2026-11-07
Collection
Models and datasets created on 2026-11-07 (Arizona time). • 23 items • Updated
Dataset Viewer
The dataset viewer is not available for this split.
Cannot load the dataset split (in streaming mode) to extract the first rows.
Error code: StreamingRowsError
Exception: CastError
Message: Couldn't cast
name: string
frames_received: int64
frames_written: int64
frames_remuxed: int64
frames_metadata_written: int64
frames_dropped_invalid: int64
frames_dropped_queue: int64
first_recv_ns: int64
last_recv_ns: int64
remux_status: string
remux_error: null
metadata_error: null
raw_write_error: null
source: string
video_remux_status: string
robot_type: string
device_serial: string
episode_index: int64
timestamp: timestamp[s]
task_instruction: string
task_num: string
format_version: string
transcoding_status: string
sim_source: struct<set: string, episode: int64, bottle_init_xy_yaw: list<item: double>, cap_torque_Nm: double, c (... 455 chars omitted)
child 0, set: string
child 1, episode: int64
child 2, bottle_init_xy_yaw: list<item: double>
child 0, item: double
child 3, cap_torque_Nm: double
child 4, cap_turn_deg: double
child 5, strokes_deg: list<item: double>
child 0, item: double
child 6, program_params: struct<l_tilt: double, grip_h: double, lift: double, station: list<item: double>, psi0: double, stro (... 97 chars omitted)
child 0, l_tilt: double
child 1, grip_h: double
child 2, lift: double
child 3, station: list<item: double>
child 0, item: double
child 4, psi0: double
child 5, stroke: double
child 6, speed: double
child 7, pad_clear: double
child 8, r_open: double
child 9, close_mid: double
child 10, r_close: double
child 7, release_deg: double
child 8, tail_deg: double
child 9, tilt_deg: double
child 10, tilt_az_deg: double
child 11, tilt_deg_achieved: double
child 12, station_xyh: list<item: double>
child 0, item: double
child 13, psi0_deg: double
child 14, success: bool
child 15, sim_fps: int64
segments: list<item: struct<sub_task_instruction: string, frame_duration: list<item: double>>>
child 0, item: struct<sub_task_instruction: string, frame_duration: list<item: double>>
child 0, sub_task_instruction: string
child 1, frame_duration: list<item: double>
child 0, item: double
task_name: string
transcoding_cameras_failed: struct<>
to
{'task_instruction': Value('string'), 'task_num': Value('string'), 'task_name': Value('string'), 'robot_type': Value('string'), 'device_serial': Value('string'), 'episode_index': Value('int64'), 'timestamp': Value('timestamp[s]'), 'format_version': Value('string'), 'segments': List({'sub_task_instruction': Value('string'), 'frame_duration': List(Value('float64'))}), 'transcoding_status': Value('string'), 'video_remux_status': Value('string'), 'transcoding_cameras_failed': {}, 'sim_source': {'set': Value('string'), 'episode': Value('int64'), 'bottle_init_xy_yaw': List(Value('float64')), 'cap_torque_Nm': Value('float64'), 'cap_turn_deg': Value('float64'), 'strokes_deg': List(Value('float64')), 'program_params': {'l_tilt': Value('float64'), 'grip_h': Value('float64'), 'lift': Value('float64'), 'station': List(Value('float64')), 'psi0': Value('float64'), 'stroke': Value('float64'), 'speed': Value('float64'), 'pad_clear': Value('float64'), 'r_open': Value('float64'), 'close_mid': Value('float64'), 'r_close': Value('float64')}, 'release_deg': Value('float64'), 'tail_deg': Value('float64'), 'tilt_deg': Value('float64'), 'tilt_az_deg': Value('float64'), 'tilt_deg_achieved': Value('float64'), 'station_xyh': List(Value('float64')), 'psi0_deg': Value('float64'), 'success': Value('bool'), 'sim_fps': Value('int64')}}
because column names don't match
Traceback: Traceback (most recent call last):
File "/src/services/worker/src/worker/utils.py", line 149, in get_rows_or_raise
return get_rows(
dataset=dataset,
...<4 lines>...
