drcsim_docker

Docker images for gazebo simulation of Atlas using IHMC controllers version 0.11

Prerequisites

• Ubuntu 16.04
• Nvidia GPU > Fermi2.1
• Nvidia Drivers >= 384

Instructions

1. Install docker-ce using the instructions given below.

sudo apt-get update

# to uninstall older versions of docker
sudo apt-get remove docker docker-engine docker.io

sudo apt-get update
	
sudo apt-get install \
     apt-transport-https \
     ca-certificates \
     curl \
     software-properties-common
 	
curl -fsSL https://download.docker.com/linux/ubuntu/gpg | sudo apt-key add -
 	
sudo apt-key fingerprint 0EBFCD88
#Verify that you now have the key with the fingerprint 9DC8 5822 9FC7 DD38 854A E2D8 8D81 803C 0EBF CD88, by searching 	  	   the last 8 characters of the fingerprint.
 	
sudo add-apt-repository \
   "deb [arch=amd64] https://download.docker.com/linux/ubuntu \
   $(lsb_release -cs) \
   stable"
	
sudo apt-get update
	
sudo apt-get install docker-ce

2. Install nvidia-docker plugin. Please refer to prerequisites in that link.

docker volume ls -q -f driver=nvidia-docker | xargs -r -I{} -n1 docker ps -q -a -f volume={} | xargs -r docker rm -f
sudo apt-get purge nvidia-docker

sudo apt-get install nvidia-docker2
sudo pkill -SIGHUP dockerd

3. Setup environment variables for NVIDIA binaries and libraries. If these variables are not set correct, pointcloud from lidar wont be visible

# find out the version of nvidia drivers installed
glxinfo | grep "OpenGL core profile version" 
# output of this should be something similar to 
# OpenGL core profile version string: 4.5.0 NVIDIA 384.130
# In this case, the driver version is 384.130

# search for the bin, lib and lib32 directories
dpkg -L nvidia-384 # change nvidia-384 to the major number of installed version
# based on the output, relevant directories on my machine are /usr/lib/nvidia-384/bin, /usr/lib/nvidia-384, and /usr/lib32/nvidia-384

# set environment variables in ~/.bashrc. Use output from previous commands to get the correct directory
export NVIDIA_BIN="/usr/lib/nvidia-384/bin"
export NVIDIA_LIB="/usr/lib/nvidia-384"
export NVIDIA_LIB32="/usr/lib32/nvidia-384"

# source ~/.bashrc
source ~/.bashrc

#(Advanced users modify and source zsh :P)

4. Run the script to build docker image and run the container. Docker needs sudo access by default. If you see an error while running the docker image, use sudo. Refer to this if you wish to allow docker commands without sudo.

# clone the repository if you have not done that already
git clone https://github.com/WPI-Humanoid-Robotics-Lab/drcsim_docker.git  -b master --single-branch
	
# Go the drcsim_docker directory
cd drcsim_docker/
	
# When running it for the first time, it might take around 30m minutes or more depending on your internet speed
# The scripts have IHMC controller version number in the name
bash runDrcsimDocker.sh

5. Connect gazebo client on host machine (in a new terminal)

DUID=$((UID%256))
IP=${IPADDR:-172.16.$DUID.$DUID}
export GAZEBO_MASTER_URI=http://${IP}:11345  
gzclient

6. To run code on docker image export the following variables. You can add these to ~/.bashrc for ease of use.

DUID=$((UID%256))
IP=${IPADDR:-172.16.$DUID.$DUID}
export ROS_MASTER_URI=http://${IP}:11311
export ROS_IP=172.16.232.1 # Confirm this from ifconfig results

7. To stop the docker run docker stop drcsim_${USER}

Useful Aliases

DUID=$((UID%256))
export IP=${IPADDR:-172.16.$DUID.$DUID}
alias source_dock="export ROS_MASTER_URI=http://${IP}:11311 && \
                  export ROS_IP=172.16.232.1" # Confirm this from ifconfig results

export DRCSIM_DOCKER_DIR="~/drcsim_docker" # change this based on your configuration
alias start_dock="cd $DRCSIM_DOCKER_DIR && bash runDrcsimDocker.sh"
alias stop_dock="docker stop drcsim_${USER}"
alias gazebo_dock="GAZEBO_MASTER_URI=http://${IP}:11345 gzclient"

Troubleshooting

1. If docker doesn't start without errors, please run these commands.

cd 
wget https://raw.githubusercontent.com/WPI-Humanoid-Robotics-Lab/drcsim_docker/rbe-595/fix_docker.sh
chmod +x fix_docker.sh 
./fix_docker.sh 

2. Gazebo client starts fine, but atlas model is missing.
   Solution: You need atlas model resources on local machine to see atlas in gazebo client. Create a new workspace using atlas_gazebo_ws.yaml file and the instructions available here. Run catkin_make install and source install/share/drcsim/setup.sh and start gzclient from that terminal. you can modify the alias given above as alias gazebo_dock="source ~/drcsim/install/share/drcsim/setup.sh && \ GAZEBO_MASTER_URI=http://${IP}:11345 gzclient"

3. Camera and/or lidar topic are not available. Solution: Follow step 3 in instructions.