This Healthy Pets New Zealand funded study aimed at improving the understanding of domestic dog behaviour.
Domestic dogs (Canis familiaris) have become an integral part of human society, serving various purposes such as companionship, working, hunting, research, and military service roles. Assessing their behaviour is critical for predicting suitability and targeting specific temperament characteristics required for these specific roles. Importantly, behavioural assessments not only facilitate selection of desirable traits, but also enhance our understanding of the overall welfare state of dogs. Common methods of behavioural assessment include test batteries, observational methods, and questionnaires . These methods, however, can be time consuming, labour intensive, and subject to bias, making large-scale implementation challenging.
There is a need for methods of behavioural assessment that are objective, practical and do not require extensive time or training from human observers. Recent technological advances in accelerometer technology have facilitated remote and automated behavioural measurement.
In this study, researchers used triaxial accelerometers (small, wearable sensors) and machine learning algorithms to track and identify specific behaviours in dogs. The goal was to develop a more efficient and accurate method for monitoring dogs' activities, such as walking, resting, or playing, using technology that can be easily applied in everyday settings. These accelerometers are capable of measuring movement across three different directions, making it possible to capture detailed data on a dog’s movements throughout the day.
The study involved attaching accelerometers to the dogs and using machine learning algorithms to analyse the data. The researchers trained the system by feeding it information from various behaviours that the dogs exhibited during the study. They then validated the system by testing it on different dogs and ensuring that the machine learning model could accurately identify the different types of behaviour based on the sensor data. The results were promising, showing that the system could reliably classify the dogs' activities with a high degree of accuracy.
This technology has the potential to greatly enhance the way we monitor and understand dog behaviour in various settings, including at home or in veterinary clinics. It could also be useful in research, helping scientists track behavioural patterns in dogs for studies on animal welfare, health, and even the impact of environmental changes on their behaviour. By using wearable technology and machine learning, this method provides a non-invasive, efficient, and scalable solution to behaviour tracking in domestic dogs, offering new insights into their daily lives.
You can find the full paper here