A Context-Aware Multi-Agent System for Ambient Intelligence - PowerPoint PPT Presentation

� � � � A Context-Aware Multi-Agent System for Ambient Intelligence Environments � ——————————————————————— � � Andrei Olaru � University ”Politehnica” of Bucharest University Pierre et Marie Curie, Paris � � 18.05.2010 � � 1/ 17 Computer . . Andrei Olaru Science . & Engineering . K.U.Leuven, 18.05.2010 Department

� A & C � AI-MAS � PhD � AmI A Context-Aware Multi-Agent � Scenario System for Ambient � Challenges Intelligence Environments ——————————————- � Layers � Approach overview � Self-Organization � Context � Results � Future 2/ 17 Computer . . Andrei Olaru Science . & Engineering . K.U.Leuven, 18.05.2010 Department

Faculty of Automatic Control And Computers (A&C) � A Context-Aware Multi-Agent � System for Ambient Intelligence http://acs.pub.ro/ � Environments � A & C Departments: � AI-MAS ◮ Department of Automatic Control and Systems Engineering � PhD Head: prof. Ioan Dumitrache � AmI ◮ Department of Automatic Control and Industrial Informatics � Scenario Head: prof. Radu Dobrescu ◮ Department of Computer Science and Engineering � Challenges Head: prof. Valentin Cristea http://cs.pub.ro � Layers � Approach � Self-Organization � Context � Results � Future 3/ 17 Computer . . Andrei Olaru Science . & Engineering . K.U.Leuven, 18.05.2010 Department

Artificial Intelligence and Multi Agent Systems Laboratory � A Context-Aware Multi-Agent Head: prof. Adina Magda Florea � System for Ambient Intelligence � Environments http://aimas.cs.pub.ro � A & C · Research areas: · Application areas: � AI-MAS ◮ Multi-agent systems ◮ e-commerce � PhD ◮ Autonomous agents ◮ e-learning � AmI ◮ Coordination ◮ SCM � Scenario mechanisms, negotiation ◮ semantic web services � Challenges ◮ Emotional agents for business processes ◮ Self-organization ◮ Ambient Intelligence � Layers � Approach � Self-Organization � Context � Results � Future 4/ 17 Computer . . Andrei Olaru Science . & Engineering . K.U.Leuven, 18.05.2010 Department

� A Context-Aware Multi-Agent � System for Ambient Intelligence � Environments A Context-Aware Multi-Agent System for Ambient � A & C Intelligence Environments � AI-MAS · develop a multi-agent system based model for Ambient � PhD Thesis Intelligence, considering various AmI scenarios and verifying � AmI it by means of simulation. � Scenario · advisor: prof. Adina Magda Florea � Challenges � Layers · in cotutelle with Laboratoire d’Informatique de Paris 6 � Approach (LIP6), UPMC, advised by prof. Amal El Fallah Seghrouchni � Self-Organization � Context � Results � Future 5/ 17 Computer . . Andrei Olaru Science . & Engineering . K.U.Leuven, 18.05.2010 Department

� A Context-Aware Multi-Agent Ubiquitous electronic environment that supports people � System for Ambient Intelligence � Environments in their daily tasks, in a proactive, but ”invisible” and non-intrusive manner [Ramos et al., 2008, Weiser, 1993] � A & C � AI-MAS � PhD � Ambient Intelligence � Scenario � Challenges � Layers � Approach � Self-Organization � Context � Results � Future 6/ 17 Computer . . Andrei Olaru Science . & Engineering . K.U.Leuven, 18.05.2010 Department

� A Context-Aware Multi-Agent Ubiquitous electronic environment that supports people � System for Ambient Intelligence � Environments in their daily tasks, in a proactive, but ”invisible” and non-intrusive manner [Ramos et al., 2008, Weiser, 1993] � A & C � AI-MAS � PhD � Ambient Intelligence � Scenario � Challenges � Layers � Approach � Self-Organization � Context � Results � Future 6/ 17 Computer . . Andrei Olaru Science . & Engineering . K.U.Leuven, 18.05.2010 Department

� A Context-Aware Multi-Agent Ubiquitous electronic environment that supports people � System for Ambient Intelligence � Environments in their daily tasks, in a proactive, but ”invisible” and non-intrusive manner [Ramos et al., 2008, Weiser, 1993] � A & C � AI-MAS � PhD � Ambient Intelligence � Scenario � Challenges � Layers � Approach � Self-Organization � Context People � Results � Future 6/ 17 Computer . . Andrei Olaru Science . & Engineering . K.U.Leuven, 18.05.2010 Department

