First steps - CARLA Simulator
The CARLA simulator is a comprehensive solution for producing synthetic training data for applications in autonomous driving (AD) and also other robotics applications. CARLA simulates a highly realistic environment emulating real world towns, cities and highways and the vehicles and other objects that occupy these driving spaces. The CARLA simulator is further useful as an evaluation and testing environment. You can deploy the AD agents you have trained within the simulation to test and evaluate their performance and safety, all within the security of a simulated environment, with no risk to hardware or other road users. In this tutorial, we will cover a standard workflow in CARLA, from launching the server and connecting the client, through to adding vehicles, sensors and generating training data to use for machine learning. This tutorial is meant to be light on details and go as efficiently as possible through the key steps in using CARLA to produce machine learning training data. Fo
First steps - CARLA Simulator Getting started First steps Edit on GitHub First steps with CARLA The CARLA simulator is a comprehensive solution for producing synthetic training data for applications in autonomous driving (AD) and also other robotics applications. CARLA simulates a highly realistic environment emulating real world towns, cities and highways and the vehicles and other objects that occupy these driving spaces. The CARLA simulator is further useful as an evaluation and testing environment. You can deploy the AD agents you have trained within the simulation to test and evaluate the
Explore this link on the map →related reading
- The Waymo World Model: A New Frontier For Autonomous Driving Simulationwaymo.com
- Isaac Sim - Robotics Simulation and Synthetic Data Generation | NVIDIA Developerdeveloper.nvidia.com
- Simulators — LessWronglesswrong.com
- Building reliable sim driving agents by scaling self-playarxiv.org
- Seoul World Model: Grounding World Simulation Models in a Real-World Metropolisseoul-world-model.github.io
- State of Robot Learning, December 2025vedder.io
- Andrej Karpathy — AGI is still a decade awaydwarkesh.com
- A Functional Taxonomy of World Models - Dr. Fei-Fei Lidrfeifei.substack.com
- Human-compatible driving partners through data-regularized self-play reinforcement learningarxiv.org
- [2603.08546] Interactive World Simulator for Robot Policy Training and Evaluationarxiv.org
- A VLA with Open-World Generalizationpi.website
- State of Vision-Language-Action (VLA) Research at ICLR 2026 – Moritz Reussmbreuss.github.io