[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"project-96238":3},{"id":4,"name":5,"fullName":6,"owner":7,"repo":5,"description":8,"homepage":9,"htmlUrl":10,"language":11,"languages":9,"totalLinesOfCode":9,"stars":12,"forks":13,"watchers":14,"openIssues":15,"contributorsCount":9,"subscribersCount":16,"size":16,"stars1d":16,"stars7d":16,"stars30d":16,"stars90d":16,"forks30d":16,"starsTrendScore":16,"compositeScore":17,"rankGlobal":9,"rankLanguage":9,"license":9,"archived":18,"fork":18,"defaultBranch":19,"hasWiki":18,"hasPages":18,"topics":9,"createdAt":9,"pushedAt":9,"updatedAt":20,"readmeContent":21,"aiSummary":22,"trendingCount":16,"starSnapshotCount":16,"syncStatus":23,"lastSyncTime":24,"discoverSource":25},96238,"ProjectAirSim","iamaisim\u002FProjectAirSim","iamaisim","Project AirSim is Microsoft's evolution of AirSim, an advanced simulation platform for building, training, and testing autonomous systems in high-fidelity virtual environments",null,"https:\u002F\u002Fgithub.com\u002Fiamaisim\u002FProjectAirSim","C++",846,156,36,37,0,10.59,false,"main","2026-09-21 02:04:31","# Project AirSim\n\n\u003Cdiv align=\"center\">\n\n[![Build and Deploy Sphinx Docs](https:\u002F\u002Fgithub.com\u002Fiamaisim\u002FProjectAirSim\u002Factions\u002Fworkflows\u002Fsphinx-docs.yml\u002Fbadge.svg?branch=main)](https:\u002F\u002Fgithub.com\u002Fiamaisim\u002FProjectAirSim\u002Factions\u002Fworkflows\u002Fsphinx-docs.yml)\n[![Linux SimLibs Release Tests](https:\u002F\u002Fgithub.com\u002Fiamaisim\u002FProjectAirSim\u002Factions\u002Fworkflows\u002Ftest_linux_simlibs_release.yml\u002Fbadge.svg?branch=main)](https:\u002F\u002Fgithub.com\u002Fiamaisim\u002FProjectAirSim\u002Factions\u002Fworkflows\u002Ftest_linux_simlibs_release.yml)\n[![Windows Build and Tests](https:\u002F\u002Fgithub.com\u002Fiamaisim\u002FProjectAirSim\u002Factions\u002Fworkflows\u002Ftest_windows.yml\u002Fbadge.svg?branch=main)](https:\u002F\u002Fgithub.com\u002Fiamaisim\u002FProjectAirSim\u002Factions\u002Fworkflows\u002Ftest_windows.yml)\n[![C++ Client and ROS 2 CI](https:\u002F\u002Fgithub.com\u002Fiamaisim\u002FProjectAirSim\u002Factions\u002Fworkflows\u002Ftest_cpp_client.yml\u002Fbadge.svg?branch=main)](https:\u002F\u002Fgithub.com\u002Fiamaisim\u002FProjectAirSim\u002Factions\u002Fworkflows\u002Ftest_cpp_client.yml)\n\n[![Latest release](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fv\u002Frelease\u002Fiamaisim\u002FProjectAirSim?label=release)](https:\u002F\u002Fgithub.com\u002Fiamaisim\u002FProjectAirSim\u002Freleases\u002Flatest)\n[![Ubuntu](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FUbuntu-22.04-E95420?logo=ubuntu&logoColor=white)](docs\u002Fdevelopment\u002Fdev_setup_linux.md)\n[![Unreal Engine](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FUnreal%20Engine-5.2%20%7C%205.7-0E1128?logo=unrealengine&logoColor=white)](docs\u002Fdevelopment\u002Fuse_source.md)\n[![ROS 2](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FROS%202-Humble-22314E?logo=ros&logoColor=white)](docs\u002Fros\u002Fros2.md)\n[![Python](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPython-%3E3.10-3776AB?logo=python&logoColor=white)](docs\u002Fclient_setup.md)\n[![License: MIT](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FLicense-MIT-2EA043.svg)](docs\u002Flicense.md)\n[![IAMAI Simulations](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FIAMAI-Simulations-1F6FEB)](https:\u002F\u002Fiamaisim.com\u002F)\n[![Discord](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FDiscord-Join%20the%20community-5865F2?logo=discord&logoColor=white)](https:\u002F\u002Fdiscord.gg\u002FXprQ2w64uj)\n\n\u003C\u002Fdiv>\n\nProject AirSim is an open-source, extensible, engine-independent simulation\nplatform for autonomous systems. Its simulation core and APIs can run in the\nlightweight [Project AirSim Runtime](samples\u002Fprojectairsim_runtime\u002FREADME.md)\nwithout Unreal Engine, or with [Unreal Engine 5](https:\u002F\u002Fwww.unrealengine.com\u002F)\nwhen a 3D world, rendered sensors, and environment geometry are required.