Milind Tambe. Agent architectures for flexible, practical teamwork. National Conference on AI (AAAI97), pages 22--28, 1997.  Home/Search   Document Not in Database   Summary   Related Articles   Check

.... studied in the fields of Distributed AI [3, 10, 4, 5, 7, 15] Much success has been achieved in defining theoretical foundations for guiding agent cooperation and coordination in course of teamwork activities [15, 3, 9, 5] and in developing pragmatic framework for programming teamwork agents [11, 12]. Numerous teamwork domains that often involve highly complex group activities have been explored in software simulation environment environment, such as RoboCup97 soccer games [10, 2, 11, 13] teamwork in military helicopter flying simulation [11] cooking [5] and distributed industrial ....

.... course of teamwork activities [15, 3, 9, 5] and in developing pragmatic framework for programming teamwork agents [11, 12] Numerous teamwork domains that often involve highly complex group activities have been explored in software simulation environment environment, such as RoboCup97 soccer games [10, 2, 11, 13], teamwork in military helicopter flying simulation [11] cooking [5] and distributed industrial applications [6] With the advent of ubiquitous computing technology, in particular in the fields of wireless communication and mobile computing, agent teamwork research is pushed to its limit, facing ....

[Article contains additional citation context not shown here]

M. Tambe. Agent architectures for flexible, practical teamwork. In National Conference on Artificial Intelligence (AAAI-97), 1997.

....work in the domain of robotic soccer, in which it is potentially advantageous to be able to recognize, model, and predict the other agents behaviors. Robotic soccer has been developed in simulation and with real robots [7] Important aspects of opponent modeling have been developed in simulation [11]. The real robots are physical robotic agent computer controlled either on board or off board through radio communication. In both simulation and real robots, the agents must be fully autonomous, as human operators are not allowed to interfere once the game has started. In general, robotic or ....

Milind Tambe. Agent architectures for flexible, practical teamwork. In Proceedings of AAAI-97, Menlo Park, California, 1997. AAAI Press.

....have argued that in the design of multiagent systems, to support rich collaborative behaviour it is essential to provide individual agents with various forms of social awareness. Technical concepts related to notions of social awareness that have been considered include: joint and shared plans [8, 13, 18], conventions and social responsibility [10, 11] social commitment [2] social laws [16] spheres of commitment [17] reasoning with obligations [1] etcetera. In this paper we present a new approach to social reasoning which integrates prior work on norms and obligations [3, 5, 20] with the now ....

....e.g. a norm 7 not to delete files might be easily violated when the file is known to contain a virus. A strong motivation for our work is to find a framework for building agents that can exhibit a wide range of collaborative behaviours [7] which may include tightly coordinated teamwork, e.g. [18], or could just involve the exploitation of a variety of more flexible social behaviours [2] In this context, obligations include support for efficient execution of joint plans [8, 18] and for characterising certain forms of social commitment [4] They provide additional stability to joint plans ....

[Article contains additional citation context not shown here]

Tambe, M. Agent architectures for flexible, practical teamwork. In Proceedings of the National Conference on Artificial Intelligence (AAAI-97) , 1997.

....fighter pilots need to be aware of the plans of both their enemies and their allies. In particular, when pilots are flying in formation, they need to reason about their team s mission, their role in that mission, and what they can expect the other pilots around them to do in pursuing the mission [74, 50]. For instance, without a notion of team behavior during reconnaissance missions, the researchers found that an artificial pilot would often miss a signal or fail to observe a landmark and end up wandering o# by itself. RESC team was developed to extend RESC to reason about teams and team member s ....

Milind Tambe, Agent architectures for flexible, practical teamwork, 14th National Conference on Artificial Intelligence, 1997.

....CNP based protocols[23] or BDI approaches where agents compromise to achieve the individual collective goals ( 13] 14] 24] Another perspective is based on the elaboration of conventions and or rules. Shoham [22] proposed social behaviors as a way to program multiagent systems. In STEAM [25], coordination rules are designed in order to facilitate the cohesion of the group. Cooperation for achieving independent goals has been mostly addressed in the framework of application specific techniques such as multi robot cooperative navigation [27,5] 3 Cooperation for Plan Enhancement In ....

