One of the fundamental problems of mobile robotics is to navigate autonomously. To fulfill this task, a robot must know its position relatively to the position of its goal. Moreover, a robot has to take in consideration the dangers of the surrounding environment and adjust its actions to maximize the chance to reach the destination. Putting is simply, robot navigation can be reduced to three questioidns: Where am I ? Where am I going ? How do I get there ? These three questions are answered by localization, mapping and motion planning respectively. In the context of mobile robotics, localization means determination of the robot position in the environment. Acquiring spatial information of the environment through mobile robot sensors is called mapping. The process of generating inputs for robot effectors to reach the desired destination is referred to as motion planning. Although one can address these problems separately, they are closely related and especially localization and mapping are often studied together.

Our aim is to research methods solving the problem of mobile robot navigation in real environments. These methods should use only the robot onboard sensors and should be independent on external positioning systems or other similar infrastructure. These methods should be able to deal with real world conditions and satisfy real time constraints with off the shelf sensors and computational hardware. The methods should work in large-scale environments for extended periods of time.

People involved:

Libor Přeučil
Luis Gomez Camara


Libor Přeučil


(:includefile demos/demos_navigation.html:) see more demos ...

Selected publications:

  • Camara, L.G. - Gäbert, C. - Přeučil, L.: "Highly Robust Visual Place Recognition Through Spatial Matching of CNN Features". Accepted at the 2020 International Conference on Robotics and Automation (ICRA 2020), Paris.
  • Camara. L.G. - Přeučil, L.: Spatio-semantic Convnet-based Visual Place Recognition. In 2019 European Conference on Mobile Robots (ECMR), Prague: IEEE, 2019, p. 1-8. ISBN 978-1-7281-3605-9.
  • Krajník, T. - Faigl, J. - Vonásek, V. - Košnar K. - Kulich, M. - and Přeučil, L.: Simple, yet stable bearing-only navigation, J. Field Robot., 2010.
  • Krajník, T. - Vonásek, V. - Fišer, D. - Faigl, J.:AR-Drone as a Platform for Robotic Research and Education. In International Conference on Research and Education in Robotics. Heidelberg: Springer, 2011, p. 172-186. ISBN 978-3-642-21974-0.
  • Krajník, T. - Přeučil, L.: A Simple Visual Navigation System with Convergence Property. In European Robotics Symposium 2008. Heidelberg: Springer, 2008, p. 283-292. ISBN 978-3-540-78315-2.
  • Šváb,J. - Krajník,T. - Faigl,J. - Přeučil,L.: FPGA-based Speeded Up Robust Features. In 2009 IEEE International Conference on Technologies for Practical Robot Applications [CD-ROM]. Boston: IEEE, 2009, p.35-41. ISBN 978-1-4244-4992-7.
  • Košnar, K. - Krajník, T. - Přeučil, L.: Visual Topological Mapping. In European Robotics Symposium 2008. Heidelberg: Springer, 2008, p. 333-342. ISBN 978-3-540-78315-2.

Projects and funding:

Replicator - European Union grant
Symbrion-Enlarged - European Union grant no. 216342
Page last modified on January 22, 2020, at 10:14 AM EST
Powered by PmWiki