R. Kumar and V. K. Garg, Modeling and control of logical discrete event systems, Kluwer Academic Publishers, 1995.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....(Denmark) Part of this work was done during a visit to LSV, ENS de Cachan (France) y This research was supported by NSF award CCR99 70925 and NSF award ITR SY 0121431. full observability (in terms of two player games of complete information) Tho02] and for partial observability [KG95,Rei84,KV97]. In parallel, there has been a growing importance of verification for real time systems, and this leads to the natural question of whether techniques developed in the untimed setting for the controller synthesis problem can be generalized to timed systems (see for example the papers ....

R. Kumar and V.K. Garg. Modeling and Control of Logical Discrete Event Systems. Kluwer Academic Publishers, 1995.

....of Theorem 6 requires a test for Lm (G) closedness of the fixed point K. However, the fixed point calculation can be modified to ensure Lm (G) closedness a priori. Let (Z) denote the lattice of sublanguages of Z that are Lm (G) closed and let G (Z) Z) G (Z) It follows (e.g. [7]) that (Z) is closed under arbitrary unions and that (Z) Z (Lm (G) Z)# # . By replacing G (Z) with G (Z) in Theorem 6, we find that there exists a solution to RNSCP whenever K #= #. The proof that sup G (Z) exists and is well defined is similar to the proof of Lemma 4. ....

R. Kumar and V.K. Garg. Modeling and Control of Logical Discrete Event Systems. Kluwer Academic Publishers, Boston, 1995.

....8(c 1 ; c 2 ) 2 C : c 1 v c 2 ) f(c 1 ) v f(c 2 ) 3) Therefore, we can show that Theorem 2 f has a supremal fixed point Proof: All we have to show is that ffl f is a monotone function (see (3) ffl f is defined over a complete lattice (C; v) see Lemma 1) Then, according to Theorem 2. 1 in [9], f admits a supremal fixed point. At this point, we have shown that our iterative verification process will converge to a fixed point. We still have to show that this fixed point corresponds to a correct analysis. Theorem 3 After termination of the iterative process, all relevant paths have ....

R. Kumar and V. K. Garg. Modeling and Control of Logical Discrete Event Systems. Kluwer Academic, 1995.

....continuous and hybrid control systems. We start by recalling some know interpretations of continuous and discrete control systems to gain some motivation for the general de nitions. 2.1. Discrete Control Systems. One of the usual models for discrete control systems are nite state automata [11, 20], de ned by a triple (Q; where: Q is a nite set of states, is a nite set of input symbols, Q Q is the next state function. We regard the partially de ned map as de ning the controlled dynamics, in the sense that for each q 2 Q there exists a set of choices (the ....

R. Kumar and V.K. Glarg. Modeling and Control of Logical Discrete Event Systems. Kluwer Academic Publishers, 1995.

....and its environment and the controller cannot observe certain variables of the system and the environment and also cannot observe certain internal actions of the the environment. This issue is a very important aspect of control, and has been studied in great detail in the control theory community [KG95]. While this aspect is well understood in the case of untimed systems (see [Rei84] it has not been studied for timed plants as described above. An important and new issue in the timed framework is related to the resources allowed to the controller. By a controller s resources we mean a ....

R. Kumar and V.K. Garg. Modeling and control of logical discrete event systems. Kluwer Academic Publishers, 1995.

.... CTL and CTL has been solved for maximal environments as cited above and more recently also for the calculus [KV00] As for controller synthesis, as mentioned above, most of the settings considered in the literature are linear time ones and often involve dealing with incomplete information [KG95,KS95,PR89,Var95]. As for branching time settings, control problems for settings with maximal environment is studied in [Ant95] Also, for maximal environments, the control problem can be transformed (by ipping the role of the system and the environment) into module checking problems solved in [KV96,KV97a] Yet ....

R. Kumar and V.K. Garg. Modeling and control of logical discrete event systems. Kluwer Academic Publishers, 1995.

....correct and consistent operation. Theresu#83j2 analytical problems are collectively characterized as struN u al analysis and control of sequ N ial RAS [63] and from a methodological perspective, they constitu#3 a su#xx a of the logical analysis and control of discrete event systems (DESs) [30, 10]. 1 It is envisioned th t othN industries such asph rmaceuticals and MEMS manufacturing are also likely to face similar control requirements. 5 One of the basic problems in DES logical control theory is the synthesis of su pervisors for DES withu controllable events i.e. events that cannot ....

.... [4, 85, 16, 86] mathematical frameworks, iii) the formal characterization of the logical control problem for discrete event systemsu#tem variou# control requ#KT 2V ts, and the transfer of classical control theoretic concepts, like controllability and observability, in this new modeling framework [56, 24, 38, 39, 20, 21, 40, 30, 26, 69, 42, 10], iv) the systematic stu# of a nu# ber of conflict resolu#TTx problems arising in contemporary produ#2 T# systems as stru#3T 2V control problems in theu#j18 T2V8 behavioral spaces [4, 76, 85, 80, 5, 17, 61, 82, 62, 18, 59] and (v) the emergence of PNstru#8 32 l theory as a powerfu l analytical ....

[Article contains additional citation context not shown here]

R. Ku# ar and V. Garg, Modeling and Control of Logical Discrete Event Systems. Klu wer Academic Pu#-'3 hers, 1995.

.... CTL and CTL has been solved for maximal environments as cited above and more recently also for the calculus [KV00] As for controller synthesis, as mentioned above, most of the settings considered in the literature are linear time ones and often involve dealing with incomplete information [KG95,KS95,PR89,Var95]. As for branching time settings, synthesis of memoryless controllers for settings with maximal environments is studied in [Ant95] Also, for maximal environments, the control problem can be transformed (by ipping the role of the system and the environment) into module checking problems solved in ....

R. Kumar and V.K. Garg. Modeling and control of logical discrete event systems. Kluwer Academic Publishers, 1995.

....of P satisfy , and if is branching then the computation tree of P satisfies ) It is known how to cope with incomplete information in the linear paradigm. In particular, the approach used in [PR89] can be extended to handle synthesis with incomplete information for the linear specifications [KG95, KS95, Var95] Coping with incomplete information is more difficult in the branching paradigm, where the methods used in the linear paradigm are not applicable [KV97] In [KV97] we solved the problem of synthesis with incomplete information for specification in the branching temporal logic CTL ....

R. Kumar and V.K. Garg. Modeling and control of logical discrete event systems. Kluwer Academic Publishers, 1995.

No context found.

R. Kumar and V. K. Garg. Modeling and Control of Logical Discrete Event Systems. Kluwer, 1995.

No context found.

R. Kumar and V. K. Garg. Modeling and Control of Logical Discrete Event Systems. Kluwer Academic Publishers, Boston, MA, 1995.

No context found.

R. Kumar and V. K. Garg. Modeling and Control of Logical Discrete Event Systems. Kluwer Academic Publishers, Boston, MA, 1995.

....languages The research was supported in part by the National Science Foundation under the grants NSF ECS9709796 and NSF ECS 0099851, a DoD EPSCoR grant through the Oce of Naval Research under the grant N000140110621, and a KYDEPSCoR grant. 1 Introduction The theory of supervisory control [14, 9, 6] addresses the problem of synthesizing supervisors for discrete event systems (DESs) called a plant. The aim of supervision is to enforce a given speci cation by minimally restricting the plant behavior. The supervisory role is characterized by the fact that at any given plant state, the ....

....be achieved in totality, one nds the maximal sublanguage that can be achieved under suitable supervision. This maximal sublanguage is the supremal controllable and relative closed sublanguage of the speci cation language. Ecient algorithms exist for the computation of such a supremal language [17, 2, 9]. The controller we design further restricts the behavior of the given plant. Unlike a supervisor however, a controller permits the execution of only one controllable event at a state, whenever at least one such event is permitted by the maximally permissive supervisor at that state. Obviously, ....

R. Kumar and V. K. Garg. Modeling and Control of Logical Discrete Event Systems. Kluwer Academic Publishers, Boston, MA, 1995.

....Introduction Discrete event systems (DESs) are systems with discrete states that evolve in response to the occurrence of certain discrete qualitative changes, called events. A theory for the control of qualitative behaviors of such systems was rst proposed in [15] and is described in detail in [7, 2]. The controller, also called a supervisor, restricts the behavior of the given DES, also called a plant, by dynamically disabling the occurrence of certain controllable events, based upon its observations of the event sequences executed by the plant. When the plant is physically distributed, ....

.... locally enable it [19] On the other hand, under the disjunctive fusion rule, 2 c is globally enabled if at least one local supervisor locally enables it [14] It is useful to de ne a set of weaker observation masks i [ f g (i 2 I) that mask all but the last event of a trace [18, 7]: if s = M i (s ) if s = s where 2 . Then the map (i 2 I) is de ned as follows [18, 7] M i (s) ft 2 M i (t) M i (s)g f g if s = ft 2 j M i (t ) M i (s )g if s = s where 2 . A language L L(G) is said to be ....

[Article contains additional citation context not shown here]

R. Kumar and V. K. Garg, Modeling and Control of Logical Discrete Event Systems. Boston, MA: Kluwer, 1995.

.... M OST man made systems are discrete event systems (DESs) owing to the manner in which they evolve: In response to events that are spontaneous, instantaneous, asynchronous (thus discrete in nature) Ramadge and Wonham [9] introduced the theory of supervisory control of discrete event systems [8], 3] where they employed an automaton based model of the system, called a plant, and studied how another automaton, called a supervisor, can be employed to restrict its behavior. The control speci cations which express the constrains that one wishes to impose on the system s behavior are ....

R. Kumar and V. K. Garg. Modeling and Control of Logical Discrete Event Systems. Kluwer Academic Publishers, Boston, MA, 1995.

....This paper is set in the supervisory control framework for discrete event systems developed by Ramadge and Wonham [7] For the readers convenience, some background results from the cited references are rst provided in this section. For a detailed introduction of the theory, readers may refer to [3]. An uncontrolled discrete event system is modeled as an automaton G = Q; q 0 ; Qm ) where Q is a set of states, is a set of event labels, q 0 2 Q is the initial state, Qm Q is the set of marked states, and : Q Q, the transition function, is a partial function de ned at each ....

....all the events, the concept of observation mask is introduced. An observation mask is a function M : S f g, where 62 , and is called the set of observed events. The mask function can be extended to the set of strings in a natural way. A supervisor S for G is said to be mask compatible [3] if it observes only M(L(G) In this paper, it is assumed that all supervisors for partially observed systems must be mask compatible supervisors. A sublanguage K of L(G) is said to be (L(G) M) observable [4] if the conditions s; s 0 2 pr(K) M(s) M(s 0 ) s 2 pr(K) and s 0 2 L(G) ....

R. Kumar and V. K. Garg. Modeling and Control of Logical Discrete Event Systems. Kluwer Academic Publishers, Boston, MA, 1995.

....in the same order as it was transmitted. Many extensions of the basic supervisory control problem such as control with partial observations, decentralized control, modular control, control of non deterministic systems, control of in nite behaviors represented by languages, have been studied [13, 3]. In the supervisory control framework for discrete event systems, an uncontrolled discrete event system, called plant, is modeled as a state machine, the event set of which is nite and is partitioned into the set of controllable and uncontrollable events. The language generated by such a state ....

R. Kumar and V. K. Garg. Modeling and Control of Logical Discrete Event Systems. Kluwer Academic Publishers, Boston, MA, 1995.

....architecture is a cascade of four subsystems: the input interface, the perceptor, the response controller, and the output interface. The perceptor extracts the relevant symbolic information from the incoming continuous sensor signals, while the response controller is a discrete event system [7, 12] that computes discrete control actions in response to the discrete inputs from the perceptor. Our approach to the design of intelligent controllers is behavioral. Behaviors are certain high level activities (that are independent of each of other) that determine the manner in which the system ....

R. Kumar and V. K. Garg. Modeling and Control of Logical Discrete Event Systems. Kluwer Academic Publishers, Boston, MA, 1995.

.... in the following ways: systems can interact with each other through interfaces; driven events are included; solutions for control problem at the interface level as well as at the plant level have been obtained; no control compatibility (completeness) and observation compatibility [12] re4 quirements for the supervisor exists 2. There is no increase in computational complexity for the supervisor synthesis in the setting of MPSC as compared to that in the setting of SSC with partial observations. 3. The main result presented in [8] that reduces the control problem of ....

