In the directories "dlr-spatial_cognition-[a-h]" you will find the data for each
run. map.lines is the data set with line landmarks, map.circles for the circular
landmarks. The format of the data set it documented in the file itself.


For a complete documentation of the dataset have a look at the technical report
included in the directory "techreport".

in the "src" directory you will find some python code to import and display the
data sets, which may be used for own derivations. We also added some additional
code which we developed for ourselves, like a Levenberg-Marquardt Optimizer and
EKF SLAM. They can also be used for your own works, if you like.

The files robot.jpg and robot2.jpg give you an impression of the real robot
and the scene.

map.png is a real map of the building where the data set was recorded.
The red line is the path the robot has driven throughout the experiment.

The whole experiment took place in the labs from the DLR (Deutsches Zentrum
für Luft- und Raumfahrt) on 2003/08/24. The experiment was used for the
Ph.D. Thesis from Udo Frese on the treemap algorithm.

Webpage: http://www.informatik.uni-bremen.de/agebv/en/DlrSpatialCognitionDataSet


