M. Georgeff. A theory of action for multi-agent planning. In Proceedings of the National Conference on Artificial Intelligence, pages 121--125, Austin, Texas, August 1984.  Home/Search   Document Not in Database   Summary   Related Articles   Check

.... includes Centralized Planning, where one central agent generates the global plan and then hands out pieces of that plan to be performed by the participating agents [48, 9, 67, 69] Within this centralized approach, some work falls within the category of synchronization of pre existing plans [29, 30]. The assumption there is that individual plans are first created, and then submitted to a central planner that is responsible for coordination. In [49] a fast probabilistic approach to solving this kind of coordination is suggested. There are many methods for relaxing these centralized solutions ....

M. Georgeff. A theory of action for multi-agent planning. In Proceedings of the National Conference on Artificial Intelligence, pages 121--125, Austin, Texas, August 1984. 43

....to generate task commands to the next lower level BG processes. Plans and Planning Df: plan a set of subtasks and subgoals that are designed to accomplish a task or job A plan consists of one or more paths through state space from an anticipated starting state to a desired goal state [16, 17, 18]. A plan also may include a list of required resources such as tools or materials, and a set of conditions, constraints, tolerances, and priorities that must be satisfied. Df: job plan a plan for a single agent to accomplish a job In general, a job plan can be defined by a state graph wherein ....

Georgeff, M. (1984) "A Theory of Action for Multiagent Planning," Proc AAAI 84, Texas

....agreements, ii) exchange of information about the knowledge each agent has about other agents, iii) exchange of parameters about the state of the agents. d) Distributed Multiagent Planning 4 The model removes the limitation of the previous model of having only a planner. The approach (see [6] [20], 26] works providing each agent with a model of other agents. In a particular application of this field, called Deductive Believe Model (developed by Konolige and Pollack [27] there are two classes of agents: observers and actors. The former class identifies those agents which give intentions ....

M. Georgeff, "A theory of action for multiagent planning", Proceedings of National Conference on Artificial Intelligence, Austin, TX, August

....and [34] Under the general definition of multiagent planning there are several interaction paradigms in which global plans that specify the actions of the agents are built. In this way, there is no conflict among agents. The global plan can be formed in a centralized [24] or in a distributed way [36, 60, 82, 84]. Eaton et al. 50] propose different interaction paradigms as adequate for different situations: assistance, cooperation, and competition. In particular, when an agent (called the user agent) acquires knowledge from another agent (called the provider agent) a cycle of suggestion correction ....

M. Georgeff. A theory of action for multiagent planning. In Proceedings of National Conference on Artificial Intelligence, pages 121--125, Austin, USA, August 1984.

....so in the light of what other agents in his scheme were likely to choose themselves in various circumstances. But this ability still only covers the situation in which one agent is superordinate to all others in deciding which goals to pursue. While systems that model such situations, such as [56, 97], are truly multi agent as they allow agents to rely on other agents to solve goals independently, they make no attempt at plan recognition. The process of plan recognition requires that agents come to believe potential causes for the actions they observe other agents performing. This is the ....

M. P. Georgeff. A theory of action for multiagent planning. In Proceedings of AAAI84, 1984.

.... 1988; Kraus, 1996; Kraus et al. 1995; Rosenschein and Zlotkin, 1994; Smith, 1980) However, the main focus of DAI research from a logical perspective has been that of planning for multiple agents , which considers issues inherent in centrally directed multi agent execution (Georgeff, 1983; Georgeff, 1984; Georgeff, 1986; Rao et al. 1992; Rosenschein, 1982) We propose a logical framework to represent an incentive negotiation procotocol about task distribution in bilateral Multi Agent Systems. We first extend Bell s logic (Bell, 1995) to represent autonomous agents by adding utilities to his ....

Georgeff, M. (1984). A theory of action for multiagent planning. In Proc. AAAI-84, pages 121--125, Los Altos, CA. Morgan Kaufmann.

....and rules determining what interactions can or might hold among asynchronously executing plans are proven to support the construction of sound and complete coordination mechanisms for concurrent hierarchical planning agents. Introduction The study of concurrent action in relation to planning (Georgeff 1984) has improved our understanding of how agents can reason about their interactions in order to avoid conflicts during concurrent plan execution. Conflicts can be avoided by reducing or eliminating interactions by localizing plan effects to particular agents (Lansky 1990) and by merging the ....

