Papers from the AAAI-02 Workshop "Automation as Caregiver"

These papers were published in the AAAI technical report WS-02-02, which can be ordered from AAAI Press, Menlo Park, CA.

Abowd, Bobick, Essa, Mynatt, Rogers The Aware Home: Developing Technologies for Successful Aging

Pages 1-7
We describe our ongoing research in the area of developing and testing technologies for successful aging. Sensing and perception technologies can enable a home environment to be aware of the whereabouts and activities of its occupants. Motivated by the desire to use such an awareness to help maintain independence and quality of life for an aging population, we describe the technological, design and engineering research challenges inherent in this problem domain. Our work is situated in the Georgia Tech Broadband Institute’s Residential Laboratory, a unique living laboratory for this exploration of ubiquitous computing in a domestic setting.
Campo & Chan Detecting Abnormal Behaviour by Real-time Monitoring of Patients

Pages 8-12
This paper deals with a follow-up system that passively observes elderly people moving about in a hospital setting and records automatically various patterns of activity. The aim is to propose a support tool to help the medical staff make decisions. The system is based on a sensor network connected to a processing unit. The software developed automatically follows the patient's behaviour and plots statistical curves related to a number of characteristics such as getting up, going to bed, going to the bathroom, etc, ... Thus real-time monitoring can be displayed on a graphical interface for the medical staff. Experimental platform using this system is presented along with examples of behaviour monitored during the night.
Geib Problems in Intent Recognition for Elder Care

Pages 13-17
Providing contextually appropriate help for elders requires the ability to identify the activities they are engaged in. However, inferring agent intent in these kinds of very general domains requires answers to a number of problems that existing AI research on intent recognition/task tracking has not addressed. This paper identifies some of these limitations, sketches some of the solutions and points at some promising directions for future work.
Green A Virtual World for Coaching Caregivers of Persons with Alzheimer's Disease

Pages 18-23
We present the design of a system for coaching caregivers on techniques for having social conversations with persons with Alzheimer's Disease (PAD). Through an avatar in a virtual world, the caregiver practices interventions by eliciting autobiographical stories from a virtual PAD character. Our system extends previous AI research in interactive drama and conversational agents in two areas: generation of autobiographical stories and modeling the effect on the conversation of successful and unsuccessful interventions by the caregiver.
Guralnik & Haigh Learning Models of Human Behaviour with Sequential Patterns

Pages 24-30
The Independent Lifestyle Assistant (TM) (I.L.S.A.) is an agent-based system to aid elderly people to live longer in their homes. One of the biggest challenges for I.L.S.A. is that every installation will (1) be in a home with a different layout and suite of sensor and actuator capabilities, and (2) supporting a technophobic client with unique capabilities, needs and care-giving support network. I.L.S.A. therefore must be able to adapt to its environment over time. This paper describes our approach to modelling one particular aspect of the I.L.S.A. domain: using sequential pattern learning to learn models of human behaviour. We describe the problem, our enhancements to the basic algorithm, and present experimental results collected from four test sites.
Haigh, Geib, Miller, Phelps, Wagner Agents for Recognizing and Responding to the Behaviour of an Elder

Pages 31-38
We are building an agent-oriented system to aid elderly people to live longer in their homes, increasing the duration of their independence from round-the-clock care while maintaining important social connectedness and reducing caregiver burden. Independent LifeStyle Assistant (TM), also known as I.L.S.A., is a multiagent system that incorporates a unified sensing model, probabilistically derived situation awareness, hierarchical task network response planning, real-time action selection control, complex coordination, and machine learning. This paper describes the I.L.S.A.'s agents and present a scenario of the information flow.
Haigh & Yanco Automation as Caregiver: A Survey Of Issues and Technologies

Pages 39-53
In the United States, the number of people over 65 will double between now and 2030 to 69.4 million. Providing care for this increasing population becomes increasingly difficult as the cognitive and physical health of elders deteriorates. This survey article describes some of the factors that contribute to the institutionalization of elders, and then presents some of the work done towards providing technological support for this vulnerable community.
Intille, Larson, Kukla Just-in-time Context-Sensitive Questioning for Preventative Health Care

Pages 54-59
This paper argues that there is an emerging opportunity for artificial intelligence researchers: the development of technologies that enable a new type of “proactive health care system” that continuously monitors “healthy patients” in their homes and motivates lifelong healthy behavior and health self-awareness. The components of the system, which would augment but not replace traditional health care, could be funded entirely by consumers without reliance on the financially-burdened health care industry. A research prototype of one com-ponent of such a system is described: a preventative monitoring system for congestive heart failure that uses “just-in-time,” context-sensitive questioning. The system illustrates how sensors in the home used for entertainment and communication purposes might also be employed – at a negligible cost – for preventative health care applications.
Kautz, Fox, Etzioni, Borriello, Arnstein An Overview of the Assisted Cognition Project

