Ambient Intelligence and Future Trends - (eBook)

Title Page 2
Preface 6
Organization 7
Contents 10
Long Papers 10
A Study on Autonomic Decision Method for Smart Gas Environments in Korea 14
Introduction 14
Collecting Status Data 15
The Necessity of Improving the Decision Approach 16
The Autonomic Decision Method for Smart City Gas 18
Conclusions and Future Works 20
References 21
Multiagent Systems and Wearable Devices: Helping People Live Healthier 23
Introduction 23
SWeDe Multiagent System 24
Case Study: Geriatric Residences Monitoring 27
Results and Conclusions 28
References 29
Accurate Temporal Relationships in Sequences of User Behaviours in Intelligent Environments 31
Introduction 31
Sequential Patterns of User Behaviour System 32
Related Work 34
Identifying Time Relations 34
Data Collection for Identifying Time Relations 35
‘Basic Algorithm’ for Identifying Time Relations 36
‘EM Algorithm’ for Identifying Time Relations 37
Validation and Results 37
Conclusions 38
References 38
A Framework to Enable Two-Layer Inference for Ambient Intelligence 40
Introduction 40
Standards and Technologies 41
Designing the Framework 42
AUseCase 44
Discussion 46
References 46
GLSS – Group Learning in Shared Spaces Considering Aspects Like Emotion and Personality 48
Introduction 48
Background 49
Learning Styles 49
Intelligent Tutoring 50
Emotion 50
Personality 51
Proposed Architecture 51
Architecture 51
Scenario 53
Conclusions 53
References 54
Rewiring Strategies for Changing Environments 56
Introduction 56
Scenario: Parental Control on Media Content in Dynamic Pervasive Environment 57
Conceptual Models for Parental Control Application 58
Adopting MERODE Models in ReWiRe 61
Conclusions, Limitations and Future Research 62
References 63
Toward a Behavioral Decomposition for Context-Awareness and Continuity of Services 65
Introduction 65
Related Works: Context and Context-Awareness 66
Context-Awareness 66
Limitations 67
Reactivity and Behavioral Decomposition of Context-Awareness Mechanism for Application’s Adaption 69
Towards an Hybrid Approach 70
Conclusion 71
References 71
A User-Oriented and Context-Aware Service Orchestration Framework for Dynamic Home Automation Systems 73
Introduction 73
Related Work 74
Motivating Scenario and Requirements 75
Mapping of User Settings to Low-Level Instructions 76
Architectural Overview of the Home Automation Framework 78
Conclusions and Future Work 79
References 80
A Multi-user Smart System for Reminding Medication 81
Introduction 81
System Overview 82
System Description 82
System Architecture and Technologies 83
Research Objectives 85
Methodology and Associated Work Plan 86
Final Considerations 87
References 87
Ontology and SWRL-Based Learning Model for Home Automation Controlling 89
Introduction 89
Existent Works about the Use of Ontologies on AmI Systems 90
IntelliDomo’s Learning Model 91
SWRL Rules Management 91
Learning Module 92
Conclusions and Future Lines 95
References 95
Model-Driven Development for Ubiquitous MAS 97
Introduction 97
Using MDD to Develop Embedded Agents 98
Implementing an Application Example 101
References 104
Accurate Human Limb Angle Measurement in Telerehabilitation: Sensor Fusion through Kalman, LMS and RLS Adaptive Filtering 106
Introduction 106
Sensor Fusion Approaches 108
Kalman Filter Sensor Fusion Approach 108
LMS and RLS Filters Sensor Fusion Approach 109
Theoretical Simulations: Application on Synthesized Signals 110
Results from Application on Real Signals 111
Conclusions and Future Work 113
References 113
An Agent-Based Framework for Context-Aware Digital Signage 114
Introduction 114
Basic Approach 116
Design and Implementation 116
Location Information Server 117
Agent Runtime System 117
Mobile Agent-Based Multimedia Content 118
Remarks 119
Application 119
Conclusion 120
References 121
What Context Matters? Towards Multidimensional Context Awareness 122
Introduction 122
State-of-Art in Context Relevancy 123
Context Relevancy 124
Context Dimensions 125
Application Scenarios 127
Conclusions 128
References 129
A Framework for Modeling and Analysis of Ambient Agent Systems: Application to an Emergency Case 130
Introduction 130
Case Study 131
Modeling Approach 132
Simulation and Analysis Results 134
Conclusions 137
References 137
Autonomous and Context-Aware Scheduling for Public Displays Using Place-Based Tag Clouds 139
Introduction 139
Related Work 140
A Place-Based Content Scheduler 141
Evaluation and Discussion 143
Experimental work 143
Results and Discussion 144
Conclusions 145
References 146
