Open a door to assist in the development of smart homes
Everyone walking around campus has the opportunity to participate in a project that combines architecture, civil engineering, automation, information technology and signal processing and offers new insights into the simple act of opening the door to our home. The data is used to develop smart homes and later perhaps even smart cities.
All the members of the campus community are invited to lend a helping hand to researchers collecting data in the Tietotalo building. The project involves researchers Pasi Pertilä (left), Harry Edelman, David Hästbacka and Mikko Parviainen.
For smart homes to become more widespread, they need to be easy to use, safe and genuinely useful. All the members of the campus community are encouraged to pass through a hallway that has been set up in the Tietotalo building and will remain in place until September. Everyone who walks through the ‘3A Self-learning Buildings’ hallway for test purposes are recorded on video. The data is used to teach human behaviour to future smart homes.
“We’re aiming to collect extensive visual information of people entering and exiting through the door to support the development of autonomous control solutions that employ machine learning. We hope to collect as much data as possible to improve the reliability of machine learning. To ensure the richness of the data, we also need to observe variations in people’s behaviour as they walk through the door. While our main focus is the typical sequence of events associated with opening the door to our home, participants are also very welcome to add a personal touch to the process,” says Harry Edelman from the Laboratory of Civil Engineering.
The traffic through the door has been heavy.
“We’ve had to extend the delay between the recorded events to ensure that the different visits are not combined into one,” says Edelman.
Self-learning smart homes anticipate our needs
Future smart homes offer an overall connected home experience. Research into self-learning buildings is conducted at TUT by Harry Edelman, David Hästbacka, Mikko Parviainen and Pasi Pertilä, who envision our future homes to be automatically energy saving, comfortable and safe.
“Smart homes need to be able to complete tasks independently. They must anticipate and fulfil our needs automatically and, for example, adjust heating as needed without us having to control the process via a smartphone,” says Edelman.
The data collected on campus can also be used to further develop the autonomous control of buildings.
Self-learning smart homes learn to identify the family members and even pets and are therefore better able to understand their needs.
“First we’ll focus on cutting energy consumption by developing heating and cooling solutions that adjust to changing circumstances and our daily routine. Qualitatively and quantitatively rich data has broad application areas. With the help of machine learning, it can be used to anticipate unexpected events, such as the use of physical space.”
“Self-learning smart homes learn to identify the family members and even pets and are therefore better able to understand their needs. We also have, for example, home security applications in the pipeline,” says Harry Edelman.
AI harnessed for urban development
The researchers at TUT have set their sights both on and beyond self-learning smart homes.
“In addition to homes, we’re keen to explore self-learning buildings, cities and urban functions, such as traffic, urban green spaces and commercial solutions,” says Edelman.
We need to tap into the potential of artificial intelligence (AI) to meet the challenges of rapid urbanization.
“Cities are facing great changes and opportunities. AI opens up new opportunities for the development of not only physical structures but also urban cultures and business models. For example, AI enables the development of effective distribution solutions and the optimization of the use of resources, such as money, facilities, vehicles and energy. It also facilitates the collection and combination of data that is necessary for the production of different services in our new urban economy. Autonomous AI-based solutions could also increase our understanding of sociological questions and help develop better urban environments for us all."