Technology for the Aging
A Collaborative Effort
UMassAmherst
Edward Riseman
Allen Hanson
Roderic Grupen
Erik Learned-Miller
Phebe Sessions
Julie Abramson
Mary Olson
MERL
Candy Sidner
Supported by the National Science Foundation
Collaborative Effort


Collaborative Research Effort

Social scientists and computer scientists

Elderly representatives and their stakeholders

caregivers

family

local and state agencies, etc.
Two Part Talk: Social Science and Technology
Enhancing Control and
Empowerment for the Elderly
through Assistive Technology
Dr. Phebe Sessions
Smith College School for Social Work
Social Science Components
November 4, 2005
Washington, D.C.
Outline

Issues in collaboration

Theoretical framework: ecosystemic and
social constructionist

Gerontological literature and traditions

Research plan
Issues in Collaboration
UMass Computer Sciences/Smith SSW
Why social work?
Pragmatism
 Commitment to underserved populations
 Community-building skills

Challenges of collaboration
 Novelty and not knowing
 Different languages
Theoretical framework
Ecosystemic and Social Constructionist
 Strengths perspective
 Collaboration, dialogue
 Qualitative methods in research

Gerontological literature and traditions

Aging and the deficit perspective: “declinism”

Positive aging and core significance of control

Learning issues and aging

Social networks and aging

Ethical considerations
Research plan




Qualitative methods
Year One: Focus groups with elderly,
families, and service providers
Year Two: Implementation of technology
in common room
Year Three: Revision of technology based
on Year Two findings
Technology Components

Umass Vision and Robotics groups




Professor Emeritus Edward Riseman (PI)
Professor Allen Hanson (co-PI)
Professor Roderic Grupen (co-PI)
Mitsubishi Electronic Research Laboratory

Dr. Candace Sidner
Technology Overview
ISSUES:
privacy
access
client control
Human Services
Interface
SMITH
Assisted-Living Clientele
Medical Structure
Interface
Speech
Synthesis
Tracking
Graphics
Scripts
Speech
Understanding
Vision
Speech
Conversational
Interaction
Activity
Modeling
Lifestyle/Event
Coordination
MERL
UMASS
Family
Family/Community
Interface
Interaction:
Appointment Management
IA: Hi, Maisie.
Senior: Hi.
IA: later this morning at 11 am you have a dentist appt.
Senior: Ok
IA: You will have to take the bus so you need to leave here an
hour before and walk to the bus stop. That means leaving about
10 am.
Senior: I want to leave later.
IA: Well, the bus takes 20 minutes. How long do you need to
get to the bus stop?
Senior: 15 minutes
IA: The bus comes every 15 minutes, so when do you think you
should leave?
Senior: 10:15
IA: Okay. How about I remind you at 10 am to get ready?
Senior: Good.
IA: Good bye for now.
Technical Goals

An effective API to the client’s environment:








Independent Living
Social/family
Medical
Financial
Social services
Legal
…..
Plug and play for rapid development and deployment



Interface driven user interactions
System interfaces
Nimble, reconfigurable, robust, minimally intrusive,
controllable
The Client-Care Sandbox
Nimble & Reconfigurable
Component Technologies

Fault tolerant script based interfaces: Collagen plus GUI



Speech and speech understanding



Minimal intrusion
Robust
Command and control
Voice Analysis - e.g. stress
Visual interfaces



Virtual Reality
Tracking
Change Detection
 Environment



Object locations
Motion/expected tracks, deviation from normal activity
People


Gait and postural models
Speech patterns
Collagen: Collaborative Agents

Principals: Candace L. Sidner and Charles Rich
Mitsubishi Electric Research Laboratory



Java middleware for collaborative interface agents
User and agent collaborate to complete predefined tasks
Speech or text-based discourse with agent



mouse, touch screen, others
Provides friendlier, more familiar interface for
inexperienced users
Many existing applications

http://www.merl.com/projects/collagen
Collaborative Interface Agent
*
focus stack
plan tree
Collagen
communicate
observe
interact
observe
interact
* SharedPlans per Grosz, Sidner, Lochbaum, Kraus, et al.
Collagen Example:
Calendar Application Dialog Fragment
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
User: “Let’s create a new appointment.”
Agent: activates Create Appointment window.
User: enters the appointment’s date and description.
User: “What’s Next?”
Agent: “What time is the appointment?”
User: “11 AM.”
Agent: sets the time of the appointment to 11 AM.
Agent: “I’m going to save this appointment.”
User: “OK.”
Agent: “We have created a new appointment!”
Speech Recognition

Available engines




IBM Viavoice
Microsoft Speech SDK
CMU Sphinx Engine 2 and 4
Mode of application

Continuous speech recognition mode



Speaker dependent
Low accuracy
Voice commands mode



Speaker independent
Higher Accuracy
Still susceptible to noise
Managing Speech Recognition
Technology

Reduce noise



Continuous recognition



Training is essential
Do not expect high accuracy
Approach




Pressure-gradient/Noise-canceling Microphone
Sound card/USB pod
Design dialog script carefully
Build in a fault tolerant mechanism
Always have a “take me to an operator” option
Constrained uses:


voice command mode
voice stress analysis
Local Experimental Environment

“Smart” Room



Distributed sensor network
Human and robotic residents
Complex environment
: camera locations
Tracking


Distributed sensor network
Sensor management


Selection, reasoning across sensors
Real-Time fault-tolerant approach

Fault containment units



Wrappers around a resource policy



Hardware, Software
Hierarchical
Augmented by fault and context
reporting mechanisms
Dynamic restructuring
Supports



Activity modeling
Identification
Unusual event detection
QuickTime™ and a
TIFF (LZW ) dec ompressor
are needed to s ee this picture.
Tracking Supports Virtual Worlds
Immersive Virtual Worlds


Provide several levels of interaction styles
Client control over interaction style and level of
access of remote visitors (privacy)




Avatar plus virtual world
Avatar plus real world
Person plus virtual world
Person plus real world
(Registered)
Transparent
Objects
Rapid Viewpoint Changes


Camera context switches e.g. surveillance
Disconcerting, difficult for user
QuickTime™ and a
H.264 decompressor
are needed to see this picture.
“Snap” Transitions
QuickTime™ and a
H.264 decompressor
are needed to see this picture.
“VR Smoothed” Transitions
Mixing Virtual and Real Worlds




Events in real world mapped to virtual world
Temporal events synchronized to virtual world
Transparency - look through objects
Form of access control
QuickTime™ and a
H.264 decompressor
are needed to see this picture.
Conclusions

Starting up



Empowering the elderly: they decide
Social etiquette enforced through virtual worlds


Client controls level of access and degree of privacy
Built on



Conjectures on the future
Modern social science theory
Existing technology base
Client-care Sandbox model for rapid prototyping

Particulars determined through focus groups
Descargar

Enhancing Control and Empowerment for the Elderly …