[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"similar-jslee02--awesome-robotics-libraries":3,"tool-jslee02--awesome-robotics-libraries":64},[4,17,27,35,43,56],{"id":5,"name":6,"github_repo":7,"description_zh":8,"stars":9,"difficulty_score":10,"last_commit_at":11,"category_tags":12,"status":16},3808,"stable-diffusion-webui","AUTOMATIC1111\u002Fstable-diffusion-webui","stable-diffusion-webui 是一个基于 Gradio 构建的网页版操作界面，旨在让用户能够轻松地在本地运行和使用强大的 Stable Diffusion 图像生成模型。它解决了原始模型依赖命令行、操作门槛高且功能分散的痛点，将复杂的 AI 绘图流程整合进一个直观易用的图形化平台。\n\n无论是希望快速上手的普通创作者、需要精细控制画面细节的设计师，还是想要深入探索模型潜力的开发者与研究人员，都能从中获益。其核心亮点在于极高的功能丰富度：不仅支持文生图、图生图、局部重绘（Inpainting）和外绘（Outpainting）等基础模式，还独创了注意力机制调整、提示词矩阵、负向提示词以及“高清修复”等高级功能。此外，它内置了 GFPGAN 和 CodeFormer 等人脸修复工具，支持多种神经网络放大算法，并允许用户通过插件系统无限扩展能力。即使是显存有限的设备，stable-diffusion-webui 也提供了相应的优化选项，让高质量的 AI 艺术创作变得触手可及。",162132,3,"2026-04-05T11:01:52",[13,14,15],"开发框架","图像","Agent","ready",{"id":18,"name":19,"github_repo":20,"description_zh":21,"stars":22,"difficulty_score":23,"last_commit_at":24,"category_tags":25,"status":16},1381,"everything-claude-code","affaan-m\u002Feverything-claude-code","everything-claude-code 是一套专为 AI 编程助手（如 Claude Code、Codex、Cursor 等）打造的高性能优化系统。它不仅仅是一组配置文件，而是一个经过长期实战打磨的完整框架，旨在解决 AI 代理在实际开发中面临的效率低下、记忆丢失、安全隐患及缺乏持续学习能力等核心痛点。\n\n通过引入技能模块化、直觉增强、记忆持久化机制以及内置的安全扫描功能，everything-claude-code 能显著提升 AI 在复杂任务中的表现，帮助开发者构建更稳定、更智能的生产级 AI 代理。其独特的“研究优先”开发理念和针对 Token 消耗的优化策略，使得模型响应更快、成本更低，同时有效防御潜在的攻击向量。\n\n这套工具特别适合软件开发者、AI 研究人员以及希望深度定制 AI 工作流的技术团队使用。无论您是在构建大型代码库，还是需要 AI 协助进行安全审计与自动化测试，everything-claude-code 都能提供强大的底层支持。作为一个曾荣获 Anthropic 黑客大奖的开源项目，它融合了多语言支持与丰富的实战钩子（hooks），让 AI 真正成长为懂上",140436,2,"2026-04-05T23:32:43",[13,15,26],"语言模型",{"id":28,"name":29,"github_repo":30,"description_zh":31,"stars":32,"difficulty_score":23,"last_commit_at":33,"category_tags":34,"status":16},2271,"ComfyUI","Comfy-Org\u002FComfyUI","ComfyUI 是一款功能强大且高度模块化的视觉 AI 引擎，专为设计和执行复杂的 Stable Diffusion 图像生成流程而打造。它摒弃了传统的代码编写模式，采用直观的节点式流程图界面，让用户通过连接不同的功能模块即可构建个性化的生成管线。\n\n这一设计巧妙解决了高级 AI 绘图工作流配置复杂、灵活性不足的痛点。用户无需具备编程背景，也能自由组合模型、调整参数并实时预览效果，轻松实现从基础文生图到多步骤高清修复等各类复杂任务。ComfyUI 拥有极佳的兼容性，不仅支持 Windows、macOS 和 Linux 全平台，还广泛适配 NVIDIA、AMD、Intel 及苹果 Silicon 等多种硬件架构，并率先支持 SDXL、Flux、SD3 等前沿模型。\n\n无论是希望深入探索算法潜力的研究人员和开发者，还是追求极致创作自由度的设计师与资深 AI 绘画爱好者，ComfyUI 都能提供强大的支持。其独特的模块化架构允许社区不断扩展新功能，使其成为当前最灵活、生态最丰富的开源扩散模型工具之一，帮助用户将创意高效转化为现实。",107662,"2026-04-03T11:11:01",[13,14,15],{"id":36,"name":37,"github_repo":38,"description_zh":39,"stars":40,"difficulty_score":23,"last_commit_at":41,"category_tags":42,"status":16},3704,"NextChat","ChatGPTNextWeb\u002FNextChat","NextChat 是一款轻量且极速的 AI 助手，旨在为用户提供流畅、跨平台的大模型交互体验。它完美解决了用户在多设备间切换时难以保持对话连续性，以及面对众多 AI 模型不知如何统一管理的痛点。无论是日常办公、学习辅助还是创意激发，NextChat 都能让用户随时随地通过网页、iOS、Android、Windows、MacOS 或 Linux 端无缝接入智能服务。\n\n这款工具非常适合普通用户、学生、职场人士以及需要私有化部署的企业团队使用。对于开发者而言，它也提供了便捷的自托管方案，支持一键部署到 Vercel 或 Zeabur 等平台。\n\nNextChat 的核心亮点在于其广泛的模型兼容性，原生支持 Claude、DeepSeek、GPT-4 及 Gemini Pro 等主流大模型，让用户在一个界面即可自由切换不同 AI 能力。此外，它还率先支持 MCP（Model Context Protocol）协议，增强了上下文处理能力。针对企业用户，NextChat 提供专业版解决方案，具备品牌定制、细粒度权限控制、内部知识库整合及安全审计等功能，满足公司对数据隐私和个性化管理的高标准要求。",87618,"2026-04-05T07:20:52",[13,26],{"id":44,"name":45,"github_repo":46,"description_zh":47,"stars":48,"difficulty_score":23,"last_commit_at":49,"category_tags":50,"status":16},2268,"ML-For-Beginners","microsoft\u002FML-For-Beginners","ML-For-Beginners 是由微软推出的一套系统化机器学习入门课程，旨在帮助零基础用户轻松掌握经典机器学习知识。这套课程将学习路径规划为 12 周，包含 26 节精炼课程和 52 道配套测验，内容涵盖从基础概念到实际应用的完整流程，有效解决了初学者面对庞大知识体系时无从下手、缺乏结构化指导的痛点。\n\n无论是希望转型的开发者、需要补充算法背景的研究人员，还是对人工智能充满好奇的普通爱好者，都能从中受益。课程不仅提供了清晰的理论讲解，还强调动手实践，让用户在循序渐进中建立扎实的技能基础。其独特的亮点在于强大的多语言支持，通过自动化机制提供了包括简体中文在内的 50 多种语言版本，极大地降低了全球不同背景用户的学习门槛。此外，项目采用开源协作模式，社区活跃且内容持续更新，确保学习者能获取前沿且准确的技术资讯。如果你正寻找一条清晰、友好且专业的机器学习入门之路，ML-For-Beginners 将是理想的起点。",84991,"2026-04-05T10:45:23",[14,51,52,53,15,54,26,13,55],"数据工具","视频","插件","其他","音频",{"id":57,"name":58,"github_repo":59,"description_zh":60,"stars":61,"difficulty_score":10,"last_commit_at":62,"category_tags":63,"status":16},3128,"ragflow","infiniflow\u002Fragflow","RAGFlow 是一款领先的开源检索增强生成（RAG）引擎，旨在为大语言模型构建更精准、可靠的上下文层。它巧妙地将前沿的 RAG 技术与智能体（Agent）能力相结合，不仅支持从各类文档中高效提取知识，还能让模型基于这些知识进行逻辑推理和任务执行。\n\n在大模型应用中，幻觉问题和知识滞后是常见痛点。RAGFlow 通过深度解析复杂文档结构（如表格、图表及混合排版），显著提升了信息检索的准确度，从而有效减少模型“胡编乱造”的现象，确保回答既有据可依又具备时效性。其内置的智能体机制更进一步，使系统不仅能回答问题，还能自主规划步骤解决复杂问题。\n\n这款工具特别适合开发者、企业技术团队以及 AI 研究人员使用。无论是希望快速搭建私有知识库问答系统，还是致力于探索大模型在垂直领域落地的创新者，都能从中受益。RAGFlow 提供了可视化的工作流编排界面和灵活的 API 接口，既降低了非算法背景用户的上手门槛，也满足了专业开发者对系统深度定制的需求。作为基于 Apache 2.0 协议开源的项目，它正成为连接通用大模型与行业专有知识之间的重要桥梁。",77062,"2026-04-04T04:44:48",[15,14,13,26,54],{"id":65,"github_repo":66,"name":67,"description_en":68,"description_zh":69,"ai_summary_zh":69,"readme_en":70,"readme_zh":71,"quickstart_zh":72,"use_case_zh":73,"hero_image_url":74,"owner_login":75,"owner_name":76,"owner_avatar_url":77,"owner_bio":78,"owner_company":78,"owner_location":78,"owner_email":78,"owner_twitter":79,"owner_website":80,"owner_url":81,"languages":82,"stars":87,"forks":88,"last_commit_at":89,"license":90,"difficulty_score":91,"env_os":92,"env_gpu":93,"env_ram":94,"env_deps":95,"category_tags":99,"github_topics":100,"view_count":113,"oss_zip_url":78,"oss_zip_packed_at":78,"status":16,"created_at":114,"updated_at":115,"faqs":116,"releases":147},1580,"jslee02\u002Fawesome-robotics-libraries","awesome-robotics-libraries",":sunglasses: A curated list of robotics libraries and software","awesome-robotics-libraries 是一份精心整理的机器人领域开源库与仿真器清单，旨在为开发者提供一站式的资源导航。在机器人研发过程中，从动力学仿真、运动规划到视觉感知、强化学习，往往涉及众多分散的技术栈，寻找高质量且维护良好的代码库耗时费力。这份清单通过系统化的分类（如逆运动学、SLAM、抓取算法等）和清晰的状态标记（活跃、缓慢更新或已归档），有效解决了资源筛选难、版本过时等痛点。\n\n它特别适合机器人工程师、学术研究人员以及高校师生使用。无论是需要快速原型验证的初创团队，还是深耕底层算法的研究者，都能在此找到适配的工具。清单收录了如 Google 的 Brax、MIT 的 Drake、NVIDIA 的 Flex 等知名项目，涵盖了从刚体\u002F柔体物理引擎到多物理场仿真等前沿技术。其独特的亮点在于不仅罗列名称，还直观展示了各项目的活跃度与社区热度，帮助用户规避“坑爹”的废弃项目，直接锁定当前最主流、最可靠的解决方案，从而大幅提升研发效率。","# Awesome Robotics Libraries\n\n[![Awesome](https:\u002F\u002Fawesome.re\u002Fbadge.svg)](https:\u002F\u002Fawesome.re)\n\nA curated list of robotics libraries and simulators.\n\n## Contents\n* [Libraries](#libraries)\n  * [Dynamics Simulation](#dynamics-simulation)\n  * [Inverse Kinematics](#inverse-kinematics)\n  * [Machine Learning](#machine-learning)\n  * [Motion Planning and Control](#motion-planning-and-control)\n  * [Optimization](#optimization)\n  * [Robot Modeling](#robot-modeling)\n  * [Robot Platform](#robot-platform)\n  * [Reinforcement Learning for Robotics](#reinforcement-learning-for-robotics)\n  * [SLAM](#slam)\n  * [Vision](#vision)\n  * [Fluid](#fluid)\n  * [Grasping](#grasping)\n  * [Humanoid Robotics](#humanoid-robotics)\n  * [Multiphysics](#multiphysics)\n  * [Math](#math)\n  * [ETC](#etc)\n* [Simulators](#simulators)\n* [Other Awesome Lists](#other-awesome-lists)\n\n> **Legend**: 🟢 Active (\u003C1yr) · 🟡 Slow (1-2yr) · 🔴 Stale (>2yr) · 💀 Archived\n\n## [Libraries](#contents)\n\n### [Dynamics Simulation](#contents)\n\n_Physics engines and rigid\u002Fsoft body dynamics libraries. See also [Comparisons](COMPARISONS.md)._\n\n* [ARCSim](http:\u002F\u002Fgraphics.berkeley.edu\u002Fresources\u002FARCSim\u002Findex.html) - Adaptive remeshing cloth and shell simulator for thin deformable objects.\n* 🟢 [Brax](https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fbrax) - Massively parallel differentiable rigid body physics engine in JAX for robotics and RL. [⭐ 3.1k](https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fbrax)\n* 🟢 [Bullet](https:\u002F\u002Fpybullet.org\u002F) - Real-time physics simulation for games, visual effects, and robotics. [⭐ 14.2k](https:\u002F\u002Fgithub.com\u002Fbulletphysics\u002Fbullet3)\n* 🟢 [CHRONO::ENGINE](https:\u002F\u002Fprojectchrono.org\u002F) - Multi-physics simulation of rigid and flexible bodies, granular, and fluid systems. [⭐ 2.7k](https:\u002F\u002Fgithub.com\u002Fprojectchrono\u002Fchrono)\n* 🟢 [DART](http:\u002F\u002Fdartsim.github.io\u002F) - Dynamic Animation and Robotics Toolkit for multibody simulation and planning. [⭐ 1.1k](https:\u002F\u002Fgithub.com\u002Fdartsim\u002Fdart)\n* 🟢 [Drake](https:\u002F\u002Fdrake.mit.edu\u002F) - Planning, control, and analysis toolbox for nonlinear dynamical systems. [⭐ 3.9k](https:\u002F\u002Fgithub.com\u002FRobotLocomotion\u002Fdrake)\n* 💀 [Flex](https:\u002F\u002Fdeveloper.nvidia.com\u002Fflex) - GPU-based particle simulation for rigid bodies, fluids, and deformables. [⭐ 788](https:\u002F\u002Fgithub.com\u002FNVIDIAGameWorks\u002FFleX)\n* 🔴 [FROST](https:\u002F\u002Fayonga.github.io\u002Ffrost-dev\u002Findex.html) - Fast Robot Optimization and Simulation Toolkit for hybrid dynamical systems in MATLAB. [⭐ 170](https:\u002F\u002Fgithub.com\u002Fayonga\u002Ffrost-dev)\n* 🟢 [Genesis](https:\u002F\u002Fgenesis-world.readthedocs.io) - Generative and universal physics platform for robotics with GPU-accelerated parallel simulation. [⭐ 28.1k](https:\u002F\u002Fgithub.com\u002FGenesis-Embodied-AI\u002FGenesis)\n* [IBDS](http:\u002F\u002Fwww.interactive-graphics.de\u002Findex.php\u002Fdownloads\u002F12-ibds) - Impulse-based dynamics simulation for rigid bodies and particle systems.\n* 🟢 idyntree - Library for estimation and whole-body dynamics of floating-base robots. [⭐ 224](https:\u002F\u002Fgithub.com\u002Fgbionics\u002Fidyntree)\n* 🟢 [KDL](https:\u002F\u002Fwww.orocos.org\u002Fkdl.html) - Orocos Kinematics and Dynamics Library for kinematic chains. [⭐ 859](https:\u002F\u002Fgithub.com\u002Forocos\u002Forocos_kinematics_dynamics)\n* 🟢 kindr - Kinematics and dynamics library for rigid body transformations. [⭐ 607](https:\u002F\u002Fgithub.com\u002FANYbotics\u002Fkindr)\n* 🟢 [Klampt](https:\u002F\u002Fklampt.org\u002F) - Robot planning, control, and simulation with visualization support. [⭐ 427](https:\u002F\u002Fgithub.com\u002Fkrishauser\u002FKlampt)\n* 🔴 [LibrePilot](http:\u002F\u002Fwww.librepilot.org\u002Fsite\u002Findex.html) - Open-source autopilot for UAVs and other autonomous vehicles. [⭐ 348](https:\u002F\u002Fgithub.com\u002Flibrepilot\u002FLibrePilot)\n* 🟢 [MARS](http:\u002F\u002Frock-simulation.github.io\u002Fmars\u002F) - Machina Arte Robotum Simulans — a cross-platform simulation environment. [⭐ 67](https:\u002F\u002Fgithub.com\u002Frock-simulation\u002Fmars)\n* [MBDyn](https:\u002F\u002Fwww.mbdyn.org\u002F) - General-purpose multibody dynamics analysis software. [[code](https:\u002F\u002Fwww.mbdyn.org\u002F?Software_Download)]\n* 🟢 [MBSim](https:\u002F\u002Fwww.mbsim-env.de\u002F) - Multi-body simulation environment for flexible and rigid systems. [⭐ 51](https:\u002F\u002Fgithub.com\u002Fmbsim-env\u002Fmbsim)\n* 🔴 [MBSlib](http:\u002F\u002Fwww.sim.informatik.tu-darmstadt.de\u002Fres\u002Fsw\u002Fmbslib) - Lightweight multibody system dynamics library. [⭐ 11](https:\u002F\u002Fgithub.com\u002FSIM-TU-Darmstadt\u002Fmbslib)\n* 💀 metapod - Template-based robot dynamics library using spatial algebra. [⭐ 14](https:\u002F\u002Fgithub.com\u002Flaas\u002Fmetapod)\n* 🔴 [Moby](http:\u002F\u002Fphyssim.sourceforge.net\u002Findex.html) - Multi-body dynamics simulation for rigid bodies with contact. [⭐ 37](https:\u002F\u002Fgithub.com\u002FPositronicsLab\u002FMoby)\n* 🟢 [mrpt](https:\u002F\u002Fwww.mrpt.org\u002F) - Mobile Robot Programming Toolkit for SLAM, navigation, and computer vision. [⭐ 2.1k](https:\u002F\u002Fgithub.com\u002FMRPT\u002Fmrpt)\n* 🟢 [MuJoCo](https:\u002F\u002Fmujoco.org\u002F) - Multi-joint dynamics with contact for physics-based simulation and control. [⭐ 12k](https:\u002F\u002Fgithub.com\u002Fgoogle-deepmind\u002Fmujoco)\n* 🟢 [mvsim](http:\u002F\u002Fwiki.ros.org\u002Fmvsim) - Lightweight