Adaptive systems research group

The PRIO project

Aim:
To generate an intelligent, interactive partner agent (PRIO) for the elderly and the visually impaired.

Brief description
In this project, we are developing a software agent capable of interacting with its user in an intelligent way, and thereby help the user with various daily tasks, such as (for example) accessing information, contacting relatives or caretakers etc.

Background and motivation
The fraction of elderly people in the developed world is increasing rapidly. For example, the fraction of people above the age of 80 will rise from around 5% of the population to around 12% of the population in 2050. This increase is already putting strong pressure on health and elderly care systems, particularly since the working-age population is in (relative) decline. Thus, strong efforts are being made to develop a variety of technological tools, which will be able to provide assistance in simple tasks that do not require human intervention, thus giving caretakers more time to focus on their primary tasks, namely to provide support and comfort for elderly people.

Detailed description
Partner agents constitute a special case of the more general concept of software agents, i.e. intelligent, autonomous computer programs. A partner agent's function is to interact with its user in various (mostly non-critical) tasks. In this particular project, the agent is intended both to help its primary user (e.g. an elderly person) and to function as an easily accessible and reliable communication hub for communication between the primary user, the user's relatives, and caretakers (such as home service personnel or the staff at a nursing home).

In the project, we are using various methods from the wide fields of machine learning and artificial intelligence in order to build an intelligent partner agent, focusing on ease of use (for the primary user, which typically is an elderly person without much experience of computers or mobile phones).

More information will be added to this page in the Autumn of 2015.

Participants

Mattias Wahde


Last update: 20150804, 08.00