mcity-digital-twin

Digital twin of the University of Michigan’s Mcity Test Facility to support research and development of transportation technology, both academic and commercial.

https://github.com/mcity/mcity-digital-twin

Science Score: 26.0%

This score indicates how likely this project is to be science-related based on various indicators:

  • CITATION.cff file
  • codemeta.json file
    Found codemeta.json file
  • .zenodo.json file
    Found .zenodo.json file
  • DOI references
  • Academic publication links
  • Academic email domains
  • Institutional organization owner
  • JOSS paper metadata
  • Scientific vocabulary similarity
    Low similarity (13.4%) to scientific vocabulary

Keywords

autonomous-driving carla-simulator connected-vehicle digital-twin mcity open-source research university-of-michigan unreal-engine
Last synced: 11 months ago · JSON representation

Repository

Digital twin of the University of Michigan’s Mcity Test Facility to support research and development of transportation technology, both academic and commercial.

Basic Info
  • Host: GitHub
  • Owner: mcity
  • License: mit
  • Language: Python
  • Default Branch: main
  • Homepage: https://mcity.umich.edu/
  • Size: 1.57 GB
Statistics
  • Stars: 33
  • Watchers: 4
  • Forks: 4
  • Open Issues: 1
  • Releases: 1
Topics
autonomous-driving carla-simulator connected-vehicle digital-twin mcity open-source research university-of-michigan unreal-engine
Created almost 2 years ago · Last pushed about 1 year ago
Metadata Files
Readme License Citation

README.md

Mcity Digital Twin

Version 1.6.1

image

Flythrough

https://github.com/user-attachments/assets/015a96be-20f2-46a6-8d56-70d5f74790f4

This digital twin was developed for the University of Michigans Mcity Test Facility to support research and development of transportation technology, both academic and commercial. It may be used freely, under the MIT License. To leverage the full set of digital twin capabilities, including live integration with the Mcity Test Facility, please see the Mcity 2.0 initiative, funded by the National Science Foundation.

https://mcity.umich.edu/

SYSTEM REQUIREMENTS

OMNIVERSE

  • GeForce RTX 3070+
  • 16GB RAM
  • An Omniverse Blueprint, application or service. #### CARLA
  • Unreal 4.26
  • CARLA 0.9.12+
  • Ubuntu 20+ or Windows 10+
  • Git LFS
  • 6GB+ GPU

In order to download the files properly, you will need to have Git LFS installed. Once installed, you can run the installation commands to clone or pull the available content.

OMNIVERSE DIGITAL TWIN INSTALLATION INSTRUCTIONS

Copy the Omniverse/Collected_Mcity_NSR_v4_1_6 folder to your desired location. Open McityMap_Main.usdc stage in any Omniverse Blueprint, application or service and enjoy!

CARLA DIGITAL INSTALLATION INSTRUCTIONS

Download the content for the first time. git lfs clone https://github.com/mcity/mcity-digital-twin.git

Pull in new content. git lfs pull

Source Version of CARLA

Copy the McityMap folder inside the CARLA/source_version folder and paste it in this location of your source version of CARLA Unreal/CarlaUE4/Content/.

Launch your source version of CARLA and navigate to Content/McityMap. Open up the McityMap_Main level and enjoy! image

Packaged Version of CARLA

Create an McityMap folder inside your packaged version of CARLA at this location CarlaUE4/Content. Then copy the contents inside of the CARLA/packaged_version/windows or CARLA/packaged_version/linux directory and paste it here: CarlaUE4/Content/McityMap.

Paste the following lines at the bottom of the CarlaUE4/Config/DefaultGame.ini file under the [/Script/UnrealEd.ProjectPackagingSettings] section.

+MapsToCook=(FilePath="/Game/Carla/Maps/McityMap") +DirectoriesToAlwaysStageAsUFS=(Path="McityMap/OpenDrive") +DirectoriesToAlwaysStageAsUFS=(Path="McityMap/Nav")

Launch your packaged version of CARLA, run the CARLA/scripts/load_mcity_digital_twin.py script inside your CARLA python environment and enjoy!

KNOWN ISSUES

Omniverse Digital Twin

  • When using the Asset Validator (v0.16.2), you will get the following error for the GroundTruthCapabilityChecker:
    • "SemanticsAPI schema is applied but QCode is not authored." This can be ignored.

