René Mandiau

Trends in distributed artificial intelligence

Distributed artificial intelligence (DAI) is a subfield of artificial intelligence that deals with interactions of intelligent agents. Precisely, DAI attempts to construct intelligent agents that make decisions that allow them to achieve their goals in a world populated by other intelligent agents with their own goals. This paper discusses major concepts used in DAI today. To do this, a taxonomy of DAI is presented, based on the social abilities of an individual agent, the organization of agents, and the dynamics of this organization through time. Social abilities are characterized by the reasoning about other agents and the assessment of a distributed situation. Organization depends on the degree of cooperation and on the paradigm of communication. Finally, the dynamics of organization is characterized by the global coherence of the group and the coordination between agents. A reasonably representative review of recent work done in DAI field is also supplied in order to provide a better appreciation of this vibrant AI field. The paper concludes with important issues in which further research in DAI is needed.

A behavioral multi-agent model for road traffic simulation

Multi-agent systems allow the simulation of complex phenomena that cannot easily be described analytically. Multi-agent approaches are often based on coordinating agents whose actions and interactions are related to the emergence of the phenomenon to be simulated. In this article, we focus on road traffic simulation, specifically the design of a road traffic simulation tool able to deal realistically with road junctions. We propose a multi-agent behavioral model based on (i) the opportunistic individual behaviors that describe the norm violation and (ii) the anticipatory individual abilities of simulated drivers that allow critical situations to be detected. Our proposition has been validated for different traffic scenarios. Specifically, we simulated the traffic in a real intersection and then compared the simulated traffic flow with the real flow to highlight the relevance of our approach.

Behaviour based on decision matrices for a coordination between agents in a urban traffic simulation

Abstract This paper describes a multi-agent coordination mechanism applied to intersection simulation situations. In a goal of urban traffic simulation, we must consider the dynamic interactions between autonomous vehicles. The field of multi-agent systems provides us some studies for such systems, in particular on the coordination mechanisms. Conflicts between vehicles (i.e. agents) are very frequent in such applications, and they may cause deadlocks, particularly at intersections such as crossroads. Our approach is based on the solving of two player games/decision matrices which characterize three basic situations. An aggregation method generalizes to n-player games for complex crossroads. The objective of this approach consists in searching basic two-player matrices for solving n-agent problems. To explain the principle, we describe our approach for a particular case of crossroad with three agents. Finally, the obtained results have been examined via a tool of road traffic simulation, ARCHISIM. We assume also that the global traffic replicates the behavior of agents in different situations.

Anticipation based on constraint processing in a multi-agent context

Anticipation is a general concept used and applied in various domains. Many studies in the field of artificial intelligence have investigated the capacity for anticipation. In this article, we focus on the use of anticipation in multi-agent coordination, particularly preventive anticipation which consists of anticipating undesirable future situations in order to avoid them. We propose to use constraint processing to formalize preventive anticipation in the context of multi-agent coordination. The resulting algorithm allows any action that may induce an undesirable future state to be detected upstream of any multi-agent coordination process. Our proposed method is instantiated in a road traffic simulation tool. For the specific question of simulating traffic at road junctions, our results show that taking anticipation into account allows globally realistic behaviors to be reproduced without provoking gridlock between the simulated vehicles.

Simulation in Contexts Involving an Interactive Table and Tangible Objects

By using an interactive table, it is possible to interact with several people (decision-makers) in a simultaneous and collaborative way, around the table, during a simulation session. Thanks to the RFID technology with which the table is fitted, it is possible to give tangible objects a unique identity to include and to consider them in the simulation. The paper describes a context model, which takes into consideration the specificities related to interactive tables. The TangiSense interactive table is presented; it is connected to a multi-agent system making it possible to give the table a certain level of adaptation: each tangible object can be associated to an agent which can bring roles to the object (i.e., the roles are the equivalent of a set of behaviors). The multi-agent system proposed in this paper is modeled according to an architecture adapted to the exploitation of tangible and virtual objects during simulation on an interactive table. A case study is presented; it concerns a simulation of road traffic management. The illustrations give an outline of the potentialities of the simulation system as regards the context-awareness aspect, following both the actions of the decision-makers implied in simulation, and the agents composing the road traffic simulation.

Non-normative Behaviour in Multi-agent System: Some Experiments in Traffic Simulation

Most of the works related to norms and multi-agent systems focus on the design of normative agents systems making the assumption that agents always respect norms. Our aims in this article are (i) to discuss the relevance of this assumption in some specific contexts and to highlight some benefits of designing non-normative behaviour agents, (ii) to expound the methodology followed in a concrete application which consists in traffic simulation at junction. In particular, based on statistical traffic results, we show how non-normative behaviours contribute to improving the realism of simulation.

A Multi-Agent System Approach for Interactive Table Using RFID

This paper presents a model of Multi-Agent System dedicated to the management of a tabletop that detects traceable objects using RFID technology, which are moved on the table by a set of users during application. In addition, virtual objects can interact with tangible objects. We propose a new model (MAM4IT) to manage this kind of object with situated agents (that we call tangible agents) based on the notion of dynamic roles. The roles will be able to evolve during the application and adapt their behaviors according to the environment and the unpredictable actions of users. In this situation the agents are able to compose new roles. A first concrete case study based on road traffic shows the interaction between tangible and virtual agents.

HOMASCOW: a holonic multi-agent system for cooperative work

Assistance tools for actors of cooperative systems, which are often centralized are increasingly designed to adhere to the system organization using a distributed architecture. However, generally these systems have difficulty in managing data coherency at the global level. In answer to this problem, we turned towards a particular organizational model: holonic systems. This concept has guided us in the design of the Holonic Multi-Agent System (HOLOMAS), which offers a good compromise between the distribution of knowledge and control centralization. In fact, the use of the HOLOMAS appeared to be necessary following the study of the procedures used in the patent department of a large company.

Flexible Roles in a Holonic Multi-Agent System

Holonic Multi-Agent organisations are particular pyramidal organisations where agents of a layer (having the same coordinator) are able to communicate and to negotiate directly between them. In order to benefit from control allowed by the hierarchical structure of a holonic organisation and from dynamic roles, we propose a holonic multi-agent organisation inspired from working mechanisms of the complex human organisations. We have implemented this architecture by offering a role management capacity to a hierarchical multi-agent platform.

Roles and hierarchy in multi-agent organizations

Holonic Multi-Agent organisations are particular pyramidal organisations where agents of a layer (having the same coordinator) are able to communicate and to negotiate directly between them. Holonic agents are generally structured by services (for examples: to search information, to interact with the user). Some holonic agents, distributed in the system, can have same roles without being in a same layer. However, it is essential to interact with agents according to theirs roles: most of multi-agent methods and platforms are based on the roles management. In order to beneficiate of the two advantages which are the control allowed by the holonic architecture and the roles management, we propose a particular holonic architecture based on the workings mechanisms of human organisations. We have implemented this architecture by offering a role management capacity to a hierarchical multi-agent platform.