Catálogo de publicaciones - libros

Roles are an important concept used for different purposes as the modeling of the organizational structure of multi-agent systems, the modeling of protocols, and as basic building blocks for defining the behavior of agentsModeling interactions by roles brings several advantages, the most important of which is the separation of concerns by distinguishingthe agent-level and system-level with regard to interaction. However, in open MASs, the composition of independently developed roles can lead to unexpected emergent interaction among agents. This paper identifies requirements for modeling role-based interactions, and presents a formal specification model of roles for complex interactions. Our approach aims to integrate specification and verification of roles into the Component Based Development approach. An interaction protocol example is given to illustrate our formal framework.

Multi-agent systems are distributed systems of loosely coupled agents. Description and construction of these systems are eased by separating their structure from their dynamic behavior. ANote is a modeling language for multi-agent system analysis that supports this approach. It provides a notation language which supports multi-agent system analysis through its decomposition into structural and behavioral views. Each view is responsible for picturing an important aspect, while ignoring less important details. This paper describes the ANote notation language and its views. The notation language is described and illustrated by an example, an e-insurance system.

If agents are to negotiate automatically with one another they must share a negotiation mechanism, specifying what possible actions each party can take at any given time, when negotiation terminates, and what is the structure of the resulting agreements. Current standardization activities such as FIPA [2] and WS-Agreement [3] represent this as a negotiation protocol specifying the flow of messages. However, they omit other aspects of the rules of negotiation (such as obliging a participant to improve on a previous offer), requiring these to be represented implicitly in an agent’s design, potentially resulting incompatibility, maintenance and re-usability problems. In this chapter, we propose an alternative approach, allowing all of a mechanism to be formal and explicit. We present (i) a taxonomy of declarative rules which can be used to capture a wide variety of negotiation mechanisms in a principled and well-structured way; (ii) a simple interaction protocol, which is able to support any mechanism which can be captured using the declarative rules; (iii) a software framework for negotiation that allows agents to effectively participate in negotiations defined using our rule taxonomy and protocol and (iv) a language for expressing aspects of the negotiation based on OWL-Lite [4]. We provide examples of some of the mechanisms that the framework can support.

In open multi-agent systems, agents are mobile and may leave or enter the system. This dynamicity results in two closely related agent communication problems, namely, efficient message passing and service agent discovery. This paper describes how these problems are addressed in the (). Agents in AA obey the operational semantics of actors, and the architecture is designed to support large-scale open multi-agent systems. Efficient message passing is facilitated by the use of dynamic names: a part of the mobile agent name is a function of the platform that currently hosts the agent. To facilitate service agent discovery, middle agents support application agent-oriented matchmaking and brokering services. The middle agents may accept search objects to enable customization of searches; this reduces communication overhead in discovering service agents when the matching criteria are complex. The use of mobile search objects creates a security threat, as codes developed by different groups may be moved to the same middle agent. This threat is mitigated by restricting which operations a migrated object is allowed to perform. We describes an empirical evaluation of these ideas using a large scale multi-agent UAV (Unmanned Aerial Vehicle) simulation that was developed using AA.

The increased pervasiveness of mobile devices like cell phones, PDAs, and laptops draws attention to the need for coordination among these networked devices. The very nature of the environment requires devices to interact opportunistically when resources are available. Such interactions occur unpredictably as device users have no advance knowledge of others they will encounter. The openness of these environments also requires users to protect themselves and their data from unwanted interactions while maintaining desired, yet unscripted, coordination. As the ubiquity of communicating mobile devices increases, the number of applications supported by the network grows drastically and managing access control is crucial to such systems. Application agents must directly manipulate and examine access policies because these networks are often decoupled from a fixed infrastructure, rendering reliance on centralized servers for authentication and access policies impractical. In this paper, we explore context-aware access control policies tailored to the needs of agent coordination in open environments that exhibit mobility. We propose and evaluate novel constructs to support such policies, especially in the presence of large numbers of highly dynamic application agents.

Multi-agent systems present a promising paradigm for coping with the complexity of intelligent mechatronic applications, particularly where purposeful behavior and complex structures emerge from the interactions of seemingly simple elements. The safety of mechatronic systems relies on predictability, which is apparently at odds with the concept of emergent behavior. When designing complex mechatronic multi-agent systems, the main challenge thus lies in achieving predictability without ruling out the desired emergent behavior. We propose to achieve this by decomposing the requirements and design into largely independent concerns, represented by social structures with behavioral norms, which are reconciled at the agent level. An explicit grounding of all constructs in observable entities from the mechatronic system’s environment model makes them amenable to formal analysis and enables rapid prototyping.