Objectives

The purpose of this assignment is to get comfortable writing ROS code to control the Turtlebot.

Requirements

For this project you will create a ROS package named "approach" that will contain code to move the robot so that it maintains a desired distance to the nearest object in front of it.

Your package must contain a python-based ROS node named approach that meets the following specifications:

• Subscribes to the /scan topic in order to determine the distance to the nearest object.
• Subscribes to the /approach/dist topic to determine the current goal distance. The message type for this topic is Float32.
• Publishes appropriate Twist messages to the topic /mobile_base/commands/velocity. The PID class should be used to control the velocity. You must also enforce a maximum velocity (maybe .3 meters/second?) to avoid toppling the robot.
• Publishes current PID errors to the topic /approach/pid_error. See below for the appropriate message type. Both the error and twist messages should be published at a rate of 10HZ.

You must also create two launch files:

• approach.launch This launch file should start the Kinect and start the approach node.
• approach_plot.launch This launch file should do everything that approach.launch does as well as bring up an rqt_plot node to display all three error terms.

I should be able to test your submitted code by checking it out and running

roslaunch approach approach.launch

or

roslaunch approach approach_plot.launch

Your launch file does not need to bring up the Turtlebot. You may assume that minimal.launch has already been run.

I'm providing the following code to get you started:

• pid.py This module contains a simple PID controller class.
• approach_teleop.py This is a teleop node that will be used to send messages to your controller and to display error terms. MODIFIED 1/28
• PIDError.msg. This is the message type for sending PID errors.

Submitting and Grading (MODIFIED 1/28)

You will submit this project the same way you submitted Lab3. I will checkout your code as follows:

 $ svn checkout https://svn.cs.jmu.edu/spragunr/GROUP/submit/approach/ cs480_ws/approach

Acknowledgments

This is based on a project originally designed by Bill Smart at Washington University in St. Louis.