Rapid specification and automated generation of prompting systems to assist people with dementia

Jesse Hoey, Thomas Pltz, Dan Jackson, Andrew Monk, Cuong Pham, Patrick Olivier

Research output: Contribution to journalArticleResearchpeer-review

Abstract

Activity recognition in intelligent environments could play a key role for supporting people in their activities of daily life. Partially observable Markov decision process (POMDP) models have been used successfully, for example, to assist people with dementia when carrying out small multistep tasks such as hand washing. POMDP models are a powerful, yet flexible framework for modeling assistance that can deal with uncertainty and utility in a theoretically well-justified manner. Unfortunately, POMDPs usually require a very labor-intensive, manual set-up procedure. This paper describes a knowledge-driven method for automatically generating POMDP activity recognition and context-sensitive prompting systems for complex tasks. It starts with a psychologically justified description of the task and the particular environment in which it is to be carried out that can be generated from empirical data. This is then combined with a specification of the available sensors and effectors to build a working prompting system. The method is illustrated by building a system that prompts through the task of making a cup of tea in a real-world kitchen. The case is made that, with further development and tool support, the method could feasibly be used in a clinical or industrial setting.

Original languageEnglish
Pages (from-to)299-318
Number of pages20
JournalPervasive and Mobile Computing
Volume7
Issue number3
DOIs
Publication statusPublished - 1 Jan 2011

Keywords

  • Activity recognition
  • Behavior analysis
  • Dementia
  • POMDP
  • Prompting systems

Cite this

Hoey, Jesse ; Pltz, Thomas ; Jackson, Dan ; Monk, Andrew ; Pham, Cuong ; Olivier, Patrick. / Rapid specification and automated generation of prompting systems to assist people with dementia. In: Pervasive and Mobile Computing. 2011 ; Vol. 7, No. 3. pp. 299-318.
@article{c405858494574ffca89ef55ad8e87afc,
title = "Rapid specification and automated generation of prompting systems to assist people with dementia",
abstract = "Activity recognition in intelligent environments could play a key role for supporting people in their activities of daily life. Partially observable Markov decision process (POMDP) models have been used successfully, for example, to assist people with dementia when carrying out small multistep tasks such as hand washing. POMDP models are a powerful, yet flexible framework for modeling assistance that can deal with uncertainty and utility in a theoretically well-justified manner. Unfortunately, POMDPs usually require a very labor-intensive, manual set-up procedure. This paper describes a knowledge-driven method for automatically generating POMDP activity recognition and context-sensitive prompting systems for complex tasks. It starts with a psychologically justified description of the task and the particular environment in which it is to be carried out that can be generated from empirical data. This is then combined with a specification of the available sensors and effectors to build a working prompting system. The method is illustrated by building a system that prompts through the task of making a cup of tea in a real-world kitchen. The case is made that, with further development and tool support, the method could feasibly be used in a clinical or industrial setting.",
keywords = "Activity recognition, Behavior analysis, Dementia, POMDP, Prompting systems",
author = "Jesse Hoey and Thomas Pltz and Dan Jackson and Andrew Monk and Cuong Pham and Patrick Olivier",
year = "2011",
month = "1",
day = "1",
doi = "10.1016/j.pmcj.2010.11.007",
language = "English",
volume = "7",
pages = "299--318",
journal = "Pervasive and Mobile Computing",
issn = "1574-1192",
publisher = "Elsevier",
number = "3",

}

Rapid specification and automated generation of prompting systems to assist people with dementia. / Hoey, Jesse; Pltz, Thomas; Jackson, Dan; Monk, Andrew; Pham, Cuong; Olivier, Patrick.

In: Pervasive and Mobile Computing, Vol. 7, No. 3, 01.01.2011, p. 299-318.

Research output: Contribution to journalArticleResearchpeer-review

TY - JOUR

T1 - Rapid specification and automated generation of prompting systems to assist people with dementia

AU - Hoey, Jesse

AU - Pltz, Thomas

AU - Jackson, Dan

AU - Monk, Andrew

AU - Pham, Cuong

AU - Olivier, Patrick

PY - 2011/1/1

Y1 - 2011/1/1

N2 - Activity recognition in intelligent environments could play a key role for supporting people in their activities of daily life. Partially observable Markov decision process (POMDP) models have been used successfully, for example, to assist people with dementia when carrying out small multistep tasks such as hand washing. POMDP models are a powerful, yet flexible framework for modeling assistance that can deal with uncertainty and utility in a theoretically well-justified manner. Unfortunately, POMDPs usually require a very labor-intensive, manual set-up procedure. This paper describes a knowledge-driven method for automatically generating POMDP activity recognition and context-sensitive prompting systems for complex tasks. It starts with a psychologically justified description of the task and the particular environment in which it is to be carried out that can be generated from empirical data. This is then combined with a specification of the available sensors and effectors to build a working prompting system. The method is illustrated by building a system that prompts through the task of making a cup of tea in a real-world kitchen. The case is made that, with further development and tool support, the method could feasibly be used in a clinical or industrial setting.

AB - Activity recognition in intelligent environments could play a key role for supporting people in their activities of daily life. Partially observable Markov decision process (POMDP) models have been used successfully, for example, to assist people with dementia when carrying out small multistep tasks such as hand washing. POMDP models are a powerful, yet flexible framework for modeling assistance that can deal with uncertainty and utility in a theoretically well-justified manner. Unfortunately, POMDPs usually require a very labor-intensive, manual set-up procedure. This paper describes a knowledge-driven method for automatically generating POMDP activity recognition and context-sensitive prompting systems for complex tasks. It starts with a psychologically justified description of the task and the particular environment in which it is to be carried out that can be generated from empirical data. This is then combined with a specification of the available sensors and effectors to build a working prompting system. The method is illustrated by building a system that prompts through the task of making a cup of tea in a real-world kitchen. The case is made that, with further development and tool support, the method could feasibly be used in a clinical or industrial setting.

KW - Activity recognition

KW - Behavior analysis

KW - Dementia

KW - POMDP

KW - Prompting systems

UR - http://www.scopus.com/inward/record.url?scp=79958236542&partnerID=8YFLogxK

U2 - 10.1016/j.pmcj.2010.11.007

DO - 10.1016/j.pmcj.2010.11.007

M3 - Article

VL - 7

SP - 299

EP - 318

JO - Pervasive and Mobile Computing

JF - Pervasive and Mobile Computing

SN - 1574-1192

IS - 3

ER -