Localisation is one of the most important tasks to be accomplished in order to realize the complete autonomy of a mobile robot. In this paper, a new strategy for global localisation is proposed. Applying this method a robot is able to safely initialise its position or relocalise itself in case of recovery of pose tracking failure. The algorithm presented adopts a hybrid approach. First a particle filter is used to generate hypotheses on the possible pose supposing that no movements are allowed to avoid collisions. Thereafter safe trajectories are planned and executed to reduce the remaining ambiguities while the hypotheses are monitored and validated by a set of parallel extended Kalman filters. The novelty of this approach stands on the ability to generate the pose hypotheses without any feature-based knowledge. As a consequence, a landmark-based description of the environment is no longer required for the algorithm execution.

Gasparri, A., Panzieri, S., Pascucci, F., Ulivi, G. (2007). A Hybrid Active Global Localisation Algorithm for Mobile Robots. In IEEE Int. Conference on Robotics and Automation (ICRA 2007) (pp.3148-3153). Roma : IEEE [10.1109/ROBOT.2007.363958].

A Hybrid Active Global Localisation Algorithm for Mobile Robots

GASPARRI, ANDREA;PANZIERI, Stefano;PASCUCCI, Federica;ULIVI, Giovanni
2007

Abstract

Localisation is one of the most important tasks to be accomplished in order to realize the complete autonomy of a mobile robot. In this paper, a new strategy for global localisation is proposed. Applying this method a robot is able to safely initialise its position or relocalise itself in case of recovery of pose tracking failure. The algorithm presented adopts a hybrid approach. First a particle filter is used to generate hypotheses on the possible pose supposing that no movements are allowed to avoid collisions. Thereafter safe trajectories are planned and executed to reduce the remaining ambiguities while the hypotheses are monitored and validated by a set of parallel extended Kalman filters. The novelty of this approach stands on the ability to generate the pose hypotheses without any feature-based knowledge. As a consequence, a landmark-based description of the environment is no longer required for the algorithm execution.
1-4244-0602-1
Gasparri, A., Panzieri, S., Pascucci, F., Ulivi, G. (2007). A Hybrid Active Global Localisation Algorithm for Mobile Robots. In IEEE Int. Conference on Robotics and Automation (ICRA 2007) (pp.3148-3153). Roma : IEEE [10.1109/ROBOT.2007.363958].
File in questo prodotto:
Non ci sono file associati a questo prodotto.

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11590/177928
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 17
  • ???jsp.display-item.citation.isi??? 10
social impact