|
Design of Intelligent Multi-Agent Systems
|
Abstract
Emerging technologies allowing two-way communication between utility
companies and their customers, as well as with smart equipment, are
changing the rules of the energy market. Deregulation makes it even more
demanding for utility companies to create new business processes for mutual
benefit. Dynamic load management of the power grid is essential to make
better and more cost-effective use of electricity production capabilities, and to
increase customer satisfaction. The compositional development method
DESIRE has been used to analyse, design, implement and verify a multi-agent
system capable of negotiation for load management.
Abstract
A compositional verification method for multi-agent systems is presented and
applied to a multi-agent system for one-to-many negotiation in the domain of
load balancing of electricity use. Advantages of the method are that the
complexity of the verification process is managed by compositionality, and
that parts of the proofs can be reused in relation to reuse of components.
Abstract
A generic model for the internal dynamic behaviour of BDI agents is
proposed. This model, a refinement of a generic agent model, explicitly
specifies beliefs and motivational attitudes such as desires, goals, intentions,
commitments, and plans, and their relations. A formal meta-language is used
to represent beliefs, motivational attitudes and strategies. Dynamic aspects of
reasoning about and revision of beliefs and motivational attitudes are modelled
in a compositional manner within the modelling framework DESIRE.
Abstract
This paper introduces a multi-agent system architecture to increase the value
of 24 hour a day call centre service. This system supports call centres in
making appointments with clients on the basis of knowledge of employees
and their schedules. Relevant activities of employees are scheduled for
employees in preparation of such appointments. The multi-agent system
architecture is based on principled design, using the compositional
development method for multi-agent systems DESIRE. To schedule
procedures in which more than one employee is involved, each employee is
represented by its own personal assistant agent, and a work manager agent
co-ordinates the schedules of the personal assistant agents, and clients
through the call centre. The multi-agent system architecture has been applied
to the banking domain, in co-operation with and partially funded by the
Rabobank.
Abstract
In many areas development methods have been developed in a conceptual
design of complex systems is specified before systems are implemented. Such
specifications focus on the semantics of systems abstracting from
implementation details, providing a basis for verification and validation of the
functionality of the systems. A dedicated development method for multi-agent
systems, requires adequate means to describe the characteristics of multi-agent
systems, in particular, the control of the dynamics of (concurrent) reasoning
behaviour and acting behaviour (e.g., guided reasoning, observation,
communication and execution of actions). A compositional multi-agent system
development method based on the principles described in this paper can
provide support to multi-agent system designers during the entire design
process: process and knowledge abstraction, compositionality, reuse,
semantics and formal analysis with respect to verification and validation. One
specific compositional multi-agent development method is DESIRE (DEsign
and Specification of Interacting REasoning components). Although the paper
addreses the principles of compositional multi-agent system development
from a generic perspective, some of the examples used to illustrate the notions
discussed are taken from this compositional development method.
Abstract
Abilities of agents and properties of their environment provide a means to
describe behaviour and functionality. These abilities also provide a basis for
re-design. In this paper an example is given of a prototype system for re-
design of a multi-agent system in which the abilities and properties are made
explicit.
Abstract
When a compositional multi-agent system is to be verified, it is natural to do
this compositionally. In this paper we explore the use of a temporal epistemic
logic to formalize specification, properties, and their proofs, of compositional
multi-agent systems. The specification of the system, its properties and their
proofs are of a compositional nature. It is shown that compositional proofs
are valid under certain restrictions. Finally, the possibility of incorporating
default persistence of information in a system into a temporal specification, is
explored.
Abstract
To support users on the World Wide Web, various types of agents can be,
and actually have been, developed. For example, to support brokering
processes in electronic commerce, personal assistant agents can be developed
that support a user offering products (or services) at the Web, or agents that
support search for information on products within a user's scope of interest,
or agents that combine both types of functionalities. Moreover, mediating
agents can be developed that communicate both with personal assistant agents
that provide information on products and with personal assistants that ask for
information on products. Recently a few applications of broker agents have
been addressed for this area. In general, applications like these are
implemented in an ad hoc fashion without an explicit design at a conceptual
level. This paper briefly describes a generic agent architecture for agents
acting in brokering processes that has been designed in a principled,
compositional manner using the compositional development method for multi-
agent systems DESIRE. This agent architecture can be instantiated by adding
specific types of knowledge to support functionalities and behaviour required.
