Smart Home TFG and TFM Dissertations
Smart Home's Control by Gesture
Users in the Smart Home can control lights and roller shutters by gesture. This is carried out thanks to Tactigon device, which gathers the acceleration of users' hands and send this information to a Raspberry Pi device that decides which gesture is caried out and acts accordingly.
Developed by: Antonio Ayala Tirado
Smart Fridge which controls the quantity of products which are stored inside it. This control is carried out by placing RFID tags and pressure sensors.
Developed by: Rafael Alejandro García Rodríguez
A medical case is incorporated in the Smart Home. It includes several devices connected by Bluetooth which measure the heart rate, glucose and blood pressure among other things. A web tool is designed to show all the medical data.
Developed by: María Teresa Martínez Rodríguez
A bed adapted to the elderly people is developed. It moves the upper and lower part as the user requires. Alexa is used to communicate with the bed. Thus, the bed incorporates four servo motors connected to an Arduino, which executes the orders given by the Alexa trough the Raspberry Pi.
Developed by: Silvia Almodóvar Fernández
Smart Home's Control by Voice
Lights (normal and dimmed), blinds and roller blinds can be controlled with Alexa.
Developed by: Juan José Escarabajal Hinojo
Smart Home's Control by Facial Recognition
Depending on the user which enters, the home is adapted to him/her thanks to person identification by cameras. The home is therefore adapted to the location of the user, regulating the lights and roller shutters accordingly. Also, the favorite TV channel is shown when the user turns on TV.
Developed by: Manuel Marín Peral
OpenHAB has been configured in a Raspberry Pi, which lets the KNX devices from the Smart Home to be controlled independently from the central server. Alexa and Telegram messages have been integrated.
Developed by: Toni Schaarschmidt
Activity Recognition System
A real-time activity recognition system is incorporated in the smart home. This system make use of binary sensors, location tags and smartwatches to monitorize users activities. A total of 14 daily activities are recognized by using a machine learning algorithm.
Developed by: Marcos Lupión Lorente