M. Tambe. Agent architectures for flexible, practical teamwork. In First International Conference on Autonomous Agents, 1998.

....(OR) any agent (one or more) can handle it. And with shared competitive responsibilities (XOR) at most one agent must carry it out, for example due to the potential for interference. These responsibility types bear some resemblance to AND and OR combinations of team operators in STEAM [16] . Each of these types of joint responsibilities would be handled differently among the agents sharing them, and these interactions often require communication. With AND responsibilities, the agents must coordinate by synchronizing to perform the act simultaneously, waiting until everybody is ....

M. Tambe. Agent architectures for flexible, practical teamwork. In Fourteenth National Conference on Artificial Intelligence, 1997.

.... or intentions and those that have an explicit representation for group intentionality [8, 5] Theories of group intentionality have proven difficult to implement in practice and still lack practical algorithms for governing not only the need for communication, but also the utility of communication [19]. Further, work on modeling group intentionality does not generally address the visual recognition of group intention, where agents are acquiring much of their information about the world through probabilistic, perceptually based action recognition. Our hypothesis is that visual goal networks and ....

M. Tambe. Agent architectures for flexible, practical teamwork. In Proc. Nat. Conf. on Artificial Intelligence, 1997.

....set of such theorems is presented. 1 Introduction Collaboration must be designed into systems from the start; it cannot be patched on. 4] Simply fitting individual agents with precomputed coordination plans will not do, for their inflexibility can cause severe failures in teamwork. [12] When a group of agents get together to work on some complex group action, whether it be a group of helicopter agents embarking on a scouting mission [12] or a group of people making dinner [5] collaboration does not just happen. It requires the existence or formation of mutual beliefs about the ....

....fitting individual agents with precomputed coordination plans will not do, for their inflexibility can cause severe failures in teamwork. 12] When a group of agents get together to work on some complex group action, whether it be a group of helicopter agents embarking on a scouting mission [12] or a group of people making dinner [5] collaboration does not just happen. It requires the existence or formation of mutual beliefs about the capabilities and commitments of the agents involved, the adoption by individual agents of various intentions (not only intentions to do various actions, ....

Milind Tambe. Agent architectures for flexible, practical teamwork. In Proceedings of the Fourteenth National Conference on Artificial Intelligence. 1997.

....and third, no individual agent has sucient competence, resources, or information to solve the entire problem. There are several frameworks, such as the BDI based agent architecture [5, 1] the Little JIL process programming language [10] the collaborative plan model [6] and the teamwork model [7], that deal with high level coordination issues. They logically reason about which task should be performed collaboratively or the resources required for certain tasks, but they do not quantitatively reason about detailed coordination issues like the concurrent scheduling and selection of multiple ....

Milind Tambe. Agent Architecture for Flexible, Practical Teamwork. Proc. of the 14th National Conf. on AI (AAAI), 1997, pages 22-28.

....work in the domain of robotic soccer, in which it is potentially advantageous to be able to recognize, model, and predict the other agents behaviors. Robotic soccer has been developed in simulation and with real robots [7] Important aspects of opponent modeling have been developed in simulation [11]. The real robots are physical robotic agent computer controlled either on board or off board through radio communication. In both simulation and real robots, the agents must be fully autonomous, as human operators are not allowed to interfere once the game has started. In general, robotic or ....

Milind Tambe. Agent architectures for flexible, practical teamwork. In Proceedings of AAAI-97, Menlo Park, California, 1997. AAAI Press.

....at hand. Other work in AI include efforts on semantics of KQML [29, 41] which is closely related to earlier work of Cohen and Levesque, but it does not include the notion of value central to our approach. While work on communication in negotiation is reported in [28, 48, 49] Recently, Tambe [42] suggested decision theoretic communication selectivity to establish mutual belief among agents in a team. This approach is similar to ours in that the focus is on whether or not an agent should transmit a given message (fact) to others. Tambe uses a decision tree containing reward and penalty ....

