README: add YAM Ultra 2

README.md

@@ -1,5 +1,5 @@
 ---
-pretty_name: "rhoyn Teleoperation Sample: Unitree G1, OpenArm and reBot"
+pretty_name: "rhoyn Teleoperation Sample: Unitree G1, OpenArm, reBot and YAM Ultra 2"
 license: cc-by-4.0
 task_categories:
   - robotics
@@ -11,6 +11,7 @@ tags:
   - unitree-g1
   - openarm
   - rebot
+  - yam
   - teleoperation
   - whole-body
   - bimanual
@@ -48,11 +49,23 @@ configs:
     data_files: rebot/socks/data/*/*.parquet
   - config_name: rebot_wipes
     data_files: rebot/wipes/data/*/*.parquet
+  - config_name: yam_eggs
+    data_files: yam/eggs/data/*/*.parquet
+  - config_name: yam_connect_four
+    data_files: yam/connect_four/data/*/*.parquet
+  - config_name: yam_jenga
+    data_files: yam/jenga/data/*/*.parquet
+  - config_name: yam_puzzle
+    data_files: yam/puzzle/data/*/*.parquet
+  - config_name: yam_socks
+    data_files: yam/socks/data/*/*.parquet
+  - config_name: yam_wipes
+    data_files: yam/wipes/data/*/*.parquet
 ---
 
-# rhoyn Teleoperation Sample: Unitree G1, OpenArm and reBot
+# rhoyn Teleoperation Sample: Unitree G1, OpenArm, reBot and YAM Ultra 2
 
-Teleoperation data from [rhoyn](https://rhoyn.com), recorded with our own low-latency VR teleoperation stack: the operator wears a PICO 4 Ultra headset and the robot's arms and grippers follow their hands precisely. The sample covers three robots: a **Unitree G1** whole-body humanoid, a bimanual **OpenArm** (two 7-DoF arms) and a bimanual **reBot B601-RS** (two 6-DoF arms). Each episode records every sensor on the robot alongside a standard LeRobot v3.0 dataset. This is a free sample of the format; [rhoyn collects data to your spec](#custom-data-collection). See also [rhoyn.com/data](https://rhoyn.com/data).
+Teleoperation data from [rhoyn](https://rhoyn.com), recorded with our own low-latency VR teleoperation stack: the operator wears a PICO 4 Ultra headset and the robot's arms and grippers follow their hands precisely. The sample covers four robots: a **Unitree G1** whole-body humanoid, a bimanual **OpenArm** (two 7-DoF arms), a bimanual **reBot B601-RS** (two 6-DoF arms) and a bimanual **YAM Ultra 2** (two 6-DoF arms). Each episode records every sensor on the robot alongside a standard LeRobot v3.0 dataset. This is a free sample of the format; [rhoyn collects data to your spec](#custom-data-collection). See also [rhoyn.com/data](https://rhoyn.com/data).
 
 ### Unitree G1
 
@@ -102,19 +115,19 @@ Each reBot preview is about 8 s of the episode, in the same layout as the OpenAr
 
