KTH Longterm Dataset

KTH Scitos G5 robot - Rosie
The data was collected autonomously by a Scitos G5 robot with an RGB-D camera on a pan-tilt, navigating through the KTH office environment over a period of approximately 30 days. Each observation consists of a set of 51 RGB-D images obtained by moving the pan-tilt in a pattern, in increments of 20 degrees horizontally and 25 vertically. Waypoints are visited between 80 and 100 times and a total of approximately 720 observations are collected at the eight waypoints that can be seen in the figure below. The data is a part of the Strands EU FP7 project.
WayPoint positions on the mapObservations overlayed on the 2D map

Dataset structure

The data is structured in folders as follows: YYYYMMDD/patrol_run_YYY/room_ZZZ , where: Each folder of the type YYYMMDD/patrol_run_YYY/room_ZZZ contains the following files: The room.xml file accompanying an observation contains the following (relevant) fields:


A parser is provided here (can be installed with sudo apt-get install ros-indigo-metaroom-xml-parser) which reads in the data and returns C++ data structures encapsulating the low-level data from the disk. Form more information please refer to the parser README ( or here for a list of supported methods). Information about how to use the Strands package repository can be found here.


This dataset is available for download in a single archive file (~ 300 GB). As an alternative, the individual folders and files can be obtained from here, and would have to be downloaded manually.

Condition of use

If you use the dataset for your research, please cite our paper that describes it:

	Unsupervised learning of spatial-temporal models of objects in a long-term autonomy scenario 
	Ambrus, Rares and Ekekrantz, Johan and Folkesson, John and Jensfelt, Patric
	Intelligent Robots and Systems (IROS), 2015 IEEE/RSJ International Conference on
We attached a bibtex record for your convenience.