multi-vehicle 2D simulator with ROS integration. [⭐ 361](https:\u002F\u002Fgithub.com\u002FMRPT\u002Fmvsim)\n* 🟢 [Newton](https:\u002F\u002Fnewton-physics.github.io\u002Fnewton\u002F) - GPU-accelerated differentiable physics engine built on NVIDIA Warp for robotics simulation. [⭐ 2.5k](https:\u002F\u002Fgithub.com\u002Fnewton-physics\u002Fnewton)\n* 🟢 [Newton Dynamics](https:\u002F\u002Fnewtondynamics.com\u002F) - Real-time physics engine for rigid body simulation. [⭐ 1k](https:\u002F\u002Fgithub.com\u002FMADEAPPS\u002Fnewton-dynamics)\n* 🔴 [nphysics](https:\u002F\u002Fnphysics.org\u002F) - 2D and 3D rigid body physics engine written in Rust. [⭐ 1.6k](https:\u002F\u002Fgithub.com\u002Fdimforge\u002Fnphysics)\n* [ODE](https:\u002F\u002Fode.org\u002F) - Open Dynamics Engine for simulating rigid body dynamics. [[bitbucket](https:\u002F\u002Fbitbucket.org\u002Fodedevs\u002Fode)]\n* 🟢 [OpenRAVE](https:\u002F\u002Fwww.openrave.org\u002F) - Open Robotics Automation Virtual Environment for planning and simulation. [⭐ 798](https:\u002F\u002Fgithub.com\u002Frdiankov\u002Fopenrave)\n* 🟢 [PhysX](https:\u002F\u002Fnvidia-omniverse.github.io\u002FPhysX\u002Fphysx\u002F5.5.0\u002Findex.html) - NVIDIA physics engine for real-time rigid body and vehicle simulation. [⭐ 4.4k](https:\u002F\u002Fgithub.com\u002FNVIDIA-Omniverse\u002FPhysX)\n* 🟢 [pinocchio](https:\u002F\u002Fstack-of-tasks.github.io\u002Fpinocchio\u002F) - Fast and flexible algorithms for rigid-body dynamics with analytical derivatives. [⭐ 3.1k](https:\u002F\u002Fgithub.com\u002Fstack-of-tasks\u002Fpinocchio)\n* 🟢 PositionBasedDynamics - Position-based methods for simulating deformable objects and fluids. [⭐ 2.2k](https:\u002F\u002Fgithub.com\u002FInteractiveComputerGraphics\u002FPositionBasedDynamics)\n* 🟢 [PyDy](https:\u002F\u002Fwww.pydy.org\u002F) - Multibody dynamics analysis with symbolic Python using SymPy. [⭐ 405](https:\u002F\u002Fgithub.com\u002Fpydy\u002Fpydy)\n* 💀 qu3e - Lightweight 3D physics engine for rigid body dynamics. [⭐ 976](https:\u002F\u002Fgithub.com\u002FRandyGaul\u002Fqu3e)\n* 💀 [RaiSim](https:\u002F\u002Fslides.com\u002Fjeminhwangbo\u002Fraisim-manual) - Cross-platform physics engine for robotics and reinforcement learning. [⭐ 329](https:\u002F\u002Fgithub.com\u002Fleggedrobotics\u002FraisimLib)\n* 🟢 [RBDL](https:\u002F\u002Frbdl.github.io\u002F) - Rigid Body Dynamics Library based on Featherstone algorithms. [⭐ 684](https:\u002F\u002Fgithub.com\u002Frbdl\u002Frbdl)\n* 🟢 RBDyn - Rigid body dynamics algorithms using spatial algebra with Eigen. [⭐ 211](https:\u002F\u002Fgithub.com\u002Fjrl-umi3218\u002FRBDyn)\n* 🟢 [ReactPhysics3d](https:\u002F\u002Fwww.reactphysics3d.com\u002F) - Open-source 3D physics engine for rigid body simulation and collision detection. [⭐ 1.7k](https:\u002F\u002Fgithub.com\u002FDanielChappuis\u002Freactphysics3d)\n* 🟡 RigidBodyDynamics.jl - Julia library for rigid body dynamics algorithms. [⭐ 308](https:\u002F\u002Fgithub.com\u002FJuliaRobotics\u002FRigidBodyDynamics.jl)\n* 🟢 [Rigs of Rods](https:\u002F\u002Fwww.rigsofrods.org\u002F) - Soft-body vehicle simulator using beam physics. [⭐ 1.2k](https:\u002F\u002Fgithub.com\u002FRigsOfRods\u002Frigs-of-rods)\n* 🔴 [Robopy](https:\u002F\u002Fadityadua24.github.io\u002Frobopy\u002F) - Python robotics toolbox inspired by Peter Corke's Robotics Toolbox. [⭐ 228](https:\u002F\u002Fgithub.com\u002Fadityadua24\u002Frobopy)\n* 🟢 [Robotics Library](https:\u002F\u002Fwww.roboticslibrary.org\u002F) - Self-contained C++ library for robot kinematics, planning, and control. [⭐ 1.2k](https:\u002F\u002Fgithub.com\u002Froboticslibrary\u002Frl)\n* [RobWork](https:\u002F\u002Frobwork.dk\u002F) - Framework for simulation and control of robot systems. [[gitlab](https:\u002F\u002Fgitlab.com\u002Fsdurobotics\u002FRobWork)]\n* 🟢 [siconos](https:\u002F\u002Fnonsmooth.gricad-pages.univ-grenoble-alpes.fr\u002Fsiconos\u002F) - Nonsmooth dynamical systems modeling and simulation platform. [⭐ 183](https:\u002F\u002Fgithub.com\u002Fsiconos\u002Fsiconos)\n* 🟢 [Simbody](https:\u002F\u002Fsimtk.org\u002Fhome\u002Fsimbody\u002F) - Multibody dynamics library for biomechanical and mechanical systems. [⭐ 2.5k](https:\u002F\u002Fgithub.com\u002Fsimbody\u002Fsimbody)\n* 🟢 [SOFA](https:\u002F\u002Fwww.sofa-framework.org\u002F) - Simulation Open Framework Architecture for medical and physics simulation. [⭐ 1.1k](https:\u002F\u002Fgithub.com\u002Fsofa-framework\u002Fsofa)\n* 🟡 Tiny Differentiable Simulator - Header-only differentiable physics engine for robotics. [⭐ 1.3k](https:\u002F\u002Fgithub.com\u002Ferwincoumans\u002Ftiny-differentiable-simulator)\n* 🔴 [trep](http:\u002F\u002Fmurpheylab.github.io\u002Ftrep\u002F) - Simulation and optimal control using variational integrators. [⭐ 20](https:\u002F\u002Fgithub.com\u002FMurpheyLab\u002Ftrep)\n\n### [Inverse Kinematics](#contents)\n\n_Libraries for computing joint configurations from end-effector poses._\n\n  * 🟢 IKBT - A python package to solve robot arm inverse kinematics in symbolic form. [⭐ 215](https:\u002F\u002Fgithub.com\u002Fuw-biorobotics\u002FIKBT)\n  * 🟢 Kinpy - A simple pure python package to solve inverse kinematics. [⭐ 179](https:\u002F\u002Fgithub.com\u002Fneka-nat\u002Fkinpy)\n  * 🔴 Lively - A highly configurable toolkit for commanding robots in mixed modalities. [⭐ 7](https:\u002F\u002Fgithub.com\u002FWisc-HCI\u002Flively)\n  * 🔴 RelaxedIK - Real-time Synthesis of Accurate and Feasible Robot Arm Motion. [⭐ 236](https:\u002F\u002Fgithub.com\u002Fuwgraphics\u002Frelaxed_ik)\n  * 🔴 [Trip](https:\u002F\u002Ftrip-kinematics.readthedocs.io\u002Fen\u002Fmain\u002Findex.html) - A python package that solves inverse kinematics of parallel-, serial- or hybrid-robots. [⭐ 44](https:\u002F\u002Fgithub.com\u002FTriPed-Robot\u002Ftrip_kinematics)\n\n### [Machine Learning](#contents)\n\n_Machine learning frameworks and tools applied to robotics._\n\n* 🟢 [AllenAct](https:\u002F\u002Fallenact.org\u002F) - Python\u002FPyTorch-based Research Framework for Embodied AI. [⭐ 376](https:\u002F\u002Fgithub.com\u002Fallenai\u002Fallenact)\n* 🟢 Any4LeRobot - A collection of utilities and tools for LeRobot. [⭐ 856](https:\u002F\u002Fgithub.com\u002FTavish9\u002Fany4lerobot)\n* 🟢 DLL - Deep Learning Library (DLL) for C++. [⭐ 687](https:\u002F\u002Fgithub.com\u002Fwichtounet\u002Fdll)\n* 🔴 [DyNet](https:\u002F\u002Fdynet.readthedocs.io\u002Fen\u002Flatest\u002F) - The Dynamic Neural Network Toolkit. [⭐ 3.4k](https:\u002F\u002Fgithub.com\u002Fclab\u002Fdynet)\n* 🔴 [Fido](http:\u002F\u002Ffidoproject.github.io\u002F) - Lightweight C++ machine learning library for embedded electronics and robotics. [⭐ 462](https:\u002F\u002Fgithub.com\u002FFidoProject\u002FFido)\n* 🟢 [Gymnasium](https:\u002F\u002Fgymnasium.farama.org\u002F) - Developing and comparing reinforcement learning algorithms. [⭐ 11.3k](https:\u002F\u002Fgithub.com\u002FFarama-Foundation\u002FGymnasium)\n  * 🔴 gym-dart - OpenAI Gym environments using the DART physics engine. [⭐ 141](https:\u002F\u002Fgithub.com\u002FDartEnv\u002Fdart-env)\n  * 💀 gym-gazebo - OpenAI Gym environments for the Gazebo simulator. [⭐ 845](https:\u002F\u002Fgithub.com\u002Ferlerobot\u002Fgym-gazebo)\n* 🟢 [Ivy](https:\u002F\u002Flets-unify.ai\u002F) - Unified Machine Learning Framework. [⭐ 14.2k](https:\u002F\u002Fgithub.com\u002Fivy-llc\u002Fivy)\n* 🟢 LeRobot - Pretrained models, datasets, and simulation environments for real-world robotics in PyTorch. [⭐ 21.5k](https:\u002F\u002Fgithub.com\u002Fhuggingface\u002Flerobot)\n* 🟢 [LeRobot Episode Scoring Toolkit](https:\u002F\u002Fgithub.com\u002FRoboticsData\u002Fscore_lerobot_episodes) - One-click tool to score, filter, and export higher-quality LeRobot datasets. [⭐ 48](https:\u002F\u002Fgithub.com\u002FRoboticsData\u002Fscore_lerobot_episodes)\n* 🔴 MiniDNN - A header-only C++ library for deep neural networks. [⭐ 431](https:\u002F\u002Fgithub.com\u002Fyixuan\u002FMiniDNN)\n* 🟢 [mlpack](https:\u002F\u002Fwww.mlpack.org\u002F) - Scalable C++ machine learning library. [⭐ 5.6k](https:\u002F\u002Fgithub.com\u002Fmlpack\u002Fmlpack)\n* 🔴 RLLib - Temporal-difference learning algorithms in reinforcement learning. [⭐ 208](https:\u002F\u002Fgithub.com\u002Fsamindaa\u002FRLLib)\n* 🟢 [robosuite](https:\u002F\u002Frobosuite.ai) - A modular simulation framework and benchmark for robot learning. [⭐ 2.2k](https:\u002F\u002Fgithub.com\u002FARISE-Initiative\u002Frobosuite)\n* 🔴 [tiny-dnn](http:\u002F\u002Ftiny-dnn.readthedocs.io\u002Fen\u002Flatest\u002F) - Header only, dependency-free deep learning framework in C++14. [⭐ 6k](https:\u002F\u002Fgithub.com\u002Ftiny-dnn\u002Ftiny-dnn)\n\n### [Motion Planning and Control](#contents)\n\n_Libraries for robot motion planning, trajectory optimization, and control._\n\n\n* 🔴 [AIKIDO](https:\u002F\u002Fgithub.com\u002Fpersonalrobotics\u002Faikido) - Solving robotic motion planning and decision making problems. [⭐ 229](https:\u002F\u002Fgithub.com\u002Fpersonalrobotics\u002Faikido)\n* 🟢 Bioptim - Bioptim, a Python Framework for Musculoskeletal Optimal Control in Biomechanics. [⭐ 114](https:\u002F\u002Fgithub.com\u002Fpyomeca\u002Fbioptim)\n* 🔴 [Control Toolbox](https:\u002F\u002Fethz-adrl.github.io\u002Fct\u002F) - Open-Source C++ Library for Robotics, Optimal and Model Predictive Control. [⭐ 1.7k](https:\u002F\u002Fgithub.com\u002Fethz-adrl\u002Fcontrol-toolbox)\n* 🟢 Crocoddyl - Optimal control library for robot control under contact sequence. [⭐ 1.2k](https:\u002F\u002Fgithub.com\u002Floco-3d\u002Fcrocoddyl)\n* [CuiKSuite](http:\u002F\u002Fwww.iri.upc.edu\u002Fpeople\u002Fporta\u002FSoft\u002FCuikSuite2-Doc\u002Fhtml) - Applications to solve position analysis and path planning problems.\n* 🟢 [cuRobo](https:\u002F\u002Fcurobo.org) - A CUDA accelerated library containing a suite of robotics algorithms that run significantly faster. [⭐ 1.3k](https:\u002F\u002Fgithub.com\u002Fnvlabs\u002Fcurobo)\n* 🟢 Fields2Cover - Robust and efficient coverage paths for autonomous agricultural vehicles. [⭐ 751](https:\u002F\u002Fgithub.com\u002Ffields2cover\u002Ffields2cover)\n* 🔴 GPMP2 - Gaussian Process Motion Planner 2. [⭐ 351](https:\u002F\u002Fgithub.com\u002Fgtrll\u002Fgpmp2)\n* [HPP](https:\u002F\u002Fhumanoid-path-planner.github.io\u002Fhpp-doc\u002F) - Path planning for kinematic chains in environments cluttered with obstacles.