Skip to main content

How to use the SO-ARM100 and SO-ARM101 robotic arm in Lerobot

Introduction​

The SO-10xARM is a fully open-source robotic arm project launched by TheRobotStudio. It includes the follower arm and the leader robotic arm, and also provides detailed 3D printing files and operation guides. LeRobot is committed to providing models, datasets and tools for real-world robotics in PyTorch. Its aim is to reduce the entry barrier of robotics, enabling everyone to contribute and benefit from sharing datasets and pretrained models. LeRobot integrates cutting-edge methodologies validated for real-world application, centering on imitation learning. It has furnished a suite of pre-trained models, datasets featuring human-gathered demonstrations, and simulation environments, enabling users to commence without the necessity of robot assembly. In the forthcoming weeks, the intention is to augment support for real-world robotics on the most cost-effective and competent robots presently accessible.

Projects Introduction​

The SO-ARM10x and reComputer Jetson AI intelligent robot kit seamlessly combine high-precision robotic arm control with a powerful AI computing platform, providing a comprehensive robot development solution. This kit is based on the Jetson Orin or AGX Orin platform, combined with the SO-ARM10x robotic arm and the LeRobot AI framework, offering users an intelligent robot system applicable to multiple scenarios such as education, research, and industrial automation. This wiki provides the assembly and debugging tutorial for the SO ARM10x and realizes data collection and training within the Lerobot framework.

Get One Now 🖱️

Main Features​

  1. Open-source and low-cost: It is an open-source, low-cost robotic arm solution from TheRobotStudio
  2. Integration with LeRobot: Designed for integration with LeRobot platform
  3. Abundant learning resources: Provides comprehensive open-source learning resources like assembly and calibration guides, and tutorials for testing, data collection, training and deployment to assist users in quickly getting started and developing robotic applications.
  4. Compatible with Nvidia: Deploy this arm kit with reComputer Mini J4012 Orin NX 16 GB.
  5. Multi-Scene Application: It is applicable to fields such as education, scientific research, automated production, and robotics, helping users achieve efficient and precise robot operations in various complex tasks.

What's news：​

  • Wiring optimization: Compared to SO-ARM100, SO-ARM101 features improved wiring that prevents disconnection issues previously seen at joint 3. The new wiring design also no longer limits the range of motion of the joints.
  • Different gear ratios for the leader arm: The leader arm now uses motors with optimized gear ratios, improving performance and eliminating the need for external gearboxes.
  • New functionality support: The leader arm can now follow the follower arm in real-time, which is crucial for the upcoming learning policy, where a human can intervene and correct the robot’s actions.
caution

Seeed Studio is only responsible for the quality of the hardware itself. The tutorials are strictly updated in accordance with the official documentation. If you encounter software issues or environment dependency problems that cannot be resolved, in addition to checking the FAQ section at the end of this tutorial, please promptly report the issue to the LeRobot platform or the LeRobot Discord channel.

Specification​

TypeSO-ARM100SO-ARM101
Arm KitArm Kit ProArm KitArm Kit Pro
Leader Arm12x ST-3215- C001 (7.4V) motors with 1:345 gear ratio for all joints12x ST-3215-C018/ST-3215-C047 (12V) motors with 1:345 gear ratio for all joints1x ST-3215- C001 (7.4V) motor with 1:345 gear ratio for joint 2 only
2x ST-3215-C044 (7.4V) motors with 1:191 gear ratio for joints 1 and 3
3x ST-3215-C046 (7.4V) motors with 1:147 gear ratio for joints 4, 5, and gripper (joint 6)
Follower ArmSame as SO-ARM100
Power Supply5.5 mm × 2.1 mm DC 5 V 4 A5.5 mm × 2.1 mm DC 12 V 2 A5.5 mm × 2.1 mm DC 5 V 4 A5.5 mm × 2.1 mm DC 12 V 2 A (Leader Arm)
5.5 mm × 2.1 mm DC 5 V 4 A (Follower Arm)
Angle Sensor12-bit magnetic encoder
Recommended Operating Temperature0 °C to 40 °C
CommunicationUART
Control MethodPC
danger

If you purchase the Arm Kit version, both power supplies are 5V. If you purchase the Arm Kit Pro version, please use the 5V power supply for the calibration and every step of the Leader robotic arm, and the 12V power supply for the calibration and every step of the Follower robotic arm.

