Paper 2603.28032
CARLA-Air: Fly Drones Inside a CARLA World -- A Unified Infrastructure for Air-Ground Embodied Intelligence
- Published
- Mar 2026
- Research lab
- Independent
- Citations
- 2
- GitHub
- 1.1K stars
01 In brief
Summary
CARLA-Air is an open-source simulation platform that integrates CARLA and AirSim within a single Unreal Engine process, enabling joint simulation of aerial and ground agents for embodied intelligence research.
It resolves the UE4 single-game-mode conflict by inheriting CARLA's ground subsystems and composing AirSim's flight actor as a world entity, preserving both native Python APIs and ROS 2 interfaces.
The platform supports up to 18 synchronized sensor modalities, photorealistic urban and natural environments with traffic and pedestrians, and an extensible asset pipeline.
Performance benchmarks show ~20 FPS under joint workloads, with a 3-hour stability run (357 reset cycles, zero crashes).
Five representative workflows validate air-ground cooperation, embodied navigation, multi-modal dataset collection, cross-view perception, and RL training.
CARLA-Air provides a sustainable evolution path for AirSim's archived flight stack and is released with prebuilt binaries and source code at https://github.com/louiszengCN/CarlaAir.
Limitations include moderate actor density and map switching requiring process restart, with future work on physics-state synchronization and ROS 2 bridging.
The platform fills the gap left by domain-segregated simulators, offering a unified foundation for low-altitude robotics and cross-domain autonomy research.
- CARLA-Air integrates CARLA and AirSim in a single UE4 process, preserving both native APIs and ROS 2 interfaces.
- It resolves the UE4 single-game-mode conflict via inheritance and composition, with minimal upstream modifications.
- Supports up to 18 synchronized sensor modalities across aerial and ground platforms per tick.
- Performance: ~20 FPS under joint workloads, 3-hour stability with zero crashes over 357 reset cycles.
- Validated on five workflows: precision landing…
02 From the paper
Abstract
The convergence of low-altitude economies, embodied intelligence, and air-ground cooperative systems creates growing demand for simulation infrastructure capable of jointly modeling aerial and ground agents within a single physically coherent environment. Existing open-source platforms remain domain-segregated: driving simulators lack aerial dynamics, while multirotor simulators lack realistic ground scenes. Bridge-based co-simulation introduces synchronization overhead and cannot guarantee strict spatial-temporal consistency. We present CARLA-Air, an open-source infrastructure that unifies high-fidelity urban driving and physics-accurate multirotor flight within a single Unreal Engine process. The platform preserves both CARLA and AirSim native Python APIs and ROS 2 interfaces, enabling zero-modification code reuse. Within a shared physics tick and rendering pipeline, CARLA-Air delivers photorealistic environments with rule-compliant traffic, socially-aware pedestrians, and aerodynamically consistent UAV dynamics, synchronously capturing up to 18 sensor modalities across all platforms at each tick. The platform supports representative air-ground embodied intelligence workloads spanning cooperation, embodied navigation and vision-language action, multi-modal perception and dataset construction, and reinforcement-learning-based policy training. An extensible asset pipeline allows integration of custom robot platforms into the shared world. By inheriting AirSim's aerial capabilities -- whose upstream development has been archived -- CARLA-Air ensures this widely adopted flight stack continues to evolve within a modern infrastructure. Released with prebuilt binaries and full source: https://github.com/louiszengCN/CarlaAir