A Context-Aware Multi-Agent System for AmI Environments - PowerPoint PPT Presentation

� � � � A Context-Aware Multi-Agent System for AmI Environments � ——————————————————————— � � Andrei Olaru � Scientific advisers: Adina Magda Florea, AI-MAS Lab, UPB � Amal El Fallah Seghrouchni, LIP6, UPMC � � 15.12.2011 � 0/ 33 Computer . Science . Andrei Olaru & Engineering . Bucharest, Romania Department . 15.12.2011

� The Problem � Objectives � Related Work � Solution A Context-Aware Multi-Agent � - Agent Behavior System for AmI Environments � - System Topology ——————————————- � - Context Representation overview � Model � A New Platform � Conclusions � Future Work � Publications 0/ 33 Computer . Science . Andrei Olaru & Engineering . Bucharest, Romania Department . 15.12.2011

� � · Ambient Intelligence – or AmI – is the vision � A Context-Aware Multi-Agent � System for AmI Environments of a future ubiquitous electronic environment � Defining the Problem (1) that supports people in their daily tasks, in a proactive and context-aware, but ”invisible” � Objectives and non-intrusive manner [Ramos et al., 2008, Weiser, 1995, � Related Work Ducatel et al., 2001] � Solution · Context is any information that can be used to characterize � - Agent Behavior the situation of an entity [Dey, 2001] . Context-awareness is the property of an application that makes it adapt its behavior � - System Topology depending on context. � - Context Representation · AmI environments are characterized by a large number of � Model interconnected heterogeneous devices with generally limited storage and performance. � A New Platform � Conclusions · we can get insights on the features of Ambient Intelligence by � Future Work means of scenarios and existing Ambient Intelligence projects. � Publications 1/ 33 Computer . Science . Andrei Olaru & Engineering . Bucharest, Romania Department . 15.12.2011

� � A Context-Aware Multi-Agent � System for AmI Environments � Defining the Problem (2) � Objectives � Related Work � Solution � - Agent Behavior � - System Topology We see AmI as a large number of People � - Context Representation � Model � A New Platform � Conclusions � Future Work � Publications 2/ 33 Computer . Science . Andrei Olaru & Engineering . Bucharest, Romania Department . 15.12.2011

� � A Context-Aware Multi-Agent � System for AmI Environments � Defining the Problem (2) � Objectives � Related Work � Solution � - Agent Behavior � - System Topology We see AmI as a large number of People · Devices � - Context Representation � Model � A New Platform � Conclusions � Future Work � Publications 2/ 33 Computer . Science . Andrei Olaru & Engineering . Bucharest, Romania Department . 15.12.2011

� � A Context-Aware Multi-Agent � System for AmI Environments � Defining the Problem (2) � Objectives � Related Work � Solution � - Agent Behavior � - System Topology We see AmI as a large number of People · Devices · Services � - Context Representation � Model � A New Platform � Conclusions � Future Work � Publications 2/ 33 Computer . Science . Andrei Olaru & Engineering . Bucharest, Romania Department . 15.12.2011

� � A Context-Aware Multi-Agent � System for AmI Environments � Defining the Problem (2) � Objectives � Related Work � Solution � - Agent Behavior � - System Topology We see AmI as a large number of People · Devices · Services · and intense Communication � - Context Representation � Model � A New Platform � Conclusions � Future Work � Publications 2/ 33 Computer . Science . Andrei Olaru & Engineering . Bucharest, Romania Department . 15.12.2011

� � A Context-Aware Multi-Agent � System for AmI Environments � Defining the Problem (2) � Objectives � Related Work � Solution � - Agent Behavior � - System Topology We see AmI as a large number of People · Devices · Services · and intense Communication � - Context Representation � Model � A New Platform � Conclusions � Future Work � Publications 2/ 33 Computer . Science . Andrei Olaru & Engineering . Bucharest, Romania Department . 15.12.2011

