Developing an ActivityAware Service Platform for Assisted Living

1
Developing an Activity-Aware Service Platform for
Assisted Living
Media and Knowledge Engineering Halldór
Fjalldal std 1334557
Daily supervisor dr. ir. M.H. Vastenburg
Faculty of Industrial Design Engineering
Supervisor dr. ir. C.A.P.G. van der
Mast Faculty of EWI, MMI Group
2
Overview

• Introduction
• Problem
• Scenarios
• Background
• Context-Awareness
• Assistive living
• Activity-Aware Home
• Summary
• Questions

3
Introduction

• Challenge
• Increasing number of elderly people in need of
  care
• Growing shortage of professional caregivers
• Possible approach
• Assistive living technology could support elderly
  in living safely and independently at their own
  home
• Could also be used to improve well-being

4
Introduction

• Project at the Faculty of Industrial Design
  Engineering aims at developing activity-aware
  service platform to support elderly at home. I_at_H
  project focus on human-product interaction.
  However, working prototypes needed to test
  concepts with users.
• Example services
• Smart-routine based calendar that will be linked
  to activities of the elderly and automatically
  updated using sensors in the home (remote
  monitoring)
• Activity-aware service that stimulates elderly
  people to improve their behaviour
• Medicine reminders linked to activities

5
Problem

• Changes in living patterns, living environment or
  user needs of the elderly requires platform to be
  easily modified to meet new requirements
• Assistive living services are usually dependent
  on hardware (sensors, products) and software from
  specific vendors -gt interoperability problem
• Difficult to invoke the elderly in design of
  service because it is time consuming to install
  and configure if possible at all
• Designers (who develop services) generally are
  good at creative design, but not so good at
  hardcore technology

6
Problem

• Research questions
• How can assistive living services be linked to
  activities of daily living(ADLs)?
• How can service designers be supported in
  prototyping and configuring ADL-aware services?

7
Scenarios Actors

• The service platform involves three actors

• External actors
• e.g. doctor, family

• Elderly person

• Designer

8
Scenario I Elderly home in the morning

• Service 1 Home automation
• Situation The elderly person is waking up
• Action Turn on the lights

Service 2 Medicine reminder Situation The
elderly person is having breakfast Action Remind
him to take his medicine after eating breakfast
Service 3 Motivation service Situation The
elderly is leaving the house Action Virtual
trainer displays how much he should walk
according to his doctor recommendations
9
Scenario II Monitoring elderly
The doctor monitors the activities of the elderly
person, e.g. if he has been taking his medicine
accordingly and doing his exercises
Family members are able to monitor activities of
the elderly e.g. via online calendar service to
see if they are doing fine
10
Scenario III Designing service
The designer can easily link the new services to
the domestic routines of the elderly
Consequently service prototype can be tested with
users more often
11
Scenario Overview
For these scenarios to take place multiple
computing devices, applications and displays, and
mechanism for collecting data and reasoning are
needed. All these different parts need to be
connected together in a framework that allows
designers to define their services on the level
of activities rather than low-level sensors,
thereby making it easier to develop services.
12
Background Context-Awareness

• Examples of existing Context-Aware frameworks
• Context-Toolkit
• Widgets hide low-level details, aggregators for
  processing context
• CoBrA (Context Broker Architecture)
• Agent-based, OWL ontologies for logic inference
  and knowledge sharing
• jCaf (The Java Context Awareness Framework)
• Java-based set of APIs for creating context-aware
  application

13
Background Context-Awareness

• Based on common architecture principles of
  existing context-aware frameworks

Context-aware framework ED Handles the process
of gathering sensor data from different
hardware/software vendors C Analysis the data
and abstracts to a higher-level B Allows
knowledge and resource sharing, provides the
high-level information to the relevant
services A Context-aware services and agents
14
Background Context-Awareness

• Existing Context-Aware framework

• Lacks the ability to easily add/remove
  context- aware services

• Could be used as the basis of an Activity-Aware
  framework to support elderly at home
• Could be extended to support rapid prototyping
  of Activity-Aware services

15
Background Assistive living

• Assistive living is for people who need help with
  everyday tasks. They may need help with
  dressing, bathing, eating, or using the
  bathroom, but they don't need full-time nursing
  care
• Research on assistive living technology is now
  taking place in lab environment
• e.g. The Gator Tech Smart House at Florida
  University, the Philips Homelab and the Aware
  Home at Georgia Tech

16
Background Assistive living

• The Aware Home aims at helping older people to
  live independently in their own homes
• The Context Toolkit middleware is being used to
  support rapid prototyping of assisting living
  services

17
Background Super Assist Project

• The Super Assist project within the MMI group
• Adaptive computer assistant for the supervision
  of patients self-care
• Tested in the Aware-Home

18
Background SA vs I_at_H

• Similarities
• Distributed actors
• Framework should be extendable
• Agents assisting the user (services)
• External monitoring and manipulation of data
• Differences
• SA focus on medical domain (doctor/patient
  issues)
• I_at_H focus on home domain (user/design issues) and
  on linking services to activities of elderly
  people

19
Background ADLs

• Activities of daily living is used by caregivers
  to measure the physical status of the elderly
• Existing research shows the importance of ADLs
  when recognizing activities. However research
  mainly focused on technology
• ADLs appear to be a suitable abstraction of
  living patterns of elderly. I_at_H aims therefore
  at linking services to living patterns based on
  a model of ADLs

20
Background ADLs
21
Activity-Aware Home

• Aim

Services can easily be linked to the activities
of the elderly
Enable fast design iterations in order to
increase user involvement in the design process
22
Summary

• How can assistive living services be linked to
  activities of daily living(ADLs)?
• User-oriented models could be applied to known
  ADLs to identify possible sequences of actions
• Actions could be recognized using an existing
  Context- Aware framework
• Existing Context-Aware framework could be
  extended to allow services to be easily linked
  to actions and act in an assistive way according
  to the behaviour of the elderly

23
Summary

• How can service designers be supported in
  prototyping and configuring ADL-aware services?
• Modifying services at a level of ADLs rather than
  low-level technology
• Easy prototyping
• Modular approach with ready-to-use building blocks

24
Thank you for your Attention!Questions?