Bill of Materials(BOM)​

PartAmountIncluded
Servo Motos12✅
Motor Control Board2✅
USB-C Cable 2 pcs1✅
Power Supply22✅
Table Clamp4✅
3D printed parts of the arm1❌

Initial System Environment​

For Ubuntu x86:

  • Ubuntu 22.04
  • CUDA 12+
  • Python 3.10
  • Torch 2.6

For Jetson Orin:

  • Jetson JetPack 6.2
  • Python 3.10
  • Torch 2.6

Table of Contents​

A. 3D Printing Guide

B. Install LeRobot

C. Configure the motors

D. Assembly

E. Calibrate

F. Teleoperate

G. Add cameras

H. Record the dataset

I. Visualize the dataset

J. Replay an episode

K. Train a policy

L. Evaluate your policy

3D Printing Guide​

caution

Following the official update of SO101, SO100 will no longer support it and the source files will be deleted as per the official, but the source files can still be found in our Makerworld. However, for users who have previously purchased SO100, the tutorials and installation methods remain compatible. The print of SO101 is fully compatible with the motor kit installation of SO100.

Step 1: Choose a printer​

The STL files provided are ready to print on many FDM printers. Below are the tested and suggested settings though others may work.

  • Material: PLA+
  • Nozzle Diameter and Precision: 0.4mm nozzle diameter at 0.2mm layer height or 0.6mm nozzle at 0.4mm layer height.
  • Infill Density: 15%

Step 2: Set up the printer​

  • Ensure that the printer is calibrated and the bed level is correctly set using the printer specific instructions.
  • Clean the print bed, making sure it is free from dust, or grease. If cleaning the bed using water, or other liquid, dry the bed.
  • If your printer recommends it, use a standard glue stick and apply a thin, even layer of glue across the print area of the bed. Avoid clumping or uneven application.
  • Load the printer filament using printer specific instructions.
  • Ensure the printer settings match the ones suggested above (most printers have multiple settings so choose the ones that most closely match).
  • Set for supports everywhere but ignore slopes greater than 45 degrees to the horizontal.
  • There should be no supports in the screw holes with horizontal axes.

Step 3: Print the parts​

All the parts for the leader or follower are for easy 3D printing already contained in a single file, correctly orientated for z upwards to minimize supports.

  • For printer bed sizes of 220mmx220mm (such as the Ender), print these files:

  • For printer bed sizes of 205mm x 250mm (such as the Prusa/Up):

Install LeRobot​

Environments such as pytorch and torchvision need to be installed based on your CUDA.

  1. Install Miniconda: For Jetson:
wget https://repo.anaconda.com/miniconda/Miniconda3-latest-Linux-aarch64.sh
chmod +x Miniconda3-latest-Linux-aarch64.sh
./Miniconda3-latest-Linux-aarch64.sh
source ~/.bashrc

Or, For X86 Ubuntu 22.04:

mkdir -p ~/miniconda3
cd miniconda3
wget https://repo.anaconda.com/miniconda/Miniconda3-latest-Linux-x86_64.sh -O ~/miniconda3/miniconda.sh
bash ~/miniconda3/miniconda.sh -b -u -p ~/miniconda3
rm ~/miniconda3/miniconda.sh
source ~/miniconda3/bin/activate
conda init --all
  1. Create and activate a fresh conda environment for lerobot
conda create -y -n lerobot python=3.10 && conda activate lerobot
  1. Clone Lerobot:
git clone https://github.com/ZhuYaoHui1998/lerobot.git ~/lerobot
  1. When using miniconda, install ffmpeg in your environment:
conda install ffmpeg -c conda-forge
tip

This usually installs ffmpeg 7.X for your platform compiled with the libsvtav1 encoder. If libsvtav1 is not supported (check supported encoders with ffmpeg -encoders), you can:

  • [On any platform] Explicitly install ffmpeg 7.X using:
conda install ffmpeg=7.1.1 -c conda-forge
  • [On Linux only] Install ffmpeg build dependencies and compile ffmpeg from source with libsvtav1, and make sure you use the corresponding ffmpeg binary to your install with which ffmpeg.
  1. Install LeRobot with dependencies for the feetech motors:
cd ~/lerobot && pip install -e ".[feetech]"

For Jetson Jetpack devices (please make sure to install Pytorch-gpu and Torchvision from step 5 before executing this step):

conda install -y -c conda-forge "opencv>=4.10.0.84"  # Install OpenCV and other dependencies through conda, this step is only for Jetson Jetpack 6.0+
conda remove opencv   # Uninstall OpenCV 
pip3 install opencv-python==4.10.0.84  # Then install opencv-python via pip3
conda install -y -c conda-forge ffmpeg
conda uninstall numpy
pip3 install numpy==1.26.0  # This should match torchvision
  1. Check Pytorch and Torchvision

Since installing the lerobot environment via pip will uninstall the original Pytorch and Torchvision and install the CPU versions of Pytorch and Torchvision, you need to perform a check in Python.

import torch
print(torch.cuda.is_available())

If the printed result is False, you need to reinstall Pytorch and Torchvision according to the official website tutorial.

