Introducing MuJoCo.jl: Julia bindings for DeepMind’s MuJoCo physics engine!
MuJoCo.jl contains a Julia wrapper for DeepMind’s general-purpose physics engine, MuJoCo. It allows users to simulate robotic systems, their controllers, and their interactions with surrounding environments, all in native Julia.
This wrapper aims to provide a simple, performant interface to the MuJoCo C library, allowing direct manipulation of the data used by the C engine for optimal performance.
The package also includes a fully-interactive visualiser, allowing the user to observe the response of MuJoCo models under passive or controlled dynamics, or to inspect saved motions as a trajectory to be replayed later.
Much of this project builds upon the work from Lyceum and their package Lyceum/MuJoCo.jl, along with their visualisation project Lyceum/LyceumMuJoCoViz.jl. We would like to thank the authors of these packages for their amazing work in originally bringing MuJoCo to Julia.
This work is joint effort between myself and @nic-barbara.
Check out the project on GitHub and the documentation to get started!