B. Burmeister and K. Sundermeyer. Cooperative problem solving guided by intentions and perception. In Werner and Y. Demazeau, editors, Decentralized AI III. Elsevier Science, 1992.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....tool, the system performance, as well as the experience gained during the implementation phase. 5. 1 Multi Agent Tools and a Test Run The scenario depicted in section 2 and the architecture described in the previous section have been implemented in the DASEDIS testbed for multi agent systems [BS92]. The main advantage of this testbed is the built in agent architecture. This architecture represents the actoric, sensoric, communicative, intentional and cognitive aspects of an agent in separate modules. Except for the module COGNITION, representing the cognitive aspects, all other modules are ....

B. Burmeister, K. Sundermeyer: "Cooperative Problem-Solving Guided by Intentions and Perception", in: E. Werner, Y. Demazeau (eds.), Decentralized A.I. 3, North-Holland, 1992, pp. 77-82

....n action intention at a situation, could be a conditional plan, as in presence of sensing and incompleteness, the unfolding may produce a conditional plan. We discuss this correspondence in further detail in [BG97b] There has been a lot of work on interacting agents [Fis93, Jen92, SBK 93, BS92] Although, we allow exogenous actions, we have not touched upon multiple cooperating agents in this paper. We hope to expand our ideas in this paper to such a case. We believe that in our future work along these lines we will be using the ideas of multi agent planning [Lan89] and multi agent ....

B. Burmeister and K. Sundermeyer. Cooperative problem solving guided by intentions and perception. Decentralized AI, 3, 1992.

....participate in a team planning phase to decide who will do what, nor do they have a principled model of cooperation on which to base their behaviour. This organisation corresponds to a typical GRATE community, it is also similar to several agent architectures based solely on individual intentions [9, 11] and to the seminal Distributed Vehicle Monitoring Testbed in which the organisational knowledge guiding the cooperating knowledge sources is specified as a set of interest areas [50] In our implicit community, the agents form individual intentions (related to their own actions) and plan which ....

B. Burmeister and K. Sundermeyer, Cooperative Problem Solving Guided by Intentions and Perception, in: E. Werner and Y. Demazeau, eds., Decentralised AI 3, (Elsevier, 1992), 77-92.

.... methodologies (including arrangements of data, algorithms, and control and data ows) for building agents has been described in the literature (see, e.g. 24, 37, 57] for an overview) Standard examples that indicate the broad structural and functional range of existing architectures are COSY [7], GRATE [28] INTERRAP [16] and TouringMachines [14] see Figures 1, 2, 3, and 4 for illustrations. Existing architectures are typically built of logically separate parts (usually called modules, components, or layers) each responsible for generating one of the three abilities reactivity, ....

B. Burmeister and K. Sundermeyer. Cooperative problem-solving guided by intentions and perception. In E. Werner and Y. Demazeau, editors, Decentralized A.I. 3, pages 77-92. North-Holland/Elsevier, Amsterdam et al., 1992.

.... Currently, there has been a growing interest in developing interoperable software agents which combine cognitive abilities to facilitate understanding of complex real situations, and thereby behave appropriately [9,10,21] The related issues which are under active research are : co operation [2,13,29,31], negotiation [16,17,33] collaboration [12, 14, 18, 19] planning [8, 11,22] and co ordination [30] Of late, another issue which has been studied in connection with agent interactions is influencing [5] The purpose of influencing is to make an agent adopt or abandon a goal. In [5] influencing ....

B. Burmeister and K. Sundermeyer, "Cooperative problem-solving guided by intentions and perception", in Decentralized AI.3, E. Werner and Y. Demazeau (Eds.), Elsevier Sc. pub., 1992, pp 77-92.

....n action intention at a 57 situation, could be a conditional plan, as in presence of sensing and incompleteness, the unfolding may produce a conditional plan. We discuss this correspondence in further detail in [BG97b] There has been a lot of work on interacting agents [Fis93,Jen92,SBK 93,BS92] Although, we allow exogenous actions, we have not touched upon multiple cooperating agents in this paper. We hope to expand our ideas in this paper to such a case. We believe that in our future work along these lines we will be using the ideas of multi agent planning [Lan89] and multi agent ....

B. Burmeister and K. Sundermeyer. Cooperative problem solving guided by intentions and perception. Decentralized AI, 3, 1992.