....hierarchical plans. However, we also wanted the model to share important aspects of plans used by PRSs, HTNs, Allen s temporal plans, and many STRIPS style plan representations. As such, this theory of action tries to distill appropriate aspects of other theories, including (Allen Koomen 1983) (Georgeff 1984), and (Fagin et al. 1995) CHiPs A concurrent hierarchical plan p is a tuple hpre, in, post, type, subplans, orderi. pre(p) in(p) and post(p) are sets of literals (v or :v for some propositional variable v) representing the preconditions, inconditions, and postconditions defined for plan p. ....

Georgeff, M. P. 1984. A theory of action for multiagent planning. In Proc. AAAI, 121--125.

....and rules determining what interactions can or might hold among asynchronously executing plans are proven to support the construction of sound and complete coordination mechanisms for concurrent hierarchical planning agents. Introduction The study of concurrent action in relation to planning (Georgeff 1984) has improved our understanding of how agents can reason about their interactions in order to avoid conflicts during concurrent plan execution. Conflicts can be avoided by reducing or eliminating interactions by localizing plan effects to particular agents (Lansky 1990) and by merging the ....

....hierarchical plans. However, we also wanted the model to share important aspects of plans used by PRSs, HTNs, Allen s temporal plans, and many STRIPS style plan representations. As such, this theory of action tries to distill appropriate aspects of other theories, including (Allen Koomen 1983) (Georgeff 1984), and (Fagin et al. 1995) CHiPs A concurrent hierarchical plan p is a tuple hpre, in, post, type, subplans, orderi. pre(p) in(p) and post(p) are sets of literals (v or :v for some propositional variable v) representing the preconditions, inconditions, and postconditions defined for plan p. ....

Georgeff, M. P. 1984. A theory of action for multiagent planning. In Proc. AAAI, 121--125.

....have been fully elaborated [60] Developing or refining the recipe is a complex activity because of the inter dependencies which exist between the agents actions. Because of this complexity a number of different planning paradigms, each with their own benefits and drawbacks, have been devised [12, 21, 25, 30] for example, the recipe may be generated before any action has started or it may be interleaved with execution, the recipe may be devised by one agent or by a group of agents, and so on. An important facet of the desired cooperation model is that it specifies criteria against which joint ....

M. P. Georgeff, A Theory of Action for Multi-Agent Planning, in: Proc. National Conf. on AI, Austin, USA, (1984) 125-129. 55

.... a given problem and integrate the results [21,20,30,24,114,36] Other lines of inquiry concerns planning for task allocation, so that effective execution will result [141,95] centralized planning for multi agent execution [113,67] or centralized planning to avoid execution time action conflicts [50,51]. Another area of research has concentrated on multiple agents mental attitudes for coordinating their activities [16,150,55,154] Another area of investigation has focused on multiple agents each of which is self motivated, i.e. has its own goals, which could be in conflict with the goals ....

M. A. Georgeff. Theory of action for multi-agent planning. In Proceedings of AAAI-84, pages 121--125, Austin, TX, 1984. AAAI. 80

....By taking actions, agents achieve their goals and fulfill some functionality for which they are responsible. Researchers have developed several process models (or action theories) to represent and reason about actions, including sequencing, selection, non determinism, iteration, and concurrency (Georgeff, 1984). By actions, we mean agents behaviors including those that are not externally observable. Internal agent activities are as important as external ones and should be considered in the design of an agent. Such activities include reasoning about behaviors, planning, resolving conflicts, and decision ....

Georgeff, M. P. 1984. A Theory of Action for Multi-Agents Planning. In Proceedings of the Proceedings of 1984 Conference of the American Association for Artificial Intelligence, 121-125.

....By taking actions, agents achieve their goals and fulfill some functionality for which they are responsible. Researchers have developed several process models (or action theories) to represent and reason about actions, including sequencing, selection, non determinism, iteration, and concurrency (Georgeff, 1984). By actions, we mean agents behaviors including those that are not externally observable. Internal agent activities are as important as external ones and should be considered in the design of an agent. Such activities include reasoning about behaviors, planning, resolving conflicts, and ....