\n\nIntegrate an autonomy stack with the Project AirSim APIs, reuse compatible\nscene and robot configurations, and select the simulation host that fits each\ntest. Use Runtime for fast controller, API, physics, automation, and CI\nworkflows; move the same integration to Unreal when the scenario requires\nvisual fidelity, cameras, LiDAR, radar, or mesh-based interaction. This lets a\nteam vary simulation cost and fidelity without maintaining a separate client\nintegration for every host.\n\nProject AirSim builds on the work of\n[AirSim](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002FAirSim) and provides a modular framework\nfor drones, fixed-wing aircraft, unmanned ground vehicles (UGVs), and other\nautonomous systems.\n\n**[Download the latest release](https:\u002F\u002Fgithub.com\u002Fiamaisim\u002FProjectAirSim\u002Freleases\u002Flatest)** ·\n**[Use a pre-built environment](docs\u002Fdevelopment\u002Fuse_prebuilt.md)** ·\n**[Build from source](docs\u002Fdevelopment\u002Fuse_source.md)** ·\n**[Read the documentation](https:\u002F\u002Fiamaisim.github.io\u002FProjectAirSim\u002F)**\n\n\u003Ctable>\n\u003Ctr>\n\u003Ctd width=\"50%\">\u003Cimg src=\"docs\u002Fimages\u002Fautonomy\u002Ftakeoff-landing-app-cam-view.gif\" width=\"100%\" alt=\"Autonomous landing application with a live camera view\">\u003Cbr>\u003Csub>\u003Cb>Autonomous Landing.\u003C\u002Fb> Perception-guided vehicle control with a live camera stream.\u003C\u002Fsub>\u003C\u002Ftd>\n\u003Ctd width=\"50%\">\u003Cimg src=\"docs\u002Fimages\u002Fadjustable_weather.gif\" width=\"100%\" alt=\"Quadrotor flying through adjustable weather conditions\">\u003Cbr>\u003Csub>\u003Cb>Adjustable Weather.\u003C\u002Fb> Change environmental conditions while the simulation is running.\u003C\u002Fsub>\u003C\u002Ftd>\n\u003C\u002Ftr>\n\u003Ctr>\n\u003Ctd width=\"50%\">\u003Cimg src=\"docs\u002Fimages\u002Fsimulate_your_swarm.gif\" width=\"100%\" alt=\"Fixed-wing swarm flying over a mountain environment\">\u003Cbr>\u003Csub>\u003Cb>Simulate Your Swarm.\u003C\u002Fb> Run multiple vehicles together in a shared simulation.\u003C\u002Fsub>\u003C\u002Ftd>\n\u003Ctd width=\"50%\">\u003Cimg src=\"docs\u002Fimages\u002Fairtaxi_in_dynamic_city.jpg\" width=\"100%\" alt=\"Air taxi flying over a dynamic city\">\u003Cbr>\u003Csub>\u003Cb>Dynamic City.\u003C\u002Fb> An air taxi operating in a dense Unreal city environment.\u003C\u002Fsub>\u003C\u002Ftd>\n\u003C\u002Ftr>\n\u003Ctr>\n\u003Ctd width=\"50%\">\u003Cimg src=\"docs\u002Fimages\u002Fwind_turbine_inspection.gif\" width=\"100%\" alt=\"Aircraft inspecting wind turbines\">\u003Cbr>\u003Csub>\u003Cb>Wind Turbine Inspection.\u003C\u002Fb> Inspect renewable-energy infrastructure in a large Unreal environment.\u003C\u002Fsub>\u003C\u002Ftd>\n\u003Ctd width=\"50%\">\u003Cimg src=\"docs\u002Fimages\u002Fvehicle_caravan.gif\" width=\"100%\" alt=\"Caravan of unmanned ground and aerial vehicles traversing a mountain village\">\u003Cbr>\u003Csub>\u003Cb>Multi-Vehicle Caravan.\u003C\u002Fb> Simulate UGVs and aerial vehicles together with multiple sensor integration.\u003C\u002Fsub>\u003C\u002Ftd>\n\u003C\u002Ftr>\n\u003C\u002Ftable>\n\n## Table of Contents\n\n- [Current Repository Capabilities](#current-repository-capabilities)\n- [Choose Your Starting Point](#choose-your-starting-point)\n- [Latest Project Updates](#latest-project-updates)\n- [Architecture](#architecture)\n- [Key Integrations and Reference Documentation](#key-integrations-and-reference-documentation)\n- [Supported Development Platforms](#supported-development-platforms)\n- [Source Build Overview](#source-build-overview)\n- [Headless Execution](#headless-execution)\n- [Community and Contributions](#community-and-contributions)\n- [Licensing](#third-party-interoperability-and-licensing)\n\n## Current Repository Capabilities\n\nThe current `main` branch lets you:\n\n- Create configurable single-vehicle and multi-vehicle simulation scenes.