Milind Tambe. Agent architectures for flexible, practical teamwork. In Proceedings of the Fourteenth National Conference on Artificial Intelligence, pages 22 -- 28, 1997.

.... stage , etc. along with plans and operators, for describing the domain and ideal team behavior. Our ontology defines the semantics of those predicates used in encoding teamwork. Although there are other methods for automatically generating coordinated teambehaviors (joint intentions [1,11]; contract nets [6] multi agent decision theory [13] we take a more knowledge based approach because: We are interested in structured domains with well defined roles and well documented procedures, The virtual agents must be able to interact with human trainees and communicate in ....

....documented procedures, The virtual agents must be able to interact with human trainees and communicate in realistic ways, The explicit representation of goal hierarchies and intentions will be important for diagnosing problems with team behavior and providing useful feedback. Tambe s work [1] focuses on establishing the joint intentions of team members in trying to achieve a joint goal, but not on the information needs of a team member in order to provide information proactively. This is a focus of our approach. Second, given a description of a team in MALLET, we will then define an ....

Tambe, M. Agent architectures for flexible, practical teamwork. In Proceedings of the 14th National Conference on Artificial Intelligence (Providence, Rhode Island 1997), 22-28.

....halts. There are two issues for the coordination in the above scenario. First, an agent need to know which action to perform in accordance with the other agent. Second, for synchronization, messages should be communicated between agents. For the first issue, I exploited Tambe s teamwork study[3]. Tambe developed flexible teamwork framework(STEAM) for battlefield simulation. The teamwork in his work is based on the idea of team goal, mutual belief, and joint commitment. Though his work included performance monitoring, repair, and decision theoretic communication, I exploited the idea of ....

Tambe, M. 1997, Agent architecture for flexible, practical teamwork, National Conference on Artificial Intelligence(AAAI-97)

....research ideas under the theme of social reasoning. All of these work deal with improving problem solving in the context of a group of agents. The following list is a sampling of this body of work: ffl the design of conventions [56] social laws [57] or domain independent rules for good teamwork [58], ffl identification and resolution of social dilemmas where greedy strategies used by everyone would lead to worse results for all [59] 9 ffl mechanisms for arriving at, monitoring, fulfilling, and retracting commitments to oneself or to others [60, 56, 61, 62] ffl how agents form shared ....

Milind Tambe. Agent architectures for flexible, practical teamwork. In Fourteenth National Conference on Artificial Intelligence, pages 22--28, Menlo Park, CA, 1997. AAAI Press/MIT Press.

No context found.

Tambe, M. 1997. Agent architectures for flexible, practical teamwork. National Conference on AI (AAAI97) 22--28.

No context found.

Milind Tambe. Agent architectures for flexible, practical teamwork. National Conference on AI (AAAI97), pages 22--28, 1997.

No context found.

Tambe, M. 1997. Agent architectures for flexible, practical teamwork. National Conference on AI (AAAI97) 22--28.

No context found.

Milind Tambe. Agent architectures for flexible, practical teamwork. National Conference on AI (AAAI97), pages 22--28, 1997.

No context found.

Milind Tambe. Agent architectures for flexible, practical teamwork. National Conference on AI (AAAI97), pages 22--28, 1997.

No context found.

Milind Tambe. Agent architectures for flexible, practical teamwork. National Conference on AI (AAAI97), pages 22--28, 1997.

No context found.

Milind Tambe. Agent architectures for flexible, practical teamwork. National Conference on AI (AAAI97), pages 22--28, 1997.

No context found.

Tambe, M. Agent architectures for flexible, practical teamwork. In Proceedings of AAAI-97), 22-28, 1997.

No context found.

Milind Tambe. Agent architectures for flexible, practical teamwork. National Conference on AI (AAAI97), pages 22--28, 1997.

No context found.

Tambe, M. Agent architectures for flexible, practical teamwork. In Proceedings of the National Conference on Artificial Intelligence (AAAI). August, 1997.

No context found.

Tambe, M. Agent Architectures for Flexible, Practical Teamwork. In Proceedings of AAAI 97 14th National Conference on Aritificial Intelligence (Providence, Rhode Island, 1997), AAAI Press/ The MIT Press, 1092.

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