 ## At a glance
 
-| | Unitree G1 | OpenArm | reBot |
-|---|---|---|---|
-| **Robot** | Unitree G1 Edu+, 29 DoF whole-body humanoid, Dex1-1 grippers | OpenArm, two 7-DoF arms (bimanual), Dex1-1 grippers | reBot B601-RS, two 6-DoF arms (bimanual) with grippers |
-| **Teleoperation** | rhoyn VR teleop, PICO 4 Ultra headset and controllers | rhoyn VR teleop, PICO 4 Ultra headset and controllers | rhoyn VR teleop, PICO headset and controllers |
-| **Episodes** | 3, one per task · 12,633 frames · 7 min | 6, one per task · 20,841 frames · 11.6 min | 6, one per task · 23,543 frames · 13.1 min |
-| **Rate** | 30 Hz rows; depth at 15 fps; lidar, IMUs and VR packets at native rates | 30 Hz rows; depth at 15 fps; VR packets at native rate | 30 Hz rows; depth at 15 fps; VR packets at native rate |
-| **Cameras** | 9 streams: RealSense D435i on the head + two wrist D405s, each color + stereo IR | 9 streams: RealSense D435F on top of the column + two wrist D405s, each color + stereo IR | 9 streams: RealSense D435F on a camera post + two wrist D405s, each color + stereo IR |
-| **Depth** | Lossless 16-bit from all three cameras | Lossless 16-bit from all three cameras | Lossless 16-bit from all three cameras |
-| **Lidar** | Livox MID-360 points, raw packets and IMU | None | None |
-| **IMUs** | Pelvis, chest, D435i | None | None |
-| **Format** | LeRobotDataset v3.0 per task folder, plus sidecar streams | LeRobotDataset v3.0 per task folder, plus sidecar streams | LeRobotDataset v3.0 per task folder, plus sidecar streams |
-| **Size** | 2.26 GB | 4.26 GB | 5.16 GB |
-| **License** | CC BY 4.0 | CC BY 4.0 | CC BY 4.0 |
+| | Unitree G1 | OpenArm | reBot | YAM Ultra 2 |
+|---|---|---|---|---|
+| **Robot** | Unitree G1 Edu+, 29 DoF whole-body humanoid, Dex1-1 grippers | OpenArm, two 7-DoF arms (bimanual), Dex1-1 grippers | reBot B601-RS, two 6-DoF arms (bimanual) with grippers | I2RT YAM Ultra 2, two 6-DoF arms (bimanual) with grippers |
+| **Teleoperation** | rhoyn VR teleop, PICO 4 Ultra headset and controllers | rhoyn VR teleop, PICO 4 Ultra headset and controllers | rhoyn VR teleop, PICO headset and controllers | rhoyn VR teleop, PICO headset and controllers |
+| **Episodes** | 3, one per task · 12,633 frames · 7 min | 6, one per task · 20,841 frames · 11.6 min | 6, one per task · 23,543 frames · 13.1 min | 6, one per task · 26,651 frames · 14.8 min |
+| **Rate** | 30 Hz rows; depth at 15 fps; lidar, IMUs and VR packets at native rates | 30 Hz rows; depth at 15 fps; VR packets at native rate | 30 Hz rows; depth at 15 fps; VR packets at native rate | 30 Hz rows; depth at 15 fps; VR packets at native rate |
+| **Cameras** | 9 streams: RealSense D435i on the head + two wrist D405s, each color + stereo IR | 9 streams: RealSense D435F on top of the column + two wrist D405s, each color + stereo IR | 9 streams: RealSense D435F on a camera post + two wrist D405s, each color + stereo IR | 9 streams: RealSense D435F on a camera post + two wrist D405s, each color + stereo IR |
+| **Depth** | Lossless 16-bit from all three cameras | Lossless 16-bit from all three cameras | Lossless 16-bit from all three cameras | Lossless 16-bit from all three cameras |
+| **Lidar** | Livox MID-360 points, raw packets and IMU | None | None | None |
+| **IMUs** | Pelvis, chest, D435i | None | None | None |
+| **Format** | LeRobotDataset v3.0 per task folder, plus sidecar streams | LeRobotDataset v3.0 per task folder, plus sidecar streams | LeRobotDataset v3.0 per task folder, plus sidecar streams | LeRobotDataset v3.0 per task folder, plus sidecar streams |
+| **Size** | 2.26 GB | 4.26 GB | 5.16 GB | 5.58 GB |
+| **License** | CC BY 4.0 | CC BY 4.0 | CC BY 4.0 | CC BY 4.0 |
 
 ## Tasks
 
@@ -156,6 +169,19 @@ Three more G1 tasks (connect four, jenga and a puzzle) are in the full six-task
 
 **Per reBot episode:** 9 synchronized camera streams (a RealSense D435F on a camera post + two wrist cameras, color and stereo IR), lossless 16-bit depth from all three cameras, the full joint state and torque pipeline (PD gains, feedforward and velocity targets, requested/capped/commanded torque, torque budget and limiter state, measured joint torque, motor temperature), and the raw PICO VR teleoperation packets. There is no lidar and there are no IMUs on the reBot.
 
+### YAM Ultra 2
+
+| Folder | Task | Duration | Frames |
+|---|---|---|---|
+| `yam/eggs/` | Put butter in the pan, crack the egg into it and scramble it | 252 s | 7,548 |
+| `yam/connect_four/` | Pick up a disc, pass it to the other gripper and drop it into the connect four board | 57 s | 1,702 |
+| `yam/jenga/` | Push a block out of the Jenga tower, pass it to the other gripper and place it on top | 139 s | 4,162 |
+| `yam/puzzle/` | Pick up the puzzle pieces and place them | 79 s | 2,366 |
+| `yam/socks/` | Pick up the two socks and pair them together | 240 s | 7,201 |
+| `yam/wipes/` | Open the pack, pull out a wet wipe and wipe the table | 122 s | 3,672 |
+
+**Per YAM Ultra 2 episode:** the same streams as the reBot: 9 synchronized camera streams (a RealSense D435F on a camera post + two wrist cameras, color and stereo IR), lossless 16-bit depth from all three cameras, the full joint state and torque pipeline, and the raw PICO VR teleoperation packets. There is no lidar and there are no IMUs on the YAM Ultra 2.
+
 Each folder holds one episode. The task column below is the episode's language instruction, exactly as stored in `meta/tasks.parquet`.
 