If you are using a Jetson device, install Pytorch and Torchvision according to this tutorial.

Configure the motors​

Designate one bus servo adapter and 6 motors for your leader arm, and similarly the other bus servo adapter and 6 motors for the follower arm. It's convenient to label them and write on each motor if it's for the follower F or for the leader L and it's ID from 1 to 6 (F1...F6 and L1...L6).

Follow steps 1 of the assembly video which illustrates the use of our scripts below.

Find USB ports associated to your arms To find the correct ports for each arm, run the utility script twice:

python lerobot/scripts/find_motors_bus_port.py

Example output when identifying the leader arm's port (e.g., /dev/tty.usbmodem575E0031751 on Mac, or possibly /dev/ttyACM0 on Linux):

Example output when identifying the follower arm's port (e.g., /dev/tty.usbmodem575E0032081, or possibly /dev/ttyACM1 on Linux):

Troubleshooting: On Linux, you might need to give access to the USB ports by running:

sudo chmod 666 /dev/ttyACM0
sudo chmod 666 /dev/ttyACM1

Configure your motors

Plug your first motor and run this script to set its ID to 1. It will also set its present position to 2048, so expect your motor to rotate:

python lerobot/scripts/configure_motor.py \
  --port /dev/ttyACM0 \
  --brand feetech \
  --model sts3215 \
  --baudrate 1000000 \
  --ID 1
note

Note: These motors are currently limitated. They can take values between 0 and 4096 only, which corresponds to a full turn. They can't turn more than that. 2048 is at the middle of this range, so we can take -2048 steps (180 degrees anticlockwise) and reach the maximum range, or take +2048 steps (180 degrees clockwise) and reach the maximum range. The configuration step also sets the homing offset to 0, so that if you misassembled the arm, you can always update the homing offset to account for a shift up to ± 2048 steps (± 180 degrees).

Then unplug your motor and plug the second motor and set its ID to 2.

python lerobot/scripts/configure_motor.py \
  --port /dev/ttyACM0 \
  --brand feetech \
  --model sts3215 \
  --baudrate 1000000 \
  --ID 2

Redo the process for all your motors until ID 6. Do the same for the 6 motors of the leader arm.

Assembly​

Assemble Leader Arm

tip
  • The dual-arm assembly process of SO-ARM101 is the same as that of SO-ARM100. The only differences are the addition of cable clips on SO-ARM101 and the different gear ratios of the joint servos on the Leader Arm. So both SO100 and SO101 can be installed by referring to the following content
  • After calibrating the servos, do not rotate them before tightening the screws. Make sure the orientation of the 3D printed parts matches the reference direction in the images and that the motors are in their middle positions.
Step 1Step 2Step 3Step 4Step 5Step 6
fig1fig2fig3fig4fig5fig6
Step 7Step 8Step 9Step 10Step 11Step 12
fig1fig2fig3fig4fig5fig6
Step 13Step 14Step 15Step 16Step 17Step 18
fig1fig2fig3fig4fig5fig6
Step 19Step 20
fig1fig2

Assemble Follower Arm

tip
  • The steps for assembling the Follower Arm are generally the same as those for the Leader Arm. The only difference lies in the installation method of the end-effector (gripper and handle) after Step 12.
Step 1Step 2Step 3Step 4Step 5Step 6
fig1fig2fig3fig4fig5fig6
Step 7Step 8Step 9Step 10Step 11Step 12
fig1fig2fig3fig4fig5fig6
Step 13Step 14Step 15Step 16Step 17
fig1fig2fig3fig4fig5

Calibrate​

tip

The SO100 and SO101 codes are compatible. Users of SO100 can directly utilize SO101's parameters and code for operation.

danger

If you purchased the SO101 Arm Kit Standard Edition, all power supplies are 5V. If you purchased the SO101 Arm Kit Pro Edition, the Leader Arm should be calibrated and operated at every step using a 5V power supply, while the Follower Arm should be calibrated and operated at every step using a 12V power supply.

