Frédéric Migeon

Model Driven Engineering for Designing Adaptive Multi-Agents Systems

A challenge for our days is to provide new efficient CASE (Computer Aided Software Engineering) tools enabling MAS designers towards Model Driven Engineering (MDE) approaches. The goal of MDE is to improve the development process and the quality of the software produced. Our work focuses on two different aspects of MAS. The functional one, which is application dependent and close to the decision process of agents, and the operational one related to elementary capabilities of agents. For each point of view, we have defined specific meta-models. Our goal in this paper is to provide a mapping from the functional meta-model to the operational that constitutes a specific platform model. As we are interested in adaptive systems, we have to deal with adaptation both at the agent and the system level. We address this problem by respectively using the JavAct flexible architecture and the Adaptive MAS principles.

Mobile Agents Based Self-Adaptive Join for Wide-Area Distributed Query Processing

In this article, optimization of decision support queries is considered in the context of wide-area distributed databases. An original approach based on the “mobile agent†paradigm is proposed and evaluated. Agents’ autonomy and reactivity allow operators of the execution plan to adapt dynamically to estimation errors on relations and to evolutions in the state of the execution system, avoiding time overheads commonly associated with centralized monitoring. We present decentralized self-adaptive algorithms for dynamic optimization of join operators, and their implementations in Java using mobile agents. Then, we evaluate performance depending on error rate on statistical information on database, and on communication bandwidth and CPU frequency. The results show that the agent-based approach can lead to a significant reduction of response time and provide decision criteria for developing an effective migration policy.

Methodology Fragments Definition in SPEM for Designing Adaptive Methodology: A First Step

The aim of this paper is to highlight how SPEM (Software and System Process Engineering Meta-model) 2.0 OMG (Object Management Group) can participate to design adaptive methodology process. The idea follows the FIPA Methodology Technical Committee (TC) one which consists in expressing a methodology in several fragments. Then, designer has to combine the relevant fragments to compose his own methodology. In this paper, we have chosen SPEM 2.0 OMG to express the fragments. The latest SPEM version improves methodology content and process re-usability, by introducing new capabilities as a clear separation between structural and dynamic methodology concerns. Those improvements in the field of methodology specification, are studied to determine their interests in the scope of Agent-Oriented Software Engineering (AOSE) and particularly, their impact on methodology fragments definition. ADELFE and PASSI methodologies have been taken as example to illustrate the use of SPEM 2.0 in the scope of fragment definition. In this paper, only the first step of the general objective consisting in expressing the fragments, is done and presented.

An API for high-level software engineering of distributed and mobile applications

This paper proposes actors as a model for mobile agents. Because of their intrinsic autonomy (thread integration) and behavior changing ability, added to asynchronous communication, actors are naturally mobile entities, and moving actors has no effect on their initial semantics. Then, we propose a standard API on top of Java and RMI, called JavAct, for distributed and mobile applications programming. JavAct is actor-based and allows nearly immediate and strong agent mobility. Mobile actors are localized by means of a forwarding chain technique. Network-level references of actors contribute to location independence of computations. Additionally, in this paper, we give an overview of some related programming systems.

Formal Modelling and Verification of Cooperative Ant Behaviour in Event-B

Multi-agent technology is a promising approach to development of complex decentralised systems that dynamically adapt to changing environmental conditions. The main challenge while designing such multi-agent systems is to ensure that reachability of the system-level goals emerges through collaboration of autonomous agents despite changing operating conditions. In this paper, we present a case study in formal modelling and verification of a colony of foraging ants. We formalise the behaviour of cooperative ants in Event-B and verify by proofs that the desired system-level properties become achievable via agent collaboration. The applied refinement-based approach weaves proof-based verification into the formal development. It allows us to rigorously define constraints on the environment and the ant behaviour at different abstraction levels and systematically explore the relationships between system-level goals, environment and autonomous ants. We believe that the proposed approach helps to structure complex system requirements, facilitates formal analysis of various system interdependencies, and supports formalisation of intricate mechanisms of agent collaboration.

A Pattern based Modelling for Self-organizing Multi-agent Systems with Event-B

Self-Organizing Multi-Agent Systems (SO-MAS) are defined as a set of autonomous entities called agents interacting together in order to achieve a given task. Generally, the development process of these systems is based on the bottom-up approach which focuses on the design of the entities individual behavior. The main question arising when developing SO-MAS is how to insure that the designed entities, when interacting together, will give rise to the desired behavior? Our proposition to deal with this question is to use formal n nmethods. We propose a correct by construction method for systematic design of SO-MAS based on the use of design patterns and formal stepwise refinements. Our work gives guidelines to assist the designer when developing the individual behavior of the entities and prove its correctness at the early stages of the design process. The method is illustrated with the foraging antsâx80x99 case study.

A Stepwise Refinement Based Development of Self-Organizing Multi-Agent Systems: Application to the Foraging Ants

This paper proposes a formal modeling for Self-Organizing Multi-Agent Systems (SOMAS) based on stepwise refinements, with the Event-B language and the Temporal Logic of Actions (TLA). This modeling allows to develop this kind of systems in a more structured manner. In addition, it enables to reason, in a rigorous way, about the correctness of the derived models both at the individual level and the global level. Our work is illustrated by the foraging ants case study.

A stepwise refinement-based development of self-organising multi-agent systems: application to the foraging ants

This paper proposes a formal modelling for self-organising multi-agent systems SOMAS based on stepwise refinements, with the Event-B language and the temporal logic of actions TLA. This modelling allows to develop this kind of systems in a more structured manner. In addition, it enables to reason, in a rigorous way, about the correctness of the derived models both at the local level and the global level. Our work is illustrated by the foraging ants case study.

Simulation Based Design for Adaptive Multi-Agent Systems with the ADELFE Methodology

Complex systems are qualified by the heterogeneity of the involved actors, the large mass of data, the distribution of the manipulated information and the dynamic of the environment in which they are immersed. Designing such systems necessitate theory and methodology. This paper is situated in the AMAS theory and the ADELFE methodology. The AMAS theory Adaptive Multi-Agent System was proposed to model complex systems. The ADELFE methodology was proposed to guide the AMAS designer. Building AMAS systems is not an easy task. The objective of the authors work is to make the task of the AMAS designer less difficult. This paper details two main contributions that the authors proposed in this field. The first contribution consists in improving the ADELFE methodology by simulation techniques. The second contribution consists in enhancing the AMAS agent behavior by giving it the ability to self-design.

Forward Self-combined Method Fragments

Developing complex systems is generally simplified if designer is guided by method from Software Engineering. However a single engineering process is often not enough to cover all the possible requirements due to different levels of expertise and systems to design. Currently, Agent Oriented Software Engineering methods aim at providing an adaptive engineering process. The method processes have been broken up into different parts called fragments, enabling the mix of different engineering processes parts to get better adequacy between the system to be done and the process. But some difficulties still remain concerning the expertise needed to compose these fragments when the amount of fragments prevents the composition to be done by hand. This paper presents an Adaptive Multi-Agent Systems (AMAS) to deal with a new paradigm of automated fragments combining. This process is made from both the characteristics of users and system and the known fragments. Thanks to their information, agents of the AMAS self-organise and design a tailored method process. The developed system is described and then usual tests are depicted.