Georgeff, M. P. 1984. A Theory of Action for Multi-Agents Planning. In Proceedings of the Proceedings of 1984 Conference of the American Association for Artificial Intelligence, 121-125.

.... requirements[177] Processes Goal based design[123] Goal based requirements negotiation[220] Program slicing[106] Goal regression for requirements [215] 266] Schema integration[ 11] Inconsistency dialog[76] Inconsistency reasoning[189] Inconsistency framework[40] Multi agent planning[85] Agent negotiation[238] A Historical Perspective of Requirements Interaction Management 13 1999 William N. Robinson Requirements Interaction Management GSU CIS 99 7 Reasoning about requirements goals has evolved concurrent with the evolution of software development goals. Multiple Attribute ....

....the agent knowledge base represents the requirement viewpoint description. Thus, when DAI agents interact to complete shared tasks, their representation and reasoning is similar to that found in the integration of multiple requirement viewpoints. DAI agents can cooperative to complete shared plans[85]. If they reach an inconsistent state, they may cooperatively negotiate to satisfy other plans[38] To do so, they may use economic models to guide their decision making in order to efficiently manage their resources[238] Such analysis can be applied to integrate multiple requirement ....

[Article contains additional citation context not shown here]

Georgeff, M. P., A theory of action for multiagent planning, Proceedings of 1984 conference of the AAAI, Morgan Kaufmann Publishers, 1984, pp. 121-125.

....Here, the structural properties of the domain play an outstanding role in determining which parts of the domain must be assigned to each agent so that their actions interfere minimally. Previous works on multi agent planning have mainly focused on the definition of new logic action theories [12][13] 14] 15] as well as negotiation strategies between agents [16] 17] Despite few notable exceptions, general action theories have scarcely produced practical multi agent planning systems. As an alternative, some temporal planners such as [18] 19] 20] 21] 22] have treated some restricted ....

M. Georgeff "A Theory of Action for Multi-agent Planning" in Proc. AAAI-84, Austin (USA), 1984, pp 121-125.

....an action or a sequence of actions that trigger events and change certain states. As far back as the 70s, researchers have developed several process models (or action theories) to represent and reason about actions, including sequencing, selection, non determinism, iteration, and concurrency [15]. Usually the term action refers to agents activities that can be observed as their external behaviors. When we model the whole problem solving activity, there are certain activities The Laboratory for Intelligent Processes and Systems TR99 UT LIPS AGENTS 01 Copyright The University of Texas at ....

M. P. Georgeff, "A Theory of Action for Multi-Agents Planning," 1984 Conference of the American Association for Artificial Intelligence, 1984.

....kinds of planning. We mention only some selected works relating TL and planning. TL approaches to be distinguished from temporal reasoning in general, which all planners must perform to some degree. One of the first applications of classical techniques in planning was identified by Georgeff [12]. Georgeff proposed process models to represent how the actions were selected by different agents. He also suggested the use of model based as opposed to axiomatic techniques for reasoning about the relationships among process models. Singh [26] and Belnap Perloff [3] define operators, but do ....

....above. We have taken some early steps in formalizing plans for multiagent environments. Much of our technical development considers the plans of a single agent, but by allowing branching and nondeterminism, it enables planning in multiagent environments. In this way, it is similar in style to [12, 13]. We leave it to future research to formalize ideas from cooperative planning to the same level of detail as carried out above. Acknowledgments The author benefited greatly from some early discussions with Allen Emerson, and from meticulous comments by the anonymous reviewers. ....

Michael P. Georgeff. A theory of action for multiagent planning. In Proceedings of the National Conference on Artificial Intelligence, pages 121--125, 1984.

.... design only our own intelligent agent, how should it interact with other intelligent agents Therefore, areas of interest in MAS research include game theoretic analysis of multi agent interactions (cf. GGR86, RG85, RZ94] reasoning about other agents goals, beliefs, and actions (cf. Geo83, Geo84, Ros82] and analysis of the complexity of social interactions [ST92] The influence of DAI on cooperative robotics has been limited. This is in part because researchers in DAI have mostly concentrated on domains where uncertainty is not as much of an issue as it is in the physical world. Work ....

