Motivation

Thanks to progresses in medicine and improved quality of life, European population is becoming increasingly older, with a life expectancy approaching 80 years. This demographic trend is having enormous economic and social implications in a number of areas (e.g. welfare, healthcare systems, social assistance), and there is a market need for products and services able to enhance the quality of life of elderly, and to mitigate the social impact of ageing population. Technological and socio-economic innovation plays a major role in this context. As far as ICT is concerned, Ambient Intelligence (AmI) is one of the most promising technologies. AmI is commonly referred to as the capability of acquiring a complete knowledge about a monitored environment, and reacting to events through automatic functions and advanced user interaction. AmI has been recognized by the IST Advisory Group of the EU Commission as a key to foster European competition, and is being considered one of the crucial technologies to develop Ambient Assisted Living (AAL) applications, helping impaired and elderly to achieve a higher quality of life, while preserving independency, privacy, and security.

Research areas

Acube will act in this framework, by conceiving a highly-developed smart environment to be deployed in nursing homes as a support to medical and assistance staff. The sophisticated features of the envisaged system will have a major impact on quality of care, quality of life of the assisted, as well as working conditions of caregivers. Besides the evident societal value, the selected scenario presents substantial scientific and technical challenges. Furthermore, it is a highly demanding testbed to develop robust and efficient solutions to be readily exported to different application domains, such as the intelligent monitoring and surveillance of public places (museums, schools, stations).

Application areas 

The research in ACube will be organized into three complementary streams: sensing, processing, and intelligence. The sensing architecture will be based on the paradigm of Distributed Sensor Networks (DSN). DSNs are auto-configuring networks of wireless nodes with sensing, processing, and communication capabilities, where the nodes cooperate to monitor a given environment. Solutions will be studied to achieve a highly reliable DSN at very low cost and intrusiveness. Processing concerns the development of advanced algorithms to recognize events, situations, activities, behaviors in complex multi-person scenarios. The objective is to enable the smart environment to understand who is doing what, where, when and how. This knowledge allows the system intelligence to make decisions (e.g. rising alarms). Processing includes adaptation capabilities, to fit different environments and users. The project aims at overcoming the severe limitations of current methodologies, specially related to multiple users, complex environments/scenarios, extended/heterogeneous sensorial systems. System intelligence will concern the symbolic reasoning required to maintain a consistent representation of the observed environment, monitor changes, discriminate among ambiguous scenarios, operate decisions. Furthermore, it will provide an autonomic optimization of the system (self-configuration, adaptation), achieved through sub-symbolic techniques.

Outcomes 

The major technical outcome of the project is an integrated, layered architecture for intelligent monitoring, able to efficiently accommodate and manage a wide range of sensors, and to combine the acquired data into high-level perception, in a highly configurable and autonomous manner. To demonstrate the real-world viability of the proposed technologies, ACube will deliver two pilot sites, located in the premises of public institutions for assisted living (a daycare center for Alzheimer disease, located in Trento, selected with the involvement of the Social Services units of the PAT, and a rehabilitation center owned by FDG, located in Milan).

Members

Faculty Members

Roberto Battiti	Francesco De Natale	Andrea Massa
Luigi Palopoli	Dario Petri	Gian Pietro Picco

Partners

Create-net
DISI - Università degli Studi di Trento
FBK-IRST
Fondazione Don Gnocchi ONLUS