....a decentralized control approach. All the agents have been designed using the same model. The latter had to cover both the different processing problems of the high and low levels of representation of a vision system. The defined agent model (cf. Fig 2) is a cognitive one inspired by [15] 13] [7]. It is composed of two submodels (cf. Fig. 2) the individual control model (in white on Fig 2) 4] and the social control model (in gray on Fig 2) 4] It is defined by the following 10 uplet : Agent = P S; RS; DS; CS; ES; KR; L; P r; Or; PM Different representation states (shown as ....

B. Burmeister and K. Sundermeyer. Cooperative problem solving guided by intentions and perception. in Decentralized Artificial Intelligence, Elsevier Science Publishers B.V. (North Holland), Y. Demazeau and J.P. Muller editors, p 77-92, 1992.

....The final transitions of the plan graph may be labelled with actions to be taken upon the success, failure or aborting of the plan. 5 Comparison and Conclusions A number of agent oriented systems based on BDI architectures have been developed, including PRS [ Georgeff and Lansky, 1986 ] COSY [ Burmeister and Sundermeyer, 1992 ] GRATE [ Jennings, 1993 ] ARCHON [ Jennings et al. 1995 ] INTERRAP [ Muller et al. 1995 ] and dMARS [ Kinny, 1993 ] Likewise, agent oriented languages, such as AGENT0 [ Shoham, 1991, Shoham, 1993 ] and PLACA [ Thomas, 1995 ] have been proposed as languages for programming ....

B. Burmeister and K. Sundermeyer. Cooperative problem-solving guided by intentions and perception. In E. Werner and Y. Demazeau, editors, Decentralized A.I. 3, Amsterdam, 1992. North Holland.

....our framework. 1 Introduction BDI architectures and theories based on the attitudes of belief, desire, and intention have attracted a lot of attention as appriopriate models of inteligent multi agent systems; see Singh [18] Rao Georgeff [15,17] Wooldridge [20] Muller [10] Sundermeyer et al. [5], and Jennings [9] Recently, a number of attempts has been made in order to formalize these mental attitudes and to show how they influence the actions of agents [18] 15] Most of these formalisms concentrate on the specification or characterization of rational multi agent systems using ....

....reality that it can be almost directly implemented after specifying the states and actions of the asynchronous automaton. Thus, it is somewhere in between Singh [18] Rao Georgeff [15] Wooldridge [20] formalisms and BDI architectures of Muller [10] Rao Georgeff [17] Sundermeyer et al. [5], and Jennings [9] There are no global states of MAS nor global time in our framework, so that local interactions are efficiently captured in the framework. This may be seen as an advantage of our approach in comparison with the above mentioned approaches that are based on global states. The ....

B. Burmaister and K. Sundermeyer. Cooperative problem-solving guided by intentions and perception. In Y. Demazeau and E. Werner, editors, Decentralized A.I., volume 3. North-Holland, 1992.

....system is a collection of agents 1 that continuously interact with the environment and one another according to their informational, motivational, and deliberative states. A number of such systems have been built and applied to complex application domains, such as traffic management [ Burmeister and Sundermeyer, 1992; Ljungberg and Lucas, 1992 ] space shuttle fault diagnosis [ Ingrand et al. 1992 ] telecommunications network management [ Ingrand et al. 1992 ] and air combat modelling [ Rao et al. 1992a ] However, there are no tools or methods available for formally verifying that these systems ....

B. Burmeister and K. Sundermeyer. Cooperative problem-solving guided by intentions and perception. In E. Werner and Y. Demazeau, editors, Decentralized A.I. 3, Amsterdam, The Netherlands, 1992. North Holland.

....to reactive recognition can be found elsewhere [ 19 ] In spite of the initial skepticism towards reactive planning, the approach has been quite successful compared to classical planning. This is due to the fact that, in a substantial class of problem domains (such as road traffic management [ 3 ] , space shuttle diagnosis [ 8 ] air traffic management [ 14 ] and air combat modelling [ 18 ] execution of actions and decision making tasks can be analyzed and codified as plans, in a relatively simple manner. These domaindependent plans can then be effectively used by an agent to react in ....

B. Burmeister and K. Sundermeyer. Cooperative problem-solving guided by intentions and perception. In E. Werner and Y. Demazeau, editors, Decentralized A.I. -- Proceedings of the Third European Workshop on Modelling Autonomous Agents and Multi-Agent Worlds, Amsterdam, The Netherlands, 1992. Elsevier Science Publishers.

