A software framework integrating various imitation learning methods and benchmark environments for robotic manipulation.
Provides easy-to-use baselines for policy training, evaluation, and deployment.
RoboManipBaselines_VideoVer200.mp4
RoboManipBaselines_VideoVer100.mp4
Start collecting data in the MuJoCo simulation, train your model, and rollout the ACT policy in just a few steps!
📄 See the Quick Start Guide.
Follow our step-by-step Installation Guide to get set up smoothly.
We provide several powerful policy architectures for manipulation tasks:
- 🔹 MLP – Simple feedforward policy
- 🔹 SARNN – Sequence-aware RNN policy
- 🔹 ACT – Transformer-based imitation policy
- 🔹 DiffusionPolicy – Diffusion-based behavior cloning
- 📂 Dataset List: Pre-collected expert demonstration datasets
- 🧠 Learned Parameters: Trained model checkpoints and configs
Use your own teleop interface to collect expert data.
See Teleop Tools for more info.
Explore diverse manipulation environments:
- 📚 Environment Catalog – Overview of all task environments
- 🔧 Env Setup – Installation guides per environment
Check out Misc Scripts for standalone tools and utilities.
See benchmarked performance across environments and policies:
📈 Evaluation Results
We welcome contributions!
Check out the Contribution Guide to get started.
This repository is licensed under the BSD 2-Clause License, unless otherwise stated.
Please check individual files or directories (especially third_party and assets) for specific license terms.
If you use RoboManipBaselines in your work, please cite us:
@software{RoboManipBaselines_GitHub2024,
author = {Murooka, Masaki and Motoda, Tomohiro and Nakajo, Ryoichi},
title = {{RoboManipBaselines}},
url = {https://github.com/isri-aist/RoboManipBaselines},
version = {1.0.0},
year = {2024},
month = dec,
}