GZWeb Docker Example for RoboMaker Sample Application Local Development

(Note: this is experimental and will be updated frequently...)

This docker file will enable you to run AWS RoboMaker sample applications in gzweb with a docker container.

NOTE: For the code of these demos to connect to AWS, the docker container will need access to an AWS Access Key ID and Secret Key.

Getting Started

Build the Docker Image

Clone this repository and build the docker image:

cd ~
git clone https://github.com/jerwallace/aws-robomaker-sample-apps-gzweb.git
cd aws-robomaker-sample-apps-gzweb
docker build gzweb-docker/. -t gzweb:latest
docker build ros-docker/. -t rosapp:latest
sudo chmod +x run*

Download an AWS RoboMaker Sample Application

If using the AWS RoboMaker Cloudwatch Sample Application, clone this repo:

git clone https://github.com/aws-robotics/aws-robomaker-sample-application-cloudwatch.git

You could also use your own ROS application.

Set your AWS Access Credentials as Environment Variables

To connect to your AWS account, remember to set the environment variables for your credentials.

export AWS_ACCESS_KEY_ID=<YOUR_ACCESS_KEY_ID>
export AWS_SECRET_ACCESS_KEY=<YOUR_SECRET_KEY>

Running

To run gzweb, use the shell script provided ./run.sh. Commands you can run are:

./run.sh --build-only # this command will use colcon in a container to build the sample app.
./run.sh --build # this command will use colcon in a container to build the sample app, then it will run the docker-compose up command in a silent shell
./run.sh -d # this command will run docker compose in detached mode (no shell output).

The default settings will work with the cloudwatch sample application, however feel free to modify them.

##======= CHANGE ME ========##
HOME=~/aws-robomaker-sample-apps-gzweb
WORKSPACE_DIR=aws-robomaker-sample-application-cloudwatch
WORLDS=(aws_robomaker_bookstore_world aws_robomaker_small_house_world)
export GZWEB_DOCKER_IMAGE=gzweb:latest
export ROS_DOCKER_IMAGE=rosapp:latest
export ROS_PACKAGE_SIM=cloudwatch_simulation
export ROS_LAUNCH_FILE_SIM=bookstore_turtlebot_navigation.launch
export ROS_PACKAGE_ROBOT=cloudwatch_robot
export ROS_LAUNCH_FILE_ROBOT=monitoring.launch 
export TURTLEBOT3_MODEL=waffle_pi
export ROBOT_APP_INSTALL=$HOME/$WORKSPACE_DIR/robot_ws
export SIM_APP_INSTALL=$HOME/$WORKSPACE_DIR/simulation_ws
##==========================##

Once running, open your browser to http://localhost:8080 to interact with GZWeb. Also, if you add a rosbridge to your launch file, you can checkout WebViz locally here: https://webviz.io/app

If you are in Cloud9, the Preview App functionality will automatically map 8080 to 80. So, simply press "preview app.".