\n* 🟢 [MoveIt!](https:\u002F\u002Fmoveit.ai\u002F) - Motion planning framework. [⭐ 2k](https:\u002F\u002Fgithub.com\u002Fmoveit\u002Fmoveit)\n* 🟢 OCS2 - Efficient continuous and discrete time optimal control implementation. [⭐ 1.3k](https:\u002F\u002Fgithub.com\u002Fleggedrobotics\u002Focs2)\n* 🟢 [OMPL](https:\u002F\u002Fompl.kavrakilab.org\u002F) - Open motion planning library. [⭐ 1.9k](https:\u002F\u002Fgithub.com\u002Fompl\u002Fompl)\n* 💀 pymanoid - Humanoid robotics prototyping environment based on OpenRAVE. [⭐ 232](https:\u002F\u002Fgithub.com\u002Fstephane-caron\u002Fpymanoid)\n* 🔴 ROS Behavior Tree - Behavior tree implementation for ROS-based robot task planning. [⭐ 362](https:\u002F\u002Fgithub.com\u002Fmiccol\u002FROS-Behavior-Tree)\n* 🟢 [Ruckig](https:\u002F\u002Fgithub.com\u002Fpantor\u002Fruckig) - Real-time, time-optimal and jerk-constrained online trajectory generation. [⭐ 1.1k](https:\u002F\u002Fgithub.com\u002Fpantor\u002Fruckig)\n* 🟢 [The Kautham Project](https:\u002F\u002Fsir.upc.es\u002Fprojects\u002Fkautham\u002F) - A robot simulation toolkit for motion planning. [⭐ 24](https:\u002F\u002Fgithub.com\u002Fiocroblab\u002Fkautham)\n* 🟢 [TOPP-RA](https:\u002F\u002Fhungpham2511.github.io\u002Ftoppra\u002F) - Time-parameterizing robot trajectories subject to kinematic and dynamic constraints. [⭐ 831](https:\u002F\u002Fgithub.com\u002Fhungpham2511\u002Ftoppra)\n* 🟡 [Ungar](https:\u002F\u002Fgithub.com\u002Ffdevinc\u002Fungar) - Expressive and efficient implementation of optimal control problems using template metaprogramming. [⭐ 109](https:\u002F\u002Fgithub.com\u002Ffdevinc\u002Fungar)\n\n###### Motion Optimizer\n\n* 🔴 TopiCo - Time-optimal Trajectory Generation and Control. [⭐ 144](https:\u002F\u002Fgithub.com\u002FAIS-Bonn\u002FTopiCo)\n* 🔴 [towr](http:\u002F\u002Fwiki.ros.org\u002Ftowr) - A light-weight, Eigen-based C++ library for trajectory optimization for legged robots. [⭐ 1k](https:\u002F\u002Fgithub.com\u002Fethz-adrl\u002Ftowr)\n* 🟢 TrajectoryOptimization - A fast trajectory optimization library written in Julia. [⭐ 388](https:\u002F\u002Fgithub.com\u002FRoboticExplorationLab\u002FTrajectoryOptimization.jl)\n* 🔴 [trajopt](http:\u002F\u002Frll.berkeley.edu\u002Ftrajopt\u002Fdoc\u002Fsphinx_build\u002Fhtml\u002F) - Framework for generating robot trajectories by local optimization. [⭐ 449](https:\u002F\u002Fgithub.com\u002Fjoschu\u002Ftrajopt)\n\n###### Nearest Neighbor\n\n* 🔴 [Cover-Tree](http:\u002F\u002Fhunch.net\u002F~jl\u002Fprojects\u002Fcover_tree\u002Fcover_tree.html) - Cover tree data structure for quick k-nearest-neighbor search. [⭐ 64](https:\u002F\u002Fgithub.com\u002FDNCrane\u002FCover-Tree)\n  * [Faster cover trees](http:\u002F\u002Fproceedings.mlr.press\u002Fv37\u002Fizbicki15.pdf) - by Mike Izbicki et al., ICML 2015.\n* 🟡 [FLANN](http:\u002F\u002Fwww.cs.ubc.ca\u002Fresearch\u002Fflann\u002F) - Fast Library for Approximate Nearest Neighbors. [⭐ 2.4k](https:\u002F\u002Fgithub.com\u002Fflann-lib\u002Fflann)\n* 🟢 [nanoflann](http:\u002F\u002Fwww.cs.ubc.ca\u002Fresearch\u002Fflann\u002F) - Nearest Neighbor search with KD-trees. [⭐ 2.6k](https:\u002F\u002Fgithub.com\u002Fjlblancoc\u002Fnanoflann)\n\n###### 3D Mapping\n\n* 🟢 Bonxai - Brutally fast, sparse, 3D Voxel Grid (formerly Treexy). [⭐ 813](https:\u002F\u002Fgithub.com\u002Ffacontidavide\u002FBonxai)\n* 🟢 [Goxel](https:\u002F\u002Fguillaumechereau.github.io\u002Fgoxel\u002F) - Free and open source 3D voxel editor. [⭐ 3.1k](https:\u002F\u002Fgithub.com\u002Fguillaumechereau\u002Fgoxel)\n* 🟢 [libpointmatcher](http:\u002F\u002Flibpointmatcher.readthedocs.io\u002Fen\u002Flatest\u002F) - Iterative Closest Point library for 2-D\u002F3-D mapping in Robotics. [⭐ 1.8k](https:\u002F\u002Fgithub.com\u002Fnorlab-ulaval\u002Flibpointmatcher)\n* 🟢 [OctoMap](http:\u002F\u002Foctomap.github.io\u002F) - Efficient Probabilistic 3D Mapping Framework Based on Octrees. [⭐ 2.3k](https:\u002F\u002Fgithub.com\u002FOctoMap\u002Foctomap)\n* 🔴 Octree - Fast radius neighbor search with an Octree. [⭐ 374](https:\u002F\u002Fgithub.com\u002Fjbehley\u002Foctree)\n* 🟢 [PCL](https:\u002F\u002Fpointclouds.org\u002F) - 2D\u002F3D image and point cloud processing. [⭐ 10.9k](https:\u002F\u002Fgithub.com\u002FPointCloudLibrary\u002Fpcl)\n* Utility Software\n* 🟡 voxblox - Flexible voxel-based mapping focusing on truncated and Euclidean signed distance fields. [⭐ 1.6k](https:\u002F\u002Fgithub.com\u002Fethz-asl\u002Fvoxblox)\n* 🟡 [wavemap](https:\u002F\u002Fprojects.asl.ethz.ch\u002Fwavemap\u002F) - Fast, efficient and accurate multi-resolution, multi-sensor 3D occupancy mapping. [⭐ 551](https:\u002F\u002Fgithub.com\u002Fethz-asl\u002Fwavemap)\n\n### [Optimization](#contents)\n\n_Numerical optimization solvers and frameworks used in robotics._\n\n* 🟢 [CasADi](https:\u002F\u002Fgithub.com\u002Fcasadi\u002Fcasadi\u002Fwiki) - Symbolic framework for algorithmic differentiation and numeric optimization. [⭐ 2.1k](https:\u002F\u002Fgithub.com\u002Fcasadi\u002Fcasadi)\n* 🟢 [Ceres Solver](http:\u002F\u002Fceres-solver.org\u002F) - Large scale nonlinear optimization library. [⭐ 4.4k](https:\u002F\u002Fgithub.com\u002Fceres-solver\u002Fceres-solver)\n* 🟢 eigen-qld - Interface to use the QLD QP solver with the Eigen3 library. [⭐ 16](https:\u002F\u002Fgithub.com\u002Fjrl-umi3218\u002Feigen-qld)\n* 🟡 EXOTica - Generic optimisation toolset for robotics platforms. [⭐ 161](https:\u002F\u002Fgithub.com\u002Fipab-slmc\u002Fexotica)\n* 🟢 hpipm - High-performance interior-point-method QP solvers (Ipopt, Snopt). [⭐ 666](https:\u002F\u002Fgithub.com\u002Fgiaf\u002Fhpipm)\n* 🟢 [HYPRE](https:\u002F\u002Fhypre.readthedocs.io\u002F) - Parallel solvers for sparse linear systems featuring multigrid methods. [⭐ 815](https:\u002F\u002Fgithub.com\u002Fhypre-space\u002Fhypre)\n* 🟢 ifopt - An Eigen-based, light-weight C++ Interface to Nonlinear Programming Solvers (Ipopt, Snopt). [⭐ 847](https:\u002F\u002Fgithub.com\u002Fethz-adrl\u002Fifopt)\n* 🟢 [Ipopt](https:\u002F\u002Fprojects.coin-or.org\u002FIpopt) - Large scale nonlinear optimization library. [⭐ 1.7k](https:\u002F\u002Fgithub.com\u002Fcoin-or\u002FIpopt)\n* 🟢 libcmaes - Blackbox stochastic optimization using the CMA-ES algorithm. [⭐ 355](https:\u002F\u002Fgithub.com\u002FCMA-ES\u002Flibcmaes)\n* 🔴 [limbo](http:\u002F\u002Fwww.resibots.eu\u002Flimbo\u002F) - Gaussian processes and Bayesian optimization of black-box functions. [⭐ 261](https:\u002F\u002Fgithub.com\u002Fresibots\u002Flimbo)\n* 🟢 lpsolvers - Linear Programming solvers in Python with a unified API. [⭐ 25](https:\u002F\u002Fgithub.com\u002Fstephane-caron\u002Flpsolvers)\n* 🟢 [NLopt](https:\u002F\u002Fnlopt.readthedocs.io\u002Fen\u002Flatest\u002F) - Nonlinear optimization. [⭐ 2.2k](https:\u002F\u002Fgithub.com\u002Fstevengj\u002Fnlopt)\n* 🟡 [OptimLib](https:\u002F\u002Fwww.kthohr.com\u002Foptimlib.html) - Lightweight C++ library of numerical optimization methods for nonlinear functions. [⭐ 886](https:\u002F\u002Fgithub.com\u002Fkthohr\u002Foptim)\n* 🟢 [OSQP](https:\u002F\u002Fosqp.org\u002F) - The Operator Splitting QP Solver. [⭐ 2.1k](https:\u002F\u002Fgithub.com\u002Fosqp\u002Fosqp)\n* 🟢 [Pagmo](https:\u002F\u002Fesa.github.io\u002Fpagmo2\u002Findex.html) - Scientific library for massively parallel optimization. [⭐ 908](https:\u002F\u002Fgithub.com\u002Fesa\u002Fpagmo2)\n* 🟢 [ProxSuite](https:\u002F\u002Fsimple-robotics.github.io\u002Fproxsuite\u002F) - The Advanced Proximal Optimization Toolbox. [⭐ 539](https:\u002F\u002Fgithub.com\u002FSimple-Robotics\u002FProxSuite)\n* 🔴 [pymoo](https:\u002F\u002Fwww.pymoo.org\u002F) - Multi-objective Optimization in Python. [⭐ 26](https:\u002F\u002Fgithub.com\u002Fmsu-coinlab\u002Fpymoo)\n* 🟢 qpsolvers - Quadratic Programming solvers in Python with a unified API. [⭐ 729](https:\u002F\u002Fgithub.com\u002Fqpsolvers\u002Fqpsolvers)\n* 🟢 [RobOptim](http:\u002F\u002Froboptim.net\u002Findex.html) - Numerical Optimization for Robotics. [⭐ 65](https:\u002F\u002Fgithub.com\u002Froboptim\u002Froboptim-core)\n* 🟢 [SCS](http:\u002F\u002Fweb.stanford.edu\u002F~boyd\u002Fpapers\u002Fscs.html) - Numerical optimization for solving large-scale convex cone problems. [⭐ 614](https:\u002F\u002Fgithub.com\u002Fcvxgrp\u002Fscs)\n* 🔴 sferes2 - Evolutionary computation. [⭐ 170](https:\u002F\u002Fgithub.com\u002Fsferes2\u002Fsferes2)\n* 🟢 SHOT - A solver for mixed-integer nonlinear optimization problems. [⭐ 130](https:\u002F\u002Fgithub.com\u002Fcoin-or\u002FSHOT)\n\n### [Robot Modeling](#contents)\n\n_Tools and formats for describing robot models._\n\n###### Robot Model Description Format\n\n* [SDF](http:\u002F\u002Fsdformat.org\u002F) - XML format that describes objects and environments for robot simulators, visualization, and control. [[bitbucket](https:\u002F\u002Fbitbucket.org\u002Fosrf\u002Fsdformat)]\n* 🟢 [urdf](http:\u002F\u002Fwiki.ros.org\u002Furdf) - XML format for representing a robot model. [⭐ 119](https:\u002F\u002Fgithub.com\u002Fros\u002Furdfdom)\n\n###### Utility to Build Robot Models\n\n* 🟢 [onshape-to-robot](https:\u002F\u002Fgithub.com\u002FRhoban\u002Fonshape-to-robot) - Converting OnShape assembly to robot definition (SDF or URDF) through OnShape API. [⭐ 489](https:\u002F\u002Fgithub.com\u002FRhoban\u002Fonshape-to-robot)\n* 🟢 phobos - Add-on for Blender creating URDF and SMURF robot models. [⭐ 857](https:\u002F\u002Fgithub.com\u002Fdfki-ric\u002Fphobos)\n\n### [Robot Platform](#contents)\n\n_Middleware and frameworks for building robot software systems._\n\n* 🔴 [AutoRally](http:\u002F\u002Fautorally.github.io\u002F) - High-performance testbed for advanced perception and control research. [⭐ 775](https:\u002F\u002Fgithub.com\u002Fautorally\u002Fautorally)\n* 🔴 [Linorobot](https:\u002F\u002Flinorobot.org\u002F) - ROS compatible ground robots. [⭐ 1.1k](https:\u002F\u002Fgithub.com\u002Flinorobot\u002Flinorobot)\n  * 🔴 onine - Service Robot based on Linorobot and Braccio Arm. [⭐ 47](https:\u002F\u002Fgithub.com\u002Fgrassjelly\u002Fonine)\n* 🟡 [Micro-ROS for Arduino](https:\u002F\u002Fgithub.com\u002Fkaiaai\u002Fmicro_ros_arduino_kaiaai) - a Micro-ROS fork available in the Arduino Library Manager. [⭐ 12](https:\u002F\u002Fgithub.com\u002Fkaiaai\u002Fmicro_ros_arduino_kaiaai)\n* [Rock](https:\u002F\u002Fwww.rock-robotics.org\u002F) - Software framework for robotic systems.\n* [ROS](https:\u002F\u002Fwww.ros.org\u002F) - Flexible framework for writing robot software.\n* 🟢 [ROS 2](https:\u002F\u002Fgithub.com\u002Fros2\u002Fros2\u002Fwiki) - Version 2.0 of the Robot Operating System (ROS) software stack. [⭐ 5k](https:\u002F\u002Fgithub.com\u002Fros2\u002Fros2)\n* [ros2_medkit](https:\u002F\u002Fselfpatch.github.io\u002Fros2_medkit\u002F) - Structured fault management for ROS 2 — persistent fault lifecycle, REST\u002FSSE API, root cause correlation, and automatic rosbag capture on fault. Inspired by SOVD (Service-Oriented Vehicle Diagnostics), the ASAM standard that brings HTTP\u002FREST diagnostics to automotive and robotics. [[github](https:\u002F\u002Fgithub.com\u002Fselfpatch\u002Fros2_medkit)]\n* 🟢 [YARP](https:\u002F\u002Fwww.yarp.it\u002F) - Communication and device interfaces applicable from humanoids to embedded devices. [⭐ 585](https:\u002F\u002Fgithub.com\u002Frobotology\u002Fyarp)\n\n### [Reinforcement Learning for Robotics](#contents)\n\n_Reinforcement learning libraries commonly used in robotic control._\n\n* 🟢 [Brax](https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fbrax) - Massively parallel differentiable rigid body physics engine in JAX for robotics and RL. [⭐ 3.1k](https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fbrax)\n* 🟢 [CleanRL](https:\u002F\u002Fgithub.com\u002Fvwxyzjn\u002Fcleanrl) - Single-file implementations of deep reinforcement learning algorithms. [⭐ 9.1k](https:\u002F\u002Fgithub.com\u002Fvwxyzjn\u002Fcleanrl)\n* 🟢 [Isaac Lab](https:\u002F\u002Fisaac-sim.github.io\u002FIsaacLab) - GPU-accelerated open-source framework for robot learning built on NVIDIA Isaac Sim. [⭐ 6.3k](https:\u002F\u002Fgithub.com\u002Fisaac-sim\u002FIsaacLab)\n* 🟢 [rl_games](https:\u002F\u002Fgithub.com\u002FDenys88\u002Frl_games) - High-performance RL library used in Isaac Gym environments. [⭐ 1.3k](https:\u002F\u002Fgithub.com\u002FDenys88\u002Frl_games)\n* 🟢 [SKRL](https:\u002F\u002Fgithub.com\u002FToni-SM\u002Fskrl) - Modular reinforcement learning library with support for multiple ML frameworks. [⭐ 991](https:\u002F\u002Fgithub.com\u002FToni-SM\u002Fskrl)\n* 🟢 [Stable-Baselines3](https:\u002F\u002Fgithub.com\u002FDLR-RM\u002Fstable-baselines3) - Reliable implementations of