L.R. Missiaen. Localized abductive planning with the event calculus. PhD thesis, Department of Computer Science, K.U.Leuven, 1991.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....order ft 2 t 3 ; t 1 t 3 g, it holds that holds(on; t 3 ) is false, while in the partial order itself, holds(on; t 3 ) is true. This problem with partial order time occurs also in the context of planning. The problem with the light switch scenario is similar to the anomalies reported in [36, 34] on the planning approaches in [14] 48] 36] In section 5, we come back to this issue. 4.3 Unknown e ects and nondeterminism Another way in which uncertainty can arise in temporal domains is by the presence of actions with indeterminate e ects: the outcome of such an action is not uniquely ....

....holds(on; t 3 ) is false, while in the partial order itself, holds(on; t 3 ) is true. This problem with partial order time occurs also in the context of planning. The problem with the light switch scenario is similar to the anomalies reported in [36, 34] on the planning approaches in [14] 48] [36]. In section 5, we come back to this issue. 4.3 Unknown e ects and nondeterminism Another way in which uncertainty can arise in temporal domains is by the presence of actions with indeterminate e ects: the outcome of such an action is not uniquely determined by the circumstances and the way in ....

[Article contains additional citation context not shown here]

L.R. Missiaen. Localized abductive planning with the event calculus. PhD thesis, Department of Computer Science, K.U.Leuven, 1991.

....cannot become ground. A procedure for planning should be able to cope with such situations. This condition is not satis ed by the abductive procedure de ned in [11] In the past, special abductive procedures for temporal reasoning with abductive event calculus have been presented ( 8] 20] [16, 17]) However, as the three authors argue, the treatment of constraints in [8] and the treatment of non ground abductive failure goals in [20] 16, 17] can be very inecient. The abductive procedure SLDNFA presented in this paper, provides an improved treatment of non ground abductive atoms. We have ....

.... de ned in [11] In the past, special abductive procedures for temporal reasoning with abductive event calculus have been presented ( 8] 20] 16, 17] However, as the three authors argue, the treatment of constraints in [8] and the treatment of non ground abductive failure goals in [20] [16, 17] can be very inecient. The abductive procedure SLDNFA presented in this paper, provides an improved treatment of non ground abductive atoms. We have proven its sound and completeness. Although the inspiration for the design of SLDNFA stems from temporal reasoning, we formulate it in full ....

[Article contains additional citation context not shown here]

L. Missiaen. Localized abductive planning with the event calculus. PhD thesis, Department of Computer Science, K.U.Leuven, 1991.

....checking on incomplete knowledge has been shown in the context of a translation from a temporal language A [27] to abductive logic programming. In the past, a number of abductive extensions of SLDNF resolution have been proposed for abductive logic programs with negation [21] 52] [46, 45, 44], 31] 8] 51] 29] 13] 54] Anticipating the discussion of these procedures in section 11, we can say that either these procedures have not been formalised and proven correct senior research associate of the Belgian National Fund for Scientific Research 1 [21] 52] or they can be ....

.... these procedures in section 11, we can say that either these procedures have not been formalised and proven correct senior research associate of the Belgian National Fund for Scientific Research 1 [21] 52] or they can be proven correct only for a restricted class of abductive logic programs [46, 45, 44], or they do not provide a way of checking the consistency of the abductive answers [8] 54] or they do not provide a treatment for floundering abduction [31] 51] 29] The floundering abduction problem is an analogue of the well known floundering negation problem which arises when a ....

[Article contains additional citation context not shown here]

L.R. Missiaen. Localized abductive planning with the event calculus. PhD thesis, Department of Computer Science, K.U.Leuven, 1991.

....A to extended logic programming [16] and proving its soundness, they show the expressivity of this formalism for representing temporal knowledge and, more in general, incomplete knowledge. In the past, another approach has been explored for temporal reasoning, based on event calculus [13] 31] [25], 10] 10] proposes solutions for the same benchmarks as in [17] This approach makes use of the formalism of abductive logic programming. One may interpret an abductive program as an open logic program in the sense that it contains only definitions for the non abducible predicates. The ....