Next, you need to connect the power supply and data cable to your SO-10x robot for calibration to ensure that the leader and follower arms have the same position values when they are in the same physical position. This calibration is essential because it allows a neural network trained on one SO-10x robot to work on another. If you need to recalibrate the robotic arm, delete the ~/lerobot/.cache/huggingface/calibration/so101 folder.

Manual calibration of follower arm

IMPORTANTLY: Now that you have your ports, update the port default values of SO101RobotConfig (lerobot/lerobot/common/robot_devices/robots/configs.py). You will find something like:

@RobotConfig.register_subclass("so101")
@dataclass
class So101RobotConfig(ManipulatorRobotConfig):
    calibration_dir: str = ".cache/calibration/so101"
    # `max_relative_target` limits the magnitude of the relative positional target vector for safety purposes.
    # Set this to a positive scalar to have the same value for all motors, or a list that is the same length as
    # the number of motors in your follower arms.
    max_relative_target: int | None = None

    leader_arms: dict[str, MotorsBusConfig] = field(
        default_factory=lambda: {
            "main": FeetechMotorsBusConfig(
                port="/dev/ttyACM0",  <-- UPDATE HERE
                motors={
                    # name: (index, model)
                    "shoulder_pan": [1, "sts3215"],
                    "shoulder_lift": [2, "sts3215"],
                    "elbow_flex": [3, "sts3215"],
                    "wrist_flex": [4, "sts3215"],
                    "wrist_roll": [5, "sts3215"],
                    "gripper": [6, "sts3215"],
                },
            ),
        }
    )

    follower_arms: dict[str, MotorsBusConfig] = field(
        default_factory=lambda: {
            "main": FeetechMotorsBusConfig(
                port="/dev//dev/ttyACM1",  <-- UPDATE HERE
                motors={
                    # name: (index, model)
                    "shoulder_pan": [1, "sts3215"],
                    "shoulder_lift": [2, "sts3215"],
                    "elbow_flex": [3, "sts3215"],
                    "wrist_flex": [4, "sts3215"],
                    "wrist_roll": [5, "sts3215"],
                    "gripper": [6, "sts3215"],
                },
            ),
        }
    )
Dual-Arm Teleoperation. (Option)

If you want to implement dual-arm teleoperation, it means you need two Leader robotic arms and two Follower robotic arms. Therefore, you need to add the class names of the robotic arms and their corresponding port numbers in the leader_arms dick and follower_arms dick, for example:

@RobotConfig.register_subclass("so101")
@dataclass
class So101RobotConfig(ManipulatorRobotConfig):
    calibration_dir: str = ".cache/calibration/so101"
    # `max_relative_target` limits the magnitude of the relative positional target vector for safety purposes.
    # Set this to a positive scalar to have the same value for all motors, or a list that is the same length as
    # the number of motors in your follower arms.
    max_relative_target: int | None = None

    leader_arms: dict[str, MotorsBusConfig] = field(
        default_factory=lambda: {
            "left": FeetechMotorsBusConfig(
                port="/dev/ttyACM0",  <-- UPDATE HERE
                motors={
                    # name: (index, model)
                    "shoulder_pan": [1, "sts3215"],
                    "shoulder_lift": [2, "sts3215"],
                    "elbow_flex": [3, "sts3215"],
                    "wrist_flex": [4, "sts3215"],
                    "wrist_roll": [5, "sts3215"],
                    "gripper": [6, "sts3215"],
                },
            ),
            "right": FeetechMotorsBusConfig(
                port="/dev/ttyACM1",  <-- UPDATE HERE
                motors={
                    # name: (index, model)
                    "shoulder_pan": [1, "sts3215"],
                    "shoulder_lift": [2, "sts3215"],
                    "elbow_flex": [3, "sts3215"],
                    "wrist_flex": [4, "sts3215"],
                    "wrist_roll": [5, "sts3215"],
                    "gripper": [6, "sts3215"],
                },
            ),
        }
    )

