Title :
Modelling and design of sensor-fusion: a novel computational paradigm based on interactions, and Karhunen-Loeve transform
Author :
Gianfelici, Francesco
Author_Institution :
Dipartimento di Elettronica, Intelligenza Artificiale e Telecomunicazioni, Universita Politecnica delle Marche, Ancona
fDate :
Sept. 30 2005-Oct. 1 2005
Abstract :
A novel computational paradigm based on the composition of interactions, and Karhunen-Loeve transform is proposed in this paper. This paradigm has been developed with the purpose of achieving a suitable representation in the modelling, and the design of sensor-fusion. The effectiveness of this approach is guaranteed by the composition of these two representations, which enables the regulation of nondeterministic behaviors of complex systems by means of several properties, and capabilities, at various levels of abstraction, and with different expressiveness degrees. This characterization represents a new contribution to the development of systematical representations, based on rigorous mathematical formulations that is defined with the purpose of reducing the empiric aspects of the modelling, and design of the sensor-fusion. The effectiveness of this approach is briefly verified by means of a case study that is based on a general representation equipped with KLT-based processing
Keywords :
Karhunen-Loeve transforms; large-scale systems; sensor fusion; Karhunen-Loeve transforms; complex systems; computational paradigm; interaction composition; nondeterministic behavior regulation; sensor-fusion modelling; Computational and artificial intelligence; Computer science; Concurrent computing; Intelligent sensors; Karhunen-Loeve transforms; Mathematical model; Sensor fusion; Sensor phenomena and characterization; Stochastic processes; Telecommunication computing;
Conference_Titel :
Robotic Sensors: Robotic and Sensor Environments, 2005. International Workshop on
Conference_Location :
Ottawa, Ont.
Print_ISBN :
0-7803-9378-3
DOI :
10.1109/ROSE.2005.1588335
