TFM: Development of a wearable device for monitoring therapy animals

Animals have long been part of the human experience, serving multiple purposes throughout history, from food to companionship. In recent years, the therapeutic potential that offers the use of animals to help people overcome illness and/or mental disorders has been increasingly recognized, leading to more healthcare facilities providing Animal-Assisted-Interventions (AAIs) to their patients.

The steadily increasing popularity of AAIs programs is supported by the fact that they deliver health benefits to the patients. A growing literature gathers testimonials of veterinarians, psychologists and other pet-therapy enthusiasts about the effectiveness of AAIs programs for humans. In contrast, very few researchers have focused on the possible ill effects that AAIs programs have on the animals themselves.

Nowadays, the present lines of research that are trying to determine both positive and negative effects on the physical and mental well-being of the animals involved in AAIs are divided in two groups:

  • Non-invasive methodologies based on the interpretation of the body language of the animals. For instance, a dog’s wagging tail may mean different things depending on the speed of the wag, and whether the full tail or just the tip is wagging. Besides, dogs also use a range of what the renowned dog trainer Turid Rugaas refers to as
    “calming signals” that they use to defuse stressful situations. For example, a dog may lick her nose, sniff the ground, yawn, turn away, or stare in response to a stressful situation. The main drawback of these methodologies is the subjectivity of the observer.
  • Invasive methodologies based on medical procedures such as blood extractions, faces analysis or saliva analysis in order to measure certain hormones levels that could have correlation with the stress that could be suffering the animals during the AAIs. Despite of the fact of the objectivity of the results, due to the nature of these procedures, these interventions by themselves could provoke stress in the animals.

Thus, the aim of this Master’s Thesis is to design and develop an electronic wearable device to collect physiological and behavioral variables in dogs participating in the AAIs in order to extract stress patterns in different scenarios and therefore determine objectively the effects of the AAIs in the animal welfare. The data gathered will be analyzed by ethologists than can
evaluate what is happening in the process of interaction of the therapy dog with the rest of the actors. This way, conclusions related to the dog state in the different stages of therapy could be obtained, allowing the modification of the routines to increase the dog’s quality of life.

It is worth mentioning that this project is being carried out in collaboration with the Escuela Técnica de Ingenieros de Telecomunicación and the animals and society chair at the Universidad Rey Juan Carlos, which will be in charge of the visualization and interpretation, respectively, of the data acquired by the system to be developed in this Master’s Thesis.

To achieve this goal, this Master’s Thesis has focused on the development of the electronic wearable device that will monitor the therapy dog. This development has covered both the design and hardware implementation of the three printed circuit boards that make up the device, as well as the software implementation of the drivers needed to control each sensor individually in addition to the application architecture at the user level.

Both software implementations are based on two existing design patterns that provide modularity to the system in order to incorporate new sensors to the device. Finally, in order to validate the design and implementation
phases at hardware and software level, functional tests of the system have been carried out which have allowed conclusions to be drawn on the development of this project as well as to propose future lines to improve its current state.



No cabe duda de que estamos siendo testigos del gran impacto que está teniendo el crecimiento del Internet de las Cosas (Internet of Things, IoT) en la actualidad. Cada vez son más el número de dispositivos wearables, de electrodomésticos inteligentes, coches sensorizados… presentes en nuestra vidas.

Se estima que para 2020 más de 250.000 vehículos y más de 245 millones de dispositivos wearables estén conectados a la red.
Es lógico pensar que tal conexión masiva de dispositivos produce y producirá más aún un uso ineficiente del espectro radioeléctrico en las bandas libres disponibles provocando un ineficiente uso de las prestaciones de la red . Por lo que un elemento clave para evitar tal fenómeno son las Redes inalámbricas de sensores cognitivas (Cognitive Wireless Sensor Network, CWSN) compuesta de nodos cognitivos capaces de modificar sus parámetros de comunicación dinámicamente para evitar interferencias dentro de la red, mejorando así las prestaciones de la misma.

Una de las principales líneas de investigación dentro del B105 Electronic Systems Lab, es el desarrollo de CWSN. Debido a ello en los últimos años se han llevado cabo en diferentes proyectos tanto el desarrollo hardware de un nodo cognitivo (cognitive Next Generation Device, cNGD) como la implementación software de la pila de protocolos que permite al nodo comunicarse en tres bandas diferentes y libres del espectro radioeléctrico, que cumplen con la legislación vigente en Europa.

Este Trabajo de Fin de Grado se centra en realizar el diseño y despliegue de un banco de pruebas (test-bed) de una CWSN. Esto implica realizar un estudio previo de los proyectos mencionados anteriormente para poder realizar un correcto montaje hardware de los nodos cognitivos que integrarán la CWSN así como el diseño, implementación y realización de una serie de pruebas para verificar el correcto montaje de los mismos. Para poder desplegar el test-bed ha sido necesario hacer un previo diseño de la CWSN teniendo en cuenta las limitaciones que presenta el protocolo de comunicaciones, las distintas funcionalidades que puede tener un cNGD y las pruebas a realizar sobre la red. Una vez realizado el despliegue se ha podido caracterizar ciertos parámetros de la red tales como el alcance, la latencia o el consumo de los nodos. Para medir este último parámetro se realizará el diseño e implementación de un módulo de expansión de la placa cNGD para medir el consumo del nodo en sus distintas funcionalidades.