Skip to content

UnrealRoboticsLab

UnrealRoboticsLab (URLab) embeds the MuJoCo physics engine directly inside Unreal Engine 5. You author and simulate robots with MuJoCo's accurate contact dynamics, and render them with Unreal's lighting, materials, and cameras.

URLab simulating a robot in Unreal Engine

What you can do

  • Bring MuJoCo models into Unreal. Drag an MJCF .xml file into the Content Browser and URLab builds a ready-to-simulate Blueprint, or turn any static mesh into a physics body with one component.
  • Simulate with MuJoCo, render with Unreal. Physics runs on MuJoCo on its own thread; Unreal handles rendering, so you get accurate contacts and photorealistic output at the same time.
  • Control robots however you like. Drive actuators from Blueprints, from the in-editor dashboard, or from Python and ROS 2 over the network.
  • Capture data. Render RGB, depth, and segmentation from robot cameras, and record and replay full simulation episodes.

Who it is for

URLab pairs MuJoCo physics with what Unreal does well: photorealistic rendering, scene generation, NPCs, and the rest of the editor toolset. It is built for evaluation, data generation, experimentation, and simulation.

You can train in URLab, but it runs one simulation at a time rather than parallelising across thousands of environments on the GPU, so it is not the fastest route for large-scale reinforcement learning. For that, train with a dedicated massively parallel MuJoCo framework such as mjlab or mujoco_warp, then evaluate the resulting policies directly in URLab. Because both use MuJoCo, the Python bridge lets you move between training there and evaluating, recording, and visualising here.

Get started

  • Install URLab sets up the plugin and its dependencies on Windows or Linux.
  • Quickstart imports a robot and runs your first simulation in a few minutes.
  • Python & External Control drives the simulation from Python, for scripted control and running policies.
  • Architecture explains how the pieces fit together under the hood.

Project

URLab is open source under the Apache 2.0 license. See the Roadmap for where it is headed. If you use URLab in research, please cite the ICRA 2026 paper.