim python version path. I’m not 100% on this so I’d maybe seek other advice on this too.

I’ve been haviing very similar issues with this. I highly suggest you want to run your SDG pipleine in a standalone script. So ultimately you will need to do it anyway. You arent going to run the scripts manually everytime you want to randomise so its something that you need to look at. Its something that I need to get advice on myself. I dont use ROS so cant help you there.

I highly recommend this thread, some of the code has changed but in general this is what you want to get working! @ahaidu does a fantastic job of laying it out in code.

Point cloud data to render product SDG

Probably time stamped as I need to reference the xform data I mentioned as my loss function. So I can have a separate folders for the data

I used a work around by having a on load trigger to fire a script node and then a onplayback tick to handle my SDG. I’m modifing my data for another process too.

4. becareful with creating the same script as it has issues further down the line its something i had issues with. As Isaac sim will make new naming conventions and You end up having more than one pipeline.

Hope this helps

PS one thing I did notice was.

“name”: “My Lidar”,

and

lidar_config = “My_lidar”