    follower_arms: dict[str, MotorsBusConfig] = field(
        default_factory=lambda: {
            "left": FeetechMotorsBusConfig(
                port="/dev//dev/ttyACM2",  <-- UPDATE HERE
                motors={
                    # name: (index, model)
                    "shoulder_pan": [1, "sts3215"],
                    "shoulder_lift": [2, "sts3215"],
                    "elbow_flex": [3, "sts3215"],
                    "wrist_flex": [4, "sts3215"],
                    "wrist_roll": [5, "sts3215"],
                    "gripper": [6, "sts3215"],
                },
            ),
            "right": FeetechMotorsBusConfig(
                port="/dev//dev/ttyACM3",  <-- UPDATE HERE
                motors={
                    # name: (index, model)
                    "shoulder_pan": [1, "sts3215"],
                    "shoulder_lift": [2, "sts3215"],
                    "elbow_flex": [3, "sts3215"],
                    "wrist_flex": [4, "sts3215"],
                    "wrist_roll": [5, "sts3215"],
                    "gripper": [6, "sts3215"],
                },
            ),
        }
    )

caution

You need to correctly match the left and right names of the dual arms and ensure that each robotic arm’s serial port number on the device is correctly assigned.

In the next step, when calibrating the robotic arms, you need to calibrate all four arms individually. The command is as follows:

sudo chmod 666 /dev/ttyACM*
python lerobot/scripts/control_robot.py \
  --robot.type=so101 \
  --robot.cameras='{}' \
  --control.type=calibrate \
  --control.arms='["left_follower"]'
  #  --control.arms='["right_follower"]'
  #  --control.arms='["left_leader"]'
  #  --control.arms='["right_leader"]'

After the calibration is completed, you can view the results in the .cache/calibration/so101 directory.

`-- calibration
    `-- so101
        |-- left_follower.json
        |-- left_leader.json
        |-- right_follower.json
        `-- right_leader.json

The subsequent steps are the same as for the single-arm setup.

sudo chmod 666 /dev/ttyACM*

Manual calibration of Follower arm

python lerobot/scripts/control_robot.py \
  --robot.type=so101 \
  --robot.cameras='{}' \
  --control.type=calibrate \
  --control.arms='["main_follower"]'

Manual calibration of leader arm

python lerobot/scripts/control_robot.py \
  --robot.type=so101 \
  --robot.cameras='{}' \
  --control.type=calibrate \
  --control.arms='["main_leader"]'
Follower Middle PositionFollower Zero PositionFollower Rotated PositionFollower Rest Position
fig7fig1fig2fig3
LeaderMiddle PositionLeader Zero PositionLeader Rotated PositionLeader Rest Position
fig8fig4fig5fig6

Teleoperate​

Simple teleop Then you are ready to teleoperate your robot! Run this simple script (it won't connect and display the cameras):

python lerobot/scripts/control_robot.py \
  --robot.type=so101 \
  --robot.cameras='{}' \
  --control.type=teleoperate

Add cameras​

tip

The SO100 and SO101 codes are compatible. Users of SO100 can directly utilize SO101's parameters and code for operation.

After inserting your two USB cameras, run the following script to check the port numbers of the cameras, It is important to remember that the camera must not be connected to a USB Hub; instead, it should be plugged directly into the device. The slower speed of a USB Hub may result in the inability to read image data.

python lerobot/common/robot_devices/cameras/opencv.py \
    --images-dir outputs/images_from_opencv_cameras

The terminal will print out the following information.

Mac or X86 Ubuntu detected. Finding available camera indices through scanning all indices from 0 to 60
[...]
Camera found at index 2
Camera found at index 4
[...]
Connecting cameras
OpenCVCamera(2, fps=30.0, width=640, height=480, color_mode=rgb)
OpenCVCamera(4, fps=30.0, width=640, height=480, color_mode=rgb)
Saving images to outputs/images_from_opencv_cameras
Frame: 0000 Latency (ms): 39.52
[...]
Frame: 0046 Latency (ms): 40.07
Images have been saved to outputs/images_from_opencv_cameras

You can find the pictures taken by each camera in the outputs/images_from_opencv_cameras directory, and confirm the port index information corresponding to the cameras at different positions. Then complete the alignment of the camera parameters in the lerobot/lerobot/common/robot_devices/robots/configs.py file.

@RobotConfig.register_subclass("so101")
@dataclass
class So101RobotConfig(ManipulatorRobotConfig):
    calibration_dir: str = ".cache/calibration/so101"
    ''''''''''''''''
          .
          .
    ''''''''''''''''
    cameras: dict[str, CameraConfig] = field(
        default_factory=lambda: {
            "laptop": OpenCVCameraConfig(
                camera_index=0,             ##### UPDATE HEARE
                fps=30,
                width=640,
                height=480,
            ),
            "phone": OpenCVCameraConfig(
                camera_index=1,             ##### UPDATE HEARE
                fps=30,
                width=640,
                height=480,
            ),
        }
    )

