build details

# Calibration - Line Follower Sensors

Modified 2020-03-04 by Timothy Scott

A Duckiebot set up according to Unit E-1 - Setup

Line follower sensors

Line follower sensors are calibrated.

The line follower sensors are the 4 sensors on the bottom of the front bumper of DBV2. They detect the brightness of the surface over which they are positioned. This can be used to differentiate between the black road surface, the white outer lines, the yellow center lines, and red intersection lines.

The sensors are very sensitive to their exact height above the ground, and this calibration procedure accounts for that. After calibration is done, all 4 sensors will give the same brightness values for the same color.

Step 1: Use Portainer to stop the `duckiebot-interface-dbv2` container.

Step 2: Use the following command to launch the calibration procedure:

``````$docker -H duckiebot name.local run --privileged -it -v /data:/data \ --name duckiebot-interface-calibration --network=host --rm \ duckietown/dt-duckiebot-interface-dbv2:daffy-arm32v7 \ roslaunch sensor_suite line_following_calibration.launch `````` Step 3: Follow the instructions on the command line. You will be asked to place the Duckiebot such that all 4 line follower sensors are on a particular color. The results of the calibration will be stored in `/data/config/calibrations/sensor_suite/line_follower/duckiebot name.yaml`. ## Validation Modified 2022-11-14 by Andrea Censi To validate that the calibration worked, do the following: Step 1: Start the `duckiebot-interface-dbv2` container in Portainer. Step 2: Start GUI tools on your computer using Duckietown Shell: ``````$ dts start_gui_tools duckiebot name --base_image duckietown/dt-core:daffy
``````

Step 3: Within GUI tools, run the command `rqt_plot`. A plot window should open.

Step 4: In the Plot window, select the topic `/duckiebot name/line_following_node/line_follower` and click on the ‘plus’ button.

Step 5: Place the Duckiebot such that the sensors are over various colors on a city tile, and observe the values in the plot. The values measured should be within +/- 0.2 of the ideal values given below.

## Demo

Modified 2022-11-14 by Andrea Censi