A networked transferable belief model approach for distributed data aggregation

This paper focuses on the extension of the transferable belief model (TBM) to a multiagent-distributed context where no central aggregation unit is available and the information can be exchanged only locally among agents. In this framework, agents are assumed to be independent reliable sources which collect data and collaborate to reach a common knowledge about an event of interest. Two different scenarios are considered: In the first one, agents are supposed to provide observations which do not change over time (static scenario), while in the second one agents are assumed to dynamically gather data over time (dynamic scenario). A protocol for distributed data aggregation, which is proved to converge to the basic belief assignment given by an equivalent centralized aggregation schema based on the TBM, is provided. Since multiagent systems represent an ideal abstraction of actual networks of mobile robots or sensor nodes, which are envisioned to perform the most various kind of tasks, we believe that the proposed protocol paves the way to the application of the TBM in important engineering fields such as multirobot systems or sensor networks, where the distributed collaboration among players is a critical and yet crucial aspect.     

Autonomic configuration and recovery in a mobile agent‐based distributed event monitoring system

In this paper we present a framework for building policy‐based autonomic distributed agent systems. The autonomic mechanisms of configuration and recovery are supported through a distributed event processing model and a set of policy enforcement mechanisms embedded in an agent framework. Policies are event‐driven rules derived from the system's functional and non‐functional requirements. Agents in the network monitor the system state for policy violation conditions, generate appropriate events, and communicate them to other agents for cooperative filtering, aggregation, and handling. A set of agents perform policy enforcement actions whenever events signifying any policy violation conditions occur. Policies are defined using a specification framework based on XML. The policy enforcement agents interpret the policies given in XML. We illustrate the utility of this framework in the context of an agent‐based distributed network monitoring application. We also present an experimental evaluation of our approach. Copyright © 2006 John Wiley & Sons, Ltd.     

Rule-based inference and decomposition for distributed in-network processing in wireless sensor networks

Wireless sensor networks are application specific and necessitate the development of specific network and information processing architectures that can meet the requirements of the applications involved. A common type of application for wireless sensor networks is the event-driven reactive application, which requires reactive actions to be taken in response to events. In such applications, the interest is in the higher-level information described by complex event patterns, not in the raw sensory data of individual nodes. Although the central processing of information produces the most accurate results, it is not an energy-efficient method because it requires a continuous flow of raw sensor readings over the network. As communication operations are the most expensive in terms of energy usage, the distributed processing of information is indispensable for viable deployments of applications in wireless sensor networks. This method not only helps in reducing the total amount of packets transmitted in the network and the total energy consumed by the sensor nodes, but also produces scalable and fault-tolerant networks. For this purpose, we present two schemes that distribute information processing to appropriate nodes in the network. These schemes use reactive rules, which express relations between event patterns and actions, in order to capture reactive behavior. We also share the results of the performance of our algorithms and the simulations based on our approach that show the success of our methods in decreasing network traffic while still realizing the desired functionality.     

Reasoning about Commitments in the Event Calculus: An Approach for Specifying and Executing Protocols

Commitments among agents are widely recognized as an important basis for organizing interactions in multiagent systems. We develop an approach for formally representing and reasoning about commitments in the event calculus. We apply and evaluate this approach in the context of protocols, which represent the interactions allowed among communicating agents. Protocols are essential in applications such as electronic commerce where it is necessary to constrain the behaviors of autonomous agents. Traditional approaches, which model protocols merely in terms of action sequences, limit the flexibility of the agents in executing the protocols. By contrast, by formally representing commitments, we can specify the content of the protocols through the agents' commitments to one another. In representing commitments in the event calculus, we formalize commitment operations and domain-independent reasoning rules as axioms to capture the evolution of commitments. We also provide a means to specify protocol-specific axioms through the agents' actions. These axioms enable agents to reason about their actions explicitly to flexibly accommodate the exceptions and opportunities that may arise at run time. This reasoning is implemented using an event calculus planner that helps determine flexible execution paths that respect the given protocol specifications.     