\n- Use built-in C++ Fast Physics, JSBSim flight dynamics, Simulink physics\n  models, or extend the C++ physics layer for custom requirements.\n- Use Simple Flight, PX4 in software-in-the-loop (SITL) or\n  hardware-in-the-loop (HITL), ArduPilot SITL, and manual control workflows.\n- Integrate custom controllers, actuators, sensors, and robot models.\n- Simulate fixed-wing aircraft with JSBSim, including Cessna 310 and Skywalker\n  X8 examples.\n- Simulate UGVs through Unreal Blueprint vehicle integration, including the\n  SimpleDrive SUV example.\n- Connect autonomy software through Python and C++ client libraries or the ROS 2\n  C++ bridge.\n- Work with cameras, LiDAR, radar, IMU, GPS, barometer, magnetometer, airspeed,\n  and other configurable sensors.\n- Run with Unreal rendering, off-screen rendering, or without rendering for\n  controller, API, and physics development.\n\nPublished binaries may trail the `main` branch. Check the\n[release notes](https:\u002F\u002Fgithub.com\u002Fiamaisim\u002FProjectAirSim\u002Freleases) to confirm\nwhich capabilities are included in a particular package, and see the\n[changelog](docs\u002Fchangelog.md) for repository changes.\n\n## Choose Your Starting Point\n\n### Run a Pre-built Environment\n\n> I want to evaluate Project AirSim, launch an environment, and control a\n> vehicle with Python.\n\nDownload a packaged environment from\n[GitHub Releases](https:\u002F\u002Fgithub.com\u002Fiamaisim\u002FProjectAirSim\u002Freleases), then\nfollow the [pre-built environment guide](docs\u002Fdevelopment\u002Fuse_prebuilt.md).\n\n### Build and Extend Project AirSim\n\n> I want to customize the simulation core, Unreal plugin, vehicles, sensors, or\n> environments.\n\nFollow the [source development guide](docs\u002Fdevelopment\u002Fuse_source.md) to build\nthe simulation libraries, plugin, Blocks environment, and client packages.\n\n### Run Without Unreal Engine\n\n> I need a fast, headless simulation host for controllers, APIs, physics,\n> automation, or CI without changing my Project AirSim client integration.\n\nUse [Project AirSim Runtime](samples\u002Fprojectairsim_runtime\u002FREADME.md), the\nlightweight, engine-independent host. It runs the same `SimServer`, Project\nAirSim physics, controllers, APIs, and non-rendered sensors without requiring\nUnreal Engine.\n\nRuntime supports Fast Physics, JSBSim, Simulink physics, flight-controller\nworkflows, and sensors including GPS, IMU, barometer, magnetometer, and\nairspeed. It does not provide cameras, LiDAR, radar, Unreal world meshes, or\ngeneral mesh and robot-to-robot collisions. Its host-side collision support is\nlimited to a flat ground plane for Fast Physics vehicles.\n\n### Migrate from AirSim\n\n> I have an existing AirSim environment or client workflow.\n\nStart with the [AirSim transition guide](docs\u002Ftransition_from_airsim.md).\n\n## Latest Project Updates\n\nThe current `main` branch contains the changes recorded for Project AirSim\n1.0.0, including:\n\n- JSBSim fixed-wing simulation with Cessna 310 and Skywalker X8 examples;\n- a standalone C++ client package and ROS 2 C++ bridge;\n- GPU LiDAR 360-degree scanning and additional LiDAR validation;\n- a Python client `step()` API; and\n- Project AirSim Runtime and modular scene configurations;\n- Blueprint vehicle integration, including SimpleDrive SUV examples; and\n- improved build, toolchain, and CI support for Unreal Engine 5.7.\n\nSee [Project AirSim releases](https:\u002F\u002Fgithub.com\u002Fiamaisim\u002FProjectAirSim\u002Freleases)\nfor the latest published binaries and release notes.