Introduction

The goal of this lab is to use numpy to improve our Turtlebot’s ability to wander without hitting obstacles.

Step 1:

Bringup the Turtlebot and rviz according to the instructions from this previous lab.

Then download and execute smart_wander.py. For now, just run this as a Python script:

python smart_wander.py

Click the “Add” button in the lower-left corner of the Rviz window to add a visualization for detected obstacles. If you select the “by topic” tab you should see an option for “obstacles -> MarkerArray”. Once you select that option, you should see a line of green spheres.

You may also want to run the following node, which will stop the Turtlebot whenever no other node is publishing movement commands: safe_cmd_vel.py.

Part 2

Open smart_wander.py and make sure you understand the provided code. Your goal is to update the timer_callback method so that it satisfies the following requirements:

• Calculate the X and Y coordinates of all detected obstacles based on the laser scan date. This must be done using numpy operations and there must be no loops in your code.
• Modify the visualization code so that obstacles are marked in red if they fall inside a “danger box” in front of the robot. Any obstacle less then one meter in front of the robot and within .2 meters to the left or right should be considered a collission hazard.
• Modify the control code so that the robot turns in place as long as there are any obstacles within the danger box. Otherwise the robot should move forward.

Hints and Suggestions:

• The first step will be to convert self.scan.ranges to a numpy array.
• You’ll also need a numpy array containing the angles associated with each scan. The arange function will be useful for creating this.
• Some of the range values will be invalid, which will result in invalid coordinates. Don’t worry about it: the visualization code is configured to discard invalid coordinates.

Part 3

Once you are happy with the behavior of your node, turn it into a ROS 2 package named smart_wander and submit through GitHub classroom.