M. Georgeff. A theory of action for multi-agent planning. In Proc. AAAI, 121--125, 1984.

....resolution of conflicts. Participating agents need to augment their individual intentions to comply with those of others because actions performed by different agents are intertwined and therefore need to be synchronised. There are several paradigms which may be used to generate the common plan [7, 31, 32, 33] it may be undertaken before any action has been started or interleaved with execution, it may be carried out by one agent or in a collaborative fashion. Assessment criteria and the concomitant causal link to behaviour are especially important when agents are situated in changing and complex ....

M. P. Georgeff, A Theory of Action for Multi-Agent Planning, Proc. of Second National Conference on Artificial Intelligence, Austin, USA, (1984) 125-129.

....fall with respect to three important axes, ability for global coordination, communication, and autonomy. In general, the closer to the origin a system lies, the better off it is with respect to the problem of the commons. Some approaches, such as multiagent planning [e.g. Cammarata et al. . 1983; Georgeff, 1984], would seem to be fairly safe; however, as we discuss below, there are important caveats. Low Global Control Low Communication between Peers High Autonomoy CNET FA C MAP AUV Figure 2: Relation of several DAI approaches systems with respect to three important characteristics. CNET=contract net ....

....the requisite information to identify common pool resources before their management becomes a problem. Existing DAI solutions. Some existing approaches in DAI can potentially mitigate the problem of managing common pool resources. Multiagent planning approaches [e.g. Cammarata et al. . 1983; Georgeff, 1984] would seem to be one solution to the tragedy of the commons. In these systems, one agent is designated as planner; it then makes resource allocation decisions for all the others and serves as arbiter should problems arise. Unfortunately, multiagent planning approaches have severe problems in ....

Georgeff, M. P. (1984). A theory of action for multi-agent planning. In Proceedings AAAI-84, pages 121--125.

....agents. Activity in multi agent worlds often requires agreement by the agents as to how they will act, and the reaching of consensus is a major concern of DAI. The formation of multi agent plans has been approached in several different ways: through the use of synchronization techniques [ Georgeff, 1984 ] such as those used in operating systems, through the distribution of single agent planners [ Corkill, 1979 ] such as This research was partially supported by the Israeli Ministry of Science and Technology (Grant 032 8284) NOAH, and through centralized planners that ensure coordination [ ....

M. Georgeff. A theory of action for multiagent planning. In Proceedings of the National Conference on Artificial Intelligence, pages 121--125, Austin, Texas, August 1984.

.... areas of interest in MAS research include game theoretic analysis of multi agent interactions (cf. Genesereth et al. 1986, Rosenschein and Genesereth, 1985, Rosenschein and Zlotkin, 1994] reasoning about other agents goals, beliefs, and actions (cf. Rosenschein, 1982, Georgeff, 1983, Georgeff, 1984] and analysis of the complexity of social interactions [Shoham and Tennenholtz, 1992] Work in MAS has tended to be theoretical and in very abstract domains. A common underlying assumption is that although the agents may be selfish, they are rational and highly deliberative. This is in stark ....

M. Georgeff. A theory of action for multi-agent planning. In Proc. AAAI, pages 121--125, 1984.

....a local node plan. Planning: Planning is often a coordination behavior (sometimes coordination with elements of the external environment) Bond and Gasser break it into two finer gradations multiagent planning, where one agent plans for several (for example, Georgeff s work [Georgeff, 1983, Georgeff, 1984] and distributed planning, where plans are constructed mutually by the agents (for example, PGP [Durfee and Lesser, 1987] distributed NOAH [Corkill, 1979] or Ephrati s work [ Ephrati and Rosenschein, 1994] See also planning in reference to open systems, Section 2.2.3. The only work on ....

Michael Georgeff. A theory of action for multiagent planning. In Proceedings of the Fourth National Conference on Artificial Intelligence, pages 121--125, August 1984.

No context found.

M. Georgeff. A theory of action for multi-agent planning. In Proceedings of the National Conference on Artificial Intelligence, pages 121--125, Austin, Texas, August 1984.

No context found.

Georgeff, M., "A Theory of Action for Multi-Agent Planning", In Proceedings of AAAI-84, pp121-125, Austin, 1984.

No context found.

REFERENCES 149 Georgeff, Michael P. 1984. A theory of action for multi agent planning. Pages 121--125 of: AAAI'84.

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