.... multi modal, temporal, action, and dynamic logics have been used to formalize some of these notions [2, 6, 8, 13, 18, 20, 21] The complexity of theorem proving and the completeness of these logics have not been clear [12, 23] On the other hand, there are a number of implementations of BDI agents [1, 3, 10, 17] that are being used successfully in critical application domains. These implementations have made a number of simplifying assumptions and modelled the attitudes of beliefs, desires, and intentions as data structures. Also, user written plans or programs speed up the computation in these systems. ....

....is an object oriented analogue of AgentSpeak(L) AgentSpeak(L) is a textual and simplified version of the language used to program the Procedural Reasoning System [3] and its successor dMARS. These implementations have been in use since the mid 1980s. Other agent oriented systems, such as COSY [1], INTERRAP [10] and GRATE [7] have been built based on the BDI architecture. The formal operational semantics given here could apply to some of these systems as well. However, a more thoroughanalysis of these systems and their relation to AgentSpeak(L) is beyond the scope of this paper. ....

B. Burmeister and K. Sundermeyer. Cooperative problem-solving guided by intentions and perception. In E. Werner and Y. Demazeau, editors, Decentralized A.I. 3, Amsterdam, The Netherlands, 1992. North Holland.

....Also their proposal concentrates on defining the behaviour of an individual in an asocial situation. There are also no provisions for collaborative activity and no joint intentions. Burmeister and Sundermeyer have defined and implemented a generic agent model based upon intentions and desires [44]. Their agents are characterised by their perceptive capabilities, the actions they can undertake, their intentions and the resources needed for executing the different types of behaviour. They identify two types of intention long term ones which correspond to high level objectives and desires ....

B. Burmeister and K. Sundermeyer, Cooperative Problem Solving Guided by Intentions and Perception, in Decentralised AI 3, eds. E. Werner and Y. Demazeau (North Holland Publishers, 1992) 77-92.

.... or optimal performance when deliberation is subject to resource bounds [ Bratman, 1987; Kinny and Georgeff, 1991 ] While much work has gone into the formalization [ Cohen and Levesque, 1990; Jennings, 1992; Kinny et al. 1994; Rao and Georgeff, 1991c; Singh and Asher, 1990 ] and implementation [ Burmeister and Sundermeyer, 1992; Georgeff and Lansky, 1986; Muller et al. 1994; Shoham, 1993 ] of BDI agents, two main criticisms have been levelled against these endeavours. First, the having of these three attitudes is attacked from both directions: classical decision theorists and planning researchers question the necessity ....

....rational behaviours required for that application. As a result, following the modal logic tradition, we have discussed how one can categorize different combinations of interactions between beliefs, desires, and intentions. A number of agent oriented systems have been built in the past few years [ Burmeister and Sundermeyer, 1992; Georgeff and Lansky, 1986; Muller et al. 1994; Shoham, 1993 ] However, while many of these appear interesting and have different strengths and weaknesses, none have yet been applied to as wide a class of complex applications as the ones discussed in this paper. Currently, there is very little ....

B. Burmeister and K. Sundermeyer. Cooperative problem-solving guided by intentions and perception. In E. Werner and Y. Demazeau, editors, Decentralized A.I. 3, Amsterdam, The Netherlands, 1992. North Holland.

....utiliser le terme de Protocole d Interaction est pour distinguer clairement ces protocoles des protocoles de communication standard qui existent dans les syst emes distribu es. 2 Vers un cadre unifi e pour l echange de connaissances 2. 1 Le Langage d Interaction Comme cela est indiqu e dans [3], si la communication est simplement d ecrite en termes d emission et de r eception de messages, chaque agent doit etre capable d inf erer l intention de l emetteur lorsqu il enonce le message. Si les messages ne sont pas structur es, cette inf erence peut etre tr es difficile, voire impossible. ....

B. Burmeister and K. Sundermeyer. Cooperative problem solving guided by intentions and perception. In Werner and Yves Demazeau, editors, Decentralized AI III. Elsevier Science, 1992.

....future planning and in which an override mechanism provides a computational realisation of conventions. This proposal is predominantly for an individual agent situated in an asocial context, but it provides a useful functional architecture which can be augmented for use in a cooperative context. Burmeister and 21 Sundermeyer (1992) specify and implement an architecture based on individual intentions which takes into account the fact that agents are situated in multi agent environments. Their representation of intention, although not based on a particular theoretical model, does include the resources required by a commitment ....

Burmeister, B, and Sundermeyer, K, 1992. "Cooperative Problem Solving Guided by Intentions and Perception" In E. Werner and Y. Demazeau (eds.) Decentralised A.I. 3, Elesevier, pp 77-92.