reinforcement learning algorithms in PyTorch. [⭐ 12.7k](https:\u002F\u002Fgithub.com\u002FDLR-RM\u002Fstable-baselines3)\n\n### [SLAM](#contents)\n\n_Simultaneous Localization and Mapping libraries._\n\n\n* 🔴 AprilSAM - Real-time smoothing and mapping. [⭐ 239](https:\u002F\u002Fgithub.com\u002Fxipengwang\u002FAprilSAM)\n* 🔴 Cartographer - Real-time SLAM in 2D and 3D across multiple platforms and sensor configurations. [⭐ 7.8k](https:\u002F\u002Fgithub.com\u002Fcartographer-project\u002Fcartographer)\n* 🟡 [DSO](https:\u002F\u002Fvision.in.tum.de\u002Fresearch\u002Fvslam\u002Fdso) - Novel direct and sparse formulation for Visual Odometry. [⭐ 2.4k](https:\u002F\u002Fgithub.com\u002FJakobEngel\u002Fdso)\n* 🟢 ElasticFusion - Real-time dense visual SLAM system. [⭐ 1.9k](https:\u002F\u002Fgithub.com\u002Fmp3guy\u002FElasticFusion)\n* 🟢 [fiducials](http:\u002F\u002Fwiki.ros.org\u002Ffiducials) - Simultaneous localization and mapping using fiducial markers. [⭐ 278](https:\u002F\u002Fgithub.com\u002FUbiquityRobotics\u002Ffiducials)\n* 🟢 GTSAM - Smoothing and mapping (SAM) in robotics and vision. [⭐ 3.3k](https:\u002F\u002Fgithub.com\u002Fborglab\u002Fgtsam)\n* 🔴 Kintinuous - Real-time large scale dense visual SLAM system. [⭐ 952](https:\u002F\u002Fgithub.com\u002Fmp3guy\u002FKintinuous)\n* 🔴 [LSD-SLAM](https:\u002F\u002Fvision.in.tum.de\u002Fresearch\u002Fvslam\u002Flsdslam) - Real-time monocular SLAM. [⭐ 2.7k](https:\u002F\u002Fgithub.com\u002Ftum-vision\u002Flsd_slam)\n* 🟡 ORB-SLAM2 - Real-time SLAM library for Monocular, Stereo and RGB-D cameras. [⭐ 10.1k](https:\u002F\u002Fgithub.com\u002Fraulmur\u002FORB_SLAM2)\n* 🟢 [RTAP-Map](http:\u002F\u002Fintrolab.github.io\u002Frtabmap\u002F) - RGB-D Graph SLAM approach based on a global Bayesian loop closure detector. [⭐ 3.6k](https:\u002F\u002Fgithub.com\u002Fintrolab\u002Frtabmap)\n* 🔴 [SRBA](http:\u002F\u002Fmrpt.github.io\u002Fsrba\u002F) - Solving SLAM\u002FBA in relative coordinates with flexibility for different submapping strategies. [⭐ 76](https:\u002F\u002Fgithub.com\u002FMRPT\u002Fsrba)\n\n#### SLAM Dataset\n\n* 🟡 [Awesome SLAM Datasets](https:\u002F\u002Fgithub.com\u002Fyoungguncho\u002Fawesome-slam-datasets) - Curated list of SLAM-related datasets. [⭐ 1.9k](https:\u002F\u002Fgithub.com\u002Fyoungguncho\u002Fawesome-slam-datasets)\n\n### [Vision](#contents)\n\n_Computer vision libraries for robotic perception._\n\n* 🔴 [BundleTrack](https:\u002F\u002Fgithub.com\u002Fwenbowen123\u002FBundleTrack) - 6D Pose Tracking for Novel Objects without 3D Models. [⭐ 678](https:\u002F\u002Fgithub.com\u002Fwenbowen123\u002FBundleTrack)\n* 🔴 [se(3)-TrackNet](https:\u002F\u002Fgithub.com\u002Fwenbowen123\u002Firos20-6d-pose-tracking) - 6D Pose Tracking for Novel Objects without 3D Models. [⭐ 420](https:\u002F\u002Fgithub.com\u002Fwenbowen123\u002Firos20-6d-pose-tracking)\n* 🟢 [ViSP](http:\u002F\u002Fvisp.inria.fr\u002F) - Visual Servoing Platform. [⭐ 849](https:\u002F\u002Fgithub.com\u002Flagadic\u002Fvisp)\n\n### [Fluid](#contents)\n\n_Fluid dynamics simulation libraries._\n\n* 🔴 [Fluid Engine Dev - Jet](https:\u002F\u002Ffluidenginedevelopment.org\u002F) - Fluid simulation engine for computer graphics applications. [⭐ 2.1k](https:\u002F\u002Fgithub.com\u002Fdoyubkim\u002Ffluid-engine-dev)\n\n### [Grasping](#contents)\n\n_Libraries and tools for robotic grasping and manipulation._\n\n* 🟢 [AnyGrasp SDK](https:\u002F\u002Fgithub.com\u002Fgraspnet\u002Fanygrasp_sdk) - SDK for AnyGrasp, a 6-DoF grasp pose detection method. [⭐ 756](https:\u002F\u002Fgithub.com\u002Fgraspnet\u002Fanygrasp_sdk)\n* 🟡 [Contact-GraspNet](https:\u002F\u002Fgithub.com\u002FNVlabs\u002Fcontact_graspnet) - 6-DoF grasp generation for parallel-jaw grippers using contact maps. [⭐ 449](https:\u002F\u002Fgithub.com\u002FNVlabs\u002Fcontact_graspnet)\n* 🔴 [GraspIt!](https:\u002F\u002Fgraspit-simulator.github.io\u002F) - Simulator for grasping research that can accommodate arbitrary hand and robot designs. [⭐ 208](https:\u002F\u002Fgithub.com\u002Fgraspit-simulator\u002Fgraspit)\n* 🟢 [GraspNet API](https:\u002F\u002Fgithub.com\u002Fgraspnet\u002FgraspnetAPI) - Python API and evaluation tools for the GraspNet benchmark. [⭐ 325](https:\u002F\u002Fgithub.com\u002Fgraspnet\u002FgraspnetAPI)\n\n### [Humanoid Robotics](#contents)\n\n_Environments and models for humanoid robot research._\n\n* 🟡 [Humanoid-Gym](https:\u002F\u002Fgithub.com\u002Froboterax\u002Fhumanoid-gym) - Reinforcement learning environment for humanoid robot locomotion. [⭐ 1.8k](https:\u002F\u002Fgithub.com\u002Froboterax\u002Fhumanoid-gym)\n* 🟢 [iCub](http:\u002F\u002Fwww.icub.org\u002F) - Open-source cognitive humanoid robotic platform for embodied cognition research. [⭐ 117](https:\u002F\u002Fgithub.com\u002Frobotology\u002Ficub-main)\n* 🟢 [Legged Gym](https:\u002F\u002Fgithub.com\u002Fleggedrobotics\u002Flegged_gym) - Isaac Gym environments for legged robot locomotion training. [⭐ 2.7k](https:\u002F\u002Fgithub.com\u002Fleggedrobotics\u002Flegged_gym)\n* 🟢 [MuJoCo Menagerie](https:\u002F\u002Fgithub.com\u002Fgoogle-deepmind\u002Fmujoco_menagerie) - Collection of well-tuned MuJoCo models for research and development. [⭐ 3k](https:\u002F\u002Fgithub.com\u002Fgoogle-deepmind\u002Fmujoco_menagerie)\n\n### [Multiphysics](#contents)\n\n_Frameworks for coupled multi-physics simulations._\n\n* 🟢 [Kratos](http:\u002F\u002Fwww.cimne.com\u002Fkratos\u002F) - Framework for building parallel multi-disciplinary simulation software. [⭐ 1.2k](https:\u002F\u002Fgithub.com\u002FKratosMultiphysics\u002FKratos)\n\n### [Math](#contents)\n\n_Mathematics libraries for spatial algebra, Lie groups, and linear algebra._\n\n* 🟢 Fastor - Light-weight high performance tensor algebra framework in C++11\u002F14\u002F17. [⭐ 830](https:\u002F\u002Fgithub.com\u002Fromeric\u002FFastor)\n* 🔴 linalg.h - Single header public domain linear algebra library for C++11. [⭐ 939](https:\u002F\u002Fgithub.com\u002Fsgorsten\u002Flinalg)\n* 🟢 manif - Small c++11 header-only library for Lie theory. [⭐ 1.7k](https:\u002F\u002Fgithub.com\u002Fartivis\u002Fmanif)\n* 🟡 Sophus - Lie groups using Eigen. [⭐ 2.4k](https:\u002F\u002Fgithub.com\u002Fstrasdat\u002FSophus)\n* 🟢 SpaceVelAlg - Spatial vector algebra with the Eigen3. [⭐ 81](https:\u002F\u002Fgithub.com\u002Fjrl-umi3218\u002FSpaceVecAlg)\n* 🟢 spatialmath-python - Python classes for pose and orientation in 2D\u002F3D with spatial operations toolbox. [⭐ 617](https:\u002F\u002Fgithub.com\u002Fbdaiinstitute\u002Fspatialmath-python)\n\n### [ETC](#contents)\n\n_Other robotics-related tools and utilities._\n\n* [Foxglove Studio](https:\u002F\u002Ffoxglove.dev) - A fully integrated visualization and debugging desktop app for your robotics data.\n* 🟢 fuse - General architecture for performing sensor fusion live on a robot. [⭐ 849](https:\u002F\u002Fgithub.com\u002Flocusrobotics\u002Ffuse)\n\n## [Simulators](#contents)\n\n_Simulation environments for testing and developing robotic systems._\n\n###### Free or Open Source\n\n* 🟢 [AI2-THOR](https:\u002F\u002Fai2thor.allenai.org\u002F) - Interactive household environment for embodied AI with Unity backend. [⭐ 1.7k](https:\u002F\u002Fgithub.com\u002Fallenai\u002Fai2thor)\n* 🟢 AirSim - Simulator based on Unreal Engine for autonomous vehicles. [⭐ 17.9k](https:\u002F\u002Fgithub.com\u002FMicrosoft\u002FAirSim)\n* 🟢 [ARGoS](https:\u002F\u002Fwww.argos-sim.info\u002F) - Physics-based simulator designed to simulate large-scale robot swarms. [⭐ 301](https:\u002F\u002Fgithub.com\u002Filpincy\u002Fargos3)\n* 🟢 [ARTE](http:\u002F\u002Farvc.umh.es\u002Farte\u002Findex_en.html) - Matlab toolbox focussed on robotic manipulators. [⭐ 101](https:\u002F\u002Fgithub.com\u002F4rtur1t0\u002FARTE)\n* 🟢 [AVIS Engine](https:\u002F\u002Favisengine.com) - Fast simulation software for autonomous vehicle development. [⭐ 21](https:\u002F\u002Fgithub.com\u002FAvisEngine\u002FAVIS-Engine-Python-API)\n* 🟢 [CARLA](https:\u002F\u002Fcarla.org\u002F) - Open-source simulator for autonomous driving research. [⭐ 13.6k](https:\u002F\u002Fgithub.com\u002Fcarla-simulator\u002Fcarla)\n* 🟢 [CoppeliaSim](https:\u002F\u002Fwww.coppeliarobotics.com\u002F) - Formaly V-REP. Virtual robot experimentation platform. [⭐ 139](https:\u002F\u002Fgithub.com\u002FCoppeliaRobotics\u002FCoppeliaSimLib)\n* 💀 [Gazebo](https:\u002F\u002Fgazebosim.org\u002F) - Dynamic multi-robot simulator. [⭐ 1.3k](https:\u002F\u002Fgithub.com\u002Fgazebosim\u002Fgazebo-classic)\n* 🟢 [Gazebo Sim](https:\u002F\u002Fgazebosim.org\u002F) - Open source robotics simulator (formerly Ignition Gazebo). [⭐ 1.2k](https:\u002F\u002Fgithub.com\u002Fgazebosim\u002Fgz-sim)\n* 🔴 [GraspIt!](http:\u002F\u002Fgraspit-simulator.github.io\u002F) - Simulator for grasping research that can accommodate arbitrary hand and robot designs. [⭐ 208](https:\u002F\u002Fgithub.com\u002Fgraspit-simulator\u002Fgraspit)\n* 🟢 [Habitat-Sim](https:\u002F\u002Faihabitat.org\u002F) - Simulation platform for research in embodied artificial intelligence. [⭐ 3.5k](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002Fhabitat-sim)\n* 🟢 [Hexapod Robot Simulator](https:\u002F\u002Fhexapod.netlify.app\u002F) - Open-source hexapod robot inverse kinematics and gaits visualizer. [⭐ 703](https:\u002F\u002Fgithub.com\u002Fmithi\u002Fhexapod)\n* 🟢 [Isaac Sim](https:\u002F\u002Fdeveloper.nvidia.com\u002Fisaac\u002Fsim) - NVIDIA's GPU-accelerated robotics simulation platform with PhysX 5 and RTX rendering. [⭐ 2.5k](https:\u002F\u002Fgithub.com\u002Fisaac-sim\u002FIsaacSim)\n* 🟢 [ManiSkill](https:\u002F\u002Fgithub.com\u002Fhaosulab\u002FManiSkill) - Robot simulation and manipulation learning package powered by SAPIEN. [⭐ 2.5k](https:\u002F\u002Fgithub.com\u002Fhaosulab\u002FManiSkill)\n* 🔴 [MORSE](http:\u002F\u002Fmorse-simulator.github.io\u002F) - Modular open robots simulation engine. [⭐ 370](https:\u002F\u002Fgithub.com\u002Fmorse-simulator\u002Fmorse)\n* [Neurorobotics Platform](https:\u002F\u002Fneurorobotics.net\u002F) - Internet-accessible simulation of robots controlled by spiking neural networks. [[bitbucket](https:\u002F\u002Fbitbucket.org\u002Fhbpneurorobotics\u002Fneurorobotics-platform)]\n* 🟢 [PyBullet](https:\u002F\u002Fdocs.google.com\u002Fdocument\u002Fd\u002F10sXEhzFRSnvFcl3XxNGhnD4N2SedqwdAvK3dsihxVUA\u002Fedit#heading=h.2ye70wns7io3) - An easy to use simulator for robotics and deep reinforcement learning. [⭐ 14.2k](https:\u002F\u002Fgithub.com\u002Fbulletphysics\u002Fbullet3)\n* 🟢 [PyBullet_Industrial](https:\u002F\u002Fpybullet-industrial.readthedocs.io\u002Fen\u002Flatest\u002F) - PyBullet extension for simulating robotic manufacturing processes like milling and 3D printing. [⭐ 49](https:\u002F\u002Fgithub.com\u002FWBK-Robotics\u002Fpybullet_industrial)\n* 🔴 [Robot Gui](http:\u002F\u002Frobot.glumb.de\u002F) - A three.js based 3D robot interface. [⭐ 384](https:\u002F\u002Fgithub.com\u002Fglumb\u002Frobot-gui)\n* 🟢 [SAPIEN](https:\u002F\u002Fsapien.ucsd.edu) - Physics-rich simulation environment for articulated objects and manipulation. [⭐ 718](https:\u002F\u002Fgithub.com\u002Fhaosulab\u002FSAPIEN)\n* [Simbad](http:\u002F\u002Fsimbad.sourceforge.net\u002F) - Java 3D robot simulator with custom controller and sensor support.