F. F. Ingrand, M. P. Georgeff, and A. S. Rao. An architecture for real-time reasoning and system control. IEEE Expert, 7(6), 1992.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....action selection and sensing. One concern one might have is that once we move to on line execution where nondeterministic choice defaults to being random, we have given up reasoning about courses of action, and that our programs are now just like the pre packaged plans found in rap [3] or prs [11]. Indeed in those systems, one normally does not search off line for a sequence of actions that would eventually lead to some future goal; execution relies instead on a user supplied plan library to achieve goals. In our case, with Sigma , we get the advantages of both worlds: we can write ....

F. F. Ingrand, M. P. Georgeff, and A. S. Rao. An architecture for real-time reasoning and system control. IEEE Expert, 7(6), 1992.

....The creation of groups of virtual humans capable of behaving and interacting realistically with each other in 3D environments requires the development of a specific agent architecture. Most current agent architectures are dedicated to the fulfilment of precise tasks problem solving, e.g. Ingrand [ 17], or to the simulation of purely emotional behaviour, e.g. Botelho [18] However, we think that pure goal oriented and emotional behaviour are not sufficient in order to fully model human interaction in groups. In real life, we daily engage in many interactions and social activities with ....

Ingrand, F.F., Georgeif, M.P., and Rao, A.S. (1992). An architecture for real-time reasoning and system control. IEEE Expert/Intelligent Systems, 7(6):34-44.

....a sociological perspective, and proposing a high level architecture based on Roles, Norms, Values and Types. Keywords: Socially Intelligent Virtual Agent, Socially Believable Agents 1. Introduction Most agent architectures are dedicated to the fulfillment of precise tasks problem solving, e.g. [17], or to the simulation of purely emotional behavior, e.g. 5] However, pure goal oriented and emotional behavior only correspond to a small part of our daily activities as human beings. Everyday, we engage in many interactions and social activities, adapting our behavior to the situation, dealing ....

....The BDI architecture has proved to be a useful abstraction to model autonomous agents. Its components (Belief, Desire, Intention) offer a convenient and intuitive way to structure the agent s action selection. Systems built on this architecture have produced interesting results (e.g. dMARS [17]) Nonetheless, we agree with Balzer [1] when he points out that this architecture is deficient for the simulation of social behavior. Based on the Bayesian decision theory, it does not take into account the socially situated nature of human behavior and doesn t allow the inclusion of such ....

Ingrand F.F, Georgeff M.P, Rao A.S. An architecture for real-time reasoning and system control. IEEE Expert/Intelligent Systems 1992; 7(6); 34--44

....on recognising situations where these linkages exist in a plan and controlling plan interleaving to ensure that an agent does not allow negative interference in the pursuit of separate parallel goals. We provide detailed mechanisms that can be easily implemented in agent platforms such as PRS [Ingrand et al. 1992] , JAM [Huber, 1999] dMARS [d Inverno et al. 1998] and JACK [Busetta et al. 1998] There has been significant work in the area of conflicts in agent systems [Tessier et al. 2000] However, the focus has been on multi agents and identifying various types of conflicts. Our work focuses ....

.... in a goal plan tree instance (e.g. in figure 1 # ######### is the instance name for plan ) The execution cycle of the agent is similar to the well known and developed BDI (Belief Desire Intention) Rao and Georgeff, 1995] style of agents that map to agent implementation systems such as PRS [Ingrand et al. 1992] , dMARS [d Inverno et al. 1998] and JACK [Busetta et al. 1998] where a plan is selected from an applicable plan set and if it fails an alternative applicable plan is tried if available. We extend this model by requiring that a goal with an in condition that is false is delayed until the ....

F. F. Ingrand, M. P. Georgeff, and A. S. Rao. An architecture for real-time reasoning and system control. IEEE Expert, 7(6), 1992.