MCITY TEST FACILITY DATASETS

Below is an example dataset from the Mcity Test Facility. It includes 6 Triton 2.3 MP Sony IMX392 CMOS on-vehicle camera sensors and 1 RoboSense P6 Solid State Lidar that publishes 6 lidar topics. If you would like access to the full dataset, please email mcity-engineering@umich.edu.

Example Dataset: https://drive.google.com/drive/folders/1N4zK63B4GQT-LRQyEyrUKwLh4ULf2Uu?usp=drivelink

image

PACKAGE CONTENTS

This package DOES NOT include any CARLA or Unreal code or executables, only content that can be loaded into an existing CarlaUE4 build.

  • McityMap16_0.uasset
    • Dummy material used purely to quickly show the current release version.
  • McityMap_Main.umap
    • Main map that contains all props, materials, textures, blueprints, and sublevels.
  • Blueprints
    • Contains custom traffic light blueprints based on CARLA's original blueprints, but use custom meshes.
  • OpenDrive
    • McityMap_Main.xodr contains the current OpenDrive file provided and maintained by Mcity.
    • xodr.text contains the current version of the OpenDrive (.xodr) file.
  • Static
    • Contains all meshes, materials, and textures, both dynamic and static.
    • Structure is based on CARLA documentation for use with their semantic segmentation sensors.
  • Sublevels
    • CarlaEnvLighting (dynamic)
    • CarlaSigns (dynamic)
    • CarlaTrafficLights (dynamic)
    • McityMap_OpenDrive (dynamic)
    • McityMap_StaticProps (static)
    • McityMap_StaticTrafficLights (static)
    • McityMap_StreetLights (static, hidden)
    • McityMap_Terrain (static)

CONTENT SOURCE NOTES

All 3D static meshes, materials, and textures (with a few exceptions listed below) were created and developed by Quantum Signal AI LLC (QSAI).

https://quantumsignalai.com/

  • All foliage and some dynamic traffic sign meshes and textures come from the CARLA asset library.
  • All other signage has been created using the MUTCD manual as reference (see credits below under LEGAL).
  • Custom traffic light blueprints are based on CARLA's original blueprints, but adapted to use custom QSAI meshes and selected textures.
  • TPerlinNoiseM.uasset and TMacroVariation.uasset come from Unreal's Starter Content.
  • Original aerial imagery obtained and provided by Mcity and used with permission.

LEGAL

  • Unreal Engine. Unreal is a trademark or registered trademark of Epic Games, Inc. in the United States of America and elsewhere. Unreal Engine, Copyright 1998 2025, Epic Games, Inc. All rights reserved.

    https://www.unrealengine.com

  • CARLA specific code is released under MIT license and specific assets are distributed under CC-BY License.

    https://carla.org/

  • University of Michigan logos and markings are trademarked by the University of Michigan and were used with permission.

    https://brand.umich.edu/trademarks-permissions/

  • ASAM OpenDrive Standard is 2025 by ASAM e.V. All Rights Reserved.

    https://www.asam.net/standards/detail/opendrive/

  • Manual on Uniform Traffic Control Devices (MUTCD), published by the United States Department of Transportation - Federal Highway Administration.

    https://mutcd.fhwa.dot.gov/kno_11th_Edition.htm. 
    

Acknowledgments

Owner

  • Name: Mcity
  • Login: mcity
  • Kind: organization
  • Location: Ann Arbor, Michigan

Repository for software used to support the Mcity autonomous/connected vehicle test facility and member system access.

GitHub Events

Total
  • Create event: 8
  • Release event: 1
  • Issues event: 4
  • Watch event: 50
  • Delete event: 1
  • Issue comment event: 3
  • Push event: 17
  • Public event: 1
  • Pull request review event: 1
  • Pull request event: 8
  • Fork event: 6
Last Year
  • Create event: 8
  • Release event: 1
  • Issues event: 4
  • Watch event: 50
  • Delete event: 1
  • Issue comment event: 3
  • Push event: 17
  • Public event: 1
  • Pull request review event: 1
  • Pull request event: 8
  • Fork event: 6