\n* 🟡 [Unity](https:\u002F\u002Funity.com\u002Fsolutions\u002Fautomotive-transportation-manufacturing\u002Frobotics) - Game engine with open-source robotics simulation tools and tutorials. [⭐ 2.5k](https:\u002F\u002Fgithub.com\u002FUnity-Technologies\u002FUnity-Robotics-Hub)\n* 🟢 [Webots](http:\u002F\u002Fwww.cyberbotics.com\u002F) - Development environment to model, program, and simulate robots and mechanical systems. [⭐ 4.1k](https:\u002F\u002Fgithub.com\u002Fcyberbotics\u002Fwebots)\n\n###### Commercial\n\n* [Actin Simulation](http:\u002F\u002Fwww.energid.com\u002F) - Real-time robot simulation and control software.\n* [Artiminds](https:\u002F\u002Fwww.artiminds.com\u002F) - Planning, programming, operation, analysis and optimization.\n* [Kineo](https:\u002F\u002Fwww.plm.automation.siemens.com\u002Fglobal\u002Fen\u002Fproducts\u002Fplm-components\u002Fkineo.html) - Path planning and trajectory optimization for industrial robotics.\n* [Robot Virtual Worlds](http:\u002F\u002Fwww.robotvirtualworlds.com\u002F) - Virtual reality software for educational robotics.\n* [RobotDK](https:\u002F\u002Frobodk.com\u002F) - Simulation and OLP for robots.\n* [RobotStudio](https:\u002F\u002Fwww.abb.com\u002Fglobal\u002Fen\u002Fareas\u002Frobotics\u002Fproducts\u002Fsoftware\u002Frobotstudio-suite) - ABB's simulation and offline programming software for robotics.\n* [Virtual Robotics Toolkit](https:\u002F\u002Fwww.virtualroboticstoolkit.com\u002F) - 3D virtual environment for programming and testing robots.\n* [Visual Components](https:\u002F\u002Fwww.visualcomponents.com\u002F) - 3D manufacturing simulation and visualization platform.\n\n###### Cloud\n\n* [AWS RoboMaker](https:\u002F\u002Faws.amazon.com\u002Frobomaker\u002F) - Service that makes it easy to develop, test, and deploy intelligent robotics applications at scale.\n\n## [Other Awesome Lists](#contents)\n\n_Related curated lists of robotics and AI resources._\n\n* [Awesome Robotics](https:\u002F\u002Fgithub.com\u002FKiloreux\u002Fawesome-robotics) - Curated list of robotics resources (Kiloreux).\n* [Awesome Robotics](https:\u002F\u002Fgithub.com\u002Fahundt\u002Fawesome-robotics) - Curated list of robotics links and software libraries (ahundt).\n* [Awesome Robotic Tooling](https:\u002F\u002Fgithub.com\u002FLy0n\u002Fawesome-robotic-tooling) - Tooling for professional robotic development in C++ and Python with ROS and autonomous driving.\n* [Awesome Artificial Intelligence](https:\u002F\u002Fgithub.com\u002Fowainlewis\u002Fawesome-artificial-intelligence) - Curated list of AI courses, books, video lectures, and papers.\n* [Awesome Collision Detection](https:\u002F\u002Fgithub.com\u002Fjslee02\u002Fawesome-collision-detection) - Curated list of collision detection libraries and resources.\n* [Awesome Computer Vision](https:\u002F\u002Fgithub.com\u002Fjbhuang0604\u002Fawesome-computer-vision) - Curated list of computer vision resources.\n* [Awesome Machine Learning](https:\u002F\u002Fgithub.com\u002Fjosephmisiti\u002Fawesome-machine-learning) - Curated list of machine learning frameworks, libraries, and software.\n* [Awesome Deep Learning](https:\u002F\u002Fgithub.com\u002FChristosChristofidis\u002Fawesome-deep-learning) - Curated list of deep learning tutorials, projects, and communities.\n* [Awesome Gazebo](https:\u002F\u002Fgithub.com\u002Ffkromer\u002Fawesome-gazebo) - Curated list of Gazebo simulation framework resources.\n* [Awesome Grasping](https:\u002F\u002Fgithub.com\u002FPo-Jen\u002Fawesome-grasping) - Curated list of grasping libraries and resources.\n* [Awesome Human Robot Interaction](https:\u002F\u002Fgithub.com\u002FPo-Jen\u002Fawesome-human-robot-interaction) - Curated list of human-robot interaction libraries and resources.\n* [Awesome Physical AI](https:\u002F\u002Fgithub.com\u002Fkeon\u002Fawesome-physical-ai) - Curated list of academic papers and resources on Physical AI.\n* [PythonRobotics](https:\u002F\u002Fgithub.com\u002FAtsushiSakai\u002FPythonRobotics) - Python sample codes for robotics algorithms.\n* [Robotics Coursework](https:\u002F\u002Fgithub.com\u002Fmithi\u002Frobotics-coursework) - A list of robotics courses you can take online.\n\n## [Contributing](#contents)\n\nContributions are very welcome! Please read the [contribution guidelines](https:\u002F\u002Fgithub.com\u002Fjslee02\u002Fawesome-robotics-libraries\u002Fblob\u002Fmain\u002FCONTRIBUTING.md) first. Also, please feel free to report any error.\n\n## [Star History](#contents)\n\n[![Star History Chart](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Fjslee02_awesome-robotics-libraries_readme_4815b16021eb.png)](https:\u002F\u002Fstar-history.com\u002F#jslee02\u002Fawesome-robotics-libraries)\n\n## [License](#contents)\n\n[![CC0](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Fjslee02_awesome-robotics-libraries_readme_b7657951a0bb.png)](http:\u002F\u002Fcreativecommons.org\u002Fpublicdomain\u002Fzero\u002F1.0\u002F)\n","# 优秀的机器人库\n\n[![Awesome](https:\u002F\u002Fawesome.re\u002Fbadge.svg)](https:\u002F\u002Fawesome.re)\n\n一份精选的机器人库和仿真器列表。\n\n## 目录\n* [库](#libraries)\n  * [动力学仿真](#dynamics-simulation)\n  * [逆运动学](#inverse-kinematics)\n  * [机器学习](#machine-learning)\n  * [运动规划与控制](#motion-planning-and-control)\n  * [优化](#optimization)\n  * [机器人建模](#robot-modeling)\n  * [机器人平台](#robot-platform)\n  * [机器人强化学习](#reinforcement-learning-for-robotics)\n  * [SLAM](#slam)\n  * [视觉](#vision)\n  * [流体](#fluid)\n  * [抓取](#grasping)\n  * [类人机器人](#humanoid-robotics)\n  * [多物理场](#multiphysics)\n  * [数学](#math)\n  * [其他](#etc)\n* [仿真器](#simulators)\n* [其他 Awesome 列表](#other-awesome-lists)\n\n> **图例**: 🟢 活跃 (\u003C1年) · 🟡 缓慢 (1-2年) · 🔴 过时 (>2年) · 💀 已归档\n\n## [库](#contents)\n\n### [动力学仿真](#contents)\n\n_物理引擎和刚体\u002F柔性体动力学库。另请参阅 [比较](COMPARISONS.md)。_\n\n* [ARCSim](http:\u002F\u002Fgraphics.berkeley.edu\u002Fresources\u002FARCSim\u002Findex.html) - 用于薄型可变形物体的自适应重网格布料与壳体模拟器。\n* 🟢 [Brax](https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fbrax) - 基于 JAX 的大规模并行可微刚体物理引擎，适用于机器人与强化学习。[⭐ 3.1k](https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fbrax)\n* 🟢 [Bullet](https:\u002F\u002Fpybullet.org\u002F) - 用于游戏、视觉特效和机器人领域的实时物理仿真。[⭐ 14.2k](https:\u002F\u002Fgithub.com\u002Fbulletphysics\u002Fbullet3)\n* 🟢 [CHRONO::ENGINE](https:\u002F\u002Fprojectchrono.org\u002F) - 刚体、柔性体、颗粒及流体系统的多物理场仿真平台。[⭐ 2.7k](https:\u002F\u002Fgithub.com\u002Fprojectchrono\u002Fchrono)\n* 🟢 [DART](http:\u002F\u002Fdartsim.github.io\u002F) - 多体仿真与规划的动态动画与机器人工具包。[⭐ 1.1k](https:\u002F\u002Fgithub.com\u002Fdartsim\u002Fdart)\n* 🟢 [Drake](https:\u002F\u002Fdrake.mit.edu\u002F) - 非线性动力学系统的规划、控制与分析工具箱。[⭐ 3.9k](https:\u002F\u002Fgithub.com\u002FRobotLocomotion\u002Fdrake)\n* 💀 [Flex](https:\u002F\u002Fdeveloper.nvidia.com\u002Fflex) - 基于 GPU 的粒子仿真，适用于刚体、流体和可变形体。[⭐ 788](https:\u002F\u002Fgithub.com\u002FNVIDIAGameWorks\u002FFleX)\n* 🔴 [FROST](https:\u002F\u002Fayonga.github.io\u002Ffrost-dev\u002Findex.html) - MATLAB 中用于混合动力系统快速机器人优化与仿真工具包。[⭐ 170](https:\u002F\u002Fgithub.com\u002Fayonga\u002Ffrost-dev)\n* 🟢 [Genesis](https:\u002F\u002Fgenesis-world.readthedocs.io) - 面向机器人领域的生成式通用物理平台，支持 GPU 加速的并行仿真。[⭐ 28.1k](https:\u002F\u002Fgithub.com\u002FGenesis-Embodied-AI\u002FGenesis)\n* [IBDS](http:\u002F\u002Fwww.interactive-graphics.de\u002Findex.php\u002Fdownloads\u002F12-ibds) - 基于冲量的刚体与粒子系统动力学仿真。\n* 🟢 idyntree - 用于浮基机器人估计与全身动力学的库。[⭐ 224](https:\u002F\u002Fgithub.com\u002Fgbionics\u002Fidyntree)\n* 🟢 [KDL](https:\u002F\u002Fwww.orocos.org\u002Fkdl.html) - Orocos 运动学与动力学库，用于运动学链。[⭐ 859](https:\u002F\u002Fgithub.com\u002Forocos\u002Forocos_kinematics_dynamics)\n* 🟢 kindr - 刚体变换的运动学与动力学库。[⭐ 607](https:\u002F\u002Fgithub.com\u002FANYbotics\u002Fkindr)\n* 🟢 [Klampt](https:\u002F\u002Fklampt.org\u002F) - 支持可视化的机器人规划、控制与仿真。[⭐ 427](https:\u002F\u002Fgithub.com\u002Fkrishauser\u002FKlampt)\n* 🔴 [LibrePilot](http:\u002F\u002Fwww.librepilot.org\u002Fsite\u002Findex.html) - 用于无人机及其他自主车辆的开源自动驾驶仪。[⭐ 348](https:\u002F\u002Fgithub.com\u002Flibrepilot\u002FLibrePilot)\n* 🟢 [MARS](http:\u002F\u002Frock-simulation.github.io\u002Fmars\u002F) - Machina Arte Robotum Simulans — 跨平台仿真环境。[⭐ 67](https:\u002F\u002Fgithub.com\u002Frock-simulation\u002Fmars)\n* [MBDyn](https:\u002F\u002Fwww.mbdyn.org\u002F) - 通用多体动力学分析软件。[[代码](https:\u002F\u002Fwww.mbdyn.org\u002F?Software_Download)]\n* 🟢 [MBSim](https:\u002F\u002Fwww.mbsim-env.de\u002F) - 用于柔性与刚性系统的多体仿真环境。[⭐ 51](https:\u002F\u002Fgithub.com\u002Fmbsim-env\u002Fmbsim)\n* 🔴 [MBSlib](http:\u002F\u002Fwww.sim.informatik.tu-darmstadt.de\u002Fres\u002Fsw\u002Fmbslib) - 轻量级多体系统动力学库。[⭐ 11](https:\u002F\u002Fgithub.com\u002FSIM-TU-Darmstadt\u002Fmbslib)\n* 💀 metapod - 基于模板的空间代数机器人动力学库。[⭐ 14](https:\u002F\u002Fgithub.com\u002Flaas\u002Fmetapod)\n* 🔴 [Moby](http:\u002F\u002Fphyssim.sourceforge.net\u002Findex.html) - 带接触的刚体多体动力学仿真。[⭐ 37](https:\u002F\u002Fgithub.com\u002FPositronicsLab\u002FMoby)\n* 🟢 [mrpt](https:\u002F\u002Fwww.mrpt.org\u002F) - 移动机器人编程工具箱，用于 SLAM、导航与计算机视觉。[⭐ 2.1k](https:\u002F\u002Fgithub.com\u002FMRPT\u002Fmrpt)\n* 🟢 [MuJoCo](https:\u002F\u002Fmujoco.org\u002F) - 基于物理的仿真与控制中的多关节接触动力学。[⭐ 12k](https:\u002F\u002Fgithub.com\u002Fgoogle-deepmind\u002Fmujoco)\n* 🟢 [mvsim](http:\u002F\u002Fwiki.ros.org\u002Fmvsim) - 轻量级多车 2D 模拟器，集成 ROS。[⭐ 361](https:\u002F\u002Fgithub.com\u002FMRPT\u002Fmvsim)\n* 🟢 [Newton](https:\u002F\u002Fnewton-physics.github.io\u002Fnewton\u002F) - 基于 NVIDIA Warp 构建的 GPU 加速可微物理引擎，用于机器人仿真。[⭐ 2.5k](https:\u002F\u002Fgithub.com\u002Fnewton-physics\u002Fnewton)\n* 🟢 [Newton Dynamics](https:\u002F\u002Fnewtondynamics.com\u002F) - 实时刚体仿真物理引擎。[⭐ 1k](https:\u002F\u002Fgithub.com\u002FMADEAPPS\u002Fnewton-dynamics)\n* 🔴 [nphysics](https:\u002F\u002Fnphysics.org\u002F) - 使用 Rust 编写的 2D 和 3D 刚体物理引擎。[⭐ 1.6k](https:\u002F\u002Fgithub.com\u002Fdimforge\u002Fnphysics)\n* [ODE](https:\u002F\u002Fode.org\u002F) - 开放式动力学引擎，用于模拟刚体动力学。[[bitbucket](https:\u002F\u002Fbitbucket.org\u002Fodedevs\u002Fode)]\n* 🟢 [OpenRAVE](https:\u002F\u002Fwww.openrave.org\u002F) - 开放式机器人自动化虚拟环境，用于规划和仿真。[⭐ 798](https:\u002F\u002Fgithub.com\u002Frdiankov\u002Fopenrave)\n* 🟢 [PhysX](https:\u002F\u002Fnvidia-omniverse.github.io\u002FPhysX\u002Fphysx\u002F5.5.0\u002Findex.html) - NVIDIA 物理引擎，用于实时刚体与车辆仿真。[⭐ 4.4k](https:\u002F\u002Fgithub.com\u002FNVIDIA-Omniverse\u002FPhysX)\n* 🟢 [pinocchio](https:\u002F\u002Fstack-of-tasks.github.io\u002Fpinocchio\u002F) - 基于解析导数的快速灵活刚体动力学算法。[⭐ 3.1k](https:\u002F\u002Fgithub.com\u002Fstack-of-tasks\u002Fpinocchio)\n* 🟢 PositionBasedDynamics - 基于位置的方法，用于模拟可变形物体和流体。[⭐ 2.2k](https:\u002F\u002Fgithub.com\u002FInteractiveComputerGraphics\u002FPositionBasedDynamics)\n* 🟢 [PyDy](https:\u002F\u002Fwww.pydy.org\u002F) - 使用 SymPy 的符号 Python 进行多体动力学分析。[⭐ 405](https:\u002F\u002Fgithub.com\u002Fpydy\u002Fpydy)\n* 💀 qu3e - 轻量级 3D 物理引擎，用于刚体动力学。[⭐ 976](https:\u002F\u002Fgithub.com\u002FRandyGaul\u002Fqu3e)\n* 💀 [RaiSim](https:\u002F\u002Fslides.com\u002Fjeminhwangbo\u002Fraisim-manual) - 跨平台物理引擎，适用于机器人与强化学习。[⭐ 329](https:\u002F\u002Fgithub.com\u002Fleggedrobotics\u002FraisimLib)\n* 🟢 [RBDL](https:\u002F\u002Frbdl.github.io\u002F) - 基于 Featherstone 算法的刚体动力学库。[⭐ 684](https:\u002F\u002Fgithub.com\u002Frbdl\u002Frbdl)\n* 🟢 RBDyn - 使用 Eigen 的空间代数进行刚体动力学计算的算法。[⭐ 211](https:\u002F\u002Fgithub.com\u002Fjrl-umi3218\u002FRBDyn)\n* 🟢 [ReactPhysics3d](https:\u002F\u002Fwww.reactphysics3d.com\u002F) - 开源 3D 物理引擎，用于刚体仿真与碰撞检测。[⭐ 1.7k](https:\u002F\u002Fgithub.com\u002FDanielChappuis\u002Freactphysics3d)\n* 🟡 RigidBodyDynamics.jl - Julia 库，用于刚体动力学算法。[⭐ 308](https:\u002F\u002Fgithub.com\u002FJuliaRobotics\u002FRigidBodyDynamics.jl)\n* 🟢 [Rigs of Rods](https:\u002F\u002Fwww.rigsofrods.org\u002F) - 使用梁理论的软体车辆模拟器。[⭐ 1.2k](https:\u002F\u002Fgithub.com\u002FRigsOfRods\u002Frigs-of-rods)\n* 🔴 [Robopy](https:\u002F\u002Fadityadua24.github.io\u002Frobopy\u002F) - 受 Peter Corke 的 Robotics Toolbox 启发的 Python 机器人工具箱。