\n\n## Architecture\n\nProject AirSim has three primary layers:\n\n1. **Simulation libraries** provide the base infrastructure for defining robot\n   structures, physics, controllers, sensors, and the simulation scene tick\n   loop.\n2. **Simulation host** provides the environment-dependent services. Project\n   AirSim Runtime offers lightweight engine-independent execution, while the\n   Unreal Engine plugin adds 3D environments, rendering, mesh interaction, and\n   rendered sensors.\n3. **Client libraries** expose network APIs for loading scenes, controlling\n   vehicles, and receiving state and sensor data.\n\n![Project AirSim architecture: clients connect to either the lightweight Runtime host or the Unreal host, both of which use the common simulation libraries](docs\u002Fimages\u002Fprojectairsim_architecture.svg)\n\nFor more detail, see the\n[Project AirSim architecture overview](docs\u002Fdevelopment\u002Fuse_source.md#airsim-v-next-architecture-overview).\n\n### Experimental Unity Host Reference\n\nThe repository also contains an\n[experimental Unity host integration](unity\u002FREADME.md) composed of a Unity\nexample project and a native wrapper around the Project AirSim simulation\nlibraries. It is retained as reference code that demonstrates how another 3D\nengine can host the common simulation core.\n\nThe Unity integration is **not currently maintained, validated, packaged, or\nsupported by IAMAI** and is not part of the supported-platform matrix. Do not\nassume compatibility with current Project AirSim or Unity releases.\n\n## Key Integrations and Reference Documentation\n\n- [Configuration overview](docs\u002Fconfig.md)\n- [Scene configuration](docs\u002Fconfig_scene.md)\n- [Robot configuration](docs\u002Fconfig_robot.md)\n- [Client API](docs\u002Fapi.md)\n- [Flight controllers](docs\u002Fcontrollers\u002Fcontrollers.md)\n- [PX4 integration](docs\u002Fcontrollers\u002Fpx4\u002Fpx4.md)\n- [ArduPilot integration](docs\u002Fcontrollers\u002Fardupilot.md)\n- [Fast Physics](docs\u002Fphysics\u002Ffast-physics.md)\n- [JSBSim physics](docs\u002Fphysics\u002Fjsbsim.md)\n- [Simulink physics](docs\u002Fphysics\u002Fmatlab.md)\n- [Unreal and UGV Blueprint vehicle integration](docs\u002Funreal_vehicle_blueprint_setup.md)\n- [ROS 2 C++ bridge](docs\u002Fros\u002Fros2.md)\n- [Sensor configuration](docs\u002Fconfig_robot.md#sensor-settings)\n- [Headless and cloud execution](docs\u002Fdevelopment\u002Fheadless_cloud.md)\n\n## Supported Development Platforms\n\nThe supported development baseline is:\n\n| Component | Supported version or behavior |\n| --- | --- |\n| Linux | **Ubuntu 22.04** is the primary supported distribution |\n| Windows | **Windows 11** with Visual Studio 2022 C++ build tools |\n| Unreal Engine | **5.2 or 5.7** |\n| CMake and C++ | CMake **3.15 or newer** and C++17 |\n| Linux compiler | Unreal's packaged toolchain when `UE_ROOT` is set; otherwise Clang 13 |\n| Windows compiler | `build.cmd` selects MSVC 14.37 for UE 5.2 and MSVC 14.44 for UE 5.7 |\n| Python client | Python **newer than 3.10**, below Python 4 |\n| ROS 2 C++ bridge | **ROS 2 Humble** on Ubuntu 22.04 |\n\n`setup_linux_dev_tools.sh` recognizes some additional Ubuntu releases, but that\ninstallation logic is not a supported-platform guarantee. Use Ubuntu 22.04 for\nthe documented and CI-tested Linux development environment.\n\nHardware requirements are primarily determined by Unreal Engine and the\nrendering workload. Review the [system specifications](docs\u002Fsystem_specs.md)\nbefore installing or building the project.\n\n## Source Build Overview\n\nThe complete and authoritative instructions are in the\n[source development guide](docs\u002Fdevelopment\u002Fuse_source.md). The basic workflow\nis summarized below.\n\n### 1. Install Unreal Engine\n\nInstall Unreal Engine 5.2 or 5.7 and set `UE_ROOT` to its installation path.