Using PCA in time-of-flight vectors for reflector recognition and 3-D localization

This paper presents a reflector recognition and localization technique in three-dimensional (3-D) environments, using only times-of-flight (TOFs) data obtained from ultrasonic transducers. The recognition and localization technique is based on the principal component analysis applied to the TOF vect...

Full description

Saved in:
Bibliographic Details
Published in:IEEE transactions on robotics 2005-10, Vol.21 (5), p.909-924
Main Authors: Jimenez, J.A., Mazo, M., Urena, J., Hernandez, A., Alvarez, F., Garcia, J.J., Santiso, E.
Format: Article
Language:English
Subjects:
Acoustic applications
Applied sciences
Computer science
control theory
systems
Control theory. Systems
Exact sciences and technology
Frequency
Humidity
Indoor environments
Localization
Mathematical analysis
Mobile robots
Motion control
Position (location)
Principal component analysis
Principal component analysis (PCA)
Receivers
Recognition
reflector classification
Reflectors
Robot sensing systems
Robots
Sensors
Temperature dependence
Testing
three-dimensional (3-D) localization
times-of-flight (TOFs)
Transducers
ultrasonic sensors
Ultrasonic technology
Ultrasonic transducers
Vectors (mathematics)
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:This paper presents a reflector recognition and localization technique in three-dimensional (3-D) environments, using only times-of-flight (TOFs) data obtained from ultrasonic transducers. The recognition and localization technique is based on the principal component analysis applied to the TOF vectors originating from a sensor that contains two emitting transducers and several receivers. The two emitters simultaneously transmit two coded pulses that are detected later on and discriminated by the receivers, after being reflected in the environment. The proposed technique allows for the possibility of not only recognizing the reflectors, but also estimating approximately its localization referred to the sensor. This technique has been tested with three types of reflectors in 3-D environments: planes, edges, and corners. The achieved results are very satisfactory for reflectors located in the range 50-350 cm.
ISSN:1552-3098
1941-0468
DOI:10.1109/TRO.2005.851375