Inertial sensor dataset for Dog Posture Recognition

This dataset contains inertial data collected from 42 healthy apprentice dogs (1 Golden Retriever, 16 Labrador Retrievers and 25 Crosses) participating in the assistance dog programme. During the video recorded data collection session, five postures (Standing, Sitting, Lying down, Walking, and Body shake) were annotated and further classified into two types (Dynamic, Static). The subjects wore inertial measurement units (IMUs) in three locations (back, neck, and chest) simultaneously. Inertial data were gathered using ActiGraph GT9X Links (Actigraph LLC, Pensacola, USA) containing three-axial accelerometer, gyroscope and magnetometer with a 100Hz sampling rate. A detailed description of the data provided in each of the .csv files is presented in the readme.md file. The authors of the dataset request researchers to cite the publication when using the data and publishing results produced using it. Ethical approval has been granted by Animal Ethics Experimentation Committee (AEEC) and Social Research Ethics Committee (SREC) at University College Cork (UCC). This publication has emanated from research supported in part by a grant from The Ireland-Wales INTERREG Programme under the CALIN project [PG; grant number 80885; https://irelandwales.eu/projects/calin], from Science Foundation Ireland (SFI) [PG, BO; grant number 12/RC/2289-P2; https://www.sfi.ie/]; from SFI and Department of Agriculture, Food and Marine [PG, BO; grant number 16/RC/3835; https://www.sfi.ie/, https://www.gov.ie/en/organisation/department-of-agriculture-food-and-the-marine/]. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.