Installation

Before installing the simulation environment, make sure your desktop is setup with a standard installation of ROS Indigo on Ubuntu 14.04. Once your APT repositories are configured, you can install the simulator:

$ sudo apt-get update
$ sudo apt-get install ros-indigo-fetch-gazebo-demo

Warning

Never run the simulator on the robot. Simulation requires that the ROS parameter use_sim_time be set to true, which will cause the robot drivers to stop working correctly. In addition, be sure to never start the simulator in a terminal that has the ROS_MASTER_URI set to your robot for the same reasons.

Starting the Simulator

The fetch_gazebo and fetch_gazebo_demo packages provide the Gazebo environment for Fetch. fetch_gazebo includes several launch files:

  • simulation.launch spawns a robot in an empty world.
  • playground.launch spawns a robot inside a lab-like test environment. This environment has some tables with items that may be picked up and manipulated. It also has a pre-made map which can be used to test out robot navigation and some simple demonstrations of object grasping.

To start the simplest environment:

$ roslaunch fetch_gazebo simulation.launch

Note that all of the environments will prepare the robot by tucking the arm and giving the head an initial command.

Note: Possibly you don't see the lidar scan as in the above picture. That is because the default set of visibility of lidar scan is false. You can manually modify it in its .xacro file.

$ roscd fetch_gazebo

$ cd robots

$ gedit fetch.gazebo.xacro

locate "laser_link" block:

modify this line:

<visualize>false</visualize>

as

<visualize>true</visualize>

Visualizing with RVIZ

Even though Gazebo has a graphical visualization, RVIZ is still the preferred tool for interacting with your robot.

$ rosrun rviz rviz

You can now manually set up your RVIZ visualization or re-run RVIZ with a configuration file using the command line. The default .rviz configuration file for Fetch can be loaded using:

$ roscd fetch_navigation/config
$ rviz -d navigation.rviz

To start with playground.launch:

$ roslaunch fetch_gazebo playground.launch

Check it in rviz:

Running the Mobile Manipulation Demo

There is a fully integrated demo showing navigation, perception and MoveIt! working together on the robot in simulation. To run the demo, start Gazebo simulator with the playground:

>$ roslaunch fetch_gazebo playground.launch

Wait until the simulator is fully running and then run the demo launch file:

>$ roslaunch fetch_gazebo_demo demo.launch

This will start:

  • fetch_nav.launch - this is the navigation stack with a pre-built map of the environment.
  • move_group.launch - this is the MoveIt configuration which can plan for the movement of the arm.
  • basic_grasping_perception - this is a simple demo found in the simple_grasping package which segments objects on tables and computes grasps for them.
  • demo.py - this our specific demo which navigates the robot from the starting pose in Gazebo to the table, raises the torso, lowers the head to look at the table, and then runs perception to generate a goal for MoveIt. The arm will then grasp the cube on the table, tuck the arm and lower the torso. Once the robot is back in this tucked configuration, the navigation stack will be once again called to navigate into the room with the countertop where the robot will place the cube on the other table.

Reference:

Fetch Robotics: Tutorial: Gazebo Simulation

 

Simulating a Fetch robot in Gazebo的更多相关文章

  1. Simulating a Freight robot in Gazebo

    Installation Before installing the simulation environment, make sure your desktop is setup with a st ...

  2. Making my own Autonomous Robot in ROS / Gazebo, Day 1: Building the static model

    Day 1: Setting up ROS: Indigo OS: Ubuntu 14.04 OS: Gazebo 7.0.0 Initialize the workspace To create t ...

  3. Gazebo機器人仿真學習探索筆記(七)连接ROS

    中文稍后补充,先上官方原版教程.ROS Kinetic 搭配 Gazebo 7 附件----官方教程 Tutorial: ROS integration overview As of Gazebo 1 ...

  4. implement "slam_karto" package on a Freight robot

    1. login ssh fetch@<robot ip or robot name> 2.  set robot master modify .bashrc in robot's com ...

  5. Gazebo Ros入门

    教程代码 First step with gazebo and ros • setup a ROS workspace • create projects for your simulated rob ...

  6. ROSCon 2016视频和幻灯片发布 ROS机器人操作系统重要参考资料

    ROSCon 2016视频和幻灯片发布 By Tully Foote on 十月19,2016 7:28 AM 全部PPT下载地址:http://pan.baidu.com/s/1gf2sn2F RO ...

  7. getting started with building a ROS simulation platform for Deep Reinforcement Learning

    Apparently, this ongoing work is to make a preparation for futural research on Deep Reinforcement Le ...

  8. ROS * 通过launch文件添加多个模型

    我添加的是dae模型,urdf文件过两天贴 方法一 : <launch> <!-- these are the arguments you can pass this launch ...

  9. ROS探索总结(六)——使用smartcar进行仿真

    转自:https://www.ncnynl.com/archives/201609/843.html 总结: 一.机器人描述文件三个: 机器人主体body文件: gazebo属性文件: 主文件 sma ...

随机推荐

  1. 161110、彻底征服 Spring AOP 之 实战篇

    Spring AOP 实战 看了上面这么多的理论知识, 不知道大家有没有觉得枯燥哈. 不过不要急, 俗话说理论是实践的基础, 对 Spring AOP 有了基本的理论认识后, 我们来看一下下面几个具体 ...

  2. centos安装apache

    查询是否已安装apache:  rpm -qa httpd 安装apache: yum install httpd -y 查询: chkconfig httpd on 启动: service http ...

  3. laravel5.1启动详解

    laravel的启动过程 如果没有使用过类似Yii之类的框架,直接去看laravel,会有点一脸迷糊的感觉,起码我是这样的.laravel的启动过程,也是laravel的核心,对这个过程有一个了解,有 ...

  4. Centos6.5和Centos7 php环境搭建如何实现呢

    首先我们先查看下centos的版本信息 代码如下: #适用于所有的linux lsb_release -a#或者cat /etc/redhat-release#又或者rpm -q centos-rel ...

  5. java正则表达式四种常用的处理方式是怎么样呢《匹配、分割、代替、获取》

    java 正则表达式高级篇,介绍四种常用的处理方式:匹配.分割.替代.获取,具体内容如下package test; import java.util.regex.Matcher; import jav ...

  6. Linux CentOS 6.5 yum安装MongoDB的操作

    安装MongoDB的方法有很多种,可以源代码安装,在CentOS也可以用yum源安装的方法.由于MongoDB更新得比较快,我比较喜欢用yum源安装的方法.64位CentOS下的安装步骤如下: 1.准 ...

  7. Codefroces Gym 100781A(树上最长路径)

    http://codeforces.com/gym/100781/attachments 题意:有N个点,M条边,问对两两之间的树添加一条边之后,让整棵大树最远的点对之间的距离最近,问这个最近距离是多 ...

  8. easyui常现错误

    1.easyui-tabs:当data-options的属性设置为true时,其tab内部的内容显示不出来. 2.设置easyui-panel的title格式及字体大小无效 解决方法:在设置title ...

  9. java UUID

    UUID是Universally Unique Identifier的缩写,它是在一定的范围内(从特定的名字空间到全球)唯一的机器生成的标识符.UUID具有以下涵义: 经由一定的算法机器生成 为了保证 ...

  10. hashMap底层put和get方法逻辑

    1.hashmap put方法的实现: public V put(K key, V value) { if (key == null) return putForNullKey(value); int ...