Semantic and Implicit HCI: A Step Forward AmI Delivery at Home for Patient Experience 147
Introduction 148
State-of-the-Art 149
Framework AmI Based on Semantic Web Technologies 150
Multi-Agent System Integration in Digital Home Environment 150
Ontology Based on Standards 151
Technological Design to Support AAL Services at Home 152
Conclusions and Future Work 153
References 153
Domotic Hardware Infrastructure in PERSONA Project 156
Introduction 156
List of Devices 158
Security Sensors 158
Environmental Sensors 158
Actuators 158
Specific Designs 159
Designed PCB 159
Generic PCB 159
Specific PCB 162
References 162
Context Aware Management Plateform to Invoke Remote or Local e Learning Services: Application to Navigation and Fishing Simulator 163
Introduction 163
Case Study 164
Web Services 166
Web Services Definition 166
The SaaS Model 167
Cloud Computing 167
Context Adaptation 168
The WComp Plate Form 168
Reviewed Architecture 168
Implementation 168
Related Works 169
Conclusion 169
References 169
EVATAR – A Prototyping Middleware Embodying Virtual Agents to Autonomous Robots 172
Introduction 172
EVATAR 173
Experimentation 176
Related Work - Discussion 178
Conclusions – Future Work 179
References 179
User Recognition in AAL Environments 181
Introduction 182
Active Ageing 182
VirtualECare 182
Recognition 184
Biometric 184
Facial Recognition 185
VirtualECare Recognition Module 185
Implementation 186
Security 187
Conclusions and Future Work 187
References 188
Short Papers 12
Patient Monitoring under an Ambient Intelligence Setting 189
Ambient Intelligence in Healthcare 189
An Ambient Intelligence Based Architecture for Patient Monitoring 190
The Architecture 191
An Implementation in a Real Environment 191
Conclusions 192
References 192
Helping Elderly People and Persons with Disability to Access the Information Society 193
Introduction 193
Research and Development in Naviga 194
Naviga Main Components 194
Expected Results 195
Conclusions 196
References 196
Context Triggered Experience Sampling 197
Introduction 197
Context Aware Toolbox 198
Discussion and Future Work 200
References 200
Adaptive Experience Sampling: Addressing the Dynamic Nature of In-Situ User Studies 201
Introduction 201
Adaptive Experience Sampling 203
Discussion and Future Work 204
References 204
An Intelligent Domotics System to Automate User Actions 205
Introduction 205
The Intelligent Domotics System 206
Related Work 208
Conclusion 208
References 208
Assessing the Usefulness and Acceptance of HERMES MyFuture System in TwoEuropean Countries 209
Introduction 209
Methodology 211
Results 211
Conclusions 212
References 212
Agency, Ambience, Assistance: A Framework for Practical AAL 213
Introduction 213
Ambient Assisted Living 214
Agent-Based Middleware 214
Home Instrumentation and Beyond 216
Conclusion and Future Work 216
References 216
Automatic Generation of Personal Maps 217
Introduction 217
Why Personal Maps? 218
Data Collection and Analysis 218
Location Data 219
Improvements Using Sensor Data 219
Other Challenges 220
Conclusion 220
References 220
Modeling and Simulating Ambient Assisted Living Environments – A Case Study 221
Introduction 221
Preliminary Works 222
Simulation of Physical Assistance in BAALL 223
Conclusion and Future Works 224
References 224
Doctoral Consortium 13
Methodology for the Integration of Information, Communication and Automation Technologies in Housing Rehabilitation 225
Introduction 225
ICAT as Ecological and Social Sustainability Strategies 226
Defining the Methodology 227
Packs of ICAT Functions 228
Methodology Application – An Example 230
Conclusion 231
References 232
Why Traders Need Ambient Intelligence 233
Introduction 233
Measuring the Stress State of the Trader 234
Ambient Intelligence Traders Network: A Real Time Stress Feeling of the Market 235
Ambient Intelligence Trader vs Traditional Trader: A Workday 236
How to Build a Trading Support AmI 238
Conclusions and Future Work 239
References 239
Awareness Models for the Development of Ubiquitous Systems 241
Introduction 241
Awareness 242
Awareness Models 243
Awareness Type Model 243
Specialized Awareness Type Model 244
Awareness Requirement Model 244
Awareness Usage Model 245
Modeling Examples 246
Conclusions and Future Work 248
References 248
A Statistical-Relational Activity Recognition Framework for Ambient Assisted Living Systems 250
Introduction 250
Markov Logic Networks for Activity Recognition 252
Activity Recognition Framework 252
Experiments 254
Dataset and Setting 255
Results 255
Discussion and Conclusion 255
References 257
Author Index 258

