Authors:
Younes Raoui
1
;
El Houssine Bouyakhf
2
and
Michel Devy
3
Affiliations:
1
University Mohamed V at Agdal, CNRS; LAAS and Université de Toulouse, Morocco
;
2
University Mohamed V at Agdal, Morocco
;
3
CNRS; LAAS and Université de Toulouse, France
Keyword(s):
Mobile Robot, Indoor, Localization, RFID, Vision.
Related
Ontology
Subjects/Areas/Topics:
Informatics in Control, Automation and Robotics
;
Robotics and Automation
;
Sensors Fusion
;
Signal Processing, Sensors, Systems Modeling and Control
;
Vehicle Control Applications
;
Vision, Recognition and Reconstruction
Abstract:
A key function required on an autonomous mobile robot is the ability to localize itself accurately. This paper reports an RFID and vision-based localization scheme, which uses new rotation and scale invariant features as natural landmarks in static environments. The invariance of these features to image translation, scaling and rotation makes them suitable landmarks for mobile robot localization. Our RFID-based localization method presented in (Duarte et al., 2010) is made active minimizing the information entropy and improving the positioning accuracy. Then methods based on the detection of a priori learnt RFID tags and visual landmarks are combined in order to make more robust the method.