The gestation period of a horse ranges between 320 to 360 days, making it challenging to predict the start of foaling. Studies show that the majority of mares foal during the night, with 10% of births requiring human intervention due to dystocia. Early detection and intervention are critical to the survival of both the foal and the mare. While human observation is ideal, it is costly and requires a lot of resources. Foaling alarms, using technology such as cameras and sensors, have become useful in providing alerts when a mare shows signs of foaling, thereby reducing the need for constant human monitoring. This thesis aims to develop an effective foaling alarm using a tail-attached tag equipped with an accelerometer to monitor tail movement, providing a practical and reliable solution for horse breeders. To identify changes in horse behaviors around the time of foaling, data were collected from multiple foalings. Due to the seasonal nature of foaling in Sweden, the majority of data was collected in late spring, when most mares are near parturition. Data from 6 foalings were processed and analyzed. Foaling behaviors such as raised tail, lying down and increased activity were determined and features were extracted from the accelerometer data. The outcome was an algorithm that could detect foaling on average 25 minutes before foal discharge, with an average of 1 false alarm. This is acceptable compared to other products on the market where a similar solution could alarm on average 10-20 minutes before foal discharge with a false alarm rate of 10%.
| Identifer | oai:union.ndltd.org:UPSALLA1/oai:DiVA.org:liu-204593 |
| Date | January 2024 |
| Creators | Hult, Amanda, Sjödin, Rebecca |
| Publisher | Linköpings universitet, Reglerteknik |
| Source Sets | DiVA Archive at Upsalla University |
| Language | English |
| Detected Language | English |
| Type | Student thesis, info:eu-repo/semantics/bachelorThesis, text |
| Format | application/pdf |
| Rights | info:eu-repo/semantics/openAccess |
Page generated in 0.0024 seconds