"Context Triggered Experience Sampling (p. 193-194)

David V. Keyson

Abstract. As products and services have become more embedded in the everyday life and routines of users, the need to understand how the context of use influences product usage over time has increased. The Experience Sampling Method (ESM) aims to capture both the context and content of the daily life of individuals. Critical to the ESM methods is the notion of asking the appropriate set of experience related questions at the right moment. The current approach to ESM is based on provided the user with a set of qualitative questions, typically using scaled measures, at pre-defined intervals during the course of the day over a period of several days or longer. In the current paper the Context Aware Toolbox (CAT) is described as a means to rapidly equip an environment with sensors and actuators which in turn can trigger user input via formal questionnaires and indirect input based on the task at hand.

The main elements of the CAT is a database of algorithms, written in MAX MSP which can be rapidly adapted to particular needs, Phidget sensors and actuators, a MAX Mini containing instructions and the database, touch screens for displaying the qualitative measures on site, and installation tools, as well as a web based infrastructure for adding new content to the database. The CAT setup is currently being applied to design research cases in the area of social and environmental sustainability; an application design case focused on sustainable living is described.

Keywords: Experience Sampling, Context Awareness, Prototyping.

1 Introduction

The emergence of products and services embedded in the everyday routine of people reflect the degree to which technology has become a part of lives. Increasingly, the only way to evaluate the usability of products and services, behind the immediate ease of use of the interface, is to study user interaction in context. For pure Internet based applications, this might mean launching a “beta” version of the software and observing what happens with large number of users and then fixing the problems over successive beta releases.

In the case of evaluating physical products in context, focus group sessions, or phone or mail based questionnaires may only provide brief glimpses into product adoption and usability issues. The two predominant methods to gather ongoing usage data are the Diary Recollection Method [1] and the Experience Sampling Method [2].

Both methods can also be combined [3]. In the case of DRMS the user may be given a diary to fill-in over a fixed time interval across several days or more, or maybe given a camera to take pictures of the context of use on a regular basis. Experience sampling may be conducted via asking the user to fill in a paper form of electronic form on the basis of timed requests which may generated via a watch, beeper, Website, or PDA, to name a few possible platforms.

It is generally argued that ESM combines the ecological validity of naturalistic behavioral observations with the nonintrusive nature of diaries and the precision of scaled questionnaire measures. While the limitations of ESM are acknowledged, in particular the demands it imposes on respondents, which may contribute to a self-selection bias and selective non-response, the method is regarded as some as the best alternative to capturing the context of use.

Central to the ESM is the notion that “by sampling experience the moment it occurs, it avoids the potential distortions associated with the use of daily or weekly retrospective diaries” [2]. While this seems at first glance a logical statement, taking a deeper look at ESM studies reveals that often ESM based questionnaires appear at pre-set time intervals, but are not necessarily triggered when the behavior under study occurs in a given context.

Furthermore the ESM assumes that experiences can only be recorded via qualitative scales, in some cases the way in which a product or service is engaged may in itself reveal experiential data. For example the degree to which a user applies force to while setting a physical control may be an indication of stress or the number of repeated efforts to make a “soft” selection on a screen could reflect a degree of confusion. Such data could be monitored in real-time and collected over an extended period. Taking things a step further the product or service itself could be remotely updated given the availability of online usability data. In some cases it may make sense deploy sensors and actuators close to the product or additionally in the vicinity of the product to capture experience data. To this extent the Context Aware Toolkit was developed."