\n\nOn Linux:\n\n```bash\nexport UE_ROOT=\u002Fpath\u002Fto\u002FUnrealEngine\n```\n\n### 2. Install Linux Development Dependencies\n\n```bash\n.\u002Fsetup_linux_dev_tools.sh\n```\n\n### 3. Build the Simulation Libraries\n\nOn Linux:\n\n```bash\n.\u002Fbuild.sh simlibs_debug\n```\n\nOn Windows, use an **x64 Native Tools Command Prompt for VS 2022**:\n\n```cmd\nbuild.cmd simlibs_debug\n```\n\n### 4. Generate Project Files\n\nOn Linux:\n\n```bash\n.\u002Fblocks_genprojfiles_vscode.sh\n```\n\nOn Windows:\n\n```cmd\nblocks_genprojfiles_vscode.bat\n```\n\nOpen the generated workspace and launch the Unreal Editor in DebugGame mode.\n\n## Headless Execution\n\nTo run a packaged Unreal environment with off-screen rendering:\n\n```text\nBlocks{.exe\u002F.sh} -RenderOffScreen\n```\n\nTo disable rendering completely:\n\n```text\nBlocks{.exe\u002F.sh} -nullrhi\n```\n\nSee [headless and cloud execution](docs\u002Fdevelopment\u002Fheadless_cloud.md) for\nadditional configuration details.\n\n## Open Source and Professional Services\n\nProject AirSim's simulation core, APIs, configuration system, extension points,\nand reference workflows are available publicly under the MIT License. We want\nthe open-source project to be useful for evaluation, research, development, and\nreal autonomy workflows.\n\n[IAMAI Consulting Corp.](https:\u002F\u002Fwww.iamaisim.com) maintains and extends the\nProject AirSim ecosystem. The team includes former Microsoft AirSim engineers\nand provides professional services for organizations that need to turn a\nprototype into a repeatable simulation or validation workflow.\n\nIAMAI can help with:\n\n- simulation-readiness and architecture assessments;\n- PX4, ROS 2, JSBSim, vehicle, sensor, and autonomy-stack integration;\n- custom Unreal Engine environments and simulation workflows;\n- reproducible scenarios and validated builds; and\n- maintained delivery, updates, and engineering support.\n\nIf you have an upcoming integration, demonstration, pilot, or validation\nmilestone, **[talk to IAMAI](https:\u002F\u002Fwww.iamaisim.com)** about a focused first\nengagement.\n\n## Community and Contributions\n\nProject AirSim grows through practical use, technical feedback, and community\ncontributions.\n\n- [Report a bug or request a feature](https:\u002F\u002Fgithub.com\u002Fiamaisim\u002FProjectAirSim\u002Fissues\u002Fnew\u002Fchoose)\n- [Start a GitHub Discussion](https:\u002F\u002Fgithub.com\u002Fiamaisim\u002FProjectAirSim\u002Fdiscussions)\n- [Join the Project AirSim Discord](https:\u002F\u002Fdiscord.gg\u002FXprQ2w64uj)\n- Submit focused pull requests for code, tests, documentation, and examples\n\nThe roadmap is managed through GitHub issues and discussions. Items labeled\n[`roadmap`](https:\u002F\u002Fgithub.com\u002Fiamaisim\u002FProjectAirSim\u002Flabels\u002Froadmap) describe\nplanned direction, while items labeled\n[`need help`](https:\u002F\u002Fgithub.com\u002Fiamaisim\u002FProjectAirSim\u002Flabels\u002Fneed%20help)\nidentify opportunities for community participation.\n\n## Third-Party Interoperability and Licensing\n\nProject AirSim interoperates with third-party engines, libraries, models, and\ntools, including JSBSim-compatible aircraft definitions. Those components and\nassets remain under their respective licenses.\n\nBefore redistributing or extending an integration, review\n[Project AirSim license information](docs\u002Flicense.md), `NOTICE.txt`, and the\nlicenses under `thirdparty\u002FLicenses\u002F`.\n\n---\n\nCopyright (C) Microsoft Corporation.  \nCopyright (C) 2025-2026 IAMAI CONSULTING CORP\n\nMIT License\n","Project AirSim 是一个开源、模块化、引擎无关的自主系统仿真平台，旨在支持无人机、固定翼飞机、无人车等智能体的开发、训练与测试。其核心支持轻量级独立运行时（无需Unreal Engine）和高保真UE5集成双模式，提供统一API访问物理引擎、传感器（相机\u002FLiDAR\u002F雷达）、环境几何与自动化接口，并原生兼容ROS 2（Humble）及Python\u002FC++客户端。适用于算法快速迭代、CI\u002FCD验证、多 fidelity 场景切换及跨平台自主系统研发。",2,"2026-09-12 02:30:11","trending"]