Distributed multi-task classification: a decentralized online learning approach

Although dispersing one single task to distributed learning nodes has been intensively studied by the previous research, multi-task learning on distributed networks is still an area that has not been fully exploited, especially under decentralized settings. The challenge lies in the fact that different tasks may have different optimal learning weights while communication through the distributed network forces all tasks to converge to an unique classifier. In this paper, we present a novel algorithm to overcome this challenge and enable learning multiple tasks simultaneously on a decentralized distributed network. Specifically, the learning framework can be separated into two phases: (i) multi-task information is shared within each node on the first phase; (ii) communication between nodes then leads the whole network to converge to a common minimizer. Theoretical analysis indicates that our algorithm achieves a \(\mathcal {O}(\sqrt{T})\) regret bound when compared with the best classifier in hindsight, which is further validated by experiments on both synthetic and real-world datasets.     

External and internal choice with event groups in Event-B

Abrial’s Event-B formalism for refinement-based system development is influenced by Back’s action system approach. Morgan has defined a CSP-like failures-divergence semantics for action systems that distinguishes internal and external choice of actions. Morgan’s semantics has the characteristic that the choice between enabled actions is external while internal choice is represented less directly through nondeterministic effect of actions. Practical experience with Event-B has demonstrated the need to be able to represent both internal and external choice between enabled events more explicitly. In this paper, Morgan’s failures semantics for action systems is modified to allow both internal and external choice to be represented directly. This is achieved by grouping events so that external choice is between event groups and internal choice is within event groups. This leads to a refinement rule for preservation of choice between event groups while allowing for reduction of choice within event groups. We also provide a refinement rule for splitting event groups in order to increase external choice. The refinement rules are justified in terms of failures refinement.     

Asymptotic behaviours of a class of threshold models for collective action in social networks

A class of dynamic threshold models is proposed for describing the upset of collective actions in social networks. The agents of the network have to decide whether to undertake certain action or not. They make their decision by comparing the activity level of their neighbours with a time-varying threshold, evolving according to a time-invariant opinion dynamic model. Key features of the model are a parameter representing the degree of self-confidence of the agents and the mechanism adopted by the agents to evaluate the activity level of their neighbours. The case in which a radical agent, initially eager to undertake the action, interacts with a group of ordinary agents, is considered. The main contribution of the paper is the complete characterisation of the asymptotic behaviours of the network, for three different graph topologies. The asymptotic activity patterns are determined as a function of the self-confidence parameter and of the initial threshold of the ordinary agents. Numerical validation on a real ego network shows that the theoretical results obtained for simple graph structures provide useful insights on the network behaviour in more complex settings.     

Event‐based asynchronous communication and sampled control for synchronization of multiagent networks with input saturation

This paper investigates the problem of event‐based synchronization of linear dynamical networks subject to input saturation. The asynchronous neighboring information transmission is triggered by distributed events. The sampled control technique is utilized to exclude both the internal Zeno behavior of each agent and the network Zeno behavior attributed to neighboring agents. Allowing the input saturation to be attained, an event‐based global synchronization algorithm is proposed for multiagent networks with neutrally stable dynamics. For general linear networks, an event‐triggered control protocol is designed using the modified algebraic Riccati equation, with a low‐gain cooperative control law proposed to achieve semiglobal synchronization. A numerical example is presented to illustrate the theoretical results.     

Distributed cooperative stabilisation of continuous-time uncertain nonlinear multi-agent systems

This paper addresses the distributed cooperative stabilisation problem of continuous-time uncertain nonlinear multi-agent systems. By approximating the uncertain dynamics using neural networks, a distributed adaptive cooperative controller, based on the state information of the neighbouring agents, is proposed. The control design is developed for any undirected connected communication topologies without requiring the accurate model of each agent. This result is further extended to the output feedback case. An observer-based distributed cooperative controller is devised and a parameter dependent Riccati inequality is employed to prove stability of the overall multi-agent systems. This design is less complex than the other design methods and has a favourable decouple property between the observer design and the controller design for uncertain nonlinear multi-agent systems. For both cases, the developed controllers guarantee that all signals in the closed-loop network are uniformly ultimately bounded, and the states of all agents cooperatively converge to a small neighbourhood of origin. A comparative study is given to show the efficacy of the proposed method.     

