Loading...
PigNet: Failure-tolerant pig activity monitoring system using structural vibration
Author(s)
Bonde, Amelie
Codling, Jesse R.
Naruethep, Kanittha
Dong, Yiwen
Siripaktanakon, Wachirawich
Ariyadech, Sripong
Sangpetch, Orathai
Pan, Shijia
Noh, Hae Young
Zhang, Pei
Date Issued
May 18, 2021
Type
Conference Paper
Abstract
Automated monitoring of livestock behavior can help farmers economically by detecting changes in animal welfare. Prior approaches use video, which requires light and high storage capability, or motion detection, which has difficulty separating subtle activities. Wearable sensors can address these issues but are vulnerable to destruction by the animals. To the best of our knowledge, we present the first system that uses structural vibration to track animal behavior, and the first system to automatically detect piglet nursing. PigNet uses vibration sensors attached to a pig pen to sense the unique vibration patterns and changes in structural response caused by the animals' movement and position within the pen. Combined with our knowledge of pig behavior, we use this physical knowledge of vibration characteristics to detect pig activities and track piglet growth in a real farm environment. Our system is designed to be robust to the harsh environment, which can create unpredictable noise, as well as physically damage or disconnect sensor nodes. When deployed in a real-world farm environment, our system was able to achieve a daily pen-level status profile of up to 90% accuracy, which tracks nursing activity, sow lying activity, and changes in piglet growth over the weeks-long pre-weaning period.
Citation
Proceedings of the 20th International Conference on Information Processing in Sensor Networks IPSN 2021 Co Located with Cps Iot Week 2021, 328-340, 2021