.... fs 2 P j P (x 0 P ; s) 6= g: Given an event set , a pre x closed language H , a set of events , and a mask function M : f g with M( a language K H is said to be (H; controllable [20] if pr(K) H pr(K) and it is said to be (H; M) normal [12] if M 1 M(pr(K) H pr(K) Controllability (resp. normality) is preserved under language union. Consequently, the supremal controllable (resp. the supremal normal) sublanguage of a given language exists. The following de nition introduces the notion of prioritized synchronization [7] De ....

R. Kumar and V. K. Garg. Modeling and Control of Logical Discrete Event Systems. Kluwer Academic Publishers, Boston, MA, 1995.

....always guaranteed to meet the control speci cations. This theory is applicable to any system which evolves in response to events that are spontaneous, instantaneous, asynchronous and thus discrete in nature. Such systems are classi ed as discrete event system (DES) and have been examined in detail [8, 14]. A DES to be controlled, also called a plant, is modeled by a nite state machine (FSM) and can equivalently be described by a language model. The speci cations which express the constraints that one wishes to impose on the plant s behavior are modeled as formal languages as well. A supervisor ....

....Using the supervisory control theory we obtain the maximally permissive supervisor for the miniature assembly line that enforces the overall speci cation. This turns out to be the automaton represented by the overall speci cation itself, since the overall speci cation is found to be controllable [8, 14]. A controller is extracted out of the supervisor as described above, and then the controller is translated into speci c code understood by the LEGO Dacta control software. This is diagramitically illustrated in Figure 1. What system can do) Plant Model (What system should do) What system ....

[Article contains additional citation context not shown here]

R. Kumar and V. K. Garg. Modeling and Control of Logical Discrete Event Systems. Kluwer Academic Publishers, Boston, MA, 1995.

....This paper is set in the supervisory control framework for discrete event systems developed by Ramadge and Wonham [8] For the readers convenience, some background results from the cited references are rst provided in this section. For a detailed introduction of the theory, readers may refer to [2]. An uncontrolled discrete event system is modeled as an automaton G = Q; q 0 ; Qm ) 4 where is a set of event labels, Q is a set of states, q 0 2 Q is the initial state, Qm Q is the set of marked states, and : Q Q, the transition function, is a partial function de ned at each ....

....the events, the concept of observation mask is introduced. An observation mask is a function M : S f g, where 62 , and is called the set of observed events. The mask function can be extended to the set of strings 5 in a natural way. A supervisor S for G is said to be mask compatible [2] if it observes only M(L(G) In this paper, it is assumed that all supervisors for partially observed systems must be mask compatible supervisors. A pre x closed sublanguage K of L(G) is said to be (L(G) M) observable [5] if the conditions s; s 0 2 K; M(s) M(s 0 ) s 2 K, and s 0 2 ....

R. Kumar and V. K. Garg. Modeling and Control of Logical Discrete Event Systems. Kluwer Academic Publishers, Boston, MA, 1995.

.... (13) where the map f M , first introduced in [22] is a modified observation mask that masks all but the last event, i.e. f M(ffl) ffl; 8s 2 Sigma ; oe 2 Sigma : f M(soe) M(s)oe: In case the supervisor is local so that it is only able to control those events that it observes [10], then the desired behavior must satisfy a condition stronger than observability, called normality. A language H L(G) is said to be normal if M(pr(H) L(G) pr(H) 14) where M : 2 Sigma 2 Sigma is the map induced by the mask function M and is defined as: 8H Sigma : M(H) fs 2 ....

R. Kumar and V. K. Garg. Modeling and Control of Logical Discrete Event Systems. Kluwer Academic Publishers, Boston, MA, 1994.

No context found.

R. Kumar and V. K. Garg, Modeling and control of logical discrete event systems, Kluwer Academic Publishers, 1995.

No context found.

R. Kumar and V.K. Garg. Modeling and Control of Logical Discrete Event Systems. Kluwer Academic Publishers, 1995.

No context found.

R. Kumar and V. K. Garg. Modeling and Control of Logical Discrete Event Systems. Kluwer Academic, 1995.

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