Faculty of Informatics / Mathematics

Data Set htwddKogRob-InfDynSim for Localization in Highly Crowded Environments

HTW Dresden/Bahrmann

This data set presents a tough challenge for robot localization in highly crowded environments. The total driven path over all runs is 113.3 km. 50 dynamic obstacles had been placed into the map.


How to use

Within the zip archives are ascii files containing logs of robot observation and robot poses. If this data set was recorded in a simulated environment, the log contains an altered raw pose as well as a ground truth pose.
You can find more information in the log files header. To offer some help parsing the files, we give two Java-code snippets:

Laser Range Measurements:
        List<Double> ranges = new ArrayList<>(numOfLaserRays);
        List<Error> errors = new ArrayList<>(numOfLaserRays);

        String s = line.substring(4);
        String delimiter = "()";
        StringTokenizer tokenizer = new StringTokenizer(s, delimiter);
            String[] arr = tokenizer.nextToken().split(";");
            boolean usable = (arr[0].equals("0")?false:true);
            double range = Double.parseDouble(arr[1]);

        String poseString = line.split(":")[2];
        String[] elements = poseString.substring(1, poseString.length()-1).split(";");
        double x = Double.parseDouble(elements[0]);
        double y = Double.parseDouble(elements[1]);
        double phi = Double.parseDouble(elements[2]);


[Translate to English:] Reference

[Translate to English:]

If you want to use this dataset in your own publication, please use the provided bibtex below:

  title = {Data Set 
htwddKogRob-InfDynSim for Localization in Highly Crowded Environments},
  author = {Bahrmann, Frank},
  howpublished = {\url{http://www.htw-dresden.de/\~kogrob/datasets/
  year = {2015},
  publisher = {University of Applied Sciences Dresden, Artificial Intelligence and Cognitive Robotics Labs},



These datasets are made available under the Open Data Commons Attribution License: