| Musliner, D. J.; Durfee, E. H.; and Shin, K. G. 1993. CIRCA: A cooperative intelligent real time control architecture. IEEE Transactions on Systems, Man, and Cybernetics 23(6):1561--1574. |
....measuring the di#erence of two UTC synchronised clocks, for example) whereas humans rely on a variety of calendar systems and intervals of variable duration such as evening , month , year and so on. 2. 3 Related Work There are several examples of real time agent architectures, both hard [6,3] and soft [10] Real time agent systems are systems in which agents are temporally constrained in achieving their goals. They combine real time (RT) and artificial intelligence (AI) characteristics by either embedding AI in RT, embedding RT in AI, or by adopting a co operative RT AI approach ....
D. Musliner, E. Durfee, and K. Shin. CIRCA: A Co-operative Intelligent Realtime Control Architecture. IEEE Transactions on Systems, Man, and Cybernetics, 23(6):1561--1574, 1993.
....practical applications have to perceive and affect the real world in a timely manner and have to become a lot more stable, more predictable, and more reliable. To attain these characteristics, they draw upon the real time field, giving rise to real time agent architectures and systems, both hard [3] and soft [4] Real time agent systems attempt to provide a solution by combining real time and AI agent characteristics. On the other hand, practical use of agent systems has seen their application in areas where there is extensive interaction with humans as well as societal or organisational ....
Musliner D., Durfee E., and Shin K., CIRCA: A Co-operative Intelligent Real-time Control Architecture, IEEE Transactions on Systems, Man, and Cybernetics, Vol. 23, No. 6, pp. 1561-1574, 1993.
....to SRTA would lead to greater quality stability, by allowing it to more frequently jump to a more appropriate set of activities. 6 Related Work It is important to note that the architecture presented here falls into the soft real time computation class. In contrast to architectures like CIRCA [26], we cannot make performance guarantees [31] about agent control. However, in contrast to CIRCA, the approach presented here operates on multiple distributed agents and the statistically fast enough model addresses the requirements of this application. In addition, action primitives are ....
David J. Musliner, Edmund H. Durfee, and Kang G. Shin. CIRCA: A cooperative intelligent real-time control architecture. IEEE Transactions on Systems, Man and Cybernetics, 23(6), 1993.
....this research, maximizing our ability to incorporate existing architectural capabilities as well as avoiding introduction of yet another member to the already overpopulated AI architectures cauldron. Instead, we built upon the Cooperative Intelligent Real time Control Architecture (CIRCA) 51] [53] which combines planning, scheduling, and plan execution to enable hard real time system response for plan execution, thus already addressing many of the integrated planning execution coordination issues demanded for a fully autonomous system. We have added new capabilities and relaxed several ....
....their success in real world applications. However, many of these architectures exhibit the coincidental real time response for which we are skeptical in safety critical systems. We conclude this chapter with a description of the Cooperative Intelligent Real time Control Architecture (CIRCA) 51] [53] which was explicitly designed to guarantee failure avoidance via integrated planning scheduling planexecution software. We have built upon the CIRCA architecture in this dissertation, thus we also discuss its limitations and preview how we address these issues in CIRCA II. Probabilistic Planning: ....
[Article contains additional citation context not shown here]
D. J. Musliner, CIRCA: The Cooperative Intelligent Real-Time Control Architecture. University of Michigan Department of EECS, CSE Division Technical Report, CSE-TR-175-93.
....of the environment s geometry in its symbolic layer and none in its subsumption or servo layers. This architecture has only been applied to navigation and so the lack of representation was reasonable. As stated previously, I believe stateless systems are not reasonable for general tasks. CIRCA [56] is an architecture concerned with making real time guarantees about its performance. The authors do not address the design of efficient representation systems, but 27 much like the RAP system, there is no reason why a well designed representation could not be used in this architecture. Certain ....
Musliner, D.J., Durfee, E.H. and Shin, K.G. 1993. CIRCA: A Cooperative Intelligent Real-Time Control Architecture. IEEE Transactions on Systems, Man and Cybernetics 23(6): 1561-1573.
