Petrie, Charles J. Jr.; 1996. Agent-based engineering, the web, and intelligence. IEEE Expert 11(6).  Home/Search   Document Not in Database   Summary   Related Articles   Check

....of this survey will identify technologies that are sufficiently far along in their development to be considered for inclusion as PAs. 2.1 What Is an agent The term agent has been used in many, often meaningless, ways. Petrie gives a good overview of various uses of this term and concludes [12]: intelligent autonomous agents is a term that, for the moment, is not of obvious utility. We take a systems engineering approach to agents, and discuss here the types of agents that are useful for building planning systems. Within Petrie s classification of agents, MPA agents are ....

C. J. Petrie. Agent-based engineering, the web, and intelligence. IEEE Expert, 11(6):24--29, December 1996.

....of a multi agent distributed information retrieval system and discusses how well these are met by the FIPA ACL. 1 Introduction Software agents have long been recognised as a promising technology for constructing complex systems as open and distributed communities of loosely coupled modules [9, 11]. A key development in the field of multi agent systems has been the specification of agent communication languages (ACLs) such as KQML [3] and FIPA ACL [7] These languages are intended to provide standard declarative mechanisms for agents to communicate knowledge and make requests of each ....

C. J. Petrie. Agent-based engineering, the web, and intelligence. IEEE Expert, 11(6):24--29, December 1996.

....Language (KQML) is used as the communication language by the agents in the Collaborative Spider system. We use JATLite (Java Agent Template, Lite) as the KQML implementation platform. JATLite, developed at the Center for Design Research at Stanford University, is a Java implementation of KQML [20,34]. Agents send and receive KQML messages in our system through a message router using TCP IP protocol. The use of agent templates facilitates agent development by aggregating common functions and services across all agent classes into a few abstract classes. 4. Sample user sessions using ....

C. Petrie, Agent-based engineering, the web, and intelligence, IEEE Expert 11 (6) (1996) 24 -- 29.

....and tradeoff analysis. Researchers have also applied CSP to problem solving in distributed contexts. Work by Yokoo [17] and others has addressed the problem of solving CSP in a distributed manner. CSPs have also been used to build lightweight agents for tasks from concurrent engineering [9] to e commerce [13] 2.1 Definition of a CSP A finite binary discrete CSP is defined as: ffl A finite set of variables V , ffl A finite domain D i of possible discrete values for each variable v i 2 V , ffl A finite set of constraints C between any pairs of variables in V . A solution to the ....

P. Petrie. Agent-Based Engineering, the Web, and Intelligence. IEEE Expert, 11(6):24--29, December 1996.

....ARPI researchers. The results of this survey were submitted in our first annual report and identified technologies to be considered for inclusion as PAs. 5.1 What Is an Agent The term agent has been used in many, often meaningless, ways. Petrie gives a good overview of various uses of this term [11]. In Petrie s terminology, MPA takes a systems engineering approach to agents, and explores the types of agents that are useful for building planning systems. Within Petrie s classification of agents, MPA agents are typed message agents, which are described as follows: typed message agents] take ....

C. J. Petrie. Agent-based engineering, the web, and intelligence. IEEE Expert, 11(6):24--29, December 1996.

....agent programming to be an agent action, or a pre specified subtask (collection of agent actions) These building blocks are then chained together by the DECAF planner. This paradigm differs from most of the well known agent toolkits, which instead use the API approach to agent construction (e.g. [17]) Functionally, DECAF is based on RETSINA [24] and taems [5] To augment the functions of the agent Planner, the control or programming of DECAF agents can be provided via a Agent Initialization DECAF Task and Control Structures Plan File Incoming KQML messages Domain Facts and Beliefs KQML ....

C. J. Petrie. Agent-based engineering, the web, and intelligence. IEEE Expert, December 1996.

....message bu ering, allowing agents to fail and recover. Agents can send and receive messages using KQML, an early ACL standard, although other languages such as FIPA s ACL can also be used. Message bu ering also supports a name and password mechanism that lets agents move freely between hosts [31]. Kasbah is a virtual market place on the Web where users can create autonomous agents to buy and sell goods on their behalf. Users can specify parameters to guide and constrain the agent overall behaviour. Any intelligent agent in Kasbah is an object (an instance of a class) and the market ....

