🦊 Robo-DM

🦊 Robo-DM : An Efficient and Scalable Data Collection and Management Framework For Robotics Learning. Support Open-X-Embodiment, 🤗HuggingFace.

🦊 Robo-DM (Former Name: fog_x) considers both speed 🚀 and memory efficiency 📈 with active metadata and lazily-loaded trajectory data. It supports flexible and distributed dataset partitioning. It provides native support to cloud storage.

Design Doc | Dataset Visualization

Note to ICRA Reviewers

We are actively developing the framework. See commit a35a6 for the version we developed.

Install

git clone https://github.com/BerkeleyAutomation/fog_x.git
cd fog_x
pip install -e .

Usage

import fog_x

path = "/tmp/output.vla"

# 🦊 Data collection: 
# create a new trajectory
traj = fog_x.Trajectory(
    path = path
)

traj.add(feature = "arm_view", value = "image1.jpg")
# Automatically time-aligns and saves the trajectory
traj.close()

# load it 
fog_x.Trajectory(
    path = path
)

Examples

• Data Collection and Loading
• Convert From Open_X
• Convert From H5
• Running Benchmarks

Development

Read the CONTRIBUTING.md file.