[⭐ 228](https:\u002F\u002Fgithub.com\u002Fadityadua24\u002Frobopy)\n* 🟢 [Robotics Library](https:\u002F\u002Fwww.roboticslibrary.org\u002F) - 自包含的 C++ 库，用于机器人运动学、规划与控制。[⭐ 1.2k](https:\u002F\u002Fgithub.com\u002Froboticslibrary\u002Frl)\n* [RobWork](https:\u002F\u002Frobwork.dk\u002F) - 用于机器人系统仿真与控制的框架。[[gitlab](https:\u002F\u002Fgitlab.com\u002Fsdurobotics\u002FRobWork)]\n* 🟢 [siconos](https:\u002F\u002Fnonsmooth.gricad-pages.univ-grenoble-alpes.fr\u002Fsiconos\u002F) - 用于非光滑动力系统建模与仿真的平台。[⭐ 183](https:\u002F\u002Fgithub.com\u002Fsiconos\u002Fsiconos)\n* 🟢 [Simbody](https:\u002F\u002Fsimtk.org\u002Fhome\u002Fsimbody\u002F) - 用于生物力学与机械系统的多体动力学库。[⭐ 2.5k](https:\u002F\u002Fgithub.com\u002Fsimbody\u002Fsimbody)\n* 🟢 [SOFA](https:\u002F\u002Fwww.sofa-framework.org\u002F) - 医疗与物理仿真的开放框架架构。[⭐ 1.1k](https:\u002F\u002Fgithub.com\u002Fsofa-framework\u002Fsofa)\n* 🟡 Tiny Differentiable Simulator - 仅头文件的可微物理引擎，专为机器人设计。[⭐ 1.3k](https:\u002F\u002Fgithub.com\u002Ferwincoumans\u002Ftiny-differentiable-simulator)\n* 🔴 [trep](http:\u002F\u002Fmurpheylab.github.io\u002Ftrep\u002F) - 使用变分积分器进行仿真与最优控制。[⭐ 20](https:\u002F\u002Fgithub.com\u002FMurpheyLab\u002Ftrep)\n\n### [逆运动学](#contents)\n\n_用于从末端执行器位姿计算关节配置的库。_\n\n  * 🟢 IKBT - 一个以符号形式求解机器人手臂逆运动学问题的 Python 包。[⭐ 215](https:\u002F\u002Fgithub.com\u002Fuw-biorobotics\u002FIKBT)\n  * 🟢 Kinpy - 一个简单纯Python包，用于求解逆运动学问题。[⭐ 179](https:\u002F\u002Fgithub.com\u002Fneka-nat\u002Fkinpy)\n  * 🔴 Lively - 一个高度可配置的工具包，用于以混合模态方式控制机器人。[⭐ 7](https:\u002F\u002Fgithub.com\u002FWisc-HCI\u002Flively)\n  * 🔴 RelaxedIK - 实时合成精确且可行的机器人手臂运动。[⭐ 236](https:\u002F\u002Fgithub.com\u002Fuwgraphics\u002Frelaxed_ik)\n  * 🔴 [Trip](https:\u002F\u002Ftrip-kinematics.readthedocs.io\u002Fen\u002Fmain\u002Findex.html) - 一个Python包，用于求解并联、串联或混合型机器人的逆运动学问题。[⭐ 44](https:\u002F\u002Fgithub.com\u002FTriPed-Robot\u002Ftrip_kinematics)\n\n### [机器学习](#contents)\n\n_应用于机器人领域的机器学习框架和工具。_\n\n* 🟢 [AllenAct](https:\u002F\u002Fallenact.org\u002F) - 基于 Python\u002FPyTorch 的具身智能研究框架。[⭐ 376](https:\u002F\u002Fgithub.com\u002Fallenai\u002Fallenact)\n* 🟢 Any4LeRobot - LeRobot 的一系列实用工具和资源。[⭐ 856](https:\u002F\u002Fgithub.com\u002FTavish9\u002Fany4lerobot)\n* 🟢 DLL - C++ 深度学习库 (DLL)。[⭐ 687](https:\u002F\u002Fgithub.com\u002Fwichtounet\u002Fdll)\n* 🔴 [DyNet](https:\u002F\u002Fdynet.readthedocs.io\u002Fen\u002Flatest\u002F) - 动态神经网络工具包。[⭐ 3.4k](https:\u002F\u002Fgithub.com\u002Fclab\u002Fdynet)\n* 🔴 [Fido](http:\u002F\u002Ffidoproject.github.io\u002F) - 面向嵌入式电子设备和机器人应用的轻量级 C++ 机器学习库。[⭐ 462](https:\u002F\u002Fgithub.com\u002FFidoProject\u002FFido)\n* 🟢 [Gymnasium](https:\u002F\u002Fgymnasium.farama.org\u002F) - 用于开发和比较强化学习算法。[⭐ 11.3k](https:\u002F\u002Fgithub.com\u002FFarama-Foundation\u002FGymnasium)\n  * 🔴 gym-dart - 使用 DART 物理引擎的 OpenAI Gym 环境。[⭐ 141](https:\u002F\u002Fgithub.com\u002FDartEnv\u002Fdart-env)\n  * 💀 gym-gazebo - Gazebo 模拟器中的 OpenAI Gym 环境。[⭐ 845](https:\u002F\u002Fgithub.com\u002Ferlerobot\u002Fgym-gazebo)\n* 🟢 [Ivy](https:\u002F\u002Flets-unify.ai\u002F) - 统一的机器学习框架。[⭐ 14.2k](https:\u002F\u002Fgithub.com\u002Fivy-llc\u002Fivy)\n* 🟢 LeRobot - 在 PyTorch 中为现实世界机器人技术提供的预训练模型、数据集和仿真环境。[⭐ 21.5k](https:\u002F\u002Fgithub.com\u002Fhuggingface\u002Flerobot)\n* 🟢 [LeRobot 任务评分工具包](https:\u002F\u002Fgithub.com\u002FRoboticsData\u002Fscore_lerobot_episodes) - 一键式工具，用于对 LeRobot 数据集进行评分、筛选和导出更高质量的数据。[⭐ 48](https:\u002F\u002Fgithub.com\u002FRoboticsData\u002Fscore_lerobot_episodes)\n* 🔴 MiniDNN - 一个仅包含头文件的 C++ 深度神经网络库。[⭐ 431](https:\u002F\u002Fgithub.com\u002Fyixuan\u002FMiniDNN)\n* 🟢 [mlpack](https:\u002F\u002Fwww.mlpack.org\u002F) - 可扩展的 C++ 机器学习库。[⭐ 5.6k](https:\u002F\u002Fgithub.com\u002Fmlpack\u002Fmlpack)\n* 🔴 RLLib - 强化学习中的时序差分学习算法。[⭐ 208](https:\u002F\u002Fgithub.com\u002Fsamindaa\u002FRLLib)\n* 🟢 [robosuite](https:\u002F\u002Frobosuite.ai) - 一个模块化的仿真框架和机器人学习基准测试平台。[⭐ 2.2k](https:\u002F\u002Fgithub.com\u002FARISE-Initiative\u002Frobosuite)\n* 🔴 [tiny-dnn](http:\u002F\u002Ftiny-dnn.readthedocs.io\u002Fen\u002Flatest\u002F) - 一个仅包含头文件、无依赖项的 C++14 深度学习框架。[⭐ 6k](https:\u002F\u002Fgithub.com\u002Ftiny-dnn\u002Ftiny-dnn)\n\n### [运动规划与控制](#contents)\n\n_用于机器人运动规划、轨迹优化和控制的库。_\n\n\n* 🔴 [AIKIDO](https:\u002F\u002Fgithub.com\u002Fpersonalrobotics\u002Faikido) - 解决机器人运动规划和决策问题。[⭐ 229](https:\u002F\u002Fgithub.com\u002Fpersonalrobotics\u002Faikido)\n* 🟢 Bioptim - 生物力学中用于肌肉骨骼最优控制的 Python 框架。[⭐ 114](https:\u002F\u002Fgithub.com\u002Fpyomeca\u002Fbioptim)\n* 🔴 [Control Toolbox](https:\u002F\u002Fethz-adrl.github.io\u002Fct\u002F) - 面向机器人、最优控制和模型预测控制的开源 C++ 库。[⭐ 1.7k](https:\u002F\u002Fgithub.com\u002Fethz-adrl\u002Fcontrol-toolbox)\n* 🟢 Crocoddyl - 适用于接触序列下机器人控制的最优控制库。[⭐ 1.2k](https:\u002F\u002Fgithub.com\u002Floco-3d\u002Fcrocoddyl)\n* [CuiKSuite](http:\u002F\u002Fwww.iri.upc.edu\u002Fpeople\u002Fporta\u002FSoft\u002FCuikSuite2-Doc\u002Fhtml) - 用于解决位置分析和路径规划问题的应用程序。\n* 🟢 [cuRobo](https:\u002F\u002Fcurobo.org) - 一个基于 CUDA 加速的库，包含一系列运行速度显著更快的机器人算法。[⭐ 1.3k](https:\u002F\u002Fgithub.com\u002Fnvlabs\u002Fcurobo)\n* 🟢 Fields2Cover - 用于自主农业车辆的鲁棒高效覆盖路径生成。[⭐ 751](https:\u002F\u002Fgithub.com\u002Ffields2cover\u002Ffields2cover)\n* 🔴 GPMP2 - 高斯过程运动规划器 2。[⭐ 351](https:\u002F\u002Fgithub.com\u002Fgtrll\u002Fgpmp2)\n* [HPP](https:\u002F\u002Fhumanoid-path-planner.github.io\u002Fhpp-doc\u002F) - 用于在障碍物密集环境中进行运动链路径规划。\n* 🟢 [MoveIt!](https:\u002F\u002Fmoveit.ai\u002F) - 运动规划框架。[⭐ 2k](https:\u002F\u002Fgithub.com\u002Fmoveit\u002Fmoveit)\n* 🟢 OCS2 - 高效的连续和离散时间最优控制实现。[⭐ 1.3k](https:\u002F\u002Fgithub.com\u002Fleggedrobotics\u002Focs2)\n* 🟢 [OMPL](https:\u002F\u002Fompl.kavrakilab.org\u002F) - 开放式运动规划库。[⭐ 1.9k](https:\u002F\u002Fgithub.com\u002Fompl\u002Fompl)\n* 💀 pymanoid - 基于 OpenRAVE 的人形机器人原型开发环境。[⭐ 232](https:\u002F\u002Fgithub.com\u002Fstephane-caron\u002Fpymanoid)\n* 🔴 ROS Behavior Tree - 基于 ROS 的机器人任务规划行为树实现。[⭐ 362](https:\u002F\u002Fgithub.com\u002Fmiccol\u002FROS-Behavior-Tree)\n* 🟢 [Ruckig](https:\u002F\u002Fgithub.com\u002Fpantor\u002Fruckig) - 实时、时间最优且加速度受限的在线轨迹生成。[⭐ 1.1k](https:\u002F\u002Fgithub.com\u002Fpantor\u002Fruckig)\n* 🟢 [The Kautham Project](https:\u002F\u002Fsir.upc.es\u002Fprojects\u002Fkautham\u002F) - 用于运动规划的机器人仿真工具包。[⭐ 24](https:\u002F\u002Fgithub.com\u002Fiocroblab\u002Fkautham)\n* 🟢 [TOPP-RA](https:\u002F\u002Fhungpham2511.github.io\u002Ftoppra\u002F) - 在运动学和动力学约束下对机器人轨迹进行时间参数化。[⭐ 831](https:\u002F\u002Fgithub.com\u002Fhungpham2511\u002Ftoppra)\n* 🟡 [Ungar](https:\u002F\u002Fgithub.com\u002Ffdevinc\u002Fungar) - 使用模板元编程高效表达并求解最优控制问题。[⭐ 109](https:\u002F\u002Fgithub.com\u002Ffdevinc\u002Fungar)\n\n###### 运动优化器\n\n* 🔴 TopiCo - 时间最优轨迹生成与控制。[⭐ 144](https:\u002F\u002Fgithub.com\u002FAIS-Bonn\u002FTopiCo)\n* 🔴 [towr](http:\u002F\u002Fwiki.ros.org\u002Ftowr) - 一个轻量级、基于 Eigen 的 C++ 库，用于足式机器人的轨迹优化。[⭐ 1k](https:\u002F\u002Fgithub.com\u002Fethz-adrl\u002Ftowr)\n* 🟢 TrajectoryOptimization - 用 Julia 编写的快速轨迹优化库。[⭐ 388](https:\u002F\u002Fgithub.com\u002FRoboticExplorationLab\u002FTrajectoryOptimization.jl)\n* 🔴 [trajopt](http:\u002F\u002Frll.berkeley.edu\u002Ftrajopt\u002Fdoc\u002Fsphinx_build\u002Fhtml\u002F) - 通过局部优化生成机器人轨迹的框架。[⭐ 449](https:\u002F\u002Fgithub.com\u002Fjoschu\u002Ftrajopt)\n\n###### 最近邻\n\n* 🔴 [Cover-Tree](http:\u002F\u002Fhunch.net\u002F~jl\u002Fprojects\u002Fcover_tree\u002Fcover_tree.html) - 用于快速 k 最近邻搜索的覆盖树数据结构。[⭐ 64](https:\u002F\u002Fgithub.com\u002FDNCrane\u002FCover-Tree)\n  * [更快的覆盖树](http:\u002F\u002Fproceedings.mlr.press\u002Fv37\u002Fizbicki15.pdf) - 由 Mike Izbicki 等人在 ICML 2015 上提出。\n* 🟡 [FLANN](http:\u002F\u002Fwww.cs.ubc.ca\u002Fresearch\u002Fflann\u002F) - 快速近似最近邻库。[⭐ 2.4k](https:\u002F\u002Fgithub.com\u002Fflann-lib\u002Fflann)\n* 🟢 [nanoflann](http:\u002F\u002Fwww.cs.ubc.ca\u002Fresearch\u002Fflann\u002F) - 使用 KD 树进行最近邻搜索。[⭐ 2.6k](https:\u002F\u002Fgithub.com\u002Fjlblancoc\u002Fnanoflann)\n\n###### 3D 地图构建\n\n* 🟢 Bonxai - 极其快速、稀疏的 3D 体素网格（前身为 Treexy）。[⭐ 813](https:\u002F\u002Fgithub.com\u002Ffacontidavide\u002FBonxai)\n* 🟢 [Goxel](https:\u002F\u002Fguillaumechereau.github.io\u002Fgoxel\u002F) - 免费开源的 3D 体素编辑器。[⭐ 3.1k](https:\u002F\u002Fgithub.com\u002Fguillaumechereau\u002Fgoxel)\n* 🟢 [libpointmatcher](http:\u002F\u002Flibpointmatcher.readthedocs.io\u002Fen\u002Flatest\u002F) - 用于机器人领域 2D\u002F3D 地图构建的迭代最近点库。[⭐ 1.8k](https:\u002F\u002Fgithub.com\u002Fnorlab-ulaval\u002Flibpointmatcher)\n* 🟢 [OctoMap](http:\u002F\u002Foctomap.github.io\u002F) - 基于八叉树的高效概率 3D 地图构建框架。[⭐ 2.3k](https:\u002F\u002Fgithub.com\u002FOctoMap\u002Foctomap)\n* 🔴 Octree - 使用八叉树进行快速半径邻域搜索。[⭐ 374](https:\u002F\u002Fgithub.com\u002Fjbehley\u002Foctree)\n* 🟢 [PCL](https:\u002F\u002Fpointclouds.org\u002F) - 用于 2D\u002F3D 图像和点云处理。[⭐ 10.9k](https:\u002F\u002Fgithub.com\u002FPointCloudLibrary\u002Fpcl)\n* 工具软件\n* 🟡 voxblox - 基于体素的灵活地图构建，专注于截断和欧几里得符号距离场。[⭐ 1.6k](https:\u002F\u002Fgithub.com\u002Fethz-asl\u002Fvoxblox)\n* 🟡 [wavemap](https:\u002F\u002Fprojects.asl.ethz.ch\u002Fwavemap\u002F) - 快速、高效且精确的多分辨率、多传感器 3D 占用地图构建。[⭐ 551](https:\u002F\u002Fgithub.com\u002Fethz-asl\u002Fwavemap)\n\n### [优化](#contents)\n\n_机器人领域中使用的数值优化求解器和框架。_\n\n* 🟢 [CasADi](https:\u002F\u002Fgithub.com\u002Fcasadi\u002Fcasadi\u002Fwiki) - 用于算法微分和数值优化的符号框架。[⭐ 2.1k](https:\u002F\u002Fgithub.com\u002Fcasadi\u002Fcasadi)\n* 🟢 [Ceres Solver](http:\u002F\u002Fceres-solver.org\u002F) - 大规模非线性优化库。[⭐ 4.4k](https:\u002F\u002Fgithub.com\u002Fceres-solver\u002Fceres-solver)\n* 🟢 eigen-qld - 使用Eigen3库调用QLD QP求解器的接口。[⭐ 16](https:\u002F\u002Fgithub.com\u002Fjrl-umi3218\u002Feigen-qld)\n* 🟡 EXOTica - 面向机器人平台的通用优化工具集。[⭐ 161](https:\u002F\u002Fgithub.com\u002Fipab-slmc\u002Fexotica)\n* 🟢 hpipm - 高性能内点法QP求解器（Ipopt、Snopt）。[⭐ 666](https:\u002F\u002Fgithub.com\u002Fgiaf\u002Fhpipm)\n* 🟢 [HYPRE](https:\u002F\u002Fhypre.readthedocs.io\u002F) - 基于多重网格方法的稀疏线性方程组并行求解器。[⭐ 815](https:\u002F\u002Fgithub.com\u002Fhypre-space\u002Fhypre)\n* 🟢 ifopt - 基于Eigen的轻量级C++非线性规划求解器接口（Ipopt、Snopt）。[⭐ 847](https:\u002F\u002Fgithub.com\u002Fethz-adrl\u002Fifopt)\n* 🟢 [Ipopt](https:\u002F\u002Fprojects.coin-or.org\u002FIpopt) - 大规模非线性优化库。[⭐ 1.7k](https:\u002F\u002Fgithub.com\u002Fcoin-or\u002FIpopt)\n* 🟢 libcmaes - 使用CMA-ES算法进行黑盒随机优化。[⭐ 355](https:\u002F\u002Fgithub.com\u002FCMA-ES\u002Flibcmaes)\n* 🔴 [limbo](http:\u002F\u002Fwww.resibots.eu\u002Flimbo\u002F) - 高斯过程与黑盒函数的贝叶斯优化。[⭐ 261](https:\u002F\u002Fgithub.com\u002Fresibots\u002Flimbo)\n* 🟢 lpsolvers - Python中的线性规划求解器，具有统一的API。[⭐ 25](https:\u002F\u002Fgithub.com\u002Fstephane-caron\u002Flpsolvers)\n* 🟢 [NLopt](https:\u002F\u002Fnlopt.readthedocs.io\u002Fen\u002Flatest\u002F) - 非线性优化。[⭐ 2.2k](https:\u002F\u002Fgithub.com\u002Fstevengj\u002Fnlopt)\n* 🟡 [OptimLib](https:\u002F\u002Fwww.kthohr.com\u002Foptimlib.html) - 轻量级C++非线性数值优化方法库。[⭐ 886](https:\u002F\u002Fgithub.com\u002Fkthohr\u002Foptim)\n* 🟢 [OSQP](https:\u002F\u002Fosqp.org\u002F) - 算子分裂QP求解器。[⭐ 2.1k](https:\u002F\u002Fgithub.com\u002Fosqp\u002Fosqp)\n* 🟢 [Pagmo](https:\u002F\u002Fesa.github.io\u002Fpagmo2\u002Findex.html) - 用于大规模并行优化的科学库。[⭐ 908](https:\u002F\u002Fgithub.com\u002Fesa\u002Fpagmo2)\n* 🟢 [ProxSuite](https:\u002F\u002Fsimple-robotics.github.io\u002Fproxsuite\u002F) - 先进的近端优化工具箱。[⭐ 539](https:\u002F\u002Fgithub.com\u002FSimple-Robotics\u002FProxSuite)\n* 🔴 [pymoo](https:\u002F\u002Fwww.pymoo.org\u002F) - Python中的多目标优化。[⭐ 26](https:\u002F\u002Fgithub.com\u002Fmsu-coinlab\u002Fpymoo)\n* 🟢 qpsolvers - Python中的二次规划求解器，具有统一的API。[⭐ 729](https:\u002F\u002Fgithub.com\u002Fqpsolvers\u002Fqpsolvers)\n* 🟢 [RobOptim](http:\u002F\u002Froboptim.net\u002Findex.html) - 用于机器人的数值优化。