Skip to content

A Simulation Environment to train Robots in Large Realistic Interactive Scenes

License

Notifications You must be signed in to change notification settings

RIVeR-Lab/iGibson

This branch is 132 commits ahead of StanfordVL/iGibson:master.

Folders and files

NameName
Last commit message
Last commit date
Feb 19, 2023
Feb 19, 2023
Oct 6, 2023
Feb 21, 2025
Oct 27, 2022
May 14, 2018
Dec 24, 2020
Oct 26, 2023
Mar 8, 2022
Apr 19, 2022
Dec 18, 2017
Mar 10, 2022
Feb 14, 2024
Oct 23, 2023
May 14, 2021
Oct 9, 2023
Mar 10, 2022
Sep 17, 2022
Oct 18, 2023

Repository files navigation

mobiman-igibson


References:

iGibson Installation:

  1. Install python library:
pip install igibson
  1. Clone the igibson repository into a folder, such as "projects":
cd
mkdir projects
cd projects
git clone git@github.com:RIVeR-Lab/iGibson.git --recursive
cd iGibson
git checkout mobiman-devel-v0
pip install -e .
  1. Clone the igibson repo into the src folder of a catkin workspace, such as "catkin_ws":
cd ~/catkin_ws/src
ln -s ~/projects/iGibson/igibson/examples/ros/igibson-ros/ .
cd ..
  1. Install ROS dependencies:
rosdep install --from-paths src --ignore-src -r -y
  1. Build and source:
catkin build
source devel/setup.bash
  1. Download assets:

6.1 Download the robot models and some small objects as in iGibson's documentation.

python -m igibson.utils.assets_utils --download_assets

6.2 Clone the sim_utility repository into a folder.

git clone git@github.com:RIVeR-Lab/sim_utility.git

6.3 Copy the jackal_jaco and ycb folders into "iGibson/igibson/data/assets/models" folder.

Example 1: Multi-Turtlebot simulation in iGibson and observations are published in ROS:

In a terminal, within igibson-ros workspace:

  1. Start ROS and simulation in iGibson:
roslaunch igibson-ros turtlebot_rgbd_multi.launch

Example 2: Multi-Turtlebot training, in iGibson simulator, using Stable-Baselines3 and observations are published in ROS:

In a seperate terminal, :

  1. In a terminal within the igibson-ros workspace, start ROS:
roslaunch igibson-ros stable_baselines3_ros_turtlebot_multi.launch
  1. In a terminal within the scripts folder in the igibson-ros workspace:

2.1 Source the ROS workspace:

source ~/catkin_ws/devel/setup.bash

2.2. Run the script:

# cd igibson/examples/ros/igibson-ros/scripts/
python stable_baselines3_ros_turtlebot.py

About

A Simulation Environment to train Robots in Large Realistic Interactive Scenes

Resources

License

Stars

Watchers

Forks

Packages

No packages published

Languages

  • Python 59.5%
  • C 21.3%
  • C++ 16.4%
  • GLSL 2.0%
  • CMake 0.3%
  • Shell 0.2%
  • Other 0.3%