 ## Loading
@@ -183,6 +209,10 @@ ds = LeRobotDataset("rhoyn-robotics/data-sample", root=f"{path}/openarm/wipes")
 # reBot
 path = snapshot_download("rhoyn-robotics/data-sample", repo_type="dataset", allow_patterns="rebot/wipes/*")
 ds = LeRobotDataset("rhoyn-robotics/data-sample", root=f"{path}/rebot/wipes")
+
+# YAM Ultra 2
+path = snapshot_download("rhoyn-robotics/data-sample", repo_type="dataset", allow_patterns="yam/wipes/*")
+ds = LeRobotDataset("rhoyn-robotics/data-sample", root=f"{path}/yam/wipes")
 ```
 
 Or from the command line:
@@ -191,13 +221,14 @@ Or from the command line:
 hf download rhoyn-robotics/data-sample --repo-type dataset --include "wipes/*" --local-dir rhoyn-sample
 hf download rhoyn-robotics/data-sample --repo-type dataset --include "openarm/wipes/*" --local-dir rhoyn-sample
 hf download rhoyn-robotics/data-sample --repo-type dataset --include "rebot/wipes/*" --local-dir rhoyn-sample
+hf download rhoyn-robotics/data-sample --repo-type dataset --include "yam/wipes/*" --local-dir rhoyn-sample
 ```
 
 Loading by repo id alone (`LeRobotDataset("rhoyn-robotics/data-sample")`) is not supported, because the repo root is not itself a LeRobot dataset. Use the `root=` form above.
 
 Videos are AV1-encoded. Decoding needs an AV1-capable decoder (dav1d or libaom) in your ffmpeg / PyAV / torchcodec build. Recent pip wheels of PyAV and torchcodec include one.
 
-The per-frame data can also be read directly with pandas or pyarrow from `<task>/data/chunk-000/*.parquet` (G1), `openarm/<task>/data/chunk-000/*.parquet` (OpenArm) or `rebot/<task>/data/chunk-000/*.parquet` (reBot).
+The per-frame data can also be read directly with pandas or pyarrow from `<task>/data/chunk-000/*.parquet` (G1), `openarm/<task>/data/chunk-000/*.parquet` (OpenArm), `rebot/<task>/data/chunk-000/*.parquet` (reBot) or `yam/<task>/data/chunk-000/*.parquet` (YAM Ultra 2).
 
 ## Control and actions
 
@@ -214,13 +245,15 @@ For policy learning on the G1, the operator-driven action is the arm and gripper
 
 **reBot.** The same as the OpenArm, with 12 arm joints: both arms track the operator's wrist poses and both grippers follow the grip triggers. `action` is the full operator-driven action, and the locomotion fields of `action.teleop` are not used.
 
+**YAM Ultra 2.** The same as the reBot: 12 arm joints and 2 grippers, all following the operator. `action` is the full operator-driven action, and the locomotion fields of `action.teleop` are not used.
+
 ## Reference
 