[⭐ 65](https:\u002F\u002Fgithub.com\u002Froboptim\u002Froboptim-core)\n* 🟢 [SCS](http:\u002F\u002Fweb.stanford.edu\u002F~boyd\u002Fpapers\u002Fscs.html) - 用于求解大规模凸锥问题的数值优化。[⭐ 614](https:\u002F\u002Fgithub.com\u002Fcvxgrp\u002Fscs)\n* 🔴 sferes2 - 进化计算。[⭐ 170](https:\u002F\u002Fgithub.com\u002Fsferes2\u002Fsferes2)\n* 🟢 SHOT - 混合整数非线性优化问题求解器。[⭐ 130](https:\u002F\u002Fgithub.com\u002Fcoin-or\u002FSHOT)\n\n### [机器人建模](#contents)\n\n_用于描述机器人模型的工具和格式。_\n\n###### 机器人模型描述格式\n\n* [SDF](http:\u002F\u002Fsdformat.org\u002F) - 一种XML格式，用于描述机器人仿真器、可视化和控制所需的对象及环境。[[bitbucket](https:\u002F\u002Fbitbucket.org\u002Fosrf\u002Fsdformat)]\n* 🟢 [urdf](http:\u002F\u002Fwiki.ros.org\u002Furdf) - 用于表示机器人模型的XML格式。[⭐ 119](https:\u002F\u002Fgithub.com\u002Fros\u002Furdfdom)\n\n###### 构建机器人模型的实用工具\n\n* 🟢 [onshape-to-robot](https:\u002F\u002Fgithub.com\u002FRhoban\u002Fonshape-to-robot) - 通过OnShape API将OnShape装配转换为机器人定义（SDF或URDF）。[⭐ 489](https:\u002F\u002Fgithub.com\u002FRhoban\u002Fonshape-to-robot)\n* 🟢 phobos - Blender插件，用于创建URDF和SMURF机器人模型。[⭐ 857](https:\u002F\u002Fgithub.com\u002Fdfki-ric\u002Fphobos)\n\n### [机器人平台](#contents)\n\n_用于构建机器人软件系统的中间件和框架。_\n\n* 🔴 [AutoRally](http:\u002F\u002Fautorally.github.io\u002F) - 高性能的先进感知与控制研究测试平台。[⭐ 775](https:\u002F\u002Fgithub.com\u002Fautorally\u002Fautorally)\n* 🔴 [Linorobot](https:\u002F\u002Flinorobot.org\u002F) - 兼容ROS的地面机器人。[⭐ 1.1k](https:\u002F\u002Fgithub.com\u002Flinorobot\u002Flinorobot)\n  * 🔴 onine - 基于Linorobot和Braccio机械臂的服务机器人。[⭐ 47](https:\u002F\u002Fgithub.com\u002Fgrassjelly\u002Fonine)\n* 🟡 [Micro-ROS for Arduino](https:\u002F\u002Fgithub.com\u002Fkaiaai\u002Fmicro_ros_arduino_kaiaai) - 可在Arduino库管理器中获取的Micro-ROS分支。[⭐ 12](https:\u002F\u002Fgithub.com\u002Fkaiaai\u002Fmicro_ros_arduino_kaiaai)\n* [Rock](https:\u002F\u002Fwww.rock-robotics.org\u002F) - 机器人系统的软件框架。\n* [ROS](https:\u002F\u002Fwww.ros.org\u002F) - 用于编写机器人软件的灵活框架。\n* 🟢 [ROS 2](https:\u002F\u002Fgithub.com\u002Fros2\u002Fros2\u002Fwiki) - 机器人操作系统（ROS）软件栈的2.0版本。[⭐ 5k](https:\u002F\u002Fgithub.com\u002Fros2\u002Fros2)\n* [ros2_medkit](https:\u002F\u002Fselfpatch.github.io\u002Fros2_medkit\u002F) - 面向ROS 2的结构化故障管理——持久化故障生命周期、REST\u002FSSE API、根本原因关联以及故障时自动捕获rosbag。灵感来源于SOVD（面向服务的车辆诊断），即ASAM标准，它将HTTP\u002FREST诊断引入汽车和机器人领域。[[github](https:\u002F\u002Fgithub.com\u002Fselfpatch\u002Fros2_medkit)]\n* 🟢 [YARP](https:\u002F\u002Fwww.yarp.it\u002F) - 适用于从人形机器人到嵌入式设备的通信与设备接口。[⭐ 585](https:\u002F\u002Fgithub.com\u002Frobotology\u002Fyarp)\n\n### [机器人强化学习](#contents)\n\n_常用于机器人控制的强化学习库。_\n\n* 🟢 [Brax](https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fbrax) - 基于JAX的大规模并行可微刚体物理引擎，适用于机器人和强化学习。[⭐ 3.1k](https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fbrax)\n* 🟢 [CleanRL](https:\u002F\u002Fgithub.com\u002Fvwxyzjn\u002Fcleanrl) - 单文件实现的深度强化学习算法。[⭐ 9.1k](https:\u002F\u002Fgithub.com\u002Fvwxyzjn\u002Fcleanrl)\n* 🟢 [Isaac Lab](https:\u002F\u002Fisaac-sim.github.io\u002FIsaacLab) - 基于NVIDIA Isaac Sim构建的GPU加速开源机器人学习框架。[⭐ 6.3k](https:\u002F\u002Fgithub.com\u002Fisaac-sim\u002FIsaacLab)\n* 🟢 [rl_games](https:\u002F\u002Fgithub.com\u002FDenys88\u002Frl_games) - 在Isaac Gym环境中使用的高性能强化学习库。[⭐ 1.3k](https:\u002F\u002Fgithub.com\u002FDenys88\u002Frl_games)\n* 🟢 [SKRL](https:\u002F\u002Fgithub.com\u002FToni-SM\u002Fskrl) - 支持多种ML框架的模块化强化学习库。[⭐ 991](https:\u002F\u002Fgithub.com\u002FToni-SM\u002Fskrl)\n* 🟢 [Stable-Baselines3](https:\u002F\u002Fgithub.com\u002FDLR-RM\u002Fstable-baselines3) - PyTorch中可靠的强化学习算法实现。[⭐ 12.7k](https:\u002F\u002Fgithub.com\u002FDLR-RM\u002Fstable-baselines3)\n\n### [SLAM](#contents)\n\n_同时定位与地图构建库。_\n\n\n* 🔴 AprilSAM - 实时平滑与建图。[⭐ 239](https:\u002F\u002Fgithub.com\u002Fxipengwang\u002FAprilSAM)\n* 🔴 Cartographer - 跨多个平台和传感器配置的2D及3D实时SLAM。[⭐ 7.8k](https:\u002F\u002Fgithub.com\u002Fcartographer-project\u002Fcartographer)\n* 🟡 [DSO](https:\u002F\u002Fvision.in.tum.de\u002Fresearch\u002Fvslam\u002Fdso) - 面向视觉里程计的新型直接稀疏公式。[⭐ 2.4k](https:\u002F\u002Fgithub.com\u002FJakobEngel\u002Fdso)\n* 🟢 ElasticFusion - 实时稠密视觉SLAM系统。[⭐ 1.9k](https:\u002F\u002Fgithub.com\u002Fmp3guy\u002FElasticFusion)\n* 🟢 [fiducials](http:\u002F\u002Fwiki.ros.org\u002Ffiducials) - 使用基准标记进行同时定位与建图。[⭐ 278](https:\u002F\u002Fgithub.com\u002FUbiquityRobotics\u002Ffiducials)\n* 🟢 GTSAM - 机器人与视觉领域的平滑与建图（SAM）。[⭐ 3.3k](https:\u002F\u002Fgithub.com\u002Fborglab\u002Fgtsam)\n* 🔴 Kintinuous - 实时大规模稠密视觉SLAM系统。[⭐ 952](https:\u002F\u002Fgithub.com\u002Fmp3guy\u002FKintinuous)\n* 🔴 [LSD-SLAM](https:\u002F\u002Fvision.in.tum.de\u002Fresearch\u002Fvslam\u002Flsdslam) - 实时单目SLAM。[⭐ 2.7k](https:\u002F\u002Fgithub.com\u002Ftum-vision\u002Flsd_slam)\n* 🟡 ORB-SLAM2 - 适用于单目、立体及RGB-D相机的实时SLAM库。[⭐ 10.1k](https:\u002F\u002Fgithub.com\u002Fraulmur\u002FORB_SLAM2)\n* 🟢 [RTAP-Map](http:\u002F\u002Fintrolab.github.io\u002Frtabmap\u002F) - 基于全局贝叶斯回环检测器的RGB-D图SLAM方法。[⭐ 3.6k](https:\u002F\u002Fgithub.com\u002Fintrolab\u002Frtabmap)\n* 🔴 [SRBA](http:\u002F\u002Fmrpt.github.io\u002Fsrba\u002F) - 以相对坐标解决SLAM\u002FBA问题，并支持多种子映射策略。[⭐ 76](https:\u002F\u002Fgithub.com\u002FMRPT\u002Fsrba)\n\n#### SLAM数据集\n\n* 🟡 [Awesome SLAM Datasets](https:\u002F\u002Fgithub.com\u002Fyoungguncho\u002Fawesome-slam-datasets) - 精选的SLAM相关数据集列表。[⭐ 1.9k](https:\u002F\u002Fgithub.com\u002Fyoungguncho\u002Fawesome-slam-datasets)\n\n### [Vision](#contents)\n\n_用于机器人感知的计算机视觉库。_\n\n* 🔴 [BundleTrack](https:\u002F\u002Fgithub.com\u002Fwenbowen123\u002FBundleTrack) - 无需3D模型即可对新物体进行6D位姿跟踪。[⭐ 678](https:\u002F\u002Fgithub.com\u002Fwenbowen123\u002FBundleTrack)\n* 🔴 [se(3)-TrackNet](https:\u002F\u002Fgithub.com\u002Fwenbowen123\u002Firos20-6d-pose-tracking) - 无需3D模型即可对新物体进行6D位姿跟踪。[⭐ 420](https:\u002F\u002Fgithub.com\u002Fwenbowen123\u002Firos20-6d-pose-tracking)\n* 🟢 [ViSP](http:\u002F\u002Fvisp.inria.fr\u002F) - 视觉伺服平台。[⭐ 849](https:\u002F\u002Fgithub.com\u002Flagadic\u002Fvisp)\n\n### [Fluid](#contents)\n\n_流体动力学仿真库。_\n\n* 🔴 [Fluid Engine Dev - Jet](https:\u002F\u002Ffluidenginedevelopment.org\u002F) - 用于计算机图形应用的流体仿真引擎。[⭐ 2.1k](https:\u002F\u002Fgithub.com\u002Fdoyubkim\u002Ffluid-engine-dev)\n\n### [Grasping](#contents)\n\n_用于机器人抓取与操作的库和工具。_\n\n* 🟢 [AnyGrasp SDK](https:\u002F\u002Fgithub.com\u002Fgraspnet\u002Fanygrasp_sdk) - AnyGrasp SDK，一种6自由度抓取姿态检测方法。[⭐ 756](https:\u002F\u002Fgithub.com\u002Fgraspnet\u002Fanygrasp_sdk)\n* 🟡 [Contact-GraspNet](https:\u002F\u002Fgithub.com\u002FNVlabs\u002Fcontact_graspnet) - 利用接触图生成平行爪夹持器的6自由度抓取姿态。[⭐ 449](https:\u002F\u002Fgithub.com\u002FNVlabs\u002Fcontact_graspnet)\n* 🔴 [GraspIt!](https:\u002F\u002Fgraspit-simulator.github.io\u002F) - 可适应任意手部和机器人设计的抓取研究模拟器。[⭐ 208](https:\u002F\u002Fgithub.com\u002Fgraspit-simulator\u002Fgraspit)\n* 🟢 [GraspNet API](https:\u002F\u002Fgithub.com\u002Fgraspnet\u002FgraspnetAPI) - GraspNet基准测试的Python API及评估工具。[⭐ 325](https:\u002F\u002Fgithub.com\u002Fgraspnet\u002FgraspnetAPI)\n\n### [Humanoid Robotics](#contents)\n\n_用于人形机器人研究的环境与模型。_\n\n* 🟡 [Humanoid-Gym](https:\u002F\u002Fgithub.com\u002Froboterax\u002Fhumanoid-gym) - 用于人形机器人步态强化学习的环境。[⭐ 1.8k](https:\u002F\u002Fgithub.com\u002Froboterax\u002Fhumanoid-gym)\n* 🟢 [iCub](http:\u002F\u002Fwww.icub.org\u002F) - 开源认知型人形机器人平台，用于具身认知研究。[⭐ 117](https:\u002F\u002Fgithub.com\u002Frobotology\u002Ficub-main)\n* 🟢 [Legged Gym](https:\u002F\u002Fgithub.com\u002Fleggedrobotics\u002Flegged_gym) - 用于四足机器人步态训练的Isaac Gym环境。[⭐ 2.7k](https:\u002F\u002Fgithub.com\u002Fleggedrobotics\u002Flegged_gym)\n* 🟢 [MuJoCo Menagerie](https:\u002F\u002Fgithub.com\u002Fgoogle-deepmind\u002Fmujoco_menagerie) - 一系列为研究与开发精心调优的MuJoCo模型集合。[⭐ 3k](https:\u002F\u002Fgithub.com\u002Fgoogle-deepmind\u002Fmujoco_menagerie)\n\n### [Multiphysics](#contents)\n\n_用于耦合多物理场仿真框架。_\n\n* 🟢 [Kratos](http:\u002F\u002Fwww.cimne.com\u002Fkratos\u002F) - 用于构建并行多学科仿真软件的框架。[⭐ 1.2k](https:\u002F\u002Fgithub.com\u002FKratosMultiphysics\u002FKratos)\n\n### [Math](#contents)\n\n_用于空间代数、李群及线性代数的数学库。_\n\n* 🟢 Fastor - 轻量级高性能张量代数框架，基于C++11\u002F14\u002F17。[⭐ 830](https:\u002F\u002Fgithub.com\u002Fromeric\u002FFastor)\n* 🔴 linalg.h - 单头公共领域线性代数库，适用于C++11。[⭐ 939](https:\u002F\u002Fgithub.com\u002Fsgorsten\u002Flinalg)\n* 🟢 manif - 小型C++11头文件库，用于李理论。[⭐ 1.7k](https:\u002F\u002Fgithub.com\u002Fartivis\u002Fmanif)\n* 🟡 Sophus - 使用Eigen实现的李群。[⭐ 2.4k](https:\u002F\u002Fgithub.com\u002Fstrasdat\u002FSophus)\n* 🟢 SpaceVelAlg - 基于Eigen3的空间向量代数。[⭐ 81](https:\u002F\u002Fgithub.com\u002Fjrl-umi3218\u002FSpaceVecAlg)\n* 🟢 spatialmath-python - 用于2D\u002F3D中位姿与方向的Python类，包含空间运算工具箱。[⭐ 617](https:\u002F\u002Fgithub.com\u002Fbdaiinstitute\u002Fspatialmath-python)\n\n### [ETC](#contents)\n\n_其他机器人相关工具与实用程序。_\n\n* [Foxglove Studio](https:\u002F\u002Ffoxglove.dev) - 一款完全集成的桌面可视化与调试应用程序，专为您的机器人数据设计。\n* 🟢 fuse - 用于在机器人上实时执行传感器融合的通用架构。[⭐ 849](https:\u002F\u002Fgithub.com\u002Flocusrobotics\u002Ffuse)\n\n## [模拟器](#contents)\n\n_用于测试和开发机器人系统的仿真环境。_\n\n###### 免费或开源\n\n* 🟢 [AI2-THOR](https:\u002F\u002Fai2thor.allenai.org\u002F) - 基于Unity引擎的具身人工智能交互式家居环境。[⭐ 1.7k](https:\u002F\u002Fgithub.com\u002Fallenai\u002Fai2thor)\n* 🟢 AirSim - 基于虚幻引擎的自动驾驶车辆仿真器。[⭐ 17.9k](https:\u002F\u002Fgithub.com\u002FMicrosoft\u002FAirSim)\n* 🟢 [ARGoS](https:\u002F\u002Fwww.argos-sim.info\u002F) - 基于物理的仿真器，专为大规模机器人集群仿真设计。[⭐ 301](https:\u002F\u002Fgithub.com\u002Filpincy\u002Fargos3)\n* 🟢 [ARTE](http:\u002F\u002Farvc.umh.es\u002Farte\u002Findex_en.html) - 专注于机器人机械臂的Matlab工具箱。[⭐ 101](https:\u002F\u002Fgithub.com\u002F4rtur1t0\u002FARTE)\n* 🟢 [AVIS Engine](https:\u002F\u002Favisengine.com) - 用于自动驾驶车辆开发的快速仿真软件。[⭐ 21](https:\u002F\u002Fgithub.com\u002FAvisEngine\u002FAVIS-Engine-Python-API)\n* 🟢 [CARLA](https:\u002F\u002Fcarla.org\u002F) - 面向自动驾驶研究的开源仿真器。[⭐ 13.6k](https:\u002F\u002Fgithub.com\u002Fcarla-simulator\u002Fcarla)\n* 🟢 [CoppeliaSim](https:\u002F\u002Fwww.coppeliarobotics.com\u002F) - 前身为V-REP。虚拟机器人实验平台。[⭐ 139](https:\u002F\u002Fgithub.com\u002FCoppeliaRobotics\u002FCoppeliaSimLib)\n* 💀 [Gazebo](https:\u002F\u002Fgazebosim.org\u002F) - 动态多机器人仿真器。[⭐ 1.3k](https:\u002F\u002Fgithub.com\u002Fgazebosim\u002Fgazebo-classic)\n* 🟢 [Gazebo Sim](https:\u002F\u002Fgazebosim.org\u002F) - 开源机器人仿真器（前身为Ignition Gazebo）。[⭐ 1.2k](https:\u002F\u002Fgithub.com\u002Fgazebosim\u002Fgz-sim)\n* 🔴 [GraspIt!](http:\u002F\u002Fgraspit-simulator.github.io\u002F) - 用于抓取研究的仿真器，可支持任意手部和机器人设计。[⭐ 208](https:\u002F\u002Fgithub.com\u002Fgraspit-simulator\u002Fgraspit)\n* 🟢 [Habitat-Sim](https:\u002F\u002Faihabitat.org\u002F) - 用于具身人工智能研究的仿真平台。[⭐ 3.5k](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002Fhabitat-sim)\n* 🟢 [Hexapod Robot Simulator](https:\u002F\u002Fhexapod.netlify.app\u002F) - 开源六足机器人逆运动学及步态可视化工具。[⭐ 703](https:\u002F\u002Fgithub.com\u002Fmithi\u002Fhexapod)\n* 🟢 [Isaac Sim](https:\u002F\u002Fdeveloper.nvidia.com\u002Fisaac\u002Fsim) - NVIDIA推出的GPU加速机器人仿真平台，配备PhysX 5和RTX渲染。[⭐ 2.5k](https:\u002F\u002Fgithub.com\u002Fisaac-sim\u002FIsaacSim)\n* 🟢 [ManiSkill](https:\u002F\u002Fgithub.com\u002Fhaosulab\u002FManiSkill) - 基于SAPIEN的机器人仿真与操控学习包。[⭐ 2.5k](https:\u002F\u002Fgithub.com\u002Fhaosulab\u002FManiSkill)\n* 🔴 [MORSE](http:\u002F\u002Fmorse-simulator.github.io\u002F) - 模块化开源机器人仿真引擎。[⭐ 370](https:\u002F\u002Fgithub.com\u002Fmorse-simulator\u002Fmorse)\n* [Neurorobotics Platform](https:\u002F\u002Fneurorobotics.net\u002F) - 可通过互联网访问、由脉冲神经网络控制的机器人仿真平台。[[bitbucket](https:\u002F\u002Fbitbucket.org\u002Fhbpneurorobotics\u002Fneurorobotics-platform)]\n* 🟢 [PyBullet](https:\u002F\u002Fdocs.google.com\u002Fdocument\u002Fd\u002F10sXEhzFRSnvFcl3XxNGhnD4N2SedqwdAvK3dsihxVUA\u002Fedit#heading=h.2ye70wns7io3) - 易于使用的机器人及深度强化学习仿真器。