� � A Context-Aware Multi-Agent � System for AmI Environments � Defining the Problem (2) � Objectives � Related Work � Solution � - Agent Behavior � - System Topology We see AmI as a large number of People · Devices · Services · and intense Communication � - Context Representation � Model We see an AmI system as organized on several layers: ◮ intelligent interfaces � A New Platform ◮ application � Conclusions ◮ interoperability � Future Work ◮ network ◮ hardware � Publications 2/ 33 Computer . Science . Andrei Olaru & Engineering . Bucharest, Romania Department . 15.12.2011

� � A Context-Aware Multi-Agent � System for AmI Environments � Defining the Problem (2) � Objectives � Related Work � Solution � - Agent Behavior � - System Topology We see AmI as a large number of People · Devices · Services · and intense Communication � - Context Representation � Model We see an AmI system as organized on several layers: ◮ intelligent interfaces � A New Platform · application-specific processing � Conclusions ◮ application ← · generic information transfer � Future Work ◮ interoperability ◮ network � Publications ◮ hardware 2/ 33 Computer . Science . Andrei Olaru & Engineering . Bucharest, Romania Department . 15.12.2011

� � � A Context-Aware Multi-Agent � System for AmI Environments � Defining the Problem (2) � Objectives � Related Work � Solution � - Agent Behavior We see AmI as a large number of People · Devices · Services � - System Topology · and intense Communication � - Context Representation We see an AmI system as organized on several layers: � Model ◮ intelligent interfaces � A New Platform · application-specific processing ◮ application ← � Conclusions · generic information transfer ◮ interoperability � Future Work ◮ network � Publications ◮ hardware 2/ 33 Computer . Science . Andrei Olaru & Engineering . Bucharest, Romania Department . 15.12.2011

� � A Context-Aware Multi-Agent � System for AmI Environments � Defining the Problem (3) AmI challenges for the application layer: � Objectives � Related Work · How to make AmI reliable and dependable? � Solution · How to manage the large quantity of information � - Agent Behavior generated by sensors and devices? � - System Topology · How to provide the relevant information to the � - Context Representation interested user? � Model � A New Platform � Conclusions � Future Work � Publications 3/ 33 Computer . Science . Andrei Olaru & Engineering . Bucharest, Romania Department . 15.12.2011

� � � A Context-Aware Multi-Agent � System for AmI Environments � Defining the Problem (3) AmI challenges for the application layer: � Objectives � Related Work · How to make AmI reliable and dependable? � Solution · How to manage the large quantity of information � - Agent Behavior generated by sensors and devices? � - System Topology · How to provide the relevant information to the � - Context Representation interested user? � Model � A New Platform Elements of a solution: ◮ software agents � Conclusions ◮ context-awareness � Future Work ◮ elements of self-organization � Publications 3/ 33 Computer . Science . Andrei Olaru & Engineering . Bucharest, Romania Department . 15.12.2011

� � A Context-Aware Multi-Agent � System for AmI Environments � The Problem � Objectives of the thesis � Related Work Our goal: build a context-aware multi-agent system for application layer of an Ambient Intelligence system. � Solution � - Agent Behavior � - System Topology � - Context Representation � Model � A New Platform � Conclusions � Future Work � Publications 4/ 33 Computer . Science . Andrei Olaru & Engineering . Bucharest, Romania Department . 15.12.2011

� � � A Context-Aware Multi-Agent � System for AmI Environments � The Problem � Objectives of the thesis � Related Work Our goal: build a context-aware multi-agent system for application layer of an Ambient Intelligence system. � Solution � - Agent Behavior Main objectives of the thesis: � - System Topology � - Context Representation ◮ propose a model of the multi-agent system; ◮ conceive scenarios that emphasize the requirements of � Model real-scale Ambient Intelligence systems; � A New Platform ◮ design and implement a simulation testbed to serve for � Conclusions experiments with AmI applications; � Future Work ◮ validation of the model by implementation and testing. � Publications 4/ 33 Computer . Science . Andrei Olaru & Engineering . Bucharest, Romania Department . 15.12.2011