....in oce hotelling environments, where the owner of an oce can change on a daily basis, the smart space needs to support a diverse set of personal applications and mobile devices. In previous work, real time system constraints have been incorporated in blackboard based intelligent control systems [7, 8, 9]. These systems focus on reasoning and planning with predictable delays. For distributed systems, Blair et al. describe an architecture for QoS support in a tuplespace model [10] It emphasizes adaptation to a changing environment through a specialized set of agents. In this paper, we focus on ....

F. F. Ingrand, M. P. George, and A. S. Rao. An architecture for real-time reasoning and system control. IEEE Expert, 7(6):34-44, 1992.

....an input port or to inhibit an output port. There are no components like plans or procedures but it is possible to define macros and use them in the definition of behaviors. 2. 4 PRS Procedural Reasoning System (PRS) is a general framework designed as a so called situated reasoning system [14]. Such systems are used for diagnosing and taking necessary measures to handle plant and process malfunctions in real time. This requires reasoning about management of tasks which includes reasoning about the criticality or urgency of tasks, potential interactions between tasks, the execution ....

F. Ingrand, M. Georgeif, and A Rao. An architecture for real-time reasoning and system control. IEEE Expert, 7(6), 1992.

....are calculated and tried. If there are no applicable plans left, the goal fails. The assumptions made correspond to a class of agent systems. One particular type of agent system that fits in with these assumptions is those based on the Belief Desire Intention (BDI) model [9] For example, PRS [7], JAM [6] dMARS [5] and JACK [1] 2.1 Example Consider a rover robot which is deployed on the Martian surface. The robot obtains energy from solar panels and is given requests to perform various experiments (Exp) on various substances (A,B) The following are some resources arising naturally ....

F. F. Ingrand, M. P. Georgeff, and A. S. Rao, `An architecture for realtime reasoning and system control', IEEE Expert, 7(6), (1992).

....and directly execute it. 7 Related work There are two main strands of work to which ours is related work on executable agent architectures and work on multi context systems. As mentioned above, most previous work which has produced formal models of agent architectures, for example dMARS [14], Agent0 [26] and GRATE [15] has failed to carry forward the clarity of the specification into the implementation there is a leap of faith required between the two. Our work, on the other hand, maintains a clear link between specification and implementation promising to allow the direct ....

F. F. Ingrand, M. P. Georgeff, and A. S. Rao. An architecture for real-time reasoning and system control. IEEE Expert, 7, 34--44, 1992.

.... languages and update languages (see [2, 3] for a more detailed discussion of the relation between the two approaches) AgentSpeak(L) 25] is a logical language for programming Belief Desire Intention (BDI) agents, originally designed by abstracting the main features of the PRS and dMARS systems [19]. Our approach shares with AgentSpeak(L) the objective of using a simple logical specification language to model the execution of an agent, rather than employing modal operators. On the other hand, while AgentSpeak(L) programs are described by means of a proof theoretic operational semantics, our ....

F. F. Ingrand, M. P. George#, and A. S. Rao. An architecture for real-time reasoning and system control. IEEE Expert, 7(6), 1992.

....actually performed by the planner (on board or at the Operator s Station) upon request of the su pervisor. Note that in this architecture, on board planning is necessary only at the second level. It is essentially a refinement using domain or task specific knowledge. For this, we use C PRS [14] which provides a suitable framework for goal driven as well as situation driven deliberation processes. Indeed, PRS implements script (called KA in PRS) selection and goal posting mechanisms. Planning can be performed through context dependent goal decomposition; situation driven reaction can be ....

F. Ingrand, M.P. Georgeif, and A.S. Rao. An architecture for real-time reasoning and system control. IEEE Expert, Intelligent Systems and Their Applications, 7:pp.34 44, 1992.

....wishes to achieve) and intentions (selected courses of action) at its core. Such an architecture was chosen because it has a proven track record of working effectively in a range of complex applications including air traffic control [20] process control [21] simulation [22] fault diagnosis [23], and transportation [24] The specific BDI implementation used in this work is dMARS [25] Fig. 10) Each dMARS agent has a set of desires, goals, beliefs, plans, and intentions. It receives percepts from its operating environment and performs actions to affect its environment. Agents use ....