C. Petrie. Agent-based engineering, the web, and intelligence. IEEE Expert, 11(6):24-29, December 1996.

....In addition, Brainstorm J provides a concrete class for communication by using KQML. The concrete class KQMLCommunicator is able to send KQML messages through the Internet by using TCP IP, SMTP POP (eMail) FTP or HTTP. Services provided by KQMLCommunicator are built by using JATLite [30]. As a consequence, all communications pass through a centralized component called Agent Message Router (AMR) The AMR also provides name services, brokering and off line operations. Brainstorm J provides several classes for responding to KQML performatives in a default manner. For example, class ....

C. Petrie. Agent-based engineering, the web, and intelligence. IEEE Expert, 11(6):24--29, December 1996. http://cdr.stanford.edu/NextLink/Expert.html.

....requirements. In addition, Brainstorm J provides a concrete class for communication by using KQML. The concrete class KQMLCommunicator is able to send and receive KQML messages through the Internet by means of TCP IP, SMTP POP (e Mail) FTP or HTTP. This class uses the services provided by JATLite (Petrie, 1996). 4.4 Coordination To validate the tailorability of the framework we have developed a multi agent solution to the n Queens problem (Barbuceanu and Fox, 1999) We choose this MAS because it requires the usage of a coordination mechanism, despite the fact that the framework had not support for ....

Petrie, C. (1996). Agent-based engineering, the web, and intelligence. IEEE Expert, 11(6):24--29. http: //cdr.stanford.edu/NextLink/Expert.html.

....Besides its role, each agent should also have a particular game strategy. Strategies are rules which tell agents what actions to choose at some specific decision point at the second round of each game, in the MSG. A modeler agent is just an 1 An introduction to JATLite can be found in [14]. agent using a strategy which builds models about the others, and uses them in order to improve its behavior. We wanted to explore a range of strategies, ranging from least informed to most informed. Our research path can be described as follows: First, we define simple basic strategies without ....

C. Petrie. Agent-based engineering, the web, and intelligence. IEEE Expert, December 1996.

....of this survey will identify technologies that are sufficiently far along in their development to be considered for inclusion as PAs. 2.1 What Is an agent The term agent has been used in many, often meaningless, ways. Petrie gives a good overview of various uses of this term and concludes [8]: intelligent autonomous agents is a term that, for the moment, is not of obvious utility. We take a systems engineering approach to agents, and discuss here the types of agents that are useful for building planning systems. Within Petrie s classification of agents, MPA agents are ....

C. J. Petrie. Agent-based engineering, the web, and intelligence. IEEE Expert, 11(6):24-- 29, December 1996.

....agent programming to be an agent action, or a pre specified subtask (collection of agent actions) These building blocks are then chained together by the DECAF planner. This paradigm differs from most of the well known agent toolkits, which instead use the API approach to agent construction (e.g. [8]) Functionally, DECAF is based on RETSINA [3] and TAEMS[2, 13] Although a traditional HTN planning component is in development, currently the control or programming of DECAF agents is provided via a picture based GUI called the Plan Editor. The Plan Editor can also be used to construct shared ....

Charles J. Petrie. Agent-based engineering, the web, and intelligence. IEEE Expert, December 1996.

....wemean that the task is handled by multiple distributed autonomous agents without global control. By cooperation we mean that the task is accomplished by those agents through information exchange and task sharing as no one agent has sufficient information to solve the entire problem [10, 17, 13, 15, 3]. To support distributed problem solving, various multi agent systems have been developed such as Odyssey[14] Voyager [20] Aglets[11] and Concordia [19] Various efforts on agent standards have been taken by Agent Interop Working Group, IETF, OMG and WWW Consortium [19] In reality, many ....

C. J. Petrie. Agent-based engineering, the web, and intelligence. IEEE Expert 96(12), 1996.

....approaches, like for example the exploitation of production rules. Some of the roles fit well in a traditional Object Oriented programming paradigm; however, others require that the components filling them are proactive and may initiate tasks although their methods are not explicitly invoked [30]. For example, the shopping cart might be monitored to check whether the customer s selections are consistent (for example, items such as a computer and a printer, which have to work together, must be compatible) in case of inconsistencies, the controller should autonomously trigger a dialog to ....

