Multi-agent system

About: Multi-agent system is a research topic. Over the lifetime, 27978 publications have been published within this topic receiving 465191 citations. The topic is also known as: multi-agent systems & multiagent system.

Neural Networks for Real-Time Traffic Signal Control

Abstract: Real-time traffic signal control is an integral part of the urban traffic control system, and providing effective real-time traffic signal control for a large complex traffic network is an extremely challenging distributed control problem. This paper adopts the multiagent system approach to develop distributed unsupervised traffic responsive signal control models, where each agent in the system is a local traffic signal controller for one intersection in the traffic network. The first multiagent system is developed using hybrid computational intelligent techniques. Each agent employs a multistage online learning process to update and adapt its knowledge base and decision-making mechanism. The second multiagent system is developed by integrating the simultaneous perturbation stochastic approximation theorem in fuzzy neural networks (NN). The problem of real-time traffic signal control is especially challenging if the agents are used for an infinite horizon problem, where online learning has to take place continuously once the agent-based traffic signal controllers are implemented into the traffic network. A comprehensive simulation model of a section of the Central Business District of Singapore has been developed using PARAMICS microscopic simulation program. Simulation results show that the hybrid multiagent system provides significant improvement in traffic conditions when evaluated against an existing traffic signal control algorithm as well as the SPSA-NN-based multiagent system as the complexity of the simulation scenario increases. Using the hybrid NN-based multiagent system, the mean delay of each vehicle was reduced by 78% and the mean stoppage time, by 85% compared to the existing traffic signal control algorithm. The promising results demonstrate the efficacy of the hybrid NN-based multiagent system in solving large-scale traffic signal control problems in a distributed manner

JACK Intelligent Agents – Summary of an Agent Infrastructure

Abstract: Intelligent Agents are being used for modelling simple rational behaviours in a wide range of distributed applications. In particular, multi-agent architectures based on the Belief-DesireIntention model have been used successfully in situations where modelling of human reasoning and team behaviour are needed, such as simulating tactical decision-making in defence operations and command and control structures. Other applications include intelligent web applications, manufacturing control, telephone call centres, and air traffic management. The JACK Intelligent Agents framework by Agent Oriented Software brings the concept of intelligent agents into the mainstream of commercial software engineering and Java. JACK Intelligent AgentsTM is a third generation agent framework, designed as a set of lightweight components with high performance and strong data typing. We present the design approach and major technical characteristics of JACK Intelligent Agents, with a focus on some of the more recent developments – modular structuring elements called Capabilities and the team-based reasoning model called SimpleTeam. Two example applications, in decision support and defence simulation are described. Also, we discuss the benefits of the component-based approach, both for the software engineer developing sophisticated distributed applications, and for the researcher exploring agent models and architectures.

Decentralized Control of Connectivity for Multi-Agent Systems

Abstract: In this paper we propose a decentralized algorithm to increase the connectivity of a multi-agent system. The connectivity property of the multi-agent system is quantified through the second smallest eigenvalue of the state dependent Laplacian of the proximity graph of agents. An exponential decay model is used to characterize the connection between agents. A supergradient algorithm is then used in conjunction with a recently developed decentralized algorithm for eigenvector computation to maximize the second smallest eigenvalue of the Laplacian of the proximity graph. A potential based control law is utilized to achieve the distances dictated by the supergradient algorithm. The algorithm is completely decentralized, where each agent receives information only from its neighbors, and uses this information to update its control law at each step of the iteration. Simulations demonstrate the effectiveness of the algorithm.

An Overview of the Multiagent Systems Engineering Methodology

Abstract: To solve complex problems, agents work cooperatively with other agents in heterogeneous environments. We are interested in coordinating the local behavior of individual agents to provide an appropriate system-level behavior. The use of intelligent agents provides an even greater amount of flexibility to the ability and configuration of the system itself. With these new intricacies, software development is becoming increasingly difficult. Therefore, it is critical that our processes for building the inherently complex distributed software that must run in this environment be adequate for the task. This paper introduces a methodology for designing these systems of interacting agents.

Developing organised multiagent systems using the MOISE + model: programming issues at the system and agent levels

Abstract: Multiagent Systems (MASs) have evolved towards the specification of global constraints that heterogeneous and autonomous agents are supposed to follow when concerning open systems A subset of these constraints is known as the MAS organisation This article describes a set of computational tools that supports the development and the programming of such systems At the system level, a middleware is provided which ensures that all agents will follow the organisational constraints At the agent level, the AgentSpeak language is extended, using Jason features, so that the agents can perceive and act upon the organisation to which they belong