    mock: bool = False
Add two or more additional cameras. (Option)
If you want to add more cameras, you can continue adding different camera names and `camera_index` values in the camera dictionary, as long as the USB input allows it. Please note that using a USB hub for cameras is not recommended.
@RobotConfig.register_subclass("so101")
@dataclass
class So101RobotConfig(ManipulatorRobotConfig):
    calibration_dir: str = ".cache/calibration/so101"
    ''''''''''''''''
          .
          .
    ''''''''''''''''
    cameras: dict[str, CameraConfig] = field(
        default_factory=lambda: {
            "laptop": OpenCVCameraConfig(
                camera_index=0,             ##### UPDATE HEARE
                fps=30,
                width=640,
                height=480,
            ),
            "phone": OpenCVCameraConfig(
                camera_index=1,             ##### UPDATE HEARE
                fps=30,
                width=640,
                height=480,
            ),
            "new_camera": OpenCVCameraConfig( ##### UPDATE HEARE
                camera_index=3,             ##### UPDATE HEARE
                fps=30,
                width=640,
                height=480,
            ),
        }
    )

    mock: bool = False

Then you will be able to display the cameras on your computer while you are teleoperating by running the following code. This is useful to prepare your setup before recording your first dataset.

python lerobot/scripts/control_robot.py \
  --robot.type=so101 \
  --control.type=teleoperate \
  --control.display_data=true

Record the dataset​

tip

The SO100 and SO101 codes are compatible. Users of SO100 can directly utilize SO101's parameters and code for operation.

Once you're familiar with teleoperation, you can record your first dataset with SO-10x.

If you want to use the Hugging Face hub features for uploading your dataset and you haven't previously done it, make sure you've logged in using a write-access token, which can be generated from the Hugging Face settings:

huggingface-cli login --token ${HUGGINGFACE_TOKEN} --add-to-git-credential

Store your Hugging Face repository name in a variable to run these commands:

HF_USER=$(huggingface-cli whoami | head -n 1)
echo $HF_USER

Record 2 episodes and upload your dataset to the hub:

python lerobot/scripts/control_robot.py \
  --robot.type=so101 \
  --control.type=record \
  --control.fps=30 \
  --control.single_task="Grasp a lego block and put it in the bin." \
  --control.repo_id=${HF_USER}/so101_test \
  --control.tags='["so101","tutorial"]' \
  --control.warmup_time_s=5 \
  --control.episode_time_s=30 \
  --control.reset_time_s=30 \
  --control.num_episodes=2 \
  --control.display_data=true \
  --control.push_to_hub=true

You will see a lot of lines appearing like this one:

INFO 2024-08-10 15:02:58 ol_robot.py:219 dt:33.34 (30.0hz) dtRlead: 5.06 (197.5hz) dtWfoll: 0.25 (3963.7hz) dtRfoll: 6.22 (160.7hz) dtRlaptop: 32.57 (30.7hz) dtRphone: 33.84 (29.5hz)

Parameter Explanations

  • wormup-time-s: It refers to the initialization time.
  • episode-time-s: It represents the time for collecting data each time.
  • reset-time-s: It is the preparation time between each data collection.
  • num-episodes: It indicates how many groups of data are expected to be collected.
  • push-to-hub: It determines whether to upload the data to the HuggingFace Hub.
tip

  • "If you want to save the data locally (--control.push_to_hub=false), replace --control.repo_id=${HF_USER}/so101_test with a custom local folder name, such as --control.repo_id=seeed_123/so101_test. It will then be stored in the system's home directory at ~/.cache/huggingface/lerobot."

  • If you uploaded your dataset to the hub with --control.push_to_hub=true, you can visualize your dataset online by copy pasting your repo id given by:

  • Press right arrow -> at any time during episode recording to early stop and go to resetting. Same during resetting, to early stop and to go to the next episode recording.

  • Press left arrow <- at any time during episode recording or resetting to early stop, cancel the current episode, and re-record it.

  • Press escape ESC at any time during episode recording to end the session early and go straight to video encoding and dataset uploading.

  • Note: You can resume recording by adding --control.resume=true. Also if you didn't push your dataset yet, add --control.local_files_only=true. You will need to manually delete the dataset directory if you want to start recording from scratch.

  • Once you're comfortable with data recording, you can create a larger dataset for training. A good starting task is grasping an object at different locations and placing it in a bin. We suggest recording at least 50 episodes, with 10 episodes per location. Keep the cameras fixed and maintain consistent grasping behavior throughout the recordings. Also make sure the object you are manipulating is visible on the camera's. A good rule of thumb is you should be able to do the task yourself by only looking at the camera images.