....actions. These actions may change the state of the environment. For example, if move is an action symbol, the robot moving from lane X to lane Y, written as move(X,Y) is an action. This action results in an environmental state where the robot is in lane Y and is no longer in lane X. Definition 4 If a is an action symbol and t 1 , t n are first order terms, then a(t 1 , t n ) or a( t) is the execution of an action. Definition 5 Goals and actions are collectively referred to as simple activities. We consider two types of complex activities: an AND activity and an OR activity. An ....

....as well as different types of plans as mental attitudes execution intention, recognition intention, joint execution intention, and joint recognition intention. The success of execution plans and execution intentions in a substantial class of problem domains (such as road traffic management [ 4 ] space shuttle diagnosis [ 9 ] air traffic management [ 14 ] and aircombat modelling [ 16, 22 ] and the early trials of using recognition plans and joint execution plans in the air combat modelling domain, offers us hope that this unified view of plans is likely to have a substantial ....

B. Burmeister and K. Sundermeyer. Cooperative problem-solving guided by intentions and perception. In E. Werner and Y. Demazeau, editors, Decentralized A.I. 3, Amsterdam, The Netherlands, 1992. North Holland.

....which goals to pursue, how best to achieve them given the limited resources it has available, and what to do in the case of goal conflict. In this paper we will focus on a particular approach to software agency that has achieved considerable maturity, the Belief Desire Intention (BDI) architecture [16, 15, 11, 27, 6, 31, 25, 28]. This architecture has its origins in the study of mental attitudes; namely, Beliefs, Desires and Intentions that represent, respectively, the agent s information, motivational, and deliberative state. These mental attitudes directly determine the system s behaviour and are critical for achieving ....

B. Burmeister and K. Sundermeyer. Cooperative problem-solving guided by intentions and perception. In Proceedings of the Third European Workshop on Modelling Autonomous Agents and Multi-Agent Worlds, MAAMAW '91, Kaiserslautern, Germany, 1991.

.... (running on the same or on a separate machine) through a suitable Agent Communication Language (ACL) Genesereth and Ketchpel, 1994] That is, we assume an intuitive general software agent model shown in Figure 1, loosely based on the more formal one described by Burmeister and Sundermeyer [Burmeister and Sundermeyer, 1992]. In our model each isolated entity possesses one or more conveniently represented information repositories. Using that information the agent may perform processing activities by means of opportune elaboration and control components, without any coordination with other agents. The information ....

Burmeister, W. and Sundermeyer, K. (1992). Cooperative problem-solving guided by intention and perception. In Proceedings of the Third European Workshop on Modelling Autonomous Agents in a Multi-Agent World, pp. 77-92, (Werner and Demazeau Eds.), Kaiserlautern, Germany.

....system agents to carry out functions such as I O, and finally a set of application specific system agents each of which carries out a task and contains some knowledge about the external world. DASEDIS (Development and Simulation Environment for Distributed Intelligent Systems) [5] is a multi agent testbed centred around the idea of long and short term intentions. The latter type are directly bound to actions. There are seperate components in each agent for performing local and social actions. Communication is by message passing. All DASEDIS agents have sensors to receive ....

B. Burmeister and K. Sundermeyer. Cooperative Problem Solving Guided by Intentions and Perception. In Decentralized A.I. 3 - Proceedings of MAAMAW'91, pages 77--92. North-Holland, 1991.

No context found.

B. Burmeister and K. Sundermeyer. Cooperative problem solving guided by intentions and perception. In Werner and Y. Demazeau, editors, Decentralized AI III. Elsevier Science, 1992.

No context found.

B. Burmeister and K. Sundermeyer. Cooperative problem-solving guided by intentions and perception. In E. Werner and Y. Demazeau, editors, Decentralized A.I. 3, Amsterdam, The Netherlands, 1992. North Holland.

No context found.

B. Burmeister and K. Sundermeyer. Cooperative problem-solving guided by intentions and perception. In E. Werner and Y. Demazeau, editors, Decentralized A.I. 3, Amsterdam, The Netherlands, 1992. North Holland.

No context found.

B. Burmeister, K. Sundermeyer, "Cooperative pRoblem Solving Guided by Intentions and Perception", Decentralized AI III, Werner & Demazeau eds, Elsevier Science Publishers B.V., 1992.

No context found.

Burmeister & al," cooperative problem solving guided by intentions and perceptions", Decentralized AI, 1992.

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