AirSim and simulator
These are complementary tools: AirSim provides physics-based vehicle dynamics and sensor simulation in a game engine, while LGSVL Simulator focuses on ROS/ROS2 integration and multi-robot coordination, so teams often use AirSim for low-level vehicle behavior and LGSVL for high-level autonomous stack testing.
About AirSim
microsoft/AirSim
Open source simulator for autonomous vehicles built on Unreal Engine / Unity, from Microsoft AI & Research
Supports software-in-the-loop simulation with PX4 and ArduPilot flight controllers, plus hardware-in-the-loop with PX4 for realistic vehicle dynamics. Provides language-agnostic RPC APIs (C++, Python, C#, Java) for programmatic vehicle control, sensor data retrieval, and training data generation, enabling code to transfer from simulation to physical vehicles. Includes specialized Computer Vision mode for camera-only workflows and weather simulation without vehicle physics.
About simulator
lgsvl/simulator
A ROS/ROS2 Multi-robot Simulator for Autonomous Vehicles
**Technical Summary:** Built on HDRP Unity, the simulator provides high-fidelity 3D environments with native ROS/ROS2 support for testing autonomous vehicle stacks end-to-end. It features direct integration with Apollo and Autoware.auto platforms, procedural HD map generation, and multi-robot orchestration without requiring custom middleware. The architecture uses a web-based cloud UI (WISE) for asset management and simulation configuration, though the core simulator engine can run locally as a standalone executable on Windows or Linux.
Related comparisons
Scores updated daily from GitHub, PyPI, and npm data. How scores work