Observation-based model for BDI-agents

2008

There is an increasing number of agent-oriented programming languages that have working interpreters and platforms, with significant progress in the quality of such platforms over the last few years. With these platforms becoming more popular, and multi-agent systems being increasingly used for safety-critical applications, the need for verification techniques that apply to systems written in such languages is proportionally intensified. Building on our previous work on model checking for a particular agent-oriented programming language, we have developed a new approach whereby model checking techniques can be used directly on a variety of such languages. The approach also supports the verification of multi-agent systems where individual agents have been programmed in different agent languages.

2008

There is an increasing number of agent-oriented programming lan- guages that have working interpreters and platforms, with signif- icant progress in the quality of such platforms over the last few years. With these platforms becoming more popular, and multi- agent systems being increasingly used for safety-critical applica- tions, the need for verification techniques that apply to systems written in such languages is proportionally intensified. Building on our previous work on model checking for a particular agent- oriented programming language, we have developed a new ap- proach whereby model checking techniques can be used directly on a variety of such languages. The approach also supports the ver- ification of multi-agent systems where individual agents have been programmed in different agent languages.

2003

This paper introduces AgentSpeak(F), a variation of the BDI logic programming language AgentSpeak(L) intended to permit the model-theoretic verification of multi-agent systems. After briefly introducing AgentSpeak(F) and discussing its relationship to AgentSpeak(L), we show how AgentSpeak(F) programs can be transformed into Promela, the model specification language for the Spin model-checking system. We also describe how specifications written in a simplified form of BDI logic can be transformed into Spin-format linear temporal logic formulae. With our approach, it is thus possible to automatically verify whether or not multi-agent systems implemented in AgentSpeak(F) satisfy specifications expressed as BDI logic formulae. We illustrate our approach with a short case study, in which we show how BDI properties of a simulated auction system implemented in AgentSpeak(F) were verified.

2005

In this paper, we present PRACTIONIST (PRACTIcal reasONIng sySTem), a new framework built on the Bratman's theory of practical reasoning to support the development of BDI agents in Java (using JADE) with a Prolog belief base. We aims at reducing the gap between the expressive power of the BDI model and the difficulty of efficiently implementing its features.

2002

A number of agent-oriented programming systems are based on a framework of beliefs, desires and intentions (BDI) and more explicitly on the BDI logic of Rao and Georgeff. In this logic, goals are a consistent set of desires, and this property is fundamental to the semantics of the logic. However, implementations based on this framework typically have no explicit representation of either desires or goals, and consequently no mechanisms for checking consistency. In this paper we address this gap between theory and practice by giving an explicit representation for a simple class of desires. The simplicity of this class makes it both straightforward and efficient to check for consistency. We provide a general framework for conflict resolution based on a preference ordering of sets of goals, and we illustrate how different rules for specifying consistent goal sets (corresponding to different preference orderings) relate to existing commitment strategies. We also report on some implementation experiments which confirm that the cost of consistency maintenance is not significant.

Intelligent software agents are becoming very popular. They are ideal for solving distributed problems that are too difficult for a non-distributed system to solve. Distributed agents can be used to retrieve, filter, and summarize information as well as provide intelligent user interfaces. However, multiagent systems are very complicated to build and must be dependable. Agent conversation protocols, a series of messages passed between agents, are the cornerstone of multiagent systems. Agents also have tasks associated with them that specify how an agent behaves. This paper introduces a formal methodology that automatically verifies the interaction between agents. Agent behavioral specifications are created graphically in the agentTool multiagent development environment. This graphical representation is then transformed into a formal modeling language called Promela that is analyzed by Spin to ensure the interaction between agents is correct. This type of verification provides the us...

Proceedings of the fifth international joint conference on Autonomous agents and multiagent systems - AAMAS '06, 2006

We present a new model of knowledge, belief, desire and intention, called the interpreted KBDI-system model (or KBDI-model for short). The key point of the interpreted KBDI-system model is that we express an agent's knowledge, belief, desire and intention as a set of runs (computing paths), which is exactly a system in the interpreted system model, a well-known agent model due to Halpern and his colleagues. Our KBDI-model is computationally grounded in that we are able to associate a KBDI-model with a computer program, and formulas, involving agents' knowledge, belief, desire (goal) and intention, can be understood as properties of program computations. We present a sound and complete proof system with respect to our KBDI-model and explore how symbolic model checking techniques can be applied to model checking multi-agent systems with KBDI-models.

Journal of Logic and Computation, 1998

The study of computational agents capable of rational behaviour has received increasing attention in recent years. A number of theoretical formalizations for such multi-agent systems have been proposed. However, most of these formalizations do not have a strong semantic basis nor a sound and complete axiomatization. Hence, it has not been clear as to how these formalizations could assist in building agents in practice. This paper explores a particular type of multi-agent system, in which each agent is viewed as having the three mental attitudes of belief (B), desire (D), and intention (I). It provides a family of multi-modal branching-time BDI logics with a possible-worlds semantics, categorizes them, provides sound and complete axiomatizations, and gives constructive tableaubased decision procedures for testing the satisfiability and validity of formulas. The computational complexity of these decision procedures is no greater than the complexity of their underlying temporal logic component.

Autonomous Agents and Multi-Agent Systems, 2006

This paper gives an overview of our recent work on an approach to verifying multi-agent programs. We automatically translate multi-agent systems programmed in the logic-based agent-oriented programming language AgentSpeak into either Promela or Java, and then use the associated Spin and JPF model checkers to verify the resulting systems. We also describe the simplified BDI logical language that is used to write the properties we want the systems to satisfy. The approach is illustrated by means of a simple case study.

2005

We present an algorithm for model checking temporal-epistemic properties of multi-agent systems, expressed in the formalism of interpreted systems. We first introduce a technique for the translation of interpreted systems into boolean formulae, and then present a model-checking algorithm based on this translation. The algorithm is based on obdd's, as they offer a compact and efficient representation for boolean formulae.