For a mobile robot to be able to work with people in their everyday environment it is essential to have the ability to localize itself using its internal representation of the environment. At the same time the robot needs to maintain this representation in response to the dynamics of the environment. This is very important because, for example, the appearance of a room in a house where a mobile robot would operate is not static over time: new objects are sometimes added, existing objects like pictures or carpets may be changed or moved, and old objects may be removed. This means that the map which the robot has built could become out-of-date after some time in a changing environment.

Our research focuses on how to update the robot’s internal representation of a changing environment during long-term operation. To this end, we introduce the following datasets which contain omni-directional images and/or laser scans along with odometry information  recorded inside changing environments over a relatively long period of time.

If you use the datasets, please cite the corresponding publications below.

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


University of Lincoln, UK

This work was carried out while Feras Dayoub was a PhD  student at the University of Lincoln, UK.

Örebro University, Sweden

Long-Term Laser Scans (5 weeks)

This work was carried out while Peter Biber was a PhD student at the University of Tübingen, during a research visit to Örebro University as an EU Marie Curie fellow. He is now based at Bosch Corporate Research.

Related Publications

These datasets were used in the following papers:

University of Lincoln dataset

Feras Dayoub, Tom Duckett and Grzegorz Cielniak, Long-Term Experiments with an Adaptive Spherical View Representation for Navigation in Changing Environments, Robotics and Autonomous Systems, Vol. 59, No. 5, pp. 285-295, May 2011.

[pdf] [BibTex]

Örebro University dataset

Peter Biber and Tom Duckett, Experimental Analysis of Sample-Based Maps for Long-Term SLAM, International Journal of Robotics Research, Vol. 28, No. 1, pp. 20-33, 2009.

[pdf] [BibTex]