....to control theory and work that attempts to bridge the gap between AI and control theory. In particular, subsumption based robots use behaviors described by finite state automata [2] and the circa system reasons about plans as finite state automata to construct provably safe control loops [13]. Discrete event system theory is directed toward building and understanding control plans for systems that can be described as finite state automata [16] These ideas have been used to build simple control systems for vision based robot navigation problems [10] The Animate Agent project uses an ....
David J. Musliner, Edmund H. Durfee, and Kang G. Shin. Circa: A cooperative intelligent real-time control architecture. IEEE Transactions on Systems, Man, and Cybernetics, 23(6), 1993.
....other hand the modeling is simpler and more realistic. However, the later versions of InteRRaP also allow separated world models, yielding this di erence non existent. Only a few of the architectures mentioned in this section address the issue of real time properties. Those that do, e.g. CIRCA [41]; usually adopt the fastenough attitude combined with anytime algorithms . A more recent architecture, explicitely addressing the hard real time requirements, is ARTIC [4] It is a modi cation of a blackboard architecture, done in the spirit of GUARDIAN [14] i.e. the hard real time response ....
D. J. Musliner. CIRCA: The Cooperative Intelligent Real-Time Control Architecture. PhD Thesis, The University of Michigan, 1993.
....task is progressively decomposed into smaller subtasks. Such hierarchical decomposition (Erol, Hendler, and Nau 1994) has been used extensively in plan generation systems. Additionally, a number of different AI systems (Firby 1987; Georgeff and Lansky 1987; Laird, Newell, and Rosenbloom 1987; Musliner, Durfee, and Shin 1993) have used this approach to execute tasks in complex, dynamic environments. Although these systems are successful, they are limited in a number of different ways. For example, each time a similar experience is encountered, all the knowledge in the decomposition must be brought to bear, taking ....
Musliner, D. J., E. H. Durfee, and K. G. Shin (1993). CIRCA: A cooperative intelligent real-time control architecture. IEEE Transactions on Systems, Man, and Cybernetics 23 (6), 1561--1574.
No context found.
Musliner, D. J., Durfee, E. H., and Shin, K. G. CIRCA: a cooperative intelligent real-time control architecture. IEEE Transactions on Systems, Man and Cybernetics 23, 6 (1993), 1561--1574.
No context found.
D. Musliner, E. Durfee, and K. Shin. CIRCA: A cooperative intelligent real time control architecture. IEEE Transactions on Systems, Man, and Cybernetics, 23(6):1561--1574, - 1993.
No context found.
David J. Musliner, Edmund H. Durfee, and Kang G. Shin. CIRCA: a cooperative intelligent real-time control architecture. IEEE Transactions on Systems, Man and Cybernetics, 23(6):1561--1574, 1993.
No context found.
D. J. Musliner, E. H. Durfee, and K. G. Shin. CIRCA: a cooperative intelligent real-time control architecture. IEEE Trans. Systems, Man, and Cybernetics, 23(6):1561--1574, 1993.
....and NASA (cooperative agreement NCC 1 399) 2 semantics of its execution environment. If the controller is found to be incorrect, information from the verification system is used to direct the search for improvements. Using the CIRCA architecture for adaptive real time control systems [6], these controllers are synthesized automatically and dynamically, on line, while the platform is operating. Unlike many other intelligent control systems, CIRCA s automatically generated control plans have strong temporal semantics and provide safety guarantees, ensuring that the controlled ....
D. Musliner, E. Durfee, and K. Shin. CIRCA: a cooperative intelligent realtime control architecture. IEEE Transactions on Systems, Man and Cybernetics 23(6):1561-1574.
....previously known elements in new ways are handled without further work by either CIRCADIA or the security engineer providing the input knowledge. 3. The Framework: Decision Theoretic CIRCA CIRCADIA is built on the foundation of the CIRCA architecture for intelligent real time control systems [12, 13] . CIRCA automatically synthesizes and executes hardreal time discrete event control systems for embedded applications. We provide a brief discussion of the CIRCA world model below. 3.1. The Original CIRCA World Model The CIRCA planner searches for a plan that is guaranteed to be safe, by ....
D. J. Musliner, E. H. Durfee, and K. G. Shin. CIRCA: a cooperative intelligent real-time control architecture. IEEE Trans. Systems, Man, and Cybernetics, 23(6):1561--1574, 1993.