Event-based broadcasting containment control for multi-agent systems under directed topology

The event-based broadcasting containment control problem for both first-order and second-order multi-agent systems under directed topology is investigated. Based on certain event, each agent decides when to transmit its current states to its neighbours and the agents’ distributed control algorithms are based on these sampled state measurements, which can significantly decrease the number of the controllers’ updates. All the agents are divided into two groups, namely, the leaders and the followers. The formation control is introduced. The leaders exchange their information to converge to a formation. The followers utilise the information from both their leader neighbours and their follower neighbours and are driven to the convex hull of the leaders using the proposed control algorithms. Numerical simulations are provided to illustrate the effectiveness of the obtained theoretical results.     

Decision support or support for situated choice: lessons for system design from effective manual systems

In a number of organisational settings where work is highly skilled but substantially routine, certain entrenched manual systems have resisted digitisation. These systems include card-based systems in emergency despatch, the paper flight progress strips system used in air traffic control, the Kanban system and whiteboard systems used in hospital wards. Research to understand or replace these systems has frequently regarded them as decision support systems (DSS). We report here a detailed case study of a manual whiteboard-based bed allocation system in the ICU of a large general hospital, which shows that the support it provides for users’ action choices cannot be validly conceived as decision support. This system and other effective manual systems may be better understood as a ‘situated choice support system’ (SCSS). Whereas DSS provide actors with a model of the action environment in order to support reasoning about the consequences of alternative actions, SCSS provide actors with structured work environments that reduce possible actions and cue-providing information resources to support a reactive choice between these limited alternatives. The findings warn of the danger of uncritically applying the DSS design paradigm to supporting action choice in skilled routine work, and provide an alternative design theory, which can potentially inform new ICT-based designs.     

FuzzyShell: a large-scale expert system shell using fuzzy logic foruncertainty reasoning

There exist in the literature today many contributions dealing with the incorporation of fuzzy logic in expert systems. However, unfortunately, much of what has been proposed can only be applied to small-scale expert systems; that is, when the number of rules is in the dozens as opposed to in the hundreds. The more traditional (nonfuzzy) expert systems are able to cope with large numbers of rules by using Rete networks for maintaining matches of all the rules and all the facts. (A Rete network obviates the need to match the rules with the facts on every cycle of the inference engine.) In this paper, we present a more general Rete network that is particularly suitable for reasoning with fuzzy logic. The generalized Rete network consists of a cascade of three networks: the pattern network, the join network, and the evidence aggregation network. The first two layers are modified versions of similar layers for the traditional Rete networks and the last, the aggregation layer, is a new concept that allows fuzzy evidence to be aggregated when fuzzy inferences are made about the same fuzzy variable by different rules     

Analyzing users’ activity in online social networks over time through a multi-agent framework

The number of people and organizations using online social networks as a new way of communication is continually increasing. Messages that users write in networks and their interactions with other users leave a digital trace that is recorded. In order to understand what is going on in these virtual environments, it is necessary systems that collect, process, and analyze the information generated. The majority of existing tools analyze information related to an online event once it has finished or in a specific point of time (i.e., without considering an in-depth analysis of the evolution of users’ activity during the event). They focus on an analysis based on statistics about the quantity of information generated in an event. In this article, we present a multi-agent system that automates the process of gathering data from users’ activity in social networks and performs an in-depth analysis of the evolution of social behavior at different levels of granularity in online events based on network theory metrics. We evaluated its functionality analyzing users’ activity in events on Twitter.     

An abstract knowledge based approach to diagnosis and recovery of plan failure in multi-agent systems

Open multi-agent systems (MAS) are decentralised and distributed systems that consist of a large number of loosely coupled autonomous agents. In the absence of centralised control they tend to be difficult to manage, especially in an open environment, which is dynamic, complex, distributed and unpredictable. This dynamism and uncertainty in an open environment gives rise to unexpected plan failures. In this paper we present an abstract knowledge based approach for the diagnosis and recovery of plan action failures. Our approach associates a sentinel agent with each problem solving agent in order to monitor the problem solving agent’s interactions. The proposed approach also requires the problem solving agents to be able to report on the status of a plan’s actions.Once an exception is detected the sentinel agents start an investigation of the suspected agents. The sentinel agents collect information about the status of failed plan abstract actions and knowledge about agents’ mental attitudes regarding any failed plan. The sentinel agent then uses this abstract knowledge and the agents’ mental attitudes, to diagnose the underlying cause of the plan failure. The sentinel agent may ask the problem solving agent to retry their failed plan based on the diagnostic result.     

