Getting Started
Learn how to set up and run Virtual Community for embodied social and physical intelligence research
What is Virtual Community?
Virtual Community is an open-world platform that simulates human-robot coexistence in shared communities, featuring a physics-based multi-agent simulator and real-world 3D scenes to study embodied social and physical intelligence at scale.
Installation
Step 1: Clone the Repository
git clone https://github.com/UMass-Embodied-AGI/Virtual-Community
cd Virtual-Community
Step 2: Install Dependencies
# We're using python 3.11, cuda 11.7 and ubuntu 24.04. If you are using a different version, please modify the env.yaml file accordingly.
conda env create -f env.yaml
# Install Genesis from the source:
cd Genesis
pip install -e .
# (Optional) To run the example tour agent, install volume grid lib:
cd agents/sg
./setup.sh
Step 3: Assets Preparation
Download the assets from the Google Drive and organize them under Genesis/genesis/assets/ViCo
.
Assets include:
scene
folder: contains the generated outdoor 3D scenesrobots
folder: contains the robot modelsobjects
folder: contains the object modelsavatars
folder: contains the human avatars’ skin and motioncars
folder: contains the vehicle models including car, bus, bike, etc
If you want to use indoor scenes from GRUTopia, please follow their instructions here to download the commercial_scenes.zip, and unzip it under Genesis/genesis/assets/ViCo/
.
After this step, the directory structure should be:
Genesis/genesis/assets/ViCo/
├── scene/
├── robots/
├── objects/
├── avatars/
├── cars/
└── commercial_scenes/ (optional)
Run the Simulation
Run the following script to test the simulation, where 15 agents will tour around the New York Scene.
./scripts/run_tour_agent.sh
Tips: Loading all indoor scenes and objects may introduce a large memory cost. If you encounter memory issues, consider adding --no_load_indoor_scene
flag to the script to disable indoor scenes.
TroubleShooting
libstdcxx. so.6: version GLIBCXX 3.4.32’not found required by sg/builder/builtin/libbuider.so
To resolve this issue, you may need to update your GCC runtime. You can do this by running:bash conda install -c conda-forge libstdcxx-ng
- Only pygltflib-1.16.0 works for now, the latest version pygltflib-1.16.1 will slower the scene loading.