 <details>
 <summary><b>Contents of each episode folder</b></summary>
 
 ```
-<task>/  (G1)        openarm/<task>/  (OpenArm)        rebot/<task>/  (reBot)
+<task>/  (G1)        openarm/<task>/  (OpenArm)        rebot/<task>/  (reBot)        yam/<task>/  (YAM Ultra 2)
 ├── meta/            info.json, stats.json, tasks.parquet, episodes/, plus sidecar metadata
 │                    (calibration.json, depth.json, lidar.json, d435_imu.json, pico.json)
 ├── data/            per-frame parquet (state, actions, IMU, torques, teleop, sync info)
@@ -233,7 +266,7 @@ For policy learning on the G1, the operator-driven action is the arm and gripper
 
 The sidecar folders (`depth/`, `lidar/`, `d435_imu/`, `pico/`) are not part of the LeRobot format. LeRobot tools such as `aggregate_datasets` ignore them, so read them directly using the paths and metadata in `meta/`.
 
-OpenArm and reBot episodes have no `lidar/` or `d435_imu/` folder and no `meta/lidar.json`. Their `meta/d435_imu.json` and the `lidar` and `d435_imu` entries (null) in `meta/calibration.json` come from the shared recording stack and point to no data.
+OpenArm, reBot and YAM Ultra 2 episodes have no `lidar/` or `d435_imu/` folder and no `meta/lidar.json`. Their `meta/d435_imu.json` and the `lidar` and `d435_imu` entries (null) in `meta/calibration.json` come from the shared recording stack and point to no data.
 
 </details>
 
@@ -242,16 +275,16 @@ OpenArm and reBot episodes have no `lidar/` or `d435_imu/` folder and no `meta/l
 
 | Key | Resolution | Description |
 |---|---|---|
-| `observation.images.d435` | 848×480 | RealSense D435i on the head (G1), D435F on top of the column, tilted 20° (OpenArm), or D435F on a camera post, tilted 20° (reBot); color |
-| `observation.images.d435_ir1`, `_ir2` | 848×480 | D435i (G1) or D435F (OpenArm, reBot), stereo IR |
+| `observation.images.d435` | 848×480 | RealSense D435i on the head (G1), D435F on top of the column, tilted 20° (OpenArm), or D435F on a camera post, tilted 20° (reBot, YAM Ultra 2); color |
+| `observation.images.d435_ir1`, `_ir2` | 848×480 | D435i (G1) or D435F (OpenArm, reBot, YAM Ultra 2), stereo IR |
 | `observation.images.d405_left`, `d405_right` | 424×240 | Wrist RealSense D405, color |
 | `observation.images.d405_{left,right}_ir1`, `_ir2` | 424×240 | Wrist D405, stereo IR |
 
-The camera keys and resolutions are the same on all three robots.
+The camera keys and resolutions are the same on all four robots.
 
 IR streams are monochrome but stored as 3-channel video (the gray value is replicated), so they load as `(3, H, W)` like the color streams.
 
-On all three robots the right wrist camera is mounted upside down and its frames are rotated upright at recording time (`frames_rotated_upright: true` in `meta/calibration.json`). The left wrist camera's frames are stored as captured.
+On all four robots the right wrist camera is mounted upside down and its frames are rotated upright at recording time (`frames_rotated_upright: true` in `meta/calibration.json`). The left wrist camera's frames are stored as captured.
 
 </details>
 
@@ -327,21 +360,32 @@ All six reBot episodes share an identical schema (`robot_type: seeed_rebot_b601_
 
 </details>
 
-**Note on NaNs:** `observation.frame_counter`, `observation.sensor_ts_us` and `observation.frame_offset_us` have three depth columns (`d435_depth`, `d405_left_depth`, `d405_right_depth`). On the G1 and OpenArm they are NaN on every row. On the reBot they are filled on the even rows, which carry a depth frame, and NaN on the odd rows. This is expected: depth is recorded at 15 fps in `depth/`, and `meta/depth.json` maps each depth frame to its data row.
+<details>
+<summary><b>Per-frame data, YAM Ultra 2</b> (<code>yam/&lt;task&gt;/data/</code>)</summary>
+
+The keys and dimensions are identical to the reBot's: 12 arm joints and 2 grippers, so `action` has 14 dimensions (12 joint position targets, 2 gripper aperture targets) and `observation.state` 26.
+
+The torque limits differ. The YAM Ultra 2 has no torque budget or temperature scale: its arm joints are limited only by their motor peak (28 N·m for the DM4340 shoulder, elbow and wrist pitch motors, 10 N·m for the DM4310 wrist yaw, wrist roll and gripper motors), and each gripper has a closing torque cap. `observation.torque_budget` and `action.torque_allowed` are present for schema compatibility but constant; their exact meaning is given per element in `meta/info.json`.
+
+All six YAM Ultra 2 episodes share an identical schema (`robot_type: i2rt_yam_ultra_2_2x6dof`) and can be merged with LeRobot's `aggregate_datasets`. Full joint names and per-element units are in `meta/info.json`.
+
+</details>
+
+**Note on NaNs:** `observation.frame_counter`, `observation.sensor_ts_us` and `observation.frame_offset_us` have three depth columns (`d435_depth`, `d405_left_depth`, `d405_right_depth`). On the G1 and OpenArm they are NaN on every row. On the reBot and YAM Ultra 2 they are filled on the even rows, which carry a depth frame, and NaN on the odd rows. This is expected: depth is recorded at 15 fps in `depth/`, and `meta/depth.json` maps each depth frame to its data row.
 