column_names=column_names,
)
File "/src/libs/libcommon/src/libcommon/utils.py", line 272, in decorator
return func(*args, **kwargs)
File "/src/services/worker/src/worker/utils.py", line 129, in get_rows
rows_plus_one = list(itertools.islice(safe_iter(ds, dataset=dataset), rows_max_number + 1))
File "/src/services/worker/src/worker/utils.py", line 544, in safe_iter
yield from ds.decode(False) if ds.features else ds
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 3052, in __iter__
for key, example in ex_iterable:
^^^^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2559, in __iter__
for key, pa_table in self._iter_arrow():
~~~~~~~~~~~~~~~~^^
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2584, in _iter_arrow
for key, pa_table in self.ex_iterable._iter_arrow():
~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 558, in _iter_arrow
for key, pa_table in iterator:
^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 440, in _iter_arrow
for key, pa_table in self.generate_tables_fn(**gen_kwags):
~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/packaged_modules/json/json.py", line 361, in _generate_tables
self._cast_table(pa_table, json_field_paths=json_field_paths),
~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/packaged_modules/json/json.py", line 132, in _cast_table
pa_table = table_cast(pa_table, self.info.features.arrow_schema)
File "/usr/local/lib/python3.14/site-packages/datasets/table.py", line 2381, in table_cast
return cast_table_to_schema(table, schema)
File "/usr/local/lib/python3.14/site-packages/datasets/table.py", line 2309, in cast_table_to_schema
raise CastError(
...<3 lines>...
)
datasets.table.CastError: Couldn't cast
name: string
frames_received: int64
frames_written: int64
frames_remuxed: int64
frames_metadata_written: int64
frames_dropped_invalid: int64
frames_dropped_queue: int64
first_recv_ns: int64
last_recv_ns: int64
remux_status: string
remux_error: null
metadata_error: null
raw_write_error: null
source: string
video_remux_status: string
robot_type: string
device_serial: string
episode_index: int64
timestamp: timestamp[s]
task_instruction: string
task_num: string
format_version: string
transcoding_status: string
sim_source: struct<set: string, episode: int64, bottle_init_xy_yaw: list<item: double>, cap_torque_Nm: double, c (... 455 chars omitted)
child 0, set: string
child 1, episode: int64
child 2, bottle_init_xy_yaw: list<item: double>
child 0, item: double
child 3, cap_torque_Nm: double
child 4, cap_turn_deg: double
child 5, strokes_deg: list<item: double>
child 0, item: double
child 6, program_params: struct<l_tilt: double, grip_h: double, lift: double, station: list<item: double>, psi0: double, stro (... 97 chars omitted)
child 0, l_tilt: double
child 1, grip_h: double
child 2, lift: double
child 3, station: list<item: double>
child 0, item: double
child 4, psi0: double
child 5, stroke: double
child 6, speed: double
child 7, pad_clear: double
child 8, r_open: double
child 9, close_mid: double
child 10, r_close: double
child 7, release_deg: double
child 8, tail_deg: double
child 9, tilt_deg: double
child 10, tilt_az_deg: double
child 11, tilt_deg_achieved: double
child 12, station_xyh: list<item: double>
child 0, item: double
child 13, psi0_deg: double
child 14, success: bool
child 15, sim_fps: int64
segments: list<item: struct<sub_task_instruction: string, frame_duration: list<item: double>>>
child 0, item: struct<sub_task_instruction: string, frame_duration: list<item: double>>
child 0, sub_task_instruction: string
child 1, frame_duration: list<item: double>
child 0, item: double
task_name: string
transcoding_cameras_failed: struct<>
to
{'task_instruction': Value('string'), 'task_num': Value('string'), 'task_name': Value('string'), 'robot_type': Value('string'), 'device_serial': Value('string'), 'episode_index': Value('int64'), 'timestamp': Value('timestamp[s]'), 'format_version': Value('string'), 'segments': List({'sub_task_instruction': Value('string'), 'frame_duration': List(Value('float64'))}), 'transcoding_status': Value('string'), 'video_remux_status': Value('string'), 'transcoding_cameras_failed': {}, 'sim_source': {'set': Value('string'), 'episode': Value('int64'), 'bottle_init_xy_yaw': List(Value('float64')), 'cap_torque_Nm': Value('float64'), 'cap_turn_deg': Value('float64'), 'strokes_deg': List(Value('float64')), 'program_params': {'l_tilt': Value('float64'), 'grip_h': Value('float64'), 'lift': Value('float64'), 'station': List(Value('float64')), 'psi0': Value('float64'), 'stroke': Value('float64'), 'speed': Value('float64'), 'pad_clear': Value('float64'), 'r_open': Value('float64'), 'close_mid': Value('float64'), 'r_close': Value('float64')}, 'release_deg': Value('float64'), 'tail_deg': Value('float64'), 'tilt_deg': Value('float64'), 'tilt_az_deg': Value('float64'), 'tilt_deg_achieved': Value('float64'), 'station_xyh': List(Value('float64')), 'psi0_deg': Value('float64'), 'success': Value('bool'), 'sim_fps': Value('int64')}}
because column names don't matchNeed help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.
open-bottle top-down (sim, scripted): 5 sim demos in the ROBOTIS MCAP layout
5 Isaac Lab demos of opening a screw-cap bottle with two arms, written in the same layout as the real Cyclo
Intelligence recordings (format_version: robotis_v2, one folder per demo with <i>_0.mcap, metadata.yaml,
episode_info.json, camera_info/, robot.urdf, videos/<i>_0/<cam>.mp4), so they can be replayed on the real robot
like the earlier *-replay5 sets. Leader command topics = the sim joint targets (16: both arms + grippers),
/joint_states = the sim joint state, both resampled from 15 Hz to 100 Hz; videos frame-repeated to 30 fps. The sim has no
right head camera, so cam_right_head.mp4 is a copy of the left head video.
