Issues in using gym-gazebo in Ubuntu 14.04

Posted 卡斯柏的博客

tags:

篇首语:本文由小常识网(cha138.com)小编为大家整理,主要介绍了Issues in using gym-gazebo in Ubuntu 14.04相关的知识,希望对你有一定的参考价值。

Overview

OpenAI-gym (Git Repo) provides a great toolkit and a playgroud for research on Deep Reinforcement Learning. Furtherly, gym-gazebo (Git Repo), based on openAI-gym, simplies ROS / Gazebo user on transferring openAI tools into Gazebo.

The following is a record when I implemented this package in my own PC, in particular how I dealt with the installation and complie issues.

Environment

  • Ubuntu 14.04
  • ROS Indigo
  • Gazebo 7
  • pip 9.0.1
  • tensorflow
  • keras

Issues

  • Dependencies
    • pip 9.0.1 (make sure it is done before automatic installation)
      • sudo apt-get install python-pip python-dev
      • sudo apt-get upgrade python-pip python-dev
      • sudo pip install --upgrade pip
  • Installation bash
    • in the bash file setup_indigo.bash, I have to modify something like below to avoid the errors
      • after "# Create catkin_wsmodify", add "cd ~"
      • ‘vcs import < ../../gazebo.repos‘ to ‘vcs import < ../../gym-gazebo/gym_gazebo/envs/installation/gazebo.repos‘
      • right after it add a line: git clone https://github.com/ethz-asl/mav_comm.git
  • Invalid tag: environment variable ‘GYM_GAZEBO_WORLD_CIRCUIT2‘ is not set.
    • simply need to set according env variables
      • cd gym-gazebo/gym-gazebo/env/installation && bash turtlebot_setup.bash
  • KeyError: ‘ROS_PORT_SIM‘
    • https://github.com/erlerobot/gym-gazebo/issues/59
      • export ROS_PORT_SIM=11311

Reference

 


以上是关于Issues in using gym-gazebo in Ubuntu 14.04的主要内容,如果未能解决你的问题,请参考以下文章

18 Issues in Current Deep Reinforcement Learning from ZhiHu

r DT中的按钮(ziyasaeed in https://github.com/rstudio/DT/issues/178)

Top ASM diskgroup mount issues in RAC Environment (文档 ID 2246762.1)

“Given final block not properly padded. Such issues can arise if a bad key is used during“错误解决

“Given final block not properly padded. Such issues can arise if a bad key is used during“错误解决

How to make your issues in GitHub more professional? [Labels feature]