 <details>
 <summary><b>Synchronization</b></summary>
 
-All streams are timestamped against one host clock. The data rows are a fixed 30 Hz grid (`observation.row_time_us`), and each camera contributes the frame nearest to each row. The cameras run freely and are not hardware-triggered, so each camera frame is within about ±16.7 ms (half a frame) of its row time, and D435i/D435F and wrist frames in the same row can be up to about 30 ms apart (G1, OpenArm). On the reBot they are typically 12–20 ms apart, and up to about 120 ms apart on the few rows around a dropped frame. The exact capture time and offset of every stream's frame are recorded per row in `observation.sensor_ts_us` and `observation.frame_offset_us`.
+All streams are timestamped against one host clock. The data rows are a fixed 30 Hz grid (`observation.row_time_us`), and each camera contributes the frame nearest to each row. The cameras run freely and are not hardware-triggered, so each camera frame is within about ±16.7 ms (half a frame) of its row time, and D435i/D435F and wrist frames in the same row can be up to about 30 ms apart (G1, OpenArm). On the reBot they are typically 12–20 ms apart, and up to about 120 ms apart on the few rows around a dropped frame. On the YAM Ultra 2 they are typically 17–20 ms apart, and up to about 110 ms apart around a dropped frame; there, 1–5% of rows have a camera frame more than 16.7 ms from its row time, by up to about 102 ms. The exact capture time and offset of every stream's frame are recorded per row in `observation.sensor_ts_us` and `observation.frame_offset_us`.
 
-Lidar points, IMU samples (G1) and Pico packets (all three robots) keep their native rates and carry their own timestamps. `meta/lidar.json`, `meta/d435_imu.json` and `meta/pico.json` describe the clocks.
+Lidar points, IMU samples (G1) and Pico packets (all four robots) keep their native rates and carry their own timestamps. `meta/lidar.json`, `meta/d435_imu.json` and `meta/pico.json` describe the clocks.
 
 </details>
 
 <details>
 <summary><b>Depth</b></summary>
 
-Depth is lossless 16-bit (FFV1, `gray16le`), 424×240 at 15 fps, for all three cameras, on all three robots. Depth frame *i* belongs to data row 2*i*. Convert to meters with `meters = raw * depth_scale` (0.001 for the D435i and D435F, 0.0001 for the D405s, given per stream in `meta/depth.json`).
+Depth is lossless 16-bit (FFV1, `gray16le`), 424×240 at 15 fps, for all three cameras, on all four robots. Depth frame *i* belongs to data row 2*i*. Convert to meters with `meters = raw * depth_scale` (0.001 for the D435i and D435F, 0.0001 for the D405s, given per stream in `meta/depth.json`).
 
 Two raw values mean "no valid depth" and should be masked:
 
@@ -357,7 +401,7 @@ The stats in `meta/depth.json` exclude both values.
 
 Camera intrinsics (all color, IR and depth streams) are RealSense factory calibration, read from each device at recording time.
 
-Camera-to-robot (and, on the G1, lidar-to-robot) extrinsics come from the robot model, not from a hand-eye calibration. On the G1, the D435i and lidar poses come from the Unitree G1 URDF; the wrist cameras' poses come from the CAD model of their 3D-printed mounts. On the OpenArm, the D435F pose comes from the CAD model of its 20° column mount and the wrist cameras' poses from their Dex1-1 brackets. On the reBot, the D435F pose comes from the same 20° mount on the B601 camera post and the wrist cameras' poses from the B601 wrist mount model. Every transform carries a `source` field saying this. These poses are suitable for coarse fusion. If you need precise depth-to-robot or lidar-to-camera alignment, plan to refine them.
+Camera-to-robot (and, on the G1, lidar-to-robot) extrinsics come from the robot model, not from a hand-eye calibration. On the G1, the D435i and lidar poses come from the Unitree G1 URDF; the wrist cameras' poses come from the CAD model of their 3D-printed mounts. On the OpenArm, the D435F pose comes from the CAD model of its 20° column mount and the wrist cameras' poses from their Dex1-1 brackets. On the reBot, the D435F pose comes from the same 20° mount on the B601 camera post and the wrist cameras' poses from the B601 wrist mount model. On the YAM Ultra 2, the D435F pose comes from the same 20° mount on its camera post and the wrist cameras' poses from the model of the D405 wrist rig on the gripper plate. Every transform carries a `source` field saying this. These poses are suitable for coarse fusion. If you need precise depth-to-robot or lidar-to-camera alignment, plan to refine them.
 
 </details>
 
@@ -367,7 +411,7 @@ This sample is released under [CC BY 4.0](https://creativecommons.org/licenses/b
 
 ```bibtex
 @misc{rhoyn2026g1sample,
-  title        = {rhoyn Teleoperation Sample: Unitree G1, OpenArm and reBot},
+  title        = {rhoyn Teleoperation Sample: Unitree G1, OpenArm, reBot and YAM Ultra 2},
   author       = {{rhoyn}},
   year         = {2026},
   howpublished = {\url{https://huggingface.co/datasets/rhoyn-robotics/data-sample}}