The rise of Alzheimer’s disease is one of the greatest health crises facing the industrialized world. Today, approximately four million Americans suffer from Alzheimer’s disease; by 2050, the number is expected to rise to 15 million people. As a result of the increasing longevity of the elderly, many sufferers are now aware that their capacities to remember, to learn, and to carry out the tasks of everyday life are slowly being lost. The Assisted Cognition Project is a new joint effort between the University of Washington’s Department of Computer Science, Medical Center, and Alzheimer’s Disease Research Center that is exploring the use of AI systems to support and enhance the independence and qual-ity of life of Alzheimer’s patients. The goal of the Assisted Cognition project is to develop novel computer systems that will enhance the quality of life of people suffer-ing from Alzheimer’s Disease and similar cognitive disorders. Assisted Cognition systems use ubiquitous computing and artificial intelligence tec
Lisetti, LeRouge, Nasoz The MAUI Project: Building MultiModal Affective User Inferfaces for Everyone

Pages 66-73
We address some of the current challenges in intelligent interfaces for universal access and present the design of an intelligent interface which is aimed at 1) inputprocessing the user's sensory modalities (or modes) via various media, 2) building (or encoding) a model of theuser's emotions (MOUE) and 3) adapting its multimedia output to provide the user with an easier and morenatural technology access and interaction. We identify key research issues relating to one particularly rich application of MOUE, namely home health care provided via telemedicine.
Mann & Helal Smart Phones for the Elders: Boosting the Intelligence of Smart Homes

Pages 74-79
The ICTA center, recently established at the University of Florida, focuses on the use of wireless and pervasive technologies to enhance the quality of life and independence of elders or those aging with or into disabilities. In this article, we describe one of the research and development activities in the center that adopts the emerging smart phones technology as an effective booster of the utility of smart homes and other smart spaces. We present our vision of using smart phones as magic wands that can be used passively as remote control devices, and proactively as intelligent companions offering advises, reminders, warnings and calls for help. We also present our smart phone/home architecture and describe our efforts in early application developments.
Miller, Haigh, Dewing First, Cause No Harm: Issues in Building Safe, Reliable and Trustworthy Elder Care Systems

Pgase 80-84
Emerging smart, adaptive, integrative reasoning and interaction management technologies-which we choose to call Interaction Design Systems-hold enormous promise to solve a growing international problem: the provision of care for elderly populations. There are, however, substantial novel challenges to providing care with this type of technology to this population. Specific challenges arise from providing safe, reliable and affordable systems for a highly diverse population that is not in a position to oversee or compensate for technologies failings. These pressures should drive us toward specific IDS architectures designed for growth, expansion and tuning-both for the individual installation and over the lifespan of the technology. Furthermore, they should also, generally drive us toward initial delivery of systems which provide minimal automation capabilities, augmenting the supervisory role of human caregivers, rather than trying to replace them. Most importantly, any such system going into final use sh
Pollack as contact Pearl: A Mobile Robotic Assistant

Pages 85-92
The Nursebot project is a multi-disciplinary, multi-university effort aimed at developing mobile robotic assistants for the elderly. In this paper, we describe one such robot, Pearl. Pearl has two primary functions: (i) reminding people about routine activities such as eating, drinking, taking medicine, and using the bathroom, and (ii) guiding them through their environments. We provide a brief overview of the hardware platform, and we sketch the major software systems that enable Pearl to perform its two main functions. A prototype version of Pearl has been completely built, with all software implemented, and preliminary testing has been done at the Longwood Retirement Community in Oakmont, PA.
Simpson, Schreckenghost, Kirsch A Task Planning, Scheduling and Sequencing Orthosis for the Cognitively Impaired

Pages 92-96
We are using planning, scheduling and task sequencing software originally developed at NASA JSC to produce a cognitive orthosis for individuals who have difficulty planning, scheduling and carrying out tasks. The cognitive orthosis decouples the user interface from the underlying planning, scheduling and sequencing software to (1) support a variety of user interfaces and (2) allow simultaneous access to the schedule by both the client and one or more caregivers.
Wagner, Guranlik, Phelps Achieving Global Coherence in Multi-Agent Caregiver Systems: Centralized versus Distributed Response Coordination in I.L.S.A.

Pages 97-102
In order for multi-agent systems to exhibit global coherence the agents must coordinate their activities or be limited to a problem space in which activities are highly independent. In the elder care domain, global coherence across the agent sys-tem responsible for giving the care requires coordination if only because the different agents in the system interact over a set of shared resources, namely the individual for whom the system provides care and the fallback human caregivers. In this paper we explore the general coordination issue in the elder care problem space and discuss the response plan-ning and coordination portions of the Independent LifeStyle Assistant (TM) agent system being developed at Honeywell.

Note that authorship was revised on 17 Sept 2003. Please use the citation: T. Wagner, V. Guralnik, J. Phelps
Whitehouse, Marling & Harvey Can a Computer be a Caregiver?

Pages 103-107
This paper critically examines the issues involved incomputer-assisted and automated caregiving for patients with Alzheimer's disease and related dementias. First, thenature of the problems experienced by those who care for dementia patients is explicated. Next, a high-level overviewof the ways in which computers might assist these caregivers, now and in the future, is presented. Finally, the ethical issuesraised by the advent of automated caregiving are explored.

Last Updated on 8/8/2002
By Karen Haigh