  • In the following sections, you’ll train your neural network. After achieving reliable grasping performance, you can start introducing more variations during data collection, such as additional grasp locations, different grasping techniques, and altering camera positions.

  • Avoid adding too much variation too quickly, as it may hinder your results.

  • On Linux, if the left and right arrow keys and escape key don't have any effect during data recording, make sure you've set the $DISPLAY environment variable. See pynput limitations.

Visualize the dataset​

tip

The SO100 and SO101 codes are compatible. Users of SO100 can directly utilize SO101's parameters and code for operation.

echo ${HF_USER}/so101_test

If you didn't upload with --control.push_to_hub=false, you can also visualize it locally with:

python lerobot/scripts/visualize_dataset_html.py \
  --repo-id ${HF_USER}/so101_test \
  --local-files-only 1

If you upload with --control.push_to_hub=false, you can also visualize it locally with:

python lerobot/scripts/visualize_dataset_html.py \
  --repo-id seeed_123/so101_test \
  --local-files-only 1

Here, seeed_123 is the custom repo_id name defined when collecting data.

Replay an episode​

tip

The SO100 and SO101 codes are compatible. Users of SO100 can directly utilize SO101's parameters and code for operation.

Now try to replay the first episode on your robot:

python lerobot/scripts/control_robot.py \
  --robot.type=so101 \
  --control.type=replay \
  --control.fps=30 \
  --control.repo_id=${HF_USER}/so101_test \
  --control.episode=0

Note: If you didn't push your dataset yet, add --control.local_files_only=true .

Train a policy​

tip

The SO100 and SO101 codes are compatible. Users of SO100 can directly utilize SO101's parameters and code for operation.

To train a policy to control your robot, use the python lerobot/scripts/train.py script. A few arguments are required. Here is an example command:

python lerobot/scripts/train.py \
  --dataset.repo_id=${HF_USER}/so101_test \
  --policy.type=act \
  --output_dir=outputs/train/act_so101_test \
  --job_name=act_so101_test \
  --policy.device=cuda \
  --wandb.enable=true

If you want to train on a local dataset, make sure the repo_id matches the one used during data collection.

Let's explain it:

  1. We provided the dataset as argument with --dataset.repo_id=${HF_USER}/so101_test.
  2. We provide the policy using policy.type=act, which will load the configuration from lerobot/lerobot/common/policies/act/configuration_act.py. Currently, ACT has been tested, but you can also try other policies such as diffusion, pi0, pi0fast, tdmpc, and vqbet.
  3. We provided policy.device=cuda since we are training on a Nvidia GPU, but you could use policy.device=mps to train on Apple silicon.
  4. We provided wandb.enable=true to use Weights and Biases for visualizing training plots. This is optional but if you use it, make sure you are logged in by running wandb login.

Training should take several hours. You will find checkpoints in outputs/train/act_so100_test/checkpoints.

To resume training from a checkpoint, below is an example command to resume from last checkpoint of the act_so101_test policy:

python lerobot/scripts/train.py \
  --config_path=outputs/train/act_so101_test/checkpoints/last/pretrained_model/train_config.json \
  --resume=true

Upload policy checkpoints Once training is done, upload the latest checkpoint with:

huggingface-cli upload ${HF_USER}/act_so101_test \
  outputs/train/act_so101_test/checkpoints/last/pretrained_model

Evaluate your policy​

tip

The SO100 and SO101 codes are compatible. Users of SO100 can directly utilize SO101's parameters and code for operation.

You can use the record function from lerobot/scripts/control_robot.py but with a policy checkpoint as input. For instance, run this command to record 10 evaluation episodes:

python lerobot/scripts/control_robot.py \
  --robot.type=so101 \
  --control.type=record \
  --control.fps=30 \
  --control.single_task="Grasp a lego block and put it in the bin." \
  --control.repo_id=${HF_USER}/eval_act_so101_test \
  --control.tags='["tutorial"]' \
  --control.warmup_time_s=5 \
  --control.episode_time_s=30 \
  --control.reset_time_s=30 \
  --control.num_episodes=10 \
  --control.push_to_hub=true \
  --control.policy.path=outputs/train/act_so101_test/checkpoints/last/pretrained_model

As you can see, it's almost the same command as previously used to record your training dataset. Two things changed:

  1. There is an additional --control.policy.path argument which indicates the path to your policy checkpoint with (e.g. outputs/train/eval_act_so100_test/checkpoints/last/pretrained_model). You can also use the model repository if you uploaded a model checkpoint to the hub (e.g. ${HF_USER}/act_so100_test).
  2. The name of dataset begins by eval to reflect that you are running inference (e.g. ${HF_USER}/eval_act_so100_test).

