ROS -Task 1 -Bagging -Turtle mouse follow session -Bag analysis -correction - Solved

For this task, you will be using the code from Lab exercise 5. Make sure you have roscore running in a separate terminal, as explained above.

a)    Run an instance of Turtle mouse following using the solution of Exercise 5, and take note of the topic names. Which command will you use to list all the topic names?

Paste the list of all topic names here, and shut down the ROS nodes.

b)    In order to record the mouse position topic, and the pose and cmd_vel topics of the turtle into a bag called turtlefollow_20xx-xx-xx-xx-xx-xx.bag, which rosbag command will you use?

_20xx-xx-xx-xx-xx-xx indicates the time of recording, which needs to be automatically added by rosbag when using the appropriate option – do not add this yourself to the bag filename.

Enter the full command line into the box.

Hint: Use the -h option to read the documentation of rosbag. You can also use -h to view the options for the record subcommand.

c)    First start recording by first running the command from b), and then run the nodes from the Turtle mouse follow exercise. Make the turtle follow the mouse for around 30 seconds, and then let it catch the mouse. Stop recording (Ctrl-C) and shut down all ROS nodes.

d)    Examine the sample script process_bag_example.py which demonstrates how to loop through all messages in a bag and write them into another bag without any changes.

e)    Write a script called process_turtle_follow.py which will take in a single command-line argument, an input bag filename, and perform the following:

i)       Calculate and display the distance covered by the turtle. To perform the covered distance calculation, sum up all Euclidean distances between sequential turtle positions, available in messages on the Turtlesim pose topic.

ii)     Calculate and display the mouse follow session duration, defined by the time difference between the first and last pose topic message of the turtle. (See the sample ellipse.py script in Task 2 for an example of calculating time differences using rospy.Time.)

iii)    Calculate and display the average velocities for the follower turtle by dividing the covered distance with the mouse follow session duration.

iv)    Write and count the messages in a new bag called processed_follow.bag, which will contain two topics:

i.      the turtle pose topic, renamed to /follower/pose;

ii.    the mouse positions topic, renamed to /mouse_positions_on_grandparents_computer, with the mouse positions scaled down to the nearest pixel on a monitor with an 800x600 resolution. (For example, on a full HD 1920x1080 display, the position (123, 456) should be mapped into (51, 253), because 51 = round(123/1920 · 800), and likewise for the other dimension.

The output of process_turtle_follow.py must look like this (the values listed are placeholder numbers for illustration purposes only and should not be considered meaningful):

$ ./process_turtle_follow.py turtlefollow_2021-12-01-20-38-01.bag

Processing input bagfile: turtlefollow_2021-12-01-20-38-01.bag

Follower turtle

Covered distance: 12.34 m

Average velocity: 56.78 m/s

Follow session duration: 34.56 s

Wrote 1234 messages to processed_follow.bag

Task 2: Laser sensor visualization

In this task, you will use Matplotlib and RViz to visualize range data from real-world 2D laser sensor data captured in the Faculty building. The data is in the bag file provided with the exercise, b-zgradaminimal.bag.

a) Run and thoroughly examine the two provided ROS node scripts, ellipse.py and visualizer.py. These two scripts are sample code to help you get started in solving this task.

The ellipse.py script calculates a set of points from an ellipse. The lengths of ellipse axes are animated to help illustrate that we are visualizing data which is changing in real time. The points are stored in the points field of a message of type visualization_msgs/Marker and published on the topic point_positions.

The visualizer.py script opens a Matplotlib window, subscribes to the point_positions topic, and plots the received points in the Matplotlib window.

Besides viewing the marker data in the Matlotlib window, visualize it in RViz as well. Run RViz (rviz), set the camera type to TopDownOrtho in the right panel, on the left panel in Global options, set the Fixed frame to match the frame id in header.frame_id in the marker message published in ellipse.py. Finally, add a Marker display using the Add button on the bottom of the left panel.

If adding display by type (first tab in the Add window), you need to set the topic to point_positions afterwards in the left panel, under the options of the Marker display you added. Alternatively, in the Add window, you can go to the second tab (By topic) and select the topic directly, which will automatically determine the display type (Marker).

Tip: After setting up everything RViz, press Ctrl-S or File -> Save Config to save the current settings as default when starting up RViz.

b)    Examine the sensor_msgs/LaserScan message (http://docs.ros.org/en/api/sensor_msgs/html/msg/ LaserScan.html) and the provided bag b-zgrada-minimal.bag using rosbag info. What is the name of the laser scan topic in the provided bag?

Helpful tips for using rostopic echo: you can use rostopic echo to examine only parts of a message. For example, to view timestamps in the header of a LaserScan message on a topic named scan, you can use the following command: rostopic echo scan/header/stamp. To print only a single (1) message, you can pass -n1. To print messages straight from a bag without having to play it using rosbag play, you can pass -b bagname.bag to rostopic echo.

c)    In the following subtasks, let j be the last two digits of your JMBAG + 100. For example, if your JMBAG is 0036465831, j = 131.

j =

d)    Write the full rostopic echo command for printing the angle_min field of the LaserScan message, for first j messages in b-zgrada-minimal.bag. Note: all angles are in radians.

What is the value of angle_min?

What is the value of angle_increment?

The formula for calculating the angle of the range measurement with an index i (where the first measurement has the index of 0) is: angle = angle min + i ∗ angle increment. What would be the angle for a range measurement with the index j as defined in c)?

e)    Note: before running any nodes (or RViz) which will be visualizing data from a played bag, make sure you run rosparam set use_sim_time true to use the simulated clock from the bag instead of the wall (system) clock. Make sure you have a roscore running in a separate tab.

(For running the sample ellipse.py from a) which animates the ellipse based on the system clock, use_sim_time should be set to false, which is the default value after after starting up roscore.)

f)     Write a script for a ROS node named laserscan_to_points.py based on the sample ellipse.py script. This ROS node must subscribe to a sensor_msgs/LaserScan topic named scan (not directly to the topic from b-zgrada-minimal.bag). Instead of publishing a marker containing an ellipse, calculate the Cartesian (x,y) points using the range readings from received laser scans. See subtask d) for calculating the angle of a range reading.

Set the marker frame id to the frame id from the received LaserScan message. (Do not hardcode it, read it from the received message.) Also copy the stamp from the received LaserScan message. You can accomplish both of these at once by simply copying the entire header.

Increase the plot extents in visualizer.py to be able to fit the range data.

Which topic remapping will you use when running laserscan_to_points.py to visualize the scans from b-zgrada-minimal.bag? Enter the full command line.

g)    Start the laserscan_to_points.py node with the command line above, along with visualizer.py. Play the provided bag using the command rosbag play --clock b-zgrada-minimal.bag. Verify that your visualization is working correctly.

Tip: you can pass --loop to keep looping the bag until you quit with Ctrl-C.

h)    What is the frame id (in the header) of the LaserScan messages in b-zgrada-minimal.bag?

How did you examine this?

Start RViz with the setup as described in a), and change the Global frame to the above frame id.

Verify that the Marker display for the point_positions topic looks identical to the visualizer.py Matplotlib window.

Finally, in RViz, add a native LaserScan display for the LaserScan topic directly. Make sure it matches the markers published by laserscan_to_points.py on the point_positions topic. If it looks different, you have incorrectly solved the subtask f) and will not receive any points for Task 2!

Feel free to adjust the LaserScan display colors in the left panel, and the Marker scale and color in laserscan_to_points.py.

Save the RViz config containing both the LaserScan and Marker displays using File -> Save Config As... in your exercise folder as exercise_06.rviz.