Depending on the choice of these requirements, an agent is created for a
specific context by including the appropriate types of knowledge.
Abstract
In this paper it is shown how animal behaviour can be simulated in an agent-
based manner. Different models are shown for different types of behaviour,
varying from purely reactive behaviour to pro-active and social behaviour.
The compositional development method for multi-agent systems DESIRE and
its software environment supports the conceptual and detailed design, and
execution of these models. Experiments reported in the literature on animal
behaviour have been simulated for a number of agent models.
Abstract
A generic broker agent architecture is introduced, designed in a principled
manner using the compositional development method for multi-agent systems
DESIRE. A flexible, easily adaptable agent architecture results in which, in
addition, facilities have been integrated that provide automated support of the
agent's own maintenance. Therefore, the agent is not only easily adaptable,
but it shows adaptive behaviour to meet new requirements, either in reaction
to communication with a maintenance agent, or in a fully autonomous
manner. Based on this architecture, applications are being developed in
Electronic Commerce, in co-operation with the Internet application company
Crisp, and in intelligent Web-sites, in co-operation with the insurance
company Delta Lloyd and the software company Ordina Utopics.
Abstract
In this paper a reusable multi-agent architecture is presented for reasoning
tasks that are supported by a system that interacts with the user at two levels:
at the level of the reasoning process itself and at the level of clarification of the
process. The multi-agent architecture distmguishes three agents: a user, a
decision support agent and a clarification support agent. The user and the
decision support agent cooperate on the basis of a shared model of the
reasoning task; the user and the clarification support agent cooperate on the
basis of a shared model of the clarification task. Interaction between the user
and the decision support agent, between the user and the clarification support
agent, and between the decision support agent and the clarification agent are
made explicit in the model. The model has been made using the compositional
modelling framework DESIRE. In cooperation with and funded by chemical
industry, it has been applied to diagnosis of chemical processes.
Abstract
In spite of the rapid spread of agent technology, there is, as yet, little evidence
of an engineering approach to the development of multi-agent systems. For
example, both development methods and verification methods for multi-agent
systems are rare. In this paper, we describe a case study aimed at comparing
two formal agent languages, namely Concurrent METATEM and DESIRE. A
version of the well known PRS architecture is developed and the approaches
are compared with respect to this application.
Abstract
This paper discusses an example of the application of a high-level modelling
framework which supports both the specification and implementation of a
system's conceptual design. This framework, DESIRE (framework for
DEsign and Specification of Interacting REasoning components), explicitly
models the knowledge, interaction, and coordination of complex tasks and
reasoning capabilities in agent systems. For the application domain addressed
in this paper, an operational multi-agent system which manages an electricity
transportation network for a Spanish electricity utility, a comprehensible
specification is presented.
Abstract
Much research concerning the design of multi-agent systems (at a conceptual
level) addresses complex agents which exhibit complex interaction patterns.
Due to this complexity, it is difficult to perform rigorous experimentation. On
the other hand, systematic experimental work regarding behaviour of societies
of more simple agents, while reporting valuable results, often lacks
conceptual specification of the system under consideration. In this paper, the
compositional multi-agent development method DESIRE is not only
successfully used to develop a conceptual specification of simple agents that
together form a society, but also to simulate the behaviour in a dynamic
environment. In DESIRE, a conceptual specification, which provides a high-
level view of an agent, has enough detail for automatic prototype generation.
The prototype implementation of the conceptual specification of the simple
agents has been used to replicate and extend one of the experiments reported
in the literature.
