机器人模拟器 [英] Robotics Simulator

问题描述

哪个是最好的机器人模拟器?到目前为止，我只是希望对势场方法，Bug算法，A *算法等进行简单的仿真！

Which is the best robotics simulator ? As of now, I am just looking to get simple simulations of Potential Field method, Bug Algorithms, A* Algorithm etc !

已知的机器人模拟器的列表是

1. 玩家项目(2D模拟器-舞台-3D模拟器-凉亭-和控制界面-开源， ROS项目的一部分)
2. MORSE (通用室内/室外3D模拟器)
3. Microsoft Robotics Studio (模拟器+控制界面)
4. KiKS (Matlab插件，仅适用于Khepera +控制界面)
5. MobotSim (适用于像机器人这样的机器人，更多的算法实现)
6. Karel (漂亮的孩子，我想这是Pascal/Logo之类的东西)
7. MRPT (看起来很不错，会尽快尝试)
8. Carmen (机器人视觉等易于实现(?))
9. Webots (开源-多平台-多语言[ROS，Python，Matlab等.]-最新外观-网络导出)
10. Simbad (Java和Jython中的2D/3D模拟器)
11. Robocode (一种Java/.NET套件)
12. 罗斯苏姆的剧场(C/C ++套装)
13. V-REP (3D，可用的源代码，Lua脚本，C/C ++ API，Python， Java，Matlab，URBI，2个物理引擎，完整的运动学求解器等)

1. Player Project (2D simulator - Stage - 3D simulator - Gazebo - and control interface - open source, part of the ROS project)
2. MORSE (general purpose indoor/outdoor 3D simulator)
3. Microsoft Robotics Studio (simulator + control interface)
4. KiKS (Matlab plugin, only for Khepera + control interface)
5. MobotSim (for point like robots, more of algorithm implementation )
6. Karel (Pretty Kiddish, I guess it is Pascal/Logo like)
7. MRPT (looks very nice, will try it soon)
8. Carmen (Robot Vision etc is easy to implement in it (?))
9. Webots (open-source - multiplatform - multilanguage [ROS, Python, Matlab, etc.] - state-of-the-art appearance - web export)
10. Simbad (2D/3D simulator in Java and Jython)
11. Robocode (A Java/ .NET suit)
12. Rossum's Playhouse (C/C++ suit)
13. V-REP (3D, source available, Lua scripting, APIs for C/C++, Python, Java, Matlab, URBI, 2 physics engines, full kinematic solver, etc.)

更多通用平台/中间件也提供了仿真工具:

Some more generic platforms/middlewares also offer simulation tools:

1. ROS (目前是此类平台的最大集成)
2. URBI
3. YARP
4. OROCOS (对此一无所知！)

1. ROS (currently the largest integration of such platforms)
2. URBI
3. YARP
4. OROCOS (Don't know anything about it !)

一些关于机器人模拟器的参考资料:

1. 自主移动机器人的开发环境:调查
2. 开源机器人技术工具包
3. 更新了对机器人软件平台的评论
4. 现有模拟器[e-mail]
5. 在Asaf Matan网站上列出

1. Development environments for autonomous mobile robots: A survey
2. Open source robotics toolkits
3. Updated review of robotics software platforms
4. Existing Simulators - [e-mail]
5. List at Asaf Matan's website

推荐答案

我会选择一个常用的框架.当您对路径规划感兴趣时，选择一个已经有模拟机器人的框架，然后编写所需的规划服务即可为其生成路径.

I would go for a commonly used framework. As you are interested in path planning, pick a framework that already has simulated robots, and just write the planning services you need to generate paths for it.

ROS非常酷，并且有很多有用的感知服务以及完整的PR2堆栈可供选择.它使用凉亭(玩家/舞台的一部分)进行仿真，并包括一些现成的机器人(对于这种情况，不稳定的比较理想).它很新，但是非常活跃.

ROS is very cool, and has quite a large selection of useful perception services, as well as the full PR2 stack. It uses gazebo (part of player/stage) for simulation, and includes a few off the shelf robots (erratic would be ideal for this). Its quite new, but very active.

MRDS有几个合适的机器人，坦率地说比ROS更好的模拟环境，但是通过高级服务却不如ROS. lego机器人模拟器非常适合您想做的事情，并且让您可以选择将来廉价地在真实硬件上进行验证.

MRDS has several suitable robots, and frankly better simulated environments then ROS, but not quite as much by way of high level services. The lego robot simulator would be fine for what you want to do, and gives you the choice of validating on real hardware cheaply in the future.

这两个人目前在开发人员中占有很大的份额.真正的决定归结为您是否更愿意在Windows下的Visual Studio或Linux平台上进行开发.两者对学生都是免费的(MSRS在ELMS/dreamspark计划下)，都有学习曲线，都有很好的社区支持.对于MRDS，请注册论坛，对于ROS，请加入邮件列表.

These two have the lions share of developers at the moment. Really the decision boils down to whether you are happier developing in visual studio under windows, or on linux platforms. Both are free for students (MSRS is under ELMS / dreamspark programs), both have a learning curve, both have pretty good community support. For MRDS, go sign up to the forum, for ROS, join the mailing list.

这篇关于机器人模拟器的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！