Improving the Welfare of Kennelled Dogs with Smart Kennels

This pilot project was conducted in 2014 as a collaboration between the ACI Lab and UK Charity Dogs Trust, to address welfare challenges faced by dogs living in kennels. The project was led by Dr Clara Mancini.


Around the world billions of managed animals live in indoor farms, research laboratories, zoos or other confining environments, which impacts on their welfare very significantly. This project therefore explored the challenges of and opportunities for employing ubicomp technologies, including sensing and monitoring devices as well as tangible and embodied interfaces, to improve the welfare of confined animals, focussing on the case of dogs living in kennelled accommodation.

Dogs are often confined to kennel environments (e.g. assistance dogs are kenneled during long training periods prior to being partnered with their assisted humans; military dogs spend a considerable proportion of their lives in kennels; homeless dogs spend variable amounts of times in kennels waiting to be homed). The necessary constraints of even the best kennel environments make meeting the welfare requirements of kenneled dogs challenging, as kennelled life affords dogs limited stimulation, as well as limited control over and predictability of salient events (e.g. meals, walks, encounters with other dogs). Furthermore, assessing the welfare of kennelled dogs is non-trivial, since the more viable measures are not necessarily the more indicative (e.g. observed behaviours may be difficult to interpret, particularly when monitoring cannot be continuous). Thus, welfare issues, whether pre-existing or induced by kennel confinement, may remain undetected and unaddressed until they become overt and possibly more difficult to resolve.

The project investigated whether advances in ubiquitous computing to enhance human wellbeing and daily experience might offer the opportunity to improve how the welfare of kenneled dogs is managed: whether ubiquitous sensor systems and ambient intelligence developed to monitor health in humans (e.g. activity levels, sleeping patterns) might be useful to record, measure, visualize and interpret non-obvious welfare-relevant phenomena, especially when it is non-viable or non-desirable for individual dogs to receive continuous attention; and whether the use of embodied and tangible interaction technologies developed to enhance human performance and experience (e.g. touch or gestural interfaces) might afford the dogs a more stimulating experience and greater control over their surroundings through forms of interaction that are accessible to them.

As a first step towards such a vision, we conducted an ethnographic study at one of the rehoming centers of Dogs Trust, the UK’s leading canine welfare charity, to identify core requirements from both canine residents and human carers who live and work in the rehoming center, in order to understand how smart technology could support these stakeholders, identify challenges and possible solutions at the level of both design and methodology. Based on our findings, we proposed a ‘three-dimensional’ welfare-centered framework for designing smart environments, integrating monitoring and interaction with information management, as well as a ubicomp-supported ethnographic approach for similar projects.

Further readings

Mancini, C., van der Linden, J., Kortuem, G., Dewsbury, G., Mills, D., Boyden, P (2014). UbiComp for Animal Welfare: Envisioning Smart Environments for Kennelled DogsProc. ACM International Conference on Pervasive and Ubiquitous Computing, UbiComp’14, ACM Press, pp. 117-128.

Mancini, C., Dewsbury, G., van der Linden, J., Kortuem, G., Mills, D., Smith, N. (2014). Towards Smart Kennels for Supporting Canine Welfare: from Design to Methodological Requirements. First Symposium on Intelligent Systems for Animal Welfare, ISAWEL’14, Proc. 50th convention on Artificial Intelligence and Simulation of Behaviour, AISB’14.