[Question] Is it possible to spawn multiple cameras per tiled Camera with different positions?

[waynemartis-iml]

Hello,

I am working on a simulation of robot with multiple onboard cameras. I want to spawn multiple cameras per tiled cameras but with different positions and rotations. This way, I could retrieve their data as single concatenated tensor rather than calling multiple instances of the cameras.

I found few snippets in the script /source/isaaclab/test/sensors, spawning multiple cameras at different prims only.

def test_frame_different_poses_multi_tiled_camera(self):
        """Test multiple tiled cameras placed at different poses render different images."""

        num_tiled_cameras = 3
        num_cameras_per_tiled_camera = 4
        positions = [(0.0, 0.0, 4.0), (0.0, 0.0, 4.0), (0.0, 0.0, 3.0)]
        rotations = [(0.0, 0.0, 1.0, 0.0), (1.0, 0.0, 1.0, 0.0), (0.0, 0.0, 1.0, 0.0)]

        tiled_cameras = []
        for i in range(num_tiled_cameras):
            for j in range(num_cameras_per_tiled_camera):
                prim_utils.create_prim(f"/World/Origin_{i}_{j}", "Xform")

            # Create camera
            camera_cfg = copy.deepcopy(self.camera_cfg)
            camera_cfg.prim_path = f"/World/Origin_{i}.*/CameraSensor"
            camera_cfg.offset = TiledCameraCfg.OffsetCfg(pos=positions[i], rot=rotations[i], convention="ros")
            camera = TiledCamera(camera_cfg)
            tiled_cameras.append(camera)

Has anyone implemented or found a workaround for spawning at different poses?

[RandomOakForest]

Thank you for posting this. I will move this post to our Discussions for follow up.