How to pubish depth in meters or greater than uint8 using ROS2 topic?

Omniverse Isaac Sim
1

I followed the 7.2.4 Publishing Camera’s Data to publish the depth topic of a created camera and found that the depth value is set according to the unit of the scene and its type is uint8. However, my scene range is beyond the 255, the depth image cannot show the far away things. I wonder how to change the dtype of the published depth to float to get a larger range, or how to publish the exact depth in meters.

Below is part of my code

def publish_depth(camera: Camera, freq):
    # The following code will link the camera's render product and publish the data to the specified topic name.
    render_product = camera._render_product_path
    step_size = int(60/freq)
    topic_name = camera.name+"_depth"
    queue_size = 1
    node_namespace = ""
    frame_id = camera.prim_path.split("/")[-1] # This matches what the TF tree is publishing.

    rv = omni.syntheticdata.SyntheticData.convert_sensor_type_to_rendervar(
                            sd.SensorType.DistanceToImagePlane.name
                        )
    writer = rep.writers.get(rv + "ROS2PublishImage")
    writer.initialize(
        frameId=frame_id,
        nodeNamespace=node_namespace,
        queueSize=queue_size,
        topicName=topic_name
    )
    writer.attach([render_product])

    # Set step input of the Isaac Simulation Gate nodes upstream of ROS publishers to control their execution rate
    gate_path = omni.syntheticdata.SyntheticData._get_node_path(
        rv + "IsaacSimulationGate", render_product
    )
    og.Controller.attribute(gate_path + ".inputs:step").set(step_size)

    return

camera = Camera(prim_path="/World/turtlebot3_burger/base_link/Camera",
                        resolution=(1280,720),
                        frequency=30)
        
camera.initialize()

publish_depth(camera, 30)
2

@user67308 out of curiosity, which version of Isaac Sim are you running?

3

2023.1.1 @Simplychenable

5

Hi there, the tutorial seems to publish a sensor_msgs/PointCloud2 message, which uses uint8 for the data:

  • #uint8[] data # Actual point data, size is (row_step*height)

The distance to image plane annotator will actually provide you with a float32 for the distance representation: