Getting lidar locaition corresponding to frame returns

[quagjustin]

Hello,
I'm working with both IsaacSim 5.1.0 and 6.0.0. For my use case, I want to have a lidar that moves constantly, getting data in new positions. I'm following the process mentioned in this guide in regards to getting frame data from the lidar, and have been changing the translation of my lidar prim. My question is, how can I know what translation and orientation of the lidar corresponds to each frame of data gotten from the lidar results?
Thanks in advance.

[hawkeyeking]

Hi @quagjustin — the IsaacCreateRTXLidarScanBuffer output dict doesn't include a transform field. But it can return a per-return timestamp (nanoseconds since sim start). Use that to look up poses you record yourself:

Initialize the annotator with timestamps enabled:
pythonsensor.attach_annotator("IsaacCreateRTXLidarScanBuffer", outputTimestamp=True)
Each sim step, record (sim_time_ns, lidar_world_transform) into a buffer — get the transform with UsdGeom.Xformable(lidar_prim).ComputeLocalToWorldTransform(Usd.TimeCode.Default()).
When you read the annotator, look up each return's timestamp in that buffer to get the matching pose.

The returned data (point cloud) is in sensor frame — transform it to world using the pose for that return's timestamp.