datamatrix: tf frames from vision the easy way

datamatrix subscribes to a stereo camera, looks for coded markers, and publishes the pose of any markers it finds as tf frames.

To use it, you:

• print out some markers with 3-letter names of your choice
• run the datamatrix node for the camera
• put the markers in a visible place

• start using the new tf frames

printing markers

Use the makeps.py script in datamatrix:

$ rosrun datamatrix makeps.py CRT | lpr

Usage:

Usage: makeps.py [options] code

Options:
  -h, --help            show this help message and exit
  -s SIZE, --size=SIZE  size of marker in m, default is .1
  -o OUT, --out=OUT     output filename, default is standard output

running the node

Takes a single parameter stereo as the base camera topic:

$ rosrun datamatrix datamatrix.py stereo:=/wide_stereo

Or in monocular mode:

$ rosrun datamatrix datamatrix.py image_stream:=/wide_stereo/left/image_rect camera_info:=/wide_stereo/left/camera_info