Sascha Ossowski

Objective versus subjective coordination in the engineering of agent systems

The governance of interaction is a critical issue in the engineering of agent systems. Research on coordination addresses this issue by providing a wide range of models, abstractions and technologies. It is often the case, however, that such a wide range of proposals could not easily find an unitary and coherent conceptual framework where all the different views and solutions can be understood and compared - and this is particularly true in the context of agent models and systems. n nIn this paper, we first discuss how all the many diverse approaches to agent coordination can be categorised in two main classes - the subjective and objective approaches -, depending on whether they adopt the agents or the engineers viewpoint, respectively. We then claim that the two approaches have a deep and different impact on the way in which agent systems are modelled and built, and show two examples rooted in different models and technologies. Finally, we advocate that both approaches play a fundamental role in the engineering of agent systems, and that any methodology for the design and development of agent systems has to exploit both objective and subjective coordination models and technologies.

Trust-based service provider selection in open environments

The problem of selecting correct counterparts to interact with is of particular relevance in open and dynamic environments. This problem increases when third parties may vary their behaviour at will. In this paper we examine the problem of service provider selection using trust and reputation techniques. Most approaches to service provider selection are based on the clients proper experiences about particular services from particular providers. A problem arises when no previous experience is available. To solve this problem, previous approaches have proposed that clients obtain the required reputation information from their acquaintances. In contrast, our work advocates an experience-based approach for service provider selection, in which clients use trust and reputation mechanisms to infer expectations of future providers behaviour from past experiences in similar situations. We present some experimental results that support our proposal.

On coordination and its significance to distributed and multi‐agent systems

Coordination is one of those words: it appears in most science and social fields, in politics, warfare, and it is even the subject of sports talks. While the usage of the word may convey different ideas to different people, the definition of coordination in all fields is quite similar—it relates to the control, planning, and execution of activities that are performed by distributed (perhaps independent) actors. Computer scientists involved in the field of distributed systems and agents focus on the distribution aspect of this concept. They see coordination as a separate field from all the others—a field that rather complements standard fields such as the ones mentioned above. This paper focuses on explaining the term coordination in relation to distributed and multi‐agent systems. Several approaches to coordination are described and put in perspective. The paper finishes with a look at what we are calling emergent coordination and its potential for efficiently handling coordination in open environments. Copyright

Coordination knowledge engineering

By adopting a structured knowledge-level approach, coordination knowledge can be ascribed to groups (societies) of system components (agents) as a whole, rather than to individuals, in order to effectively rationalise complex patterns of interaction within intelligent (multi-agent) systems. Be it either explicitly represented at the symbol-level or hard-coded within specific coordination algorithms, coordination knowledge is instrumented by a wide and heterogeneous variety of coordination models, abstractions and technologies. Coordination knowledge engineering is then about eliciting, modelling and instrumenting coordination knowledge in a principled and effective manner.In this introductory article, we briefly review two well-known frameworks to conceptualise coordination, then we discuss different dimensions along which coordination models can be classified, and analyse their impact on the design of coordination mechanisms and their supporting coordination knowledge. Finally, we sketch our view on coordination knowledge engineering and introduce the different contributions to this special issue along this line.

MULTI-AGENT SYSTEMS FOR DECISION SUPPORT: A CASE STUDY IN THE TRANSPORTATION MANAGEMENT DOMAIN

This article describes how agent and knowledge technology can be used to build advanced software systems that support operational decision making in complex domains. In particular, we present an abstract architecture and design guidelines for agent-based decision support systems. We illustrate our approach with a case study in the transportation management domain.

Dynamic coordination of ambulances for emergency medical assistance services

The main objective of emergency medical assistance (EMA) services is to attend patients with sudden diseases at any possible location within an area of influence. This usually consists in providing in situ assistance and, if necessary, the transport of the patient to a medical center. The potential of such systems to reduce mortality is directly related to the travel times of ambulances to emergency patients. An efficient coordination of the ambulance fleet of an EMA service is crucial for reducing the average travel times. In this paper we propose mechanisms that dynamically improve the allocation of ambulances to patients as well as the redeployment of available ambulances in the region under consideration. We test these mechanisms in different experiments using historical data from the EMA service of the Autonomous Region of Madrid in Spain: SUMMA112. The results empirically confirm that our proposal reduces the average response times of EMA services significantly.

A role-based support mechanism for service description and discovery

The ever-growing number of services on the WWW provides enormous business opportunities. Services can be automatically discovered and invoked, or even be dynamically composed from more simples ones. In this paper we concentrate on the problem of service discovery. Most current approaches base their search on inputs and outputs of the service. Some of them also take into account preconditions and effects, and other parameters that describe the service. We present a new approach that complements existing ones by considering the types of interactions that services can be used in. We present our proposal for a concrete application based on a real-world scenario for emergency assistance in the healthcare domain

Using Normative Markov Decision Processes for evaluating electronic contracts

Before signing electronic contracts, a rational agent should estimate the expected utilities of these contracts and calculate the violation risks related to them. In order to perform such pre-signing procedures, this agent has to be capable of computing a policy taking into account the norms and sanctions in the contracts. In relation to this, the contribution of this work is threefold. First, we present the Normative Markov Decision Process, an extension of the Markov Decision Process for explicitly representing norms. In order to illustrate the usage of our framework, we model an example in a simulated aerospace aftermarket. Second, we specify an algorithm for identifying the states of the process which characterize the violation of norms. Finally, we show how to compute policies with our framework and how to calculate the risk of violating the norms in the contracts by adopting a particular policy.

Learning and evolving combat game controllers

The design of the control mechanisms for the agents in modern video games is one of the main tasks involved in the game design process. Designing controllers grows in complexity as either the number of different game agents or the number of possible actions increase. An alternative mechanism to hard-coding agent controllers is the use of learning techniques. This paper introduces two new variants of a hybrid algorithm, named WEREWoLF and WERESARSA, that combine evolutionary techniques with reinforcement learning. Both new algorithms allow a group of different reinforcement learning controllers to be recombined in an iterative process that uses both evolution and learning. These new algorithms have been tested against different instances of predefined controllers on a one-on-one combat simulator, with underlying game mechanics similar to classic arcade games of this kind. The results have been compared with other reinforcement learning controllers, showing that WEREWoLF outperforms the other algorithms for a series of different learning conditions.

Coordination in Multi-Agent Systems: Towards a Technology of Agreement

It is commonly accepted that coordination is a key characteristic of multi-agent systems and that, in turn, the capability of coordinating with others constitutes a centrepiece of agenthood. However, the key elements of coordination models, mechanisms, and languages for multi-agent systems are still subject to considerable debate. This paper provides a brief overview of different approaches to coordination in multi-agent systems. It will then show how these approaches relate to current efforts working towards a paradigm for smart, next-generation distributed systems, where coordination is based on the concept of agreement between computational agents.