....method invocation, the other types of messages can be delivered by objects by invoking specific Voyager methods. Other systems, like JAFMAS [15] and JATLite [1] would allow our agents to be distributed on a network and communicate in a standard agent communication language based on Speech Acts [38, 20, 30]. However, following KQML, they impose that the contents of the speech acts are represented, in an appropriate content language, as strings. Therefore, they make the exchange of messages containing structured information a complex task. 3 On the other hand, in the development of our system, we ....

C.J. Petrie. Agent-based engineering, the web, and intelligence. IEEE Expert, December:24--29, 1996.

....such as multiagent systems, where there is active research underway at many research centers around the world. Unfortunately, definitions of agents abound, and almost every research group has its own. Several properties and definitions of computational agents have been studied in the literature [Petrie, 1996; Franklin Graesser, 1997] Definitions. Good definitions are not mere stipulations. They seek to capture some intuitive, pretheoretic notions. In the broad sense, definitions represent unstated consensus. There is at times a lot of interest in such definitions, but it usually dissipates in ....

Petrie, Charles J. Jr.; 1996. Agent-based engineering, the web, and intelligence. IEEE Expert 11(6):24--29.

....Information rich environments are the modern, open, large scale environments with autonomous heterogeneous information resources. They are a major target for systems based on agents and multiagent systems (MAS) technologies [6] Numerous definitions of agents are known in the literature [3, 5, 10]. Indeed, the only agreement seems to be that there is a range of definitions Some of the important properties of agents include autonomy, adaptability, and interactiveness, but exceptions reveal that an agent need not have these properties. However, we believe that agents must be capable of ....

Charles J. Petrie, Jr. Agent-based engineering, the web, and intelligence. IEEE Expert, 11(6), December 1996.

....that our use of mixed ontology expressions is a useful approach to bridging the gap between limited ontologies and broader general purpose ontologies. sofar is not the only Java based agent framework; there exist a number of others, the most notable of which are Zeus [39] JAFMAS [5] JATlite [40], fipa os [17] Ajanta [48] and JACK [1] Zeus and JAFMAS adopt a similar approach, providing both a fipa or kqml based communications infrastructure and a planning engine for handling rule based conversations by means of automata models, and as such are representative of a traditional AI ....

C. Petrie. Agent-based engineering, the web, and intelligence. IEEE Expert, December 1996.

....agent action, or a prespeci #ed subtask #collection of agent actions#. These building blocks are then chained together by the DECAF planner. This paradigm di#ers from most of the well known agent toolkits, which instead use the API approach to agent construction #e.g. #Chauhan 1997; Boloni 1996; Petrie 1996##. Functionally, DECAF is based on RETSINA #Sycara et al. 1996; Decker et al. 1997; Decker Sycara 1997; Williamson, Decker, Sycara 1996b; 1996a# and t#ms #Decker Lesser 1993; Wagner, Garvey, Lesser 1997#. In the absence of a planning component, the control or programming of DECAF ....

Petrie, C. J. 1996. Agent-based engineering, the web, and intelligence. IEEE Expert.

....distributed systems, in which there is a dynamical number of agents, most of which will not know each other. They get to know each other by performing some lookup function. This lookup function is generally performed by a separate agent called a facilitator, a broker, or a matchmaker. JATlite. (Petrie, 1996) This is apparently one of the most popular agent platforms, and has been used for some real life applications. The system is composed of ve layers. An agent may use one or more layers at will. From bottom to top: Abstract layer: describes an abstract API speci cation, providing low level ....

Petrie, C. J. (1996). Agent-based engineering, the web, and intelligence. IEEE Expert, december 1996.

No context found.

Petrie, Charles J. Jr.; 1996. Agent-based engineering, the web, and intelligence. IEEE Expert 11(6).

No context found.

Charles Petrie. Agent-based engineering, the web, and intelligence. IEEE Expert, 11(6):24-29, December 1996.

No context found.

Charles Petrie. Agent-based engineering, the web, and intelligence. IEEE Expert, December 1996.

No context found.

Petrie, Charles J. Jr.; 1996. Agent-based engineering, the web, and intelligence. IEEE Expert 11(6).

No context found.

C. Petrie, Agent-based engineering, the Web, and intelligence", in IEEE Expert, December 1996.

No context found.

C. J. Petrie. Agent-based engineering, the web, and intelligence. IEEE Expert 96#12#, 24#29, 1996.

First 50 documents

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC