address review comments

source/extensions/omni.isaac.lab/omni/isaac/lab/sensors/rtx_lidar/rtx_lidar.py

@@ -144,9 +144,6 @@ def reset(self, env_ids: Sequence[int] | None = None):
         # note: cannot do smart indexing here since we do a for loop over data.
         if env_ids is None:
             env_ids = self._ALL_INDICES
-        # reset the data
-        # note: this recomputation is useful if one performs events such as randomizations on the camera poses.
-        # self._update_poses(env_ids)
         # Reset the frame count
         self._frame[env_ids] = 0
 
@@ -274,7 +271,7 @@ def _create_buffers(self):
         # lazy allocation of data dictionary
         # since the size of the output data is not known in advance, we leave it as None
         # the memory will be allocated when the buffer() function is called for the first time.
-        self._data.output = TensorDict({}, batch_size=self._view.count, device=self.device)
+        self._data.output = {}
         self._data.info = [{name: None for name in RTX_LIDAR_INFO_FIELDS.keys()} for _ in range(self._view.count)]
 
     def _update_buffers_impl(self, env_ids: Sequence[int]):

source/extensions/omni.isaac.lab/omni/isaac/lab/sensors/rtx_lidar/rtx_lidar_cfg.py

@@ -23,7 +23,7 @@ class RtxLidarCfg(SensorBaseCfg):
     class OffsetCfg:
         """The offset pose of the sensor's frame from the sensor's parent frame."""
 
-        pos: tuple[float, float, float] = (0.0, 0.0, 1.0)
+        pos: tuple[float, float, float] = (0.0, 0.0, 0.0)
         """Translation w.r.t. the parent frame. Defaults to (0.0, 0.0, 0.0)."""
         rot: tuple[float, float, float, float] = (1.0, 0.0, 0.0, 0.0)
         """Quaternion rotation (w, x, y, z) w.r.t. the parent frame. Defaults to (1.0, 0.0, 0.0, 0.0)."""

source/extensions/omni.isaac.lab/omni/isaac/lab/sensors/rtx_lidar/rtx_lidar_data.py

@@ -39,7 +39,7 @@ class RtxLidarData:
     The product  of ticksPerScan, numChannels, and numEchos will be the same as the number of returns if you initialize
     the annotator with annotator.initialize(keepOnlyPositiveDistance=False) before attaching the render product.
     """
-    output: TensorDict = None
+    output: dict[str, torch.Tensor] = None
     """The data that changes every sample. Some fields of the out will always be returned and some are optionally
     returned when configured in RtxLidarCfg.optional_data_types.