Jean-Luc Koning

Interaction protocol engineering

Several methodologies are supplied to multiagent system designers to help them defining their agents and their multiagent systems. These methodologies focus mainly on agents and on multiagent systems and barely consider how to design interaction protocols. A problem could emerge of this lack since interaction protocols are more and more complex. The aim of this article is to present our proposal of interaction protocol engineering which is based on the communication protocol engineering [10]. Interaction protocol engineering allows designers to define protocols from scratch. Our proposal is composed of five stages: analysis, formal description, validation, protocol synthesis and conformance testing.

Extended Modeling Languages for Interaction Protocol Design

Successful development of agent interaction protocols requires modeling methods and tools that support a relatively complete development lifecycle. Agent-based systems are inherently complex but exhibit many similarities to object-oriented systems. For these reasons not only current modeling languages need to be extended, but also related tools should be provided for agent interaction protocol design to be supported. In this paper, we focus on the design stage of an agent interaction protocol development cycle. We start by giving general criteria for comparing agent modeling languages. The ones we take into consideration in this paper are extensions of Agent-UML and FIPA-UAML languages. We describe these languages and discuss some extensions on a simplified application of the Netbill electronic commerce protocol. We then briefly introduce a component-based formal specification language in order to support the protocols design stage and present a tool built upon the FIPA norm (making use of the PDN or UAML notation) which supports the analysis and design of interaction protocols.

Designing and Testing Negotiation Protocols for Electronic Commerce Applications

The most widespread and studied approach for managing negotiation between agents in a setting like electronic commerce relies on negotiation protocols. They are a set of guidelines that the agents of the system have to follow and that structure message passing. This enables agents to communicate knowing the types of messages they can receive, the types of messages they can send and at what time. Given their knowledge bases, what they perceive from the outside world, and the possible messages they receive from the other agents, they communicate among them through message passing according to a pre-established sequencing that has been possibly (and hopefully) validated. In this article, negotiation is exclusively tackled from the angle of protocols as they can be found in electronic commerce applications.

Requirement analysis for interaction protocols

Designing agent interaction protocols need first to consider what the requirements are. This is done in the requirement analysis phase. The output of this phase is an informal document written in natural language. To our best knowledge, this phase is barely considered in the literature neither in communication protocol engineering nor in interaction protocol engineering. As a consequence, it is difficult for designers to do it easily. Experience seems to be the key. In order to help designers, we propose to structure the requirement analysis document into fields. These fields gather protocols features. This paper presents such document and applies the structuration to the electronic commerce protocol NetBill.

INTRODUCTION TO POS: A PROTOCOL OPERATIONAL SEMANTICS

In this paper, we propose a system for representing interaction protocols called POS which is both Turing complete and determine a complete semantics of protocols. This work is inspired by the Structured Operational Semantics in programming languages. We precisely define POS and illustrate its power on an extended example.

State Controlled Execution for Agent-Object Hybrid Languages

This paper explores some ideas about the implementation of state controlled execution for a language centered on agent interaction protocols. Such an approach could be useful because there an increasingly strong interest on the notion of role in the multiagent system domain. The strong recurrence role and the almos spontaneous appearance of interactio protocols in most practical applications of MAS, and the strong similarities this situation has with the pre-existing notion of class on the object oriented sense, suggest the high relevance of roles and the need of having adequate notational and programming artifacts to represent them.

Formalization, implementation and validation of conversation policies using a protocol operational semantics

Coordination is crucial to the efficiency of multiagent systems. As it is traditionally achieved through message passing, agent communication languages (ACL) have become a major issue in the field. Yet, existing standard technologies, like KQML, are too rigid, and make too strong assumptions when defining their semantics. More importantly, they only allow the specification of individual messages, but hardly permit to deal with sequences of message exchanges-in other words conversations-which is unavoidable if one wants complex and yet efficient interactions to take place. Conversation policies should thus be heavily integrated with ACLs. We present a framework called POS (Protocol Operational Semantics), which is a unification and generalization of existing work on interaction protocols. It does allow the design (syntax and semantics) of dedicated ACLs together with policies or constraints over the use of messages. It is both powerful and particularly easy to implement. It is general enough to allow different types and granularities of constraints. Its similarity with traditional programming languages semantic specifications provides a basis for powerful validation of syntactic and semantic properties, some of which we present in this paper. As an illustration, we give an extended example of the formalization, implementation and validation of a non-trivial protocol, together with its dedicated ACL. Strangely enough, even though conversation policies is a hot topic in multiagent systems (numerous workshops are dedicated to this theme), very few papers deal with protocol validation. This paper makes an inventory of both syntactic and semantic properties conversation policies may satisfy and gives their possible definition. In order to highlight some of those properties we provide an application example and discuss related simulations.

Modeling Soccer-Robots Strategies through Conversation Policies

Straightforward approaches to team coordination with the expressive power of finite state automata are doomed to fail under a wide range of heterogeneity due to the combinatorial explosion of states. In this paper we propose a coordination scheme based on operational semantics, which allows an extremely compact and modular way of specifying soccer-robot team behaviors. The capabilities of our approach are demonstrated on two examples, which, though just being simple demo implementations, perform very well in a simulator tournament.

Interaction protocol design: application to an agent-based teleteaching project

In this paper we focus on the analysis and design stages of the protocol engineering development cycle. We start by sketching an application framework dedicated to a Web-based learning environment called Baghera whose aim is to teach geometry problems. We then apply our protocol engineering process to protocols for checking mathematical proofs a student happens to build. The following section discusses the analysis stage of such a protocol. We then briefly introduce our component-based formal specification language in order to then describe the protocols design stage. Finally we present a tool built upon the FIPA norm (making use of the PDN or UAML notation) which supports the analysis and design of interaction protocols.

Engineering Interaction Protocols for Multiagent Systems

Multiagent systems are becoming increasingly popular as a new programming paradigm that provides the right abstraction level and the right model to build a lot of distributed applications. Its basic components are agents which are encapsulated computer systems that are situated in some environment and that are capable of flexible, autonomous action in that environment in order to meet their design objectives [1].