F. F. Ingrand, M. P. Georgeff, and A. S. Rao, "An architecture for real time reasoning and system control," IEEE Expert Mag., vol. 7, no. 6, 1992.

....is contained in [28] 5 Related Work There are two main strands of work to which ours is related work on executable agent architectures and work on multi context systems. As mentioned above, most previous work which has produced formal models of agent architectures, for example dMARS [16], Agent0 [30] and GRATE [17] has failed to carry forward the clarity of the specification into the implementation there is a leap of faith required between the two. Our work, on the other hand, maintains a clear link between specification and implementation through the direct execution of the ....

F. F. Ingrand, M. P. Georgeff, and A. S. Rao. An architecture for real-time reasoning and system control. IEEE Expert, 7(6):34--44, 1992.

....In the future, hard real time approaches for multiple distributed agents may be possible, but, currently, the complexity of the distributed agent control problem, particularly when agents have complex activities and are situated in dynamic and uncertain environments, prevents such approaches. PRS [16] and the more recent work on UMPRS [18] both o er architectures capable of operating e ectively in unpredictable domains. Like SRTA, PRS can use context to select from among alternative goal satisfaction plans, and its continuous reevaluation of these intentions allows it to be more responsive to ....

Francois F. Ingrand, Michael P. George, and Anand S. Rao. An architecture for real-time reasoning and system control. IEEE Expert, 7(6), December 1992.

....reactive autonomous systems and are an ideal application eld for agent technology. Some applications of agents in this eld are: a system to control the distribution of electricity [17] a system that monitors and diagnoses faults in nuclear power plants [84] a controller for spacecrafts [38], and a system for climate control [14] Telecommunications. Telecommunication systems are large and distributed networks of components to be monitored and managed in real time. Agents have been applied in this eld to face problems such as feature integration, network control, transmission and ....

F. F. Ingrand, M. P. George, and A. S. Rao. An Architecture for Real-Time Reasoning and System Control. IEEE Expert, 7(6), 1992.

No context found.

F. F. Ingrand, M. P. Georgeff, and A. S. Rao. An architecture for real-time reasoning and system control. IEEE Expert, 7(6), 1992.

No context found.

F. F. Ingrand, M. P. Georgeff, and A. S. Rao. An architecture for real-time reasoning and system control. IEEE Expert, 7(6), 1992.

No context found.

F. F. Ingrand, M. P. Georgeff, and A. S. Rao. An architecture for real-time reasoning and system control. IEEE Expert, 7(6), 1992.

No context found.

F. F. Ingrand, M. P. Georgeff, and A. S. Rao. An Architecture for Real-Time Reasoning and System Control. IEEE Expert: Intelligent Systems and Their Applications, 7(6):34--44,1992.

No context found.

F. F. Ingrand, M. P. Georgeff, and A. S. Rao. An architecture for real-time reasoning and system control. IEEE Expert, 7(6), 1992.

No context found.

F. F. Ingrand, M. P. George#, and A. S. Rao. An architecture for real-time reasoning and system control. IEEE Expert, 7(6):33--44, 1992.

No context found.

F. F. Ingrand, M. P. George#, and A. S. Rao. An architecture for real-time reasoning and system control. IEEE Expert, Knowledge-Based Diagnosis in Process Engineering, 7(6):33--44, December 1992.

No context found.

F.F. Ingrand, M.P. Georgeff, and A.S. Rao. An architecture for real-time reasoning and system control. IEEE Expert, 7(6):34--44, 1992.

No context found.

F.F. Ingrand, M.P. George#, and A.S. Rao. An architecture for real-time reasoning and system control. IEEE Expert, 7(6), 1992.

No context found.

F. F. Ingrand, M. P. Georgeff, and A. S. Rao. An architecture for real-time reasoning and system control. IEEE Expert, 7(6), 1992.

No context found.

F. F. Ingrand, M. P. Georgeff, and A. S. Rao. An architecture for real-time reasoning and system control. IEEE Expert, 7(6), 1992.

First 50 documents  Next 50

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