An automated positioning system for monitoring chickens’ location: Accuracy and registration success in a free-range area
Publication Type: |
Journal Article |
Year of Publication: |
2018 |
Authors: |
Lisanne M. Stadig, Bart Ampe, T. Bas Rodenburg, Bert Reubens, Jarissa Maselyne, Shaojie Zhuang, Johan Criel, Frank A. M. Tuyttens |
Publication/Journal: |
Applied Animal Behaviour Science |
Keywords: |
accuracy, outdoor, poultry, signal reception, ultra-wideband, vegetation |
ISBN: |
0168-1591 |
Abstract:
Free-range use in chickens is often suboptimal, and the full potential of outdoor access for chicken welfare may not be achieved. Many studies use visual observations of free-range use, imposing several limitations. An automated system capable of continuously monitoring the location of multiple individual birds over a long time period has the potential to increase the amount and accuracy of the gathered data. Therefore, the aim of this study was to test a newly developed Ultra-Wideband system for monitoring the position of chickens with free-range access. This system consists of active tags (attached to the chickens) that send signals to anchors positioned at fixed locations in the field; the tags’ position can be calculated using the time of arrival of their signal. The effects of vegetation type, precipitation, tags being mounted on a chicken, tag height, angle and orientation, coverage by A-frames or mobile chicken houses, and proximity of other tags on accuracy of the registered positions (distance between the registered and the true position of the tag) and on registration success (percentage of registrations where a position could be calculated) were assessed. Overall, the median error was 0.29 m, which was below the aim of 0.5 m, and the mean percentage of successful registered positions was 68%. None of the variables had a clear effect on the accuracy of the positions. Errors were generally larger in certain areas of the experimental field, which may be due to the asymmetrical setup of the anchors. The percentage of successful registrations was negatively affected by shelter type, with lower percentages in dense vegetation (short rotation coppice willows; SRCW) than on grassland, possibly due to malfunctioning of two anchors close to the SRCW plots. Rain and placing the tags underneath a wooden A-frame, but not placing them in a mobile house, resulted in a lower percentage of successful registrations. The tag being mounted on a chicken, height and angle of the tag and proximity of other tags had no negative effect on the percentage of successful registrations. Placing more (functioning) anchors may contribute to better accuracy and registration success. Alternatively, the bias resulting from the variables that had a negative effect on registration success could be corrected for when using the system in its current setup. Overall, this system shows great promise for monitoring chickens’ free-range use.