Introduction and Resources

The goal of this lab is to gain experience using the tf system in ROS. Unfortunately, the Python API for tf is not very well documented on the ROS wiki. Here are some places you can get information:

• Official tf tutorials
• Using Python tf (A ROS wiki page. This is not very complete.)
• The complete API doesn't seem to exist online in any one place. Most information can be found in the following sources:
  • tf.Transformer (This is the base class for other tf classes, but isn't often used directly.)
  • The rest of the API (This is a local copy of the documentation, generated directly from the Python files.)
  • TransformerListener This is the main class you will interact with directly.

Warmer-Colder: Practicing with tf

• Log into one of the netbooks, check out your group repository, and connect to a Turtlebot.
• Download and unzip the warmer_colder package to your package directory: warmer_colder.zip. The package includes a launch file that will start the Turtlebot as well as the sound_play and goal_publisher nodes.
• Take a few minutes to read the code in goal_publisher.py and transform_example.py. Once you understand those files, execute the warmer_colder launch file and run transform_example.py. If everything is working correctly you should see the goal point printed in both the /odom and /base_footprint coordinate frames.

• Modify warmer_colder.py so that its behavior matches the documentation.

Warmer-Colder: Practicing with tf

• If you finish the previous steps with time remaining, you have two options:
  1. IF EVERYONE IN YOUR GROUP HAS COMPLETED PA1: Create a new Python file named navigate_to_goal.py. Add code to that file to navigate the Turtlebot to the published goal location. Your node will reuse logic from warmer_colder.py as well as you solution to PA1.
  2. IF NOT EVERYONE IN YOUR GROUP HAS COMPLETED PA1:
     • Modify goal_publisher.py so that it broadcasts a transform at the goal position in addition to publishing the goal point. You can see an example of publishing a transform in a Python node here: Writing a tf broadcaster (Python). Here is more complete documentation of the TransformBroadcaster class.
     • Use rviz to make sure that your transform is being published correctly. The game should be much easier if you drive the robot while visualizing the goal transform.

Submitting

Commit your completed code using the log message "TF LAB FINAL SUBMISSION".