feat(utils): display-independent keyboard controls for recording (Wayland / headless / macOS) (#3875)

* feat(utils): headless keyboard control

* refactor(utils): consolidate keyboard listener creation

* fix(rollout): remove import require guard for pynput

---------

Co-authored-by: Leo Toff <leo@toff.dev>
Co-authored-by: Stefano Maestri <stefano.maestri@javalinux.it>
Co-authored-by: Sahil Chande <85823961+SahilChande@users.noreply.github.com>
Co-authored-by: Vinayak Agarwal <63502278+Vinayak-Agarwal-2004@users.noreply.github.com>
Co-authored-by: Abdul Rahim Mirani <abdulrahimmirani@gmail.com>

examples/lekiwi/evaluate.py

@@ -17,7 +17,7 @@
 import logging
 import time
 
-from lerobot.common.control_utils import init_keyboard_listener, predict_action
+from lerobot.common.control_utils import predict_action
 from lerobot.datasets import LeRobotDataset
 from lerobot.policies import make_pre_post_processors
 from lerobot.policies.act import ACTPolicy
@@ -26,6 +26,7 @@ from lerobot.processor import make_default_processors
 from lerobot.robots.lekiwi import LeKiwiClient, LeKiwiClientConfig
 from lerobot.utils.constants import ACTION, OBS_STR
 from lerobot.utils.feature_utils import build_dataset_frame, hw_to_dataset_features
+from lerobot.utils.keyboard_input import init_keyboard_listener
 from lerobot.utils.robot_utils import precise_sleep
 from lerobot.utils.utils import log_say
 from lerobot.utils.visualization_utils import init_rerun, log_rerun_data

examples/lekiwi/record.py

@@ -14,7 +14,6 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
-from lerobot.common.control_utils import init_keyboard_listener
 from lerobot.datasets import LeRobotDataset
 from lerobot.processor import make_default_processors
 from lerobot.robots.lekiwi import LeKiwiClient, LeKiwiClientConfig
@@ -23,6 +22,7 @@ from lerobot.teleoperators.keyboard import KeyboardTeleop, KeyboardTeleopConfig
 from lerobot.teleoperators.so_leader import SO100Leader, SO100LeaderConfig
 from lerobot.utils.constants import ACTION, OBS_STR
 from lerobot.utils.feature_utils import hw_to_dataset_features
+from lerobot.utils.keyboard_input import init_keyboard_listener
 from lerobot.utils.utils import log_say
 from lerobot.utils.visualization_utils import init_rerun
 

examples/phone_to_so100/evaluate.py

@@ -18,7 +18,7 @@ import logging
 import time
 
 from lerobot.cameras.opencv import OpenCVCameraConfig
-from lerobot.common.control_utils import init_keyboard_listener, predict_action
+from lerobot.common.control_utils import predict_action
 from lerobot.configs import FeatureType, PolicyFeature
 from lerobot.datasets import LeRobotDataset, aggregate_pipeline_dataset_features, create_initial_features
 from lerobot.model.kinematics import RobotKinematics
@@ -41,6 +41,7 @@ from lerobot.robots.so_follower.robot_kinematic_processor import (
 from lerobot.types import RobotAction, RobotObservation
 from lerobot.utils.constants import ACTION, OBS_STR
 from lerobot.utils.feature_utils import build_dataset_frame, combine_feature_dicts
+from lerobot.utils.keyboard_input import init_keyboard_listener
 from lerobot.utils.robot_utils import precise_sleep
 from lerobot.utils.utils import log_say
 from lerobot.utils.visualization_utils import init_rerun, log_rerun_data

examples/phone_to_so100/record.py

@@ -15,7 +15,6 @@
 # limitations under the License.
 
 from lerobot.cameras.opencv import OpenCVCameraConfig
-from lerobot.common.control_utils import init_keyboard_listener
 from lerobot.datasets import LeRobotDataset, aggregate_pipeline_dataset_features, create_initial_features
 from lerobot.model.kinematics import RobotKinematics
 from lerobot.processor import (
@@ -39,6 +38,7 @@ from lerobot.teleoperators.phone.config_phone import PhoneOS
 from lerobot.teleoperators.phone.phone_processor import MapPhoneActionToRobotAction
 from lerobot.types import RobotAction, RobotObservation
 from lerobot.utils.feature_utils import combine_feature_dicts
+from lerobot.utils.keyboard_input import init_keyboard_listener
 from lerobot.utils.utils import log_say
 from lerobot.utils.visualization_utils import init_rerun
 

examples/so100_to_so100_EE/evaluate.py

@@ -18,7 +18,7 @@ import logging
 import time
 
 from lerobot.cameras.opencv import OpenCVCameraConfig
-from lerobot.common.control_utils import init_keyboard_listener, predict_action
+from lerobot.common.control_utils import predict_action
 from lerobot.configs import FeatureType, PolicyFeature
 from lerobot.datasets import LeRobotDataset, aggregate_pipeline_dataset_features, create_initial_features
 from lerobot.model.kinematics import RobotKinematics
@@ -41,6 +41,7 @@ from lerobot.robots.so_follower.robot_kinematic_processor import (
 from lerobot.types import RobotAction, RobotObservation
 from lerobot.utils.constants import ACTION, OBS_STR
 from lerobot.utils.feature_utils import build_dataset_frame, combine_feature_dicts
+from lerobot.utils.keyboard_input import init_keyboard_listener
 from lerobot.utils.robot_utils import precise_sleep
 from lerobot.utils.utils import log_say
 from lerobot.utils.visualization_utils import init_rerun, log_rerun_data

examples/so100_to_so100_EE/record.py

@@ -16,7 +16,6 @@
 
 
 from lerobot.cameras.opencv import OpenCVCameraConfig
-from lerobot.common.control_utils import init_keyboard_listener
 from lerobot.datasets import LeRobotDataset, aggregate_pipeline_dataset_features, create_initial_features
 from lerobot.model.kinematics import RobotKinematics
 from lerobot.processor import (
@@ -36,6 +35,7 @@ from lerobot.scripts.lerobot_record import record_loop
 from lerobot.teleoperators.so_leader import SO100Leader, SO100LeaderConfig
 from lerobot.types import RobotAction, RobotObservation
 from lerobot.utils.feature_utils import combine_feature_dicts
+from lerobot.utils.keyboard_input import init_keyboard_listener
 from lerobot.utils.utils import log_say
 from lerobot.utils.visualization_utils import init_rerun