[⭐ 14.2k](https:\u002F\u002Fgithub.com\u002Fbulletphysics\u002Fbullet3)\n* 🟢 [PyBullet_Industrial](https:\u002F\u002Fpybullet-industrial.readthedocs.io\u002Fen\u002Flatest\u002F) - PyBullet扩展，用于模拟铣削、3D打印等机器人制造工艺。[⭐ 49](https:\u002F\u002Fgithub.com\u002FWBK-Robotics\u002Fpybullet_industrial)\n* 🔴 [Robot Gui](http:\u002F\u002Frobot.glumb.de\u002F) - 基于three.js的3D机器人界面。[⭐ 384](https:\u002F\u002Fgithub.com\u002Fglumb\u002Frobot-gui)\n* 🟢 [SAPIEN](https:\u002F\u002Fsapien.ucsd.edu) - 针对铰接物体及操作任务的高物理仿真环境。[⭐ 718](https:\u002F\u002Fgithub.com\u002Fhaosulab\u002FSAPIEN)\n* [Simbad](http:\u002F\u002Fsimbad.sourceforge.net\u002F) - Java 3D机器人仿真器，支持自定义控制器和传感器。\n* 🟡 [Unity](https:\u002F\u002Funity.com\u002Fsolutions\u002Fautomotive-transportation-manufacturing\u002Frobotics) - 游戏引擎，提供开源机器人仿真工具和教程。[⭐ 2.5k](https:\u002F\u002Fgithub.com\u002FUnity-Technologies\u002FUnity-Robotics-Hub)\n* 🟢 [Webots](http:\u002F\u002Fwww.cyberbotics.com\u002F) - 用于建模、编程和仿真机器人及机械系统的开发环境。[⭐ 4.1k](https:\u002F\u002Fgithub.com\u002Fcyberbotics\u002Fwebots)\n\n###### 商业\n\n* [Actin Simulation](http:\u002F\u002Fwww.energid.com\u002F) - 实时机器人仿真与控制软件。\n* [Artiminds](https:\u002F\u002Fwww.artiminds.com\u002F) - 规划、编程、运行、分析及优化。\n* [Kineo](https:\u002F\u002Fwww.plm.automation.siemens.com\u002Fglobal\u002Fen\u002Fproducts\u002Fplm-components\u002Fkineo.html) - 工业机器人路径规划与轨迹优化。\n* [Robot Virtual Worlds](http:\u002F\u002Fwww.robotvirtualworlds.com\u002F) - 用于教育机器人的虚拟现实软件。\n* [RobotDK](https:\u002F\u002Frobodk.com\u002F) - 机器人仿真与离线编程。\n* [RobotStudio](https:\u002F\u002Fwww.abb.com\u002Fglobal\u002Fen\u002Fareas\u002Frobotics\u002Fproducts\u002Fsoftware\u002Frobotstudio-suite) - ABB的机器人仿真与离线编程软件。\n* [Virtual Robotics Toolkit](https:\u002F\u002Fwww.virtualroboticstoolkit.com\u002F) - 用于编程和测试机器人的3D虚拟环境。\n* [Visual Components](https:\u002F\u002Fwww.visualcomponents.com\u002F) - 3D制造仿真与可视化平台。\n\n###### 云服务\n\n* [AWS RoboMaker](https:\u002F\u002Faws.amazon.com\u002Frobomaker\u002F) - 一项服务，可轻松大规模开发、测试和部署智能机器人应用。\n\n## [其他精彩列表](#contents)\n\n_相关的机器人和人工智能资源精选列表。_\n\n* [Awesome Robotics](https:\u002F\u002Fgithub.com\u002FKiloreux\u002Fawesome-robotics) - 机器人技术资源精选列表（Kiloreux）。\n* [Awesome Robotics](https:\u002F\u002Fgithub.com\u002Fahundt\u002Fawesome-robotics) - 机器人相关链接和软件库的精选列表（ahundt）。\n* [Awesome Robotic Tooling](https:\u002F\u002Fgithub.com\u002FLy0n\u002Fawesome-robotic-tooling) - 面向专业机器人开发的工具链，支持 C++ 和 Python，涵盖 ROS 及自动驾驶领域。\n* [Awesome Artificial Intelligence](https:\u002F\u002Fgithub.com\u002Fowainlewis\u002Fawesome-artificial-intelligence) - 人工智能课程、书籍、视频讲座和论文的精选列表。\n* [Awesome Collision Detection](https:\u002F\u002Fgithub.com\u002Fjslee02\u002Fawesome-collision-detection) - 碰撞检测相关库和资源的精选列表。\n* [Awesome Computer Vision](https:\u002F\u002Fgithub.com\u002Fjbhuang0604\u002Fawesome-computer-vision) - 计算机视觉资源的精选列表。\n* [Awesome Machine Learning](https:\u002F\u002Fgithub.com\u002Fjosephmisiti\u002Fawesome-machine-learning) - 机器学习框架、库和软件的精选列表。\n* [Awesome Deep Learning](https:\u002F\u002Fgithub.com\u002FChristosChristofidis\u002Fawesome-deep-learning) - 深度学习教程、项目和社区的精选列表。\n* [Awesome Gazebo](https:\u002F\u002Fgithub.com\u002Ffkromer\u002Fawesome-gazebo) - Gazebo 仿真框架资源的精选列表。\n* [Awesome Grasping](https:\u002F\u002Fgithub.com\u002FPo-Jen\u002Fawesome-grasping) - 抓取相关库和资源的精选列表。\n* [Awesome Human Robot Interaction](https:\u002F\u002Fgithub.com\u002FPo-Jen\u002Fawesome-human-robot-interaction) - 人机交互相关库和资源的精选列表。\n* [Awesome Physical AI](https:\u002F\u002Fgithub.com\u002Fkeon\u002Fawesome-physical-ai) - 物理人工智能领域的学术论文和资源精选列表。\n* [PythonRobotics](https:\u002F\u002Fgithub.com\u002FAtsushiSakai\u002FPythonRobotics) - 用于机器人算法的 Python 示例代码。\n* [Robotics Coursework](https:\u002F\u002Fgithub.com\u002Fmithi\u002Frobotics-coursework) - 在线可学的机器人课程列表。\n\n## [贡献](#contents)\n\n非常欢迎各位贡献！请先阅读[贡献指南](https:\u002F\u002Fgithub.com\u002Fjslee02\u002Fawesome-robotics-libraries\u002Fblob\u002Fmain\u002FCONTRIBUTING.md)。如有任何错误，也欢迎随时报告。\n\n## [星标历史](#contents)\n\n[![星标历史图表](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Fjslee02_awesome-robotics-libraries_readme_4815b16021eb.png)](https:\u002F\u002Fstar-history.com\u002F#jslee02\u002Fawesome-robotics-libraries)\n\n## [许可证](#contents)\n\n[![CC0](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Fjslee02_awesome-robotics-libraries_readme_b7657951a0bb.png)](http:\u002F\u002Fcreativecommons.org\u002Fpublicdomain\u002Fzero\u002F1.0\u002F)","# awesome-robotics-libraries 快速上手指南\n\n`awesome-robotics-libraries` 并非单一的软件库，而是一个精选的机器人学库与模拟器清单。本指南将指导你如何浏览该清单，并快速安装和试用其中几个最具代表性且活跃的库（以 **Drake**、**MuJoCo** 和 **LeRobot** 为例）。\n\n## 环境准备\n\n由于清单中包含多种语言（C++, Python, Julia, Rust 等）编写的库，请根据你的目标库准备相应环境。以下是通用推荐配置：\n\n*   **操作系统**: Linux (Ubuntu 20.04\u002F22.04 推荐) 或 macOS。Windows 用户建议使用 WSL2。\n*   **核心依赖**:\n    *   Python 3.8+ (大多数现代库的首选)\n    *   CMake, build-essential, git\n    *   CUDA (可选，用于加速物理仿真如 Brax, Genesis)\n*   **包管理器**:\n    *   Python: `pip` 或 `conda`\n    *   C++: `apt` (Ubuntu) 或 `brew` (macOS)\n\n> **国内加速建议**:\n> *   **PyPI 镜像**: 安装 Python 库时，推荐使用清华或阿里镜像源。\n>     ```bash\n>     pip install -i https:\u002F\u002Fpypi.tuna.tsinghua.edu.cn\u002Fsimple \u003Cpackage_name>\n>     ```\n> *   **GitHub 加速**: 若克隆仓库速度慢，可使用国内镜像站（如 `gitclone.com`）或配置代理。\n\n## 安装步骤\n\n以下展示三个不同领域热门库的安装命令。\n\n### 1. Drake (规划、控制与分析工具箱)\n适用于非线性动态系统的建模与控制。\n\n```bash\n# 推荐使用 conda 安装（最简便）\nconda install -c conda-forge drake\n\n# 或使用 pip 安装\npip install drake\n```\n\n### 2. MuJoCo (多关节动力学物理引擎)\n适用于基于物理的仿真和控制，现由 Google DeepMind 维护并已开源。\n\n```bash\n# 安装核心库\npip install mujoco\n\n# 安装可视化辅助工具 (可选)\npip install mujoco_mjx\n```\n\n### 3. LeRobot (真实世界机器人深度学习)\nHugging Face 推出的用于真实机器人学习的 PyTorch 库，包含预训练模型和数据集。\n\n```bash\n# 克隆仓库\ngit clone https:\u002F\u002Fgithub.com\u002Fhuggingface\u002Flerobot.git\ncd lerobot\n\n# 安装依赖 (建议使用虚拟环境)\npip install -e .\n```\n\n## 基本使用\n\n### Drake 示例：创建一个简单的双摆模型\n```python\nimport numpy as np\nfrom pydrake.multibody.plant import AddMultibodyPlantSceneGraph\nfrom pydrake.systems.framework import DiagramBuilder\n\nbuilder = DiagramBuilder()\nplant, scene_graph = AddMultibodyPlantSceneGraph(builder, time_step=0.0)\n\n# 这里仅展示框架搭建，实际需加载 URDF\u002FSDF 模型\n# plant = parser.AddModelsFromUrl(\"package:\u002F\u002Fdrake\u002Fmanipulation\u002Fmodels\u002Fiiwa_description\u002Furdf\u002Fiiwa14_no_collision.urdf\")\n\nplant.Finalize()\ndiagram = builder.Build()\nprint(\"Drake diagram built successfully.\")\n```\n\n### MuJoCo 示例：加载并运行一个 XML 模型\n```python\nimport mujoco\nimport glfw\n\n# 加载预定义的 humanoids 模型 (需确保 mjpro 目录存在或在线加载)\n# 此处演示加载字符串定义的简单模型\nxml_content = \"\"\"\n\u003Cmujoco>\n  \u003Cworldbody>\n    \u003Cgeom type=\"sphere\" size=\"0.1\" pos=\"0 0 1\"\u002F>\n  \u003C\u002Fworldbody>\n\u003C\u002Fmujoco>\n\"\"\"\n\nmodel = mujoco.MjModel.from_xml_string(xml_content)\ndata = mujoco.MjData(model)\n\n# 执行一步仿真\nmujoco.mj_step(model, data)\nprint(f\"Simulation time: {data.time}, Body position: {data.geom_xpos[0]}\")\n```\n\n### LeRobot 示例：加载预训练策略\n```python\nfrom lerobot.common.policies.diffusion.modeling_diffusion import DiffusionPolicy\n\n# 从 Hugging Face Hub 加载预训练的扩散策略模型\n# 需要配置 HF_TOKEN 环境变量以访问私有模型（如有）\npolicy = DiffusionPolicy.from_pretrained(\"lerobot\u002Fdiffusion_push_tiny\")\n\nprint(\"Policy loaded successfully. Ready for inference.\")\n# input_obs = ... # 准备观测数据\n# action = policy.select_action(input_obs)\n```\n\n## 探索更多库\n\n访问 [awesome-robotics-libraries GitHub 仓库](https:\u002F\u002Fgithub.com\u002Fawesome-robotics\u002Fawesome-robotics-libraries) 查看完整列表。\n\n*   **状态图例**:\n    *   🟢 **Active**: 活跃维护中 (\u003C1 年未更新)，推荐优先使用。\n    *   🟡 **Slow**: 更新较慢 (1-2 年)。\n    *   🔴 **Stale**: 停滞 (>2 年)，需谨慎评估兼容性。\n    *   💀 **Archived**: 已归档，仅作历史参考。\n\n根据具体需求（如 SLAM、逆运动学、流体仿真等），在对应的分类章节查找带有 🟢 标记的项目进行尝试。","某初创团队正在研发一款具备复杂地形适应能力的四足机器人，急需构建高保真的动力学仿真环境以训练运动控制算法。\n\n### 没有 awesome-robotics-libraries 时\n- **选型迷茫耗时久**：团队在 GitHub 上盲目搜索\"robotics simulation\"，面对数千个仓库难以辨别哪些支持柔性体碰撞或 GPU 加速，耗费两周时间试错。\n- **工具链断裂风险**：随意选用的物理引擎缺乏文档维护（如已归档的 Flex），导致无法与现代深度学习框架（如 JAX\u002FPyTorch）对接，数据流转受阻。\n- **仿真精度不足**：早期采用的简易模拟器无法处理机器人在松软沙地上的沉陷效应，导致实机部署时步态频繁失稳。\n- **重复造轮子**：开发人员被迫手动编写基础的正逆运动学解算代码，不仅效率低下且极易引入数学误差。\n\n### 使用 awesome-robotics-libraries 后\n- **精准快速匹配**：通过分类索引直接锁定 🟢 状态的 **Brax** 和 **Genesis**，利用其 GPU 并行加速特性，将强化学习训练速度提升数十倍。\n- **生态兼容性强**：依据列表推荐选用 **Drake** 进行轨迹规划，其活跃的社区确保了与非线性控制系统及最新 Python 版本的无缝集成。\n- **多物理场覆盖**：借助列表中 **CHRONO::ENGINE** 等专用库，成功模拟了颗粒流体与刚体的相互作用，显著提升了机器人在复杂地形的鲁棒性。\n- **开箱即用模块**：直接调用 **KDL** 或 **idyntree** 等成熟库处理动力学估算，让团队将 100% 的精力聚焦于核心控制策略的创新。\n\nawesome-robotics-libraries 通过提供经过验证的工具图谱，将机器人开发从“大海捞针”式的资源搜寻转变为高效的模块化组装，极大缩短了从算法验证到实机落地的周期。","https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Fjslee02_awesome-robotics-libraries_576cd70b.png","jslee02","Jeongseok (JS) Lee","https:\u002F\u002Foss.gittoolsai.com\u002Favatars\u002Fjslee02_4883268a.jpg",null,"jeongseok_lee_","https:\u002F\u002Fjeongseok.dev","https:\u002F\u002Fgithub.com\u002Fjslee02",[83],{"name":84,"color":85,"percentage":86},"Python","#3572A5",100,2860,479,"2026-04-05T14:33:11","CC0-1.0",5,"","未说明 (列表包含多个独立库，部分如 Brax, Genesis, Newton 提及支持 GPU 加速，但无统一硬件要求)","未说明",{"notes":96,"python":97,"dependencies":98},"该仓库是一个机器人库和模拟器的精选列表（Awesome List），而非单一软件工具。因此不存在统一的运行环境需求。列表中各子项目（如 Drake, MuJoCo, Bullet, Pinocchio 等）拥有各自独立的安装要求和依赖环境。用户需根据具体选择的子项目查阅其官方文档以获取详细的系统配置信息。","未说明 (列表中包含 C++, Python, Julia, MATLAB, Rust 等多种语言编写的库)",[],[13,52,54],[101,102,103,104,105,106,107,108,109,110,111,112],"robotics-libraries","multibody-dynamics","simulation","robotics","motion-planning","slam","awesome-list","optimization","machine-learning","awesome","robotics-library","robotics-simulation",7,"2026-03-27T02:49:30.150509","2026-04-06T08:52:42.895076",[117,122,127,132,137,142],{"id":118,"question_zh":119,"answer_zh":120,"source_url":121},9526,"如何向该列表提交新的机器人库建议？","您可以直接创建 Pull Request (PR) 来添加建议。维护者明确表示欢迎用户提交 PR，只需在 Issue 中列出您建议的库（如 maniskill, spatialmath-python 等），然后按照指引创建 PR 即可。","https:\u002F\u002Fgithub.com\u002Fjslee02\u002Fawesome-robotics-libraries\u002Fissues\u002F50",{"id":123,"question_zh":124,"answer_zh":125,"source_url":126},9527,"什么样的项目会被拒绝收录？","通用数学、几何或图形库如果缺乏特定的机器人应用场景，通常会被拒绝。例如，'Geometric Tools Engine'虽然评分很高且文档完善，但因其是通用库而非专为机器人（如仿真、规划、感知、控制）构建或常用，所以不符合收录范围。除非能提供具体的机器人用例或封装器，否则不予收录。","https:\u002F\u002Fgithub.com\u002Fjslee02\u002Fawesome-robotics-libraries\u002Fissues\u002F16",{"id":128,"question_zh":129,"answer_zh":130,"source_url":131},9528,"专属于特定公司的编程语言或库会被收录吗？","通常不会。如果某个库（如 TP+）似乎是专为某一家公司（如 One Robotics Company）的特定编程语言设计的，且不具备广泛的通用性或开源社区适应性，维护者可能会认为其不符合列表的收录标准而拒绝。","https:\u002F\u002Fgithub.com\u002Fjslee02\u002Fawesome-robotics-libraries\u002Fissues\u002F2",{"id":133,"question_zh":134,"answer_zh":135,"source_url":136},9529,"提交建议后，项目是如何被评估和处理的？","提交建议后，系统会自动运行评估脚本检查项目的流行度（Stars）、活跃度（最近提交时间）、文档（README）、成熟度（项目年龄）等指标。如果自动评分通过（通常需满足多项标准），维护者会进行人工复核（特别是独特性），通过后会自动或手动创建 PR 将该项目添加到列表中。","https:\u002F\u002Fgithub.com\u002Fjslee02\u002Fawesome-robotics-libraries\u002Fissues\u002F74",{"id":138,"question_zh":139,"answer_zh":140,"source_url":141},9530,"如果发现列表中的链接失效或超时怎么办？","项目设有自动化的链接检查机制（Link Checker），会定期扫描 README 和其他文档中的链接。如果发现超时（Timeout）或错误（Error，如网络连接失败），系统会自动生成包含详细错误报告的 Issue（如 'Link Checker: broken links found'），维护者会根据报告修复失效链接或更新资源地址。","https:\u002F\u002Fgithub.com\u002Fjslee02\u002Fawesome-robotics-libraries\u002Fissues\u002F59",{"id":143,"question_zh":144,"answer_zh":145,"source_url":146},9531,"人形机器人平台（如 iCub）应该归类到哪里？","像 iCub 这样的开源认知人形机器人平台，在经过自动化评估确认其符合流行度、活跃度、文档和成熟度标准后，会被接受并添加到列表的“人形机器人”（Humanoid Robotics）章节中。","https:\u002F\u002Fgithub.com\u002Fjslee02\u002Fawesome-robotics-libraries\u002Fissues\u002F15",[]]