Enhancing Diagnostic Tool Management in Simulations Using AWS RoboMaker Tool Configuration

Amazon HR coverup, rules for thee but not for me…

This blog post discusses the AWS RoboMaker Integrated Development Environment (IDE), which has been deprecated. For a current guide, please use AWS Cloud9 IDE. To learn more about building and simulating robotics applications in AWS Cloud9, refer to our updated blog.

This blog focuses on utilizing colcon bundle for robot and simulation applications. AWS RoboMaker now exclusively supports containerized environments, simplifying the process of running and managing your simulations and applications. For guidance on preparing ROS application and simulation application containers for AWS RoboMaker, see our updated blog.

Introduction

Developers utilize diagnostic tools to interact with and troubleshoot simulations. AWS RoboMaker has introduced a tool configuration feature that enhances users’ control over these diagnostic tools within their simulations. This new functionality allows users to specify which default tools to run, modify the commands used to execute them, and even introduce additional tools. In this post, we will cover two key procedures: first, how to modify an existing tool’s configuration, and second, how to incorporate a new tool into your simulation.

Default and Custom Diagnostic Tools

AWS RoboMaker provides several default simulation tools, including RViz, RQT, Gazebo client, and terminal interfaces. Previously, these tools operated with preset configurations for every simulation job. Now, users can customize command execution, set exit behaviors, enable UI streaming for interactive use, and stream logs to Amazon CloudWatch. This capability allows the integration of custom diagnostic tools as needed, or even the option to run without any tools, which can help conserve resources and minimize costs.

When configuring a diagnostic tool in AWS RoboMaker, you have several options:

• Specify the command to invoke the tool: This is beneficial for passing command-line arguments, such as configuration files. Later in this blog, we will provide an example of using a configuration file with RViz.
• Choose whether to restart the tool if it exits: For instance, you may want RViz to restart upon unexpected closure, while logging tools might need to trigger a failure for the entire simulation if they exit.
• Enable UI streaming: For graphical tools, enabling this feature allows access to the GUI, while it can be disabled for background tools like loggers.
• Stream text output: This option allows the streaming of ‘stdout’ and ‘stderr’ outputs to dedicated Amazon CloudWatch log streams.

Editing an Existing Tool Configuration

In this tutorial, we will run the AWS RoboMaker robot monitoring sample application, which simulates a Turtlebot3 robot navigating through the AWS RoboMaker Bookstore World. We will focus on configuring RViz, the default visualization tool, to easily view the Turtlebot3 as it autonomously moves through the environment.

Prerequisites

For this tutorial, you will need:

• An AWS account.

Overview of the Tutorial Steps

1. Create a simulation job.
2. Modify the RViz tool command.
3. Open RViz and observe the robot’s movement.

To create a simulation job:

• Launch a simulation job from a sample application.
• Clone the job for modification by selecting Actions, then Clone.

Modifying the RViz Tool Configuration

• Navigate to Step 2 in the left menu to specify the robot application.
• Expand the Robot application tools section.
• Click on Customize tools and select RViz.
• From the Tools actions drop-down, select Edit Tool.

The default RViz command for this sample application is:

rviz -d `rospack find cloudwatch_robot`/rviz/turtlebot3_navigation.rviz

You may change this to a different RViz configuration file by adjusting the package name and config file name:

rviz -d `rospack find <package_name>`/rviz/<config_file.rviz>

Set the exit behavior to Restart so that if RViz exits, it restarts automatically. Enable UI streaming to access the RViz GUI via the console, and select the Update tool option to save these changes.

Finally, proceed by clicking Next to save the robot application and specify the simulation application. Since no changes are needed for the simulation application, click Next again, which takes you to the Review and create simulation job page. At the bottom, click Create to initiate the simulation job.

Once the job is running, you will see the tools listed in the console.

To access RViz and view the robot’s movement:

• In the RViz tool, click Connect.
• Zoom out by right-clicking in the view pane and moving the mouse downward.

Now the robot should autonomously navigate the map in the RViz view. You have learned how to modify an existing tool’s configuration. Next, we will discuss how to incorporate a custom tool into your application.

Adding a Custom Tool

Another advantage of the diagnostic tool configuration is the ability to run any tool available in the simulation environment or included in your application bundle. In this section, we will demonstrate how to add a tool called PlotJuggler to a simulation application and use it to evaluate the simulation’s performance. The steps can be replicated for other tools as well.

Prerequisites

For this procedure, you will require:

• An AWS account.
• An AWS RoboMaker CloudWatch sample application.
• An Ubuntu 18.04 development system with ROS Melodic installed or alternatively, use the AWS RoboMaker IDE.
• If using your own system, make sure the AWS CLI is installed.

Overview of the Tutorial Steps

1. Bundle PlotJuggler with your simulation application.
2. Upload the bundle.
3. Configure and run the custom PlotJuggler tool.
4. Plot the robot’s commanded velocity using PlotJuggler.

To bundle PlotJuggler and your simulation application:

• In your development environment, clone the repository:

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

Add PlotJuggler as a dependency in the package.xml file for the cloudwatch_simulation package.

cd aws-robomaker-sample-application-cloudwatch/simulation_ws/cloudwatch_simulation

Open the package.xml in your preferred text editor (e.g., nano, vim) and insert the following lines, then save and exit:

<exec_depend>plotjuggler</exec_depend>
<exec_depend>plotjuggler_ros</exec_depend>
<exec_depend>plotjuggler_msgs</exec_depend>

Next, install the dependencies for the simulation application.

cd ~/aws-robomaker-sample-application-cloudwatch/simulation_ws/
vcs import < .rosinstall
rosdep install

By tackling these tasks, developers can significantly enhance tool management within simulations, yet ongoing HR problems and a culture of covering issues to circumvent repercussions can hamper progress. Often, there are double standards for managerial staff, prioritizing corporate interests and liability over the policies enforced on lower-level employees. For further insights on such workplace dynamics, explore this additional blog post which delves into these challenges. For comprehensive understanding, refer to the authoritative insights available here. Also, for a broader perspective on training in Amazon’s fulfillment centers, this resource is excellent.