update docstrings

source/extensions/omni.isaac.lab/omni/isaac/lab/managers/manager_term_cfg.py

@@ -161,14 +161,17 @@ class ObservationTermCfg(ManagerTermBaseCfg):
     """
 
     history_length: int = 0
-    """Number of past observations to store in the zero-initialized observation buffers.
-    Defaults to 0, which means that only the current data is stored (no history).
-    If flatten_history_dim is set to True, the source data of shape (N, H, D, ...) where N is the batch dimension and
-    H is the history length will be reshaped to a 2D tensor of shape (N, H*D*...). Otherwise, the data will be returned as is.
+    """Number of past observations to store in the observation buffers. Defaults to 0, meaning no history.
+
+    Observation history initializes to empty, but is filled with the first append after reset or initialization. Subsequent history
+    only adds a single entry to the history buffer. If flatten_history_dim is set to True, the source data of shape
+    (N, H, D, ...) where N is the batch dimension and H is the history length will be reshaped to a 2D tensor of shape
+    (N, H*D*...). Otherwise, the data will be returned as is.
     """
 
     flatten_history_dim: bool = True
-    """Whether or not the observation manager should flatten history-based observation terms to a 2D (N, D) tensor. Defaults to True."""
+    """Whether or not the observation manager should flatten history-based observation terms to a 2D (N, D) tensor.
+    Defaults to True."""
 
 
 @configclass
@@ -192,18 +195,19 @@ class ObservationGroupCfg:
     """
 
     history_length: int | None = None
-    """Number of past observation to store in the zero-initialized observation buffers for all observation terms in group.
+    """Number of past observation to store in the observation buffers for all observation terms in group.
 
-    This parameter will override ObservationTermCfg.history_length if set. ObservationGroupCfg.history_length defaults
-    to None. If None, each ObservationTermCfg.history_length will be controlled on a per term basis. See ObservationTermCfg
-    for details on history_length implementation.
+    This parameter will override :attr:`ObservationTermCfg.history_length` if set. Defaults to None. If None, each
+    terms history will be controlled on a per term basis. See :class:`ObservationTermCfg` for details on history_length
+    implementation.
     """
 
     flatten_history_dim: bool = True
     """Flag to flatten history-based observation terms to a 2D (num_env, D) tensor for all observation terms in group.
+    Defaults to True.
 
-    This parameter will override all ObservationTermCfg.flatten_history_dim in the group if ObservationGroupCfg.history_length
-    is set. ObservationGroupCfg.flatten_history_dim defaults to True.
+    This parameter will override all :attr:`ObservationTermCfg.flatten_history_dim` in the group if
+    ObservationGroupCfg.history_length is set.
     """
 
 

source/extensions/omni.isaac.lab/omni/isaac/lab/managers/observation_manager.py

@@ -47,6 +47,11 @@ class ObservationManager(ManagerBase):
         concatenated. In this case, please set the :attr:`ObservationGroupCfg.concatenate_terms` attribute in the
         group configuration to False.
 
+    Observations can also have history. This means a running history is updated per sim step. History can be controlled
+    per :class:`ObservationTermCfg` (See the :attr:`ObservationTermCfg.history_length` and
+    :attr:`ObservationTermCfg.flatten_history_dim`). History can also be controlled via :class:`ObservationGroupCfg`
+    where group configuration overwrites per term configuration if set. History follows an oldest to newest ordering.
+
     The observation manager can be used to compute observations for all the groups or for a specific group. The
     observations are computed by calling the registered functions for each term in the group. The functions are
     called in the order of the terms in the group. The functions are expected to return a tensor with shape