FAQ​

  • If you are following this documentation/tutorial, please git clone the recommended GitHub repository https://github.com/ZhuYaoHui1998/lerobot.git. The repository recommended in this documentation is a verified stable version; the official Lerobot repository is continuously updated to the latest version, which may cause unforeseen issues such as different dataset versions, different commands, etc.

  • If you encounter the following error, you need to check whether the robotic arm connected to the corresponding port is powered on and whether the bus servos have any loose or disconnected cables.

    ConnectionError: Read failed due to comunication eror on port /dev/ttyACM0 for group key Present_Position_Shoulder_pan_Shoulder_lift_elbow_flex_wrist_flex_wrist_roll_griper: [TxRxResult] There is no status packet!

  • If you have repaired or replaced any parts of the robotic arm, please completely delete the ~/lerobot/.cache/huggingface/calibration/so100 folder and recalibrate the robotic arm.

  • If the remote control functions normally but the remote control with Camera fails to display the image interface, you can find here

  • If you encounter libtiff issues during dataset remote operation, please update the libtiff version.

    conda install libtiff==4.5.0  #for Ubuntu 22.04 is libtiff==4.5.1

  • After executing the Lerobot Installation, the GPU version of pytorch may be automatically uninstalled, so you need to manually install torch-gpu.

  • For Jetson, please first install Pytorch and Torchvsion before executing conda install -y -c conda-forge ffmpeg, otherwise, when compiling torchvision, an ffmpeg version mismatch issue may occur.

  • If the following problem occurs, it means that your computer does not support this video codec format. You need to modify line 134 in the file lerobot/lerobot/common/datasets /video_utils.py by changing the value of vcodec: str = "libsvtav1" to libx264 or libopenh264. Different computers may require different parameters, so you can try various options. Issues 705

    [vost#0:0 @ 0x13207240] Unknown encoder 'libsvtav1' [vost#0:0 @ 0x13207240] Error selecting an encoder Error opening output file /home/han/.cache/huggingface/lerobot/lyhhan/so100_test/videos/chunk-000/observation.images.laptop/episode_000000.mp4. Error opening output files: Encoder not found

  • Important!!! If during execution the servo's cable becomes loose, please restore the servo to its initial position and then reconnect the servo cable. You can also individually calibrate a servo using the Servo Initialization Command, ensuring that only one cable is connected between the servo and the driver board during individual calibration. If you encounter

    Auto-correct calibration of motor 'wrist roll' by shifting value by 1 full turns, from '-270 < -312.451171875 < 270degrees' to'-270<-312.451171875 < 270 degrees'.

    or other errors during the robotic arm calibration process related to angles and exceeding limit values, this method is still applicable.

  • Training 50 sets of ACT data on an 8G 3060 laptop takes approximately 6 hours, while on a 4090 or A100 computer, training 50 sets of data takes about 2–3 hours.

  • During data collection, ensure that the camera's position, angle, and environmental lighting remain stable, and minimize capturing excessive unstable backgrounds and pedestrians; otherwise, significant environmental changes during deployment may cause the robotic arm to fail to grasp properly.

  • Ensure that the num-episodes parameter in the data collection command is set to collect sufficient data, and do not manually pause midway. This is because the mean and variance of the data are calculated only after data collection is complete, which is necessary for training.

  • If the program prompts that it cannot read the USB camera image data, please ensure that the USB camera is not connected to a hub. The USB camera must be directly connected to the device to ensure a fast image transmission rate.

tip

If you encounter software issues or environment dependency problems that cannot be resolved, in addition to checking the FAQ section at the end of this tutorial, please promptly report the issue to the LeRobot platform or the LeRobot Discord channel.

Citation​

中文文档

TheRobotStudio Project: SO-ARM10x

Huggingface Project: Lerobot

Dnsty: Jetson Containers

Jetson AI Lab

Diffusion Policy

ACT or ALOHA

TDMPC

VQ-BeT

Tech Support & Product Discussion​

Thank you for choosing our products! We are here to provide you with different support to ensure that your experience with our products is as smooth as possible. We offer several communication channels to cater to different preferences and needs.

Loading Comments...