louiszengCN/CarlaAir

Health v2 · maintenance only

Flags: young no_license

Adoption not part of the score

1076 stars · 92 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

CarlaAir is an open-source simulation infrastructure that combines CARLA's high-fidelity urban driving environments with physics-accurate multirotor drone flight inside a single Unreal Engine process. It provides a unified foundation for air-ground embodied intelligence research, with pre-built executables for Ubuntu and Windows.

Use cases

• simulate drones flying in a CARLA urban environment
• research air-ground embodied intelligence
• train UAV navigation policies in simulation
• combine autonomous driving and drone flight in one simulator
• run drone physics simulation with Unreal Engine visuals
• test multirotor control algorithms in city scenes

When to choose

• you need both driving and drone simulation in a single Unreal Engine process
• you want pre-built CARLA+AirSim binaries without compiling from source
• you research embodied AI spanning air and ground vehicles
• you already use CARLA 0.9.16 or AirSim and need multirotor support

When to avoid

• you need a permissively licensed simulator for commercial use (license is custom non-commercial)
• you only need ground-vehicle driving simulation without drones
• you require platforms other than Ubuntu 20.04/22.04 or Windows 11
• you need a lightweight simulator without GPU/Unreal Engine dependencies

Facets

framework · maturity active

simulation robotics machine-learning computer-vision developer-tools robotics simulation autonomous-vehicles artificial-intelligence reinforcement-learning windows python cpp carla drone uav airsim unreal-engine ros2 embodied-ai multirotor air-ground-simulation non-commercial-license linux docker

Member repositories

For agents

markdown · JSON · MCP: product_card(name="louiszengCN/CarlaAir")