� A Context-Aware Multi-Agent Ubiquitous electronic environment that supports people � System for Ambient Intelligence � Environments in their daily tasks, in a proactive, but ”invisible” and non-intrusive manner [Ramos et al., 2008, Weiser, 1993] � A & C � AI-MAS � PhD � Ambient Intelligence � Scenario � Challenges � Layers � Approach � Self-Organization � Context People · Devices � Results � Future 6/ 17 Computer . . Andrei Olaru Science . & Engineering . K.U.Leuven, 18.05.2010 Department

� A Context-Aware Multi-Agent Ubiquitous electronic environment that supports people � System for Ambient Intelligence � Environments in their daily tasks, in a proactive, but ”invisible” and non-intrusive manner [Ramos et al., 2008, Weiser, 1993] � A & C � AI-MAS � PhD � Ambient Intelligence � Scenario � Challenges � Layers � Approach � Self-Organization � Context People · Devices · Communication � Results � Future 6/ 17 Computer . . Andrei Olaru Science . & Engineering . K.U.Leuven, 18.05.2010 Department

� A Context-Aware Multi-Agent � System for Ambient Intelligence � Environments � A & C � AI-MAS � PhD � AmI � An AmI Scenario � Challenges � Layers � Approach The large screen can be used to display context-aware advertisements... � Self-Organization � Context � Results � Future 7/ 17 Computer . . Andrei Olaru Science . & Engineering . K.U.Leuven, 18.05.2010 Department

� A Context-Aware Multi-Agent � System for Ambient Intelligence � Environments � A & C � AI-MAS � PhD � AmI � An AmI Scenario � Challenges � Layers � Approach ...or to draw attention of the user... � Self-Organization � Context � Results � Future 7/ 17 Computer . . Andrei Olaru Science . & Engineering . K.U.Leuven, 18.05.2010 Department

� A Context-Aware Multi-Agent � System for Ambient Intelligence � Environments � A & C � AI-MAS � PhD � AmI � An AmI Scenario � Challenges � Layers � Approach ...to show an interactive map for which the mobile phone is too small [Canut et al., 2009] ... � Self-Organization � Context � Results � Future 7/ 17 Computer . . Andrei Olaru Science . & Engineering . K.U.Leuven, 18.05.2010 Department

� A Context-Aware Multi-Agent � System for Ambient Intelligence � Environments � A & C � AI-MAS � PhD � AmI � An AmI Scenario � Challenges � Layers � Approach ...to draw attention towards important events, if the phone cannot... � Self-Organization � Context � Results � Future 7/ 17 Computer . . Andrei Olaru Science . & Engineering . K.U.Leuven, 18.05.2010 Department

� A Context-Aware Multi-Agent � System for Ambient Intelligence � Environments � A & C � AI-MAS � PhD � AmI � An AmI Scenario � Challenges � Layers � Approach ...or take part in a more complex scenario in which it facilitates the communication with another user in distress. � Self-Organization � Context � Results � Future 7/ 17 Computer . . Andrei Olaru Science . & Engineering . K.U.Leuven, 18.05.2010 Department

� A Context-Aware Multi-Agent � System for Ambient Intelligence � Environments � A & C · How to make AmI reliable and dependable? � AI-MAS � PhD � AmI · How to manage the large quantity of information � Scenario generated by sensors and devices? � AmI Challenges and Concerns � Layers · How to provide only interesting information to the user � Approach in every situation? � Self-Organization � Context · How to make AmI privacy-aware and trustable? � Results � Future 8/ 17 Computer . . Andrei Olaru Science . & Engineering . K.U.Leuven, 18.05.2010 Department

� A Context-Aware Multi-Agent � System for Ambient Intelligence � Environments � A & C � AI-MAS � PhD � AmI � Scenario � Challenges � AmI Layers � Approach � Self-Organization � Context � Results � Future 9/ 17 Computer . . Andrei Olaru Science . & Engineering . K.U.Leuven, 18.05.2010 Department

· address the application layer – above hardware, network and � A Context-Aware Multi-Agent interoperability, below intelligent user interfaces. � System for Ambient Intelligence � Environments · this is where most of the challenges and most features that � A & C make AmI ”intelligent”. � AI-MAS � PhD · answer some of these challenges, by providing a model for an AmI system’s application layer. � AmI � Scenario � Challenges � Layers � Research Approach � Self-Organization � Context � Results � Future 10/ 17 Computer . . Andrei Olaru Science . & Engineering . K.U.Leuven, 18.05.2010 Department