Sampled-data based distributed convex optimization with event-triggered communication

This paper studies the distributed convex optimization problem for multi-agent systems over undirected and connected networks. Motivated by practical considerations, we propose a new distributed optimization algorithm with event-triggered communication. The proposed event detection is decentralized, sampled-data and not requires periodic communications among agents to calculate the threshold. Based on Lyapunov approaches, we show that the proposed algorithm is asymptotically converge to the unknown optimizer if the design parameters are chosen properly. We also give an upper bound on the convergence rate. Finally, we illustrate the effectiveness of the proposed algorithm by a numerical simulation.     

Topological analysis of a two coupled evolving networks model for business systems

In multi-agent systems, the underlying networks are always dynamic and network topologies are always changing over time. Performance analyses of topologies are important for understanding the robustness of the system and also the effects of topology on the system efficiency and effectiveness. In this paper, we present an example of a real-world distributed agent system, a digital business ecosystem (DBE). It is modelled as a two coupled network system. The upper layer is the business network layer where business process between different business agents happen. The lower layer is the underlying P2P communication layer to support communications between the agents. Algorithms for multi-agent tasks negotiation and execution, interaction between agents and the underlying communication network, evolutionary network topology dynamics, are provided. These algorithms consider the two network layers evolving over time, with effects on each other. Through a comprehensive set of discrete event simulation, we investigate the effects of different evolutionary principles inspired by random graph and scale-free network in complex network theory on the topological properties and performance of the underlying network. We also find several rules to design a resilient and efficient P2P network.     

Design and evaluation of a fault-tolerant mobile-agent system

The mobile agents create a new paradigm for data exchange and resource sharing in rapidly growing and continually changing computer networks. In a distributed system, failures can occur in any software or hardware component. A mobile agent can get lost when its hosting server crashes during execution, or it can get dropped in a congested network. Therefore, survivability and fault tolerance are vital issues for deploying mobile-agent systems. This fault tolerance approach deploys three kinds of cooperating agents to detect server and agent failures and recover services in mobile-agent systems. An actual agent is a common mobile agent that performs specific computations for its owner. Witness agents monitor the actual agent and detect whether it's lost. A probe recovers the failed actual agent and the witness agents. A peer-to-peer message-passing mechanism stands between each actual agent and its witness agents to perform failure detection and recovery through time-bounded information exchange; a log records the actual agent's actions. When failures occur, the system performs rollback recovery to abort uncommitted actions. Moreover, our method uses checkpointed data to recover the lost actual agent.     

产品造型设计多阶段网络耦合决策方法研究

为综合考虑产品造型设计多阶段决策信息,引入复杂网络理论,提出了产品造型设计多阶段网络耦合决策流程.对产品造型设计多阶段决策关系进行了耦合关联建模,基于对决策者评分的一致性分析确定设计决策网络的邻接矩阵,依据网络凝聚度变化获得决策者权重,以设计决策网络凝聚度的稳定性和决策者意见一致性确定决策轮次,构建规划函数计算产品造型设计决策各阶段权重,通过线性耦合实现多阶段决策信息集结.以数控磨床造型设计决策为例验证了方法的有效性,结果表明所提方法能够实现产品造型设计多阶段决策信息的集结,提高设计决策的全局性和科学性.     

How users’ participation affects reputation management systems: The case of P2P networks

In this work human factor is explored by means of agent based simulation and analyzed in the framework of a reputation management system (RMS), within a peer-to-peer (P2P) network. Reputation is about evaluating an agent’s actions and other agents’ opinions about those actions, reporting on those actions and opinions, and reacting to that report, thus creating a feedback loop. This social mechanism has been successfully used to classify agents within normative systems. The systems rely on the feedbacks given by the members of the social network in which the RMS operates. Reputation can thus be seen as an endogenous and self produced indicator, created by the users for the users’ benefit. This implies that users’ participation and collaboration is a key factor for the effectiveness a RMS.