The Usage of Artificial Neural Networks for Intelligent Lighting Control Based on Resident’s Behavioural Pattern

  • Agne Paulauskaite-Taraseviciene Kaunas University of Technology
  • Nerijus Morkevicius Kaunas University of Technology
  • Audrone Janaviciute Kaunas University of Technology
  • Agnius Liutkevicius Kaunas University of Technology
  • Arunas Vrubliauskas Kaunas University of Technology
  • Egidijus Kazanavicius Kaunas University of Technology
Keywords: Artificial neural network, behavioural pattern, intelligent light control, resilient backpropagation, on-line learning

Abstract

Learning from the behavior of the resident is essential in order to adapt the system and to provide intelligent control based on behavior patterns. Different homes have different conditions and habits which have to be taken into account for the intelligent system to be useful. However, even deeply ingrained habits are subject to change over time. Therefore, intelligent system has to respond to changing and diverse environment. In this paper the intelligent lighting control system based on resident’s behavioural patterns is presented that employs artificial neural networks for on-line learning and adaptation. In order to manage constantly growing data and dynamic environment problems, the latter algorithms have been improved with the proposed similarity threshold based data replacement algorithm which has been experimentally tested and compared with alternative algorithms.

DOI: http://dx.doi.org/10.5755/j01.eee.21.2.8772

Published
2015-04-09
Section
SYSTEM ENGINEERING, COMPUTER TECHNOLOGY