Agent Construction Tools(5)

Bee-gent
Toshiba Corporation
Systems and Software Research Laboratories
Bee-gent is a new type of software development framework in that it is a 100%
pure agent system. As opposed to other systems which make only some use of age
nts, Bee-gent completely "Agentifies" the communication that takes place betwe
en software applications. The applications become agents, and all messages are
 carried by agents. Thus, Bee-gent allows developers to build flexible open di
stributed systems that make optimal use of existing applications.

The Bee-gent framework is comprised of two types of agents. "Agent Wrappers" a
re used to agentify existing applications, while "Mediation Agents" support in
ter-application coordination by handling all communications. The mediation age
nts move from the site of an application to another where they interact with t
he agent wrappers. The agent wrappers themselves manage the states of the appl

数据挖掘研究院

ications they are wrapped around, invoking them when necessary. Thus inter-app
lication coordination is handled through the agent wrappers generating and rec
eiving requests, which are transported around by the mediation agents. The med
iation agents do more than just transport the messages; they are able to respo
nd to the nature of the request to determine the best course of action.  


------------------------------------------------------------------------------
--

Bond Distributed Object System
Purdue University
The Bond distributed object system provides a message oriented middleware envi
ronment for developing distributed applications. Bond uses the KQML language f
or object communication.

The agent framework of the Bond system simplifies the task of developing agent
s by allowing the programmer to concentrate on the specific strategies of a ne
w agent. Bond agents have the intrinsic capability to be controlled remotely a

数据挖掘研究院

nd to cooperate with each other. The task of an application programmer is limi
ted to specify the agenda, the finite state machine of the agent, and the stra
tegies associated with each state.  


------------------------------------------------------------------------------
--

Cable - Intelligent Agents
Logica Corporation
Cable is a generic system architecture developed by Logica as part of the GRAC
E Consortium. Cable can be used to develop and execute distributed application
s that are based on the metaphor of multiple, cooperating intelligent agents.


Cable provides the user with an Agent Definition Language (ADL), for defining
agents, and a parser known as the Scribe, for compiling agent definitions writ
ten in ADL into agent applications. Agents are developed using ADL and C++. AD
L allows developers to use Cable without worrying about underlying detail, pro
viding a language with a level of abstraction close to that of the agents with 数据挖掘研究院
 which an application is designed. Inter-agent communication over a local area
 network is handled using ORBIX, an implementation of the CORBA 2.0 standard.
 


------------------------------------------------------------------------------
--

DECAF Agent Framework
University of Delaware
DECAF (Distributed Environment Centered Agent Framework) provides a platform f
or extremely rapid development of agents. This is accomplished by building an
operating environment that provides an interface, internal agent scheduling an
d monitoring in a fashion similar to operating system primitives. The agent de
veloper does not need knowledge of any of this structure and can thus focus on
 development of the agent itself. The basic DECAF architecture has been built
using the Java programming language. The rapid prototyping has been tested by
development of approximately 15 agents over a two week period by a recent clas 数据挖掘研究院
s.  


------------------------------------------------------------------------------
--

dMARS
Australian Artificial Intelligence Institute Ltd.
dMARSTM is an agent-oriented development and implementation environment for bu
ilding complex, distributed, time-critical systems. Designed for rapid configu
ration and ease of integration, it facilitates system design, maintenance and
re-engineering. This product is based on the older Procedural Reasoning System
 (PRS) developed by SRI International (California). dMARS takes advantage of t
he latest research into multi-agent, real-time reasoning.

dMARS is suited to the development of any application that requires both proac
tive goal directed behavior and reliable time-critical response to change. It
is particularly well suited to applications where a large number of complex bu
t well-defined procedures or tactics exist for accomplishing particular tasks
in a variety of situations.

数据挖掘研究院

dMARS was designed with the issues of robustness, efficiency and user-extensib
ility in mind. It provides a sophisticated suite of graphical tools for develo
pment and debugging. These tools not only provide an intuitive interface, but
address the specific issues involved with large-scale development.

dMARS provides a high-level, graphical plan language for specifying complex, c
ontext-dependent processes and tactics. dMARS is written in C and C++ and will
 execute on a variety of UNIX platforms. dMARS is available from the Australia
n Artificial Intelligence Institute Ltd., Level 6, 171 La Trobe Street, Melbou
rne 3000 Australia.  


------------------------------------------------------------------------------
--

EXCALIBUR
GMD First (German National Research Center for Information Technology,
Research Institute for Computer Architecture and Software Technology),
Technical University of Berlin,

数据挖掘研究院

EXCALIBUR is a generic architecture for autonomous agents used in a complex co
mputer-game environment. These agents must be able to find the right actions t
o pursue their given goals and adapt their behavior to new environments or opp
onents. Agents also must be able to communicate and perform coordinated group
actions.  


