RoboSynChallenge/cobotmagic_Real_manipulate_pipette

CobotMagic Real Dataset: manipulate_pipette

Generated by convert_piper_hdf5_to_lerobot.py.

RoboSynChallenge Alignment

• Task dir: manipulate_pipette
• Environment ID: ManipulatePipette
• Real dataset: RoboSynChallenge/cobotmagic_Real_manipulate_pipette
• Aligned simulation dataset: manipulate_pipette/cobotmagic_Sim_manipulate_pipette_000
• Prompt: Use the pipette on the table to transfer liquid from one beaker to another

Summary

• Robot type: aloha
• FPS: 10
• Sampling mode: full
• Leading group: slave_cam_high
• Crop strategy: LATEST_START
• Default leading tolerance: 0.1 s
• Additional constraints: []
• Arm selection: dual -> left, right
• Space ids: all -> ['joint', 'gripper', 'ee_pos', 'ee_rot']
• EE rotation format: rot6d
• Gripper format: raw
• Image preset: all -> ['observation.images.cam_high', 'observation.images.cam_left_wrist', 'observation.images.cam_right_wrist']

State / Action Index Map

action uses the exact same index order as observation.state.

indices	arm	component	names	source
0-5	left	joint	left_joint_waist, left_joint_shoulder, left_joint_elbow, left_joint_forearm_roll, left_joint_wrist_angle, left_joint_wrist_rotate	slave_left_arm/joint
6	left	gripper	left_gripper	slave_left_arm/gripper
7-9	left	ee_pos	left_ee_pos_x, left_ee_pos_y, left_ee_pos_z	slave_left_arm/ee_pose[:3]
10-15	left	ee_rot	left_ee_rot6d_0, left_ee_rot6d_1, left_ee_rot6d_2, left_ee_rot6d_3, left_ee_rot6d_4, left_ee_rot6d_5	slave_left_arm/ee_pose[3:6] (controller-scaled Piper xyz Euler) -> rot6d
16-21	right	joint	right_joint_waist, right_joint_shoulder, right_joint_elbow, right_joint_forearm_roll, right_joint_wrist_angle, right_joint_wrist_rotate	slave_right_arm/joint
22	right	gripper	right_gripper	slave_right_arm/gripper
23-25	right	ee_pos	right_ee_pos_x, right_ee_pos_y, right_ee_pos_z	slave_right_arm/ee_pose[:3]
26-31	right	ee_rot	right_ee_rot6d_0, right_ee_rot6d_1, right_ee_rot6d_2, right_ee_rot6d_3, right_ee_rot6d_4, right_ee_rot6d_5	slave_right_arm/ee_pose[3:6] (controller-scaled Piper xyz Euler) -> rot6d

Notes

• Timestamp alignment is done per HDF5 group. Each camera group and each arm group has its own timestamp dataset, which is treated like an individual topic timeline.
• The leading group is slave_cam_high. In full mode every leading-group timestep is kept; in fps mode the leading timeline is resampled first and then all other groups are bound to it by nearest timestamp.
• Piper HDF5 stores 6 arm joints plus one separate gripper scalar per arm. The historical controller saved gripper as mm / 70; this dataset stores gripper as mm / 100, i.e. a normalized physical opening ratio where 0 is closed and 1 is about 100 mm open.
• ee_pose rotation is decoded from the current PiperController convention: controller value -> multiply by 1000 -> Piper xyz Euler degrees -> convert to rot6d. When EE_ROT_FORMAT is rpy/euler, the exported values are radians.
• GRIPPER_FORMAT="01" treats values in [0, 1] as normalized opening ratio and thresholds them with GRIPPER_FULL_WIDTH=0.05 and GRIPPER_THRESHOLD=0.01.

Image Keys

dataset key	hdf5 group
observation.images.cam_high	slave_cam_high
observation.images.cam_left_wrist	slave_cam_left
observation.images.cam_right_wrist	slave_cam_right

Feature Names

observation.state / action names:

left_joint_waist, left_joint_shoulder, left_joint_elbow, left_joint_forearm_roll, left_joint_wrist_angle, left_joint_wrist_rotate, left_gripper, left_ee_pos_x, left_ee_pos_y, left_ee_pos_z, left_ee_rot6d_0, left_ee_rot6d_1, left_ee_rot6d_2, left_ee_rot6d_3, left_ee_rot6d_4, left_ee_rot6d_5, right_joint_waist, right_joint_shoulder, right_joint_elbow, right_joint_forearm_roll, right_joint_wrist_angle, right_joint_wrist_rotate, right_gripper, right_ee_pos_x, right_ee_pos_y, right_ee_pos_z, right_ee_rot6d_0, right_ee_rot6d_1, right_ee_rot6d_2, right_ee_rot6d_3, right_ee_rot6d_4, right_ee_rot6d_5