GitHub30/mujoco_mpc

MuJoCo MPC

MuJoCo MPC (MJPC) is an interactive application and software framework for real-time predictive control with MuJoCo, developed by DeepMind.

MJPC allows the user to easily author and solve complex robotics tasks, and currently supports three shooting-based planners: derivative-based iLQG and Gradient Descent, and a simple yet very competitive derivative-free method called Predictive Sampling.

• Installation
• Getting Started
• Predictive Control
• Contributing
• Known Issues
• Citation
• Acknowledgments
• License and Disclaimer

Installation

You will need CMake and a working C++20 compiler to build MJPC. We recommend using VSCode and 2 of its extensions (CMake Tools and C/C++) to simplify the build process.

1. Clone the repository: git clone https://github.com/deepmind/mujoco_mpc.git
2. Configure the project with CMake (a pop-up should appear in VSCode)
3. Build and run the mjpc target. This will open and run the graphical user interface.

Getting Started

For a video overview of MJPC, click below.

For a detailed dive of the graphical user interface, see the MJPC GUI documentation.

Predictive Control

See the Predictive Control documentation for more information.

Contributing

See the Contributing documentation for more information.

Known Issues

MJPC is not production-quality software, it is a research prototype. There are likely to be missing features and outright bugs. If you find any, please report them in the issue tracker. Below we list some known issues, including items that we are actively working on.

• We have not tested MJPC on Windows, but there should be no issues in principle.
• Task specification, in particular the setting of norms and their parameters in XML, is a bit clunky. We are still iterating on the design.
• The Gradient Descent search step is proportional to the scale of the cost function and requires per-task tuning in order to work well. This is not a bug but a property of vanilla gradient descent. It might be possible to ameliorate this with some sort of gradient normalisation, but we have not investigated this thoroughly.
• There is a subtle issue with iLQG that we have not yet been able to resolve. It manifests as jittery behaviour and increasing cost-to-go after only a single simulation step (right arrow key on the keyboard, in pause mode). We are currently investigating it and hope to resolve it in the near future.

Citation

If you use MJPC in your work, please cite our accompanying preprint:

@article{howell2022,
  author = {Howell, Taylor and Gileadi, Nimrod and Tunyasuvunakool, Saran and Zakka, Kevin and Erez, Tom and Tassa, Yuval},
  title = {{Predictive Sampling: Real-time Behaviour Synthesis with MuJoCo}},
  url = {},
  year = {2022},
}

Acknowledgments

The main effort required to make this repository publicly available was undertaken by Taylor Howell and the DeepMind Robotics Simulation team.

License and Disclaimer

All other content is Copyright 2022 DeepMind Technologies Limited and licensed under the Apache License, Version 2.0. A copy of this license is provided in the top-level LICENSE file in this repository. You can also obtain it from https://www.apache.org/licenses/LICENSE-2.0.

This is not an officially supported Google product.