Abstract
Interaction between distributed information systems can be modelled as
interaction between autonomous agents. The compositional development
method DESIRE is used to design and specify a multi-agent model for
competitive interaction, such as the interaction required to access limited
resources. A reusable generic model for a competitive agent is presented
together with a model of interaction between such agents.
Abstract
A cooperation model based on joint intentions introduced by Jennings is
formalised within the modelling framework DESIRE for compositional multi-
agent systems. By formalising the model in DESIRE, operationalisation and
reusability of the model are obtained, as DESIRE specifications are executable
and easily reusable.
******************
Abstract
In agent models often it is assumed that the agent maintains internal
representations of the material world (e.g., its beliefs). An overall model of
the agent and the material world necessarily incorporates sub-models for
physical simulation and symbolic simulation, and a formalisation of the (static
and dynamic) representation relation between the two types of sub-models. If
it is also taken into account that the agent's mind has a materialisation in the
form of a brain, the relations between mind and matter become more complex.
The question of how the different types of interaction between mind and
matter of an agent and the material world can be modelled in a semantically
sound manner is the main topic of this paper. The model can be used to
simulate a variety of phenomena in which (multiple) mind-matter interactions
occur, such as sensing, acting, (planned) birth and death, causing brain
damage, and psychosomatic diseases.
Abstract
A formal approach to the design of (meta-level) compositional architectures
for multi-agents systems is presented. A structure for reflective agents is
proposed within which reasoning about observation and communication, an
agent's own information state and reasoning processes, other agents'
information states and reasoning processes, and combinations of these types
of reflective reasoning are explicitly modelled. To illustrate the approach the
wise men's puzzle has been modelled using different types of reflection.
Abstract
A declarative compositional modelling framework, DESIRE, designed to
model knowledge-intensive multi-agent systems, is shown to provide a means
to model distributed industrial and business processes. An agent's
knowledge, reasoning processes and interaction with other agents, and the
world are explicitly specified within this framework. Electricity transportation
management is used to illustrate the characteristic elements of the approach, in
particular with respect to dynamic aspects of distributed industrial and
business processes; aspects which are of importance to knowledge
management and knowledge engineering.
Abstract
In current engineering practice often traditional management structures and
virtual organisations are combined. In addition to formal structures, informal,
dynamic organisational structures emerge in which engineers are personally
responsible for effective interaction. They decide when to exchange
information, and with whom, when to question requirements, when to
acknowledge conflicts, et cetera. In such virtual organisations, project
coordination may become quite complicated. In this paper communication and
coordination in a real-life case of concurrent design in aircraft industry has
been modelled and specified within the modelling framework DESIRE, on the
basis of Jennings informal, multi-agent model of cooperative problem solving
known in the literature.
Abstract
In this paper the modelling framework DESIRE, originally designed for
formal specification of complex reasoning systems is used to specify a real-
world multi-agent application on a conceptual level. Some extensions to
DESIRE are introduced to obtain a useful formal specification framework for
multi-agent systems.
Abstract
In this paper it is investigated how multi-agent systems with complex agents
can be designed and formally specified based on the notion of a compositional
architecture. After identifying the types of knowledge required for an agent
we formally define a multi-agent system. Moreover, a specific type of agent
with various capabilities of reasoning and acting is given. Some essential
patterns of integrated reasoning, communication and interaction with the
material world are described. Finally, we present an overview of formal
semantics for our approach.
Abstract
This paper addresses the issue of the use of formal modelling languages in
multi-agent domains. As a main question we evaluate in what respect the
formal modelling language DESIRE fulfills the requirements imposed by
these domains. To get an answer we studied two examples of multi-agent
tasks and developed formal specifications in DESIRE of large parts of these
examples. Our results are encouraging: it is quite feasible to extend the
applicability of the formal modelling language DESIRE to multi-agent
domains. Especially for the case of more complex agents the possibility to use
such a compositional formal modelling language may be valuable: not only to
model the agents themselves, but also the interactions between agents in a
conceptual and formal manner.
Up: Multi-Agent Systems publications
Last updated: March 23, 1998k