------------------------------------------------------------------------------
--

GenA
Centre de Recherche Informatique de Montreal (CRIM)
GenA (Multiplatform Generic Agents)
CRIM's generic agent model "GenA" allows agents to run on and move between dif
ferent agent platforms by using an interface called "GenAi." Programming these
 GenA agents is independent of the underlying agent platforms. An implementati
on of this agent model has been performed and validated by a small application
.
GenA (Multiplatform Generic Agents) is based on Java and allows mobile agents
to run and move between other platforms such as Voyager, Grasshopper, and Agle
ts. Programming GenA agents is independent of the underlying agent platforms.
An implementation of the agent model has been performed and validated by a sma
ll application. At the present time (March 2000), GenA has no external researc
her or students working on it. For more information, download GenA Mata '99 vf
3.ps.  


------------------------------------------------------------------------------
--

Gypsy
Technical University of Vienna
The Gypsy Project utilizes Java for the implementation of a flexible environme
nt for experimenting with mobile agent programming. It is intended for applica
tion in Internet information retrieval, Internet commerce, mobile computing, a
nd networks network management.  


------------------------------------------------------------------------------
--

Hive
The Media Lab 数据挖掘研究院
Massachusetts Institute of Technology
Hive is a Java software platform for creating distributed applications. Using
Hive, programmers can easily create systems that connect and use data from all
 over the Internet. At its heart, Hive is an environment for distributed agent
s to live, communicating and moving to fulfill applications.  


------------------------------------------------------------------------------
--

InfoSleuth
Collaboration Management Infrastructure (CMI) project at Microelectronics and
Computer Technology Corporation (MCC)
InfoSleuth is a consortial Project carried out by MCC on behalf of several par
ticipants since January 1, 1995. Current
participants are: Raytheon Company, General Dynamics Information Systems, Inc.
, SAIC , NCR Corporation, TRW, Inc.,
Schlumberger Limited, and Rafael.

InfoSleuth implements a community of cooperating agents that discovers, integr
ates and presents information on behalf of a
user or application, for which it provides a simple, consistent interface. The
 information it accesses are distributed and
heterogeneous, for example the types of information available:
• through an intranet in a large corporation, or
• on the World Wide Web.

The InfoSleuth architecture consists of a set of collaborating agents that wor
k together at the request of the user to:
1. gather information via complex queries from a changing set of databases and
 semi-structured text repositories
distributed across an internet,
2. perform rudimentary polling and notification facilities for monitoring chan
ges in data,
3. automatically route location-independent requests to update individual data
 items, and
4. analyze information using statistical data mining techniques and/or logical
 inferencing.

Users make requests to InfoSleuth from a domain-independent or domain-specific
 applet. Requests are made against an
ontology specifying his domain of interest. The applet forwards the request to
 the agent system. Within the agent system,
agents cooperate to satisfy the request on behalf of the user. Each request is
 processed by the available agents at the time of
the request. Results are presented either within the user's applet or within a
 specialized result applet.  


------------------------------------------------------------------------------
--

Infospiders
University of California San Diego - Computer Science Dept.
By Filippo Menczer and Rik Belew

InfoSpiders (aka ARACHNID: Adaptive Retrieval Agents Choosing Heuristic Neighb
orhoods for Information Discovery)

This project features an artificial life inspired model using endogenous fitne
ss for information retrieval in large, dynamic, distributed, heterogeneous dat
abases, such as the WWW. A population of agents is evolved under density depen
dent selection for the task of locating information for the user. The energy n
ecessary for survival is obtained from both environment and user in exchange f
or relevant information. By competing for energy, the agents robustly adapt to
 their environment and are allocated to efficiently exploit their shared resou
rces.  


------------------------------------------------------------------------------
--

JADE
CSELT S.p.A., University of Parma
JADE (Java Agent DEvelopment Framework) is a software framework to develop age
nt-based applications in compliance with the FIPA specifications for interoper
able intelligent multi-agent systems. The goal is to simplify the development
while ensuring standard compliance through a comprehensive set of system servi
ces and agents. JADE can then be considered an agent middle-ware that implemen
ts an Agent Platform and a development framework. It deals with all those aspe
cts that are not peculiar of the agent internals and that are independent of t 数据挖掘研究院
he applications, such as message transport, encoding and parsing, or agent lif
e-cycle.
The JADE 1.2 Agent Platform complies with FIPA97 version 1.2 specifications an
d includes all those mandatory agents that manage the platform, that is the AC
C, the AMS, and the DF. All agent communication is performed through message p
assing, where FIPA ACL is the language to represent messages.



------------------------------------------------------------------------------
--

JAFMAS
University of Cincinnati
JAFMAS provides a framework to guide the development of multiagent systems alo
ng with a set of classes for agent deployment in Java. The framework is intend
ed to help beginning and expert developers structure their ideas into concrete
 agent applications. It directs development from a speech-act perspective and
supports multicast and directed communication, KQML or other speech-act perfor
matives and analysis of multiagent system coherency and consistency. The JAFMA
S project provides a good comparison of agent tools with a particular emphasis
 on mobile agent projects.