....a missile of this type. SA CIRCA uses this information to ensure that its controllers monitor enemy radar lock ons frequently enough, and take countermeasures fast enough. SA CIRCA does not have unlimited resources at its disposal. In particular, it su#ers from the problem of bounded reactivity [6]: it can only monitor and react to a limited number of threats concurrently. To overcome this problem, SA CIRCA s Adaptive Mission Planner decomposes extended missions into multiple mission phases, each of which can have a di#erent controller tailored to its particular needs. See Figure 2. For ....
....are not faced with the problem of trading o# deliberation versus action. Because CIRCA was designed to act in hard real time environments, the CIRCA execution engine (the Real Time Subsystem) was designed to execute in parallel with the AI subsystem and not to compete for computational resources [6]. So CIRCA deliberation scheduling involves only tradeo#s between alternative planning activities, not between planning and action. 14 The first two simplifications mean that it is trivial to estimate the value of a single quantum of computation in our model. The simplifications, however, impose ....
D. J. Musliner, E. H. Durfee, and K. G. Shin, "CIRCA: A Cooperative Intelligent Real-Time Control Architecture," IEEE Transactions on Systems, Man and Cybernetics, vol. 23, no. 6, pp. 1561--1574, 1993.
....and NASA (cooperative agreement NCC 1 399) 2 semantics of its execution environment. If the controller is found to be incorrect, information from the veri cation system is used to direct the search for improvements. Using the CIRCA architecture for adaptive real time control systems [6], these controllers are synthesized automatically and dynamically, on line, while the platform is operating. Unlike many other intelligent control systems, CIRCA s automatically generated control plans have strong temporal semantics and provide safety guarantees, ensuring that the controlled ....
D. Musliner, E. Durfee, and K. Shin. CIRCA: a cooperative intelligent realtime control architecture. IEEE Transactions on Systems, Man and Cybernetics 23(6):1561-1574.
....guaranteeing the execution of jobs that are accepted. By using MARUTI to schedule and execute the reactive elements of DR, the combined system can make performance guarantees. DR MARUTI does not have mechanisms to automatically reason about or analyze its scheduling requirements. The CIRCA system [Musliner, Durfee, Shin 1993; 1995] combines a hard real time reactive executive with soft real time AI planning methods that automatically generate reactive plans given a set of high level goals. CIRCA includes its own scheduling module that allows it to reason explicitly about the real time requirements of its reactive ....
Musliner, D. J.; Durfee, E. H.; and Shin, K. G. 1993. CIRCA: A cooperative intelligent real-time control architecture. IEEE Transactions on Systems, Man, and Cybernetics 23(6):1561--1574.
No context found.
Musliner, D. J.; Durfee, E. H.; and Shin, K. G. 1993. CIRCA: A cooperative intelligent real time control architecture. IEEE Transactions on Systems, Man, and Cybernetics 23(6):1561--1574.
No context found.
D. J. Musliner. CIRCA: The Cooperative Intelligent Real-Time Control Architecture. PhD thesis, Department of Computer Science and Engineering, University of Michigan, 1993.
No context found.
D. J. Musliner. CIRCA: The Cooperative Intelligent Real-Time Control Architecture. PhD thesis, Department of Computer Science and Engineering, University of Michigan, 1993. 30
No context found.
D. J. Musliner. CIRCA: The Cooperative Intelligent Real-Time Control Architecture. PhD thesis, Department of Computer Science and Engineering, University of Michigan, 1993.
No context found.
D. J. Musliner. CIRCA: The Cooperative Intelligent RealTime Control Architecture. PhD Thesis, University of Michigan, 1993.
No context found.
D. Musliner, E. Durfee, and K. Shin, "Circa: a cooperative intelligent real-time control architecture," IEEE Trans. Syst., Man, Cybern., vol. 23, no. 6, pp. 1561--1573, 1993.
No context found.
David J. Musliner, Edmund H. Dufree, and Kang G. Shin. Circa: A cooperative intelligent real-time control architecture. IEEE trans. on computer, 23(6):1561--1574, November 1993.
No context found.
Musliner, D., Durfee, E., Shin, K.: CIRCA: a cooperative intelligent real-time control architecture. IEEE Transactions on Systems, Man and Cybernetics 23 (1993) 1561--1574
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