carla

Why we included this project

Few tools have shaped autonomous driving research as much as CARLA, an open simulator that earns its place here for what it makes possible rather than for any single feature. Teams use it to run a full virtual city with urban layouts, buildings, and traffic, then attach the sensor configuration their work needs, so perception and control models can be trained and validated against realistic camera and LiDAR feeds without ever touching real hardware. Its Python API handles scenario scripting, weather, and traffic control, which is why CARLA shows up so often in reinforcement-learning and imitation-learning papers. It also ships open digital assets and a blueprint library, saving you from building a city from scratch. If your work touches perception, planning, or control for self-driving systems, this is likely the environment you end up testing in.