Information Technology Reference
In-Depth Information
During the recent years, several approaches for modeling agent-based systems have
been proposed (e.g. [1]). Although the developed modeling languages are a step into
the right direction, they have problems with fulfilling a user's need to efficiently model
a custom application domain. An ideal modeling language would contain, beside the
core concepts of
Multiagent Systems
(MAS), specific concepts, software languages,
graphical representations, etc. for a certain target environment. The better the modeling
language covers the target concepts, the less manual customizations are necessary at
code level. Moreover, it is desirable to extend the modeling language with new concepts
from AI and agent research (e.g. new ways of modeling interaction protocols). This
paper proposes a model-driven framework for engineering agent-based systems which
addresses the described problems. The framework, called B
OCHICA
1
, is based on a
Domain Specific Language
(DSL) which covers the main concepts of MAS. It provides
several extension interfaces for integrating new concepts, AOSE methods, and 3rd party
software languages. The customizations allow the user to tailor the framework to his
needs without loosing the integration into a larger framework. The first part of this paper
provides background on AOSE (Section 2), introduces the general idea of the B
OCHICA
framework (Section 3), and provides an overview of the extension interfaces (Section 4).
The second part shows how to tailor the framework to the application domain of agents
in semantically-enhanced virtual worlds and presents the evaluation results (Section 5).
Finally, Section 6 discusses the related work and Section 7 concludes this paper.
2
Background
Raising the level of abstraction in software development was always an important driver
in computer science research. The level of abstraction of a software language can
be defined as the distance between the computer hardware and the concepts of that
language [2]. Since the invention of computer systems, the level of abstraction was
steadily increased from opcodes, assembler languages, procedural languages, up to
object-oriented languages. The question that arises is how the next higher level of
abstraction looks like. According to [2], “
The challenge for language engineers is
that the software languages that we need to create must be at a higher level of ab-
straction, a level we are not yet familiar with. They must be well designed, and soft-
ware developers must be able to intermix the use of multiple languages. Here too,
we are facing a new, unknown, and uncertain task.
” In the agent community, AOSE
has been seen as a natural successor of OOSE for a long time. Several articles dis-
cuss why AOSE has not arrived yet in mainstream software engineering [3][4][5].
Three of the identified main problems are (i) misunderstandings or wrong assumptions
by non-agent experts, (ii) agent-oriented standards and methods are not yet sufficient
for industry needs, (iii) lack of powerful tools. However, regarding the level of abstrac-
tion [3] concludes: “
With concepts such as roles and responsibilities, agent-oriented
approaches to problem and system description are much closer to the ways in which
managers conceive of business domains than are traditional software engineering de-
scriptions.
” This matches the requirements stated by Kleppe. Our research hypothesis is
1
Bochica was a semi-god of the Muisca culture who brought them living skills and showed
them how to organize their lives.
