Architecture and Prototype System

boorishadamantAI and Robotics

Oct 29, 2013 (3 years and 9 months ago)

88 views

Enhancing Learning Experiences through

Context
-
Aware Collaborative Services: Software
Architecture and Prototype System

Nikolaos Dimakis, Lazaros Polymenakos and John Soldatos

Athens Information Technology

19,5 Km Markopoulou Ave.

Peania, GR 19002

+30 2106682759

e
-
mail: {ndim, lcp, jsol}@ait.edu.gr

http://www.ait.gr/research/RG1/overview.asp

Introduction


Introduction / Motivation


Software Architecture for e
-
learning


Sensors and Actuators


Context
-
Acquisition


Service tier


The `Memory Jog’ Service


Applicable to e
-
learning


Evaluation results

Introduction


Introduction / Motivation


Software Architecture for e
-
learning


Sensors and Actuators


Context
-
Acquisition


Service tier


The `Memory Jog’ Service


Applicable to e
-
learning


Evaluation results

Pervasive
-
Ubiquitous Computing Vision


Vision:


Transform physical spaces into
computationally active and
intelligent environments


Human
-
Centric services offered
transparently regardless of time
and end
-
users location


Technology Trends
-

Enablers:


Broadband/ Wireless Networks


Web Services
-

Top Layer
Interoperability


PDA

s, Mobile Phones


Cheap Multi
-
Purpose Sensors


Recognition Algorithms


Adaptivity & User Profiling
Technologies

Main application areas


Smart house


Smart office


Health care
-

telemedicine


Ambient assisted living


Security


E
-
learning

Major Challenges


High Distribution


Context
-
Awareness


Transparency


Heterogeneity


Software / Hardware


Sensors, Terminals, Devices


Mobility


Personalization


Autonomy


Dynamism

CHIL


Computers in the Human Interaction Loop


EU FP6 Integrated Project (IP 506909)


Coordinators: Universit
ä
t Karlsruhe (TH)
Fraunhofer Institute IITB


Duration: 36 months


Total Project costs: Over 24M



Goal: Create environments in which
computers serve humans who focus on
interacting with other humans as opposed
to having to attend to and being
preoccupied with the machines
themselves


Key Research Areas:


Perceptual Technologies


Human
-
Centric Pervasive

Services


Software Infrastructure

… a little more technical


Introduction / Motivation


Software Architecture for e
-
learning


Sensors and Actuators


Context
-
Acquisition


Service tier


The `Memory Jog’ Service


Applicable to e
-
learning


Evaluation results

Software Architecture


3 tier approach


Sensors and Actuators tier


Cameras, microphones, Targeted Audio, Projectors,
Displays, etc.


Context
-
Acquisition tier


Body Tracker, Face Identification, Audio Source
Localization, Speech Activity Detection, etc.


Service tier


JADE Agent Framework.

Context
-
awareness


Dey & Abowd (1999)


Context is: “any information that can be used to
characterise the situation of an entity, where an
entity can be a person, place, physical or
computational object”


Context
-
awareness


human
-
centric
ubiquitous computing


“Smart Learning Spaces”


a realization of
the ubiquitous computing paradigm

Non
-
obtrusive interaction


Sensors


Cameras


Microphones


No wearable computers, no RFID tags

Context
-
Acquisition tier


a.k.a. Perceptual Components tier.


Exploits sensor signals to extract elementary
context.


Locations of people


Identities of people


Speaker location


Fusion of PCs


Situation Recognition


Examples

Situation Model for a lecture

Service tier


Following the JADE Framework for distributed
agents.


Each service extends ServiceAgent


Basic functionalities are ensured: Communication,
Message extraction, Database Querying etc.


Situation modeling


Enables situation recognition, situation model tracking and
event notification


Sensor and actuator control


KnowledgeBase and Database interfacing.


Each user is managed by PersonalAgent

Agent
-
hierarchy

Knowledge Base


Directory mechanism, leveraging a knowledge
base


“yellow pages” for students, courses, research etc


Registration repository for existing
components


“Smart queries”


“Which camera faces the door?”

Specialized Middleware


Glue two conceptually different ends together


IBM’s CHILiX


NIST SmartFlow


Significantly boost distribution


Architectural snapshot

… The Memory Jog service


Introduction / Motivation


Software Architecture for e
-
learning


Sensors and Actuators


Context
-
Acquisition


Service tier


The `Memory Jog’ Service


Applicable to e
-
learning


Evaluation results

The Memory Jog


A lecture, meeting support service. It provides:


Pertinent information to participants


Interfacing with Databases for past event querying


Preferred features, such as:


Agenda tracking


Intelligent Video Recording


Handles mobile participants (PDAs)



The Memory Jog (cont.)


It exploits:


The whole range of the sensors and actuators


All registered Perceptual Components


Databases for information retrieval and
Knowledgebases for high level reasoning


Both are controlled by specialized agent
-
members of the
service
-
tier.


The Memory Jog (cont..)


It looks like:

Memory Jog features for e
-
learning


The Memory Jog features:


Intelligent Video Recording.


Contextual database search.


Automated notification using Targeted Audio.


Etc.

Intelligent Video Recording

Results


Introduction / Motivation


Software Architecture for e
-
learning


Sensors and Actuators


Context
-
Acquisition


Service tier


The `Memory Jog’ Service


Applicable to e
-
learning


Evaluation results

Focus Groups Evaluation


25 people participated


Students


Engineers


Management Personnel


Faculty


Feature

Approval
Percent

Intelligent
Recording

100%

Agenda tracking

56%

What happened
while I was
away?

72%

Participants
Biography

68%

Search past
events

84%

AIT’s Smart Learning Space

The End


Questions?