ASABE Technical Library - Abstract
Member and Access Notice
Monitoring and classifying the behavior of a herd of sheep using ad hoc wireless sensor networks and artificial intelligence
Published by the American Society of Agricultural and Biological Engineers, St. Joseph, Michigan www.asabe.org
Citation: 7th World Congress on Computers in Agriculture Conference Proceedings, 22-24 June 2009, Reno, Nevada (electronic only) 711P0409e.(doi:10.13031/2013.29083)Authors: Esmaeil S Nadimi, Rasmus N Jørgensen, Victoria Blanes-Vidal, Svend Christensen
Keywords: Keywords: Wireless sensor networks, Neural networks, Behavior classification, Kaiser window
Abstract: An in-depth study of ad hoc wireless sensor networks applied to monitoring the behavior of a herd of sheep in the field was carried out. Herd motion data such as head movements were monitored by IRIS motes and MTS310 sensor boards equipped with a 2-axis accelerometer in a multi hop wireless sensor networks. Acceleration measurements were transmitted through a wireless sensor network based on ZigBee (2.4 GHz) communication protocol. After data filtering, the measurements of the head movements were used to classify the animal behavior into two classes as grazing and non-grazing. Taking the advantages of artificial neural networks compared to the other classification methods into account, a standard two-layer feed-forward network was preferred. The neural network was constructed and trained based on the measurements of the head movements of some of the animals in the herd and was used to predict the behavior of other animals in the herd. The classification results showed the possibility of determining a general decision rule which can classify the behavior of each individual in a herd of animals. The results have been confirmed by manual registration.
(Download PDF) (Export to EndNotes)