My research interests follow several main directions related to agreement technologies, semantic web services and workflows, agent-oriented software engineering and multiagent system applications. The problematic of coordination unifies my research in these different topics. From an application point of view, I am interested in the areas of medicine, e-commerce and simulations.

Agreement technologies

Open Interaction Frameworks: Towards a Governing Environment. The goal of this project is to prove that open interaction frameworks, conceived as systems of “artificial institutions” within which artificial and human agents can flexibly interact, can be used to support human collaboration in real working environments, like for example a virtual enterprise (2009-2011).

MediCoordination: Studying interoperability issues in the eHealth domain (2008-2009): http://www.medicoordination.ch/

SACoord: Semantic Agent Coordination (2008-2009)

COST action IC0801 Agreement Technologies (2008-2012): The main objective of the Action is to support and promote the harmonization of nationallyfunded high-quality research towards a new paradigm for next-generation distributed systems based on the notion of agreement between computational agents, fostering research excellence and sowing the seeds for technology transfer to industry. http://www.agreement-technologies.eu/

Semantic Web Services and Workflows

The combination of Web Services and the Semantic Web can lead to a new technology infrastructure: semantic web services. The aim of this research is to study how web services from distinct providers can be automatically composed into workflows in order to solve concrete user requirements. Current projects:

• Web Service Composition with the Workflow System OpenWFE
• Business Rules for Web Services
• Distributed Directories of Web Services (for CASCOM)
• Content Repositories for Workflows (in collaboration with Day)

Agent-oriented Software Engineering

The autonomous agent paradigm can be used as a strong abstraction to model and design complex concurrent and distributed applications. Modeling a multiagent system becomes an interaction-oriented process. To that aim, the key issue of multiagent engineering is coordination, or more concretely the handling of agent interactions, the organization of their environment, and the management of their subjective intra-agent dependencies. Software-engineering methodologies are needed to settle abstract descriptions into reality. My main focus is dedicated to requirements for agent platforms. In this context, I am especially interested in studying the notion of environment as a mean for ruling systems with e.g. security laws. Current projects:

• Objective and Subjective Coordination
• The Governing Environment

Multiagent System Applications

Multiagent systems follow the vision of an Artificial Society of Software Agents roaming and interacting over networks (such as peer-to-peer systems) in structured environments. To that aim, the study of Agent Societies has to explore new programming paradigms to support the engineering of Multi-Agent Systems (MAS) based on standard technologies. Current projects:

• Agent-based Micro-simulation of the whole Swiss Highway Network
• Agent-based Simulation of Intelligent Highways
• Smart Mobile Context-aware Timetable Application for the Public Tranports of Lausanne
• Multiuser Preference Discovery

CASCOM, a European Project of the 6th Framework Programme (2004-2007)

Since September 2004, I am the scientific manager at EPFL of the project CASCOM, a European project of the 6th Framework Programme.

The essential approach of CASCOM is the innovative combination of intelligent agent technology, semantic Web services, peer-to-peer, and mobile computing for intelligent peer-to-peer mobile service environments. Conventional peer-to-peer computing environments are extended with components for mobile and wireless communication. The services of CASCOM environments are provided by peer software agents which exploit the CASCOM coordination infrastructure to efficiently operate in highly dynamic environments. The generic CASCOM intelligent peer-to-peer infrastructure includes efficient communication means, support for context-aware adaptation techniques, as well as dynamic and secure service discovery and composition planning.

The project will deliver a full proof-of-concept implementation of the generic CASCOM support infrastructure for mobile business application service coordination for mobile users and workers, and a trialed demonstrator for selected pervasive health care application services.