The motion (user's plan, 2026-11-07). The bottle stands upright on the table. The left hand picks it exactly as in
open_bottle_replay: opens fully, grasps the bottle by its lower end (5 cm up, from the bottle's left, 30 deg down) and
lifts it 3 cm. It then carries the bottle to the opening station on the robot's right (bottle bottom at x 0.36, y -0.08,
14 cm above the table; the right arm can only twist a downward hand on its own side) and tilts it: in every demo the bottle
axis leans a random 0-15 deg off vertical (the cone of a hand-held bottle), toward a random direction in the half-cone on the
right arm's side (azimuth 225-315 deg; leaning the top toward the left arm puts the left shoulder at its joint limit, so those
directions are left out of this set). The right hand comes down along the bottle axis (perpendicular to the cap top),
half open, centres on the cap, closes to 0.70 (not yet touching), sets its pad tips 2.5 mm above the cap rim (the cap is 11 mm
tall), squeezes to 0.79, and twists the cap 45 deg about the bottle axis at 10 deg/s, rising with the thread. It lets go,
backs up along the axis, returns to the same pose above the cap and pinches again, until the thread releases: the release angle
is drawn per demo from 170-210 deg (about half a turn, as on the real cap). Because the real cap still climbs 2-3 mm after
it lets go, the hand then keeps turning a random 0-15 deg tail while rising 3 mm, and only then lifts the cap off slowly
along the axis. All 5
demos succeeded in sim: the cap came free, stayed between the right pads and was lifted off; the bottle stayed in the left
hand and both arms kept at least 5 cm of surface clearance.
The sim is set up to match the real robot: FFW SG2, ROBOTIS RH-P12-RN CAD gripper visuals with calibrated wrist cameras, 29.25 in table. The bottle is an empty BODYARMOR LYTE 28 oz (86 mm body, 212 mm with cap, 45 g) with a 38 mm 3-start screw cap (41.5 mm, 11 mm tall) modelled as a thread joint: it turns at 0.40 N m (hand-tight), rises 1.91 mm/rad and lets go after the per-demo release angle (170-210 deg).
Things to check on the real robot before replaying: the left gripper is commanded fully open (0.0, ~107 mm pad gap) for the 86 mm bottle; the right close is 0.79 on the 41.5 mm cap (the real cap torque and pad friction set how much the cap slips per stroke, so the real cap may need more or fewer of the recorded strokes).
First frames (rows = folders 0-4; columns = head, left wrist, right wrist):
| folder | sim episode | bottle x [m] | bottle y [m] | bottle yaw [deg] | bottle tilt [deg] | release after [deg] | tail turn [deg] | cap torque [N m] | strokes | cap turn per stroke [deg] | cap total [deg] | length [s] | min arm gap [mm] |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 0 | 0 | 0.376 | 0.121 | -171 | 11.9 (toward 262) | 208 | 12 | 0.40 | 6 | 6, 43, 43, 44, 42, 42 | 231 | 120.1 | 98 |
| 1 | 1 | 0.399 | 0.132 | -11 | 5.7 (toward 294) | 183 | 10 | 0.40 | 5 | 42, 34, 44, 44, 45 | 214 | 113.3 | 55 |
| 2 | 2 | 0.361 | 0.131 | +131 | 5.4 (toward 258) | 197 | 9 | 0.40 | 5 | 44, 44, 39, 44, 45 | 223 | 106.8 | 123 |
| 3 | 3 | 0.391 | 0.105 | -101 | 11.0 (toward 246) | 198 | 14 | 0.40 | 6 | 3, 44, 44, 44, 44, 45 | 254 | 128.0 | 64 |
| 4 | 4 | 0.397 | 0.133 | -174 | 0.6 (toward 269) | 208 | 12 | 0.40 | 6 | 20, 39, 44, 44, 40, 45 | 246 | 122.2 | 70 |
Source: data_collection/bottle_cap/collect_top_open.py (recorded set open_bottle_topdown_replay), exported with
data_collection/export_robotis_mcap.py; recipe validated by data_collection/bottle_cap/analysis/probe_top_open.py.
- Downloads last month
- 68