GitHub repository
An open-source simulator built on Unreal Engine for developing, training, and validating autonomous driving systems.
What is GitHub repository?
CARLA is an open-source simulator specifically designed for autonomous driving research. It provides a realistic urban environment where developers and researchers can develop, train, and validate autonomous driving systems. The platform includes open digital assets and supports flexible sensor configurations and environmental conditions.
Researchers, engineers, and students working on autonomous vehicle perception, planning, and control algorithms, as well as those in robotics and artificial intelligence fields requiring high-fidelity simulation.
Developers choose CARLA for its high-fidelity, open-source simulation built on Unreal Engine, extensive Python API for control and automation, and its dedicated focus on the complete pipeline of autonomous driving research, from training to validation.
Overview
Open-source simulator for autonomous driving research.
Use Cases
Best For
- Training and validating machine learning models for autonomous driving
- Testing perception algorithms with customizable sensor setups (cameras, LIDAR)
- Developing and benchmarking autonomous driving stacks in a safe, simulated environment
- Creating complex traffic scenarios and edge cases for system robustness testing
- Integrating autonomous systems with ROS via the provided ROS bridge
- Academic and industrial research in self-driving car technology
Not Ideal For
- Projects with limited hardware budgets or older systems, as CARLA requires high-end GPUs (e.g., RTX 3070+) and 32GB+ RAM for smooth operation.
- Real-time game development or interactive applications, since CARLA is optimized for simulation fidelity over gaming performance and lacks built-in game mechanics.
- Simple 2D robotics or lightweight simulations where urban environments are unnecessary, as CARLA's complexity adds overhead for basic tasks.
- Teams needing quick, out-of-the-box deployment, due to the multi-step build process involving Unreal Engine compilation and system dependencies.
Pros & Cons
Pros
Provides open digital assets like buildings and vehicles for authentic driving scenarios, enabling high-fidelity testing as described in the key features.
Supports customizable sensor suites including cameras and LIDAR, mimicking real-world perception systems for algorithm validation.
Offers a full Python API for programmatic control of simulations, spawning actors, and data retrieval, with extensive documentation.
Integrates with tools like Scenario Runner and ROS bridge, facilitating benchmarks and validation in autonomous driving research.
Cons
Demands NVIDIA RTX 3070 or better and 32GB+ RAM, limiting accessibility for users without high-end systems, as specified in the recommended system.
Installation involves cloning, setting up Unreal Engine with GitHub credentials, and multiple cmake steps, which can be error-prone and time-consuming.
Parallel development for UE5 and UE4 branches causes confusion and potential incompatibility, with the README warning of significant differences.
Frequently Asked Questions
Related Projects
GitHub repositoryOpen source simulator for autonomous vehicles built on Unreal Engine / Unity, from Microsoft AI & Research
WebotsWebots Robot Simulator
DrakeModel-based design and verification for robotics.
self-driving-car-simA self-driving car simulator built with Unity
Found a gem we're missing?
Open-Awesome is built by the community, for the community. Submit a project, suggest an awesome list, or help improve the catalog on GitHub.