....in it. In the past, another approach has been explored for temporal reasoning, based on event calculus [22] 13] and [31] have simplified event calculus and have extended it with abduction for the purpose of planning. 31] extended event calculus to deal with necessary preconditions of actions. [25] implemented a planning system based on this formalism. Other work has been done to extend event calculus with continuous actions [32] and time granularity [15] 26] Recently [10] applied abductive event calculus to solve a number of benchmark problems in temporal reasoning, such as the Murder ....

L.R. Missiaen. Localized abductive planning with the event calculus. PhD thesis, Department of Computer Science, K.U.Leuven, 1991.

.... involving incomplete temporal knowledge and they proposed a sound transformation to extended programs, programs with both negation as failure and classical or explicit negation [12] In the past, another approach has been explored for temporal reasoning, based on event calculus [10] 21] [17], 8] 8] proposes solutions for the same benchmarks as in [13] This approach makes use of the formalism of abductive logic programming. One may interpret an abductive program as an incomplete logic program in the sense that it contains only definitions for the non abducible predicates. The ....

....In the past, another approach has been explored for temporal reasoning, based on event calculus [15] 10] and [21] have simplified event calculus and have extended it with abduction for the purpose of planning. 21] extended event calculus to deal with necessary preconditions of actions. [17] implemented a planning system based on this formalism. Other work has been done to extend event calculus with continuous actions [22] and time granularity [11] 18] Recently [8] applied abductive event calculus to solve a number of benchmark problems in temporal reasoning, such as the Murder ....

L.R. Missiaen. Localized abductive planning with the event calculus. PhD thesis, Department of Computer Science, K.U.Leuven, 1991.

.... i is a time point constant such that i , and P is such that (EC1) EC7) T P j= G (2) or equivalently (EC1) EC7) T j= P G (3) Hence, under this formulation, planning can either be viewed as abducing sentences of the form P to add to the domain theory [4] 5] 14] [15] [16] or can be viewed as deducing theorems (of the form P G) in the spirit of Green, Reiter and others [7] 20] 21] In either case, it is necessary to ensure that the plan is consistent with the domain theory, i.e. that (EC1) EC7) T 6j= P (4) However, this need not necessarily ....

Lode Missiaen, Localized Abductive Planning with the Event Calculus, Ph.D. Thesis, Department of Computer Science, K.U.Leuven, 1991.

....Logic Programming approach to temporal reasoning, logic based learning and planning, it has been developed according to the example based methodology, that was the common methodology in late 80 s, and its range of applicability it has never been assessed. The simplification of interest [Bru95, Mis91] consists in the following two frame axioms: holds at(P; T ) happens( ET;Ev) ET T; initiates(Ev; P ) succeeds(Ev) not clipped(ET; P; T ) clipped(T 1; P; T 2) happens(BT; ET;Ev) T1 BT;ET T2; terminates(Ev; P ) succeeds(Ev) The axiom holds at(P; T ) is true when the ....

L. Missiaen. Localized abductive planning with the Event Calculus. PhD thesis, Department of Computer Science, University of Leuven, 1991.

....to represent nondeterministic actions. We assume that each action has a known set of possible outcomes, one of which will be realised 21 This problem with partial order time occurs also in the context of planning. The problem with the light switch scenario is similar to the anomalies reported in [30, 29] on the planning approaches in [13] 43] 30] In section 4.2, we come back to this issue. 18 whenever the action occurs. This can be represented at a high level by a rule of the form: A causes E 1 j : j E n if Psi in which the E i are the possible outcomes of action A if it is executed ....

....each action has a known set of possible outcomes, one of which will be realised 21 This problem with partial order time occurs also in the context of planning. The problem with the light switch scenario is similar to the anomalies reported in [30, 29] on the planning approaches in [13] 43] [30]. In section 4.2, we come back to this issue. 18 whenever the action occurs. This can be represented at a high level by a rule of the form: A causes E 1 j : j E n if Psi in which the E i are the possible outcomes of action A if it is executed when Psi holds. The symbol j denotes a ....

[Article contains additional citation context not shown here]

L.R. Missiaen. Localized abductive planning with the event calculus. PhD thesis, Department of Computer Science, K.U.Leuven, 1991.

....to the program P for all literals L in the abductible list Ab two rules of the form: 1. L not:L 2. L notL 3 Abductive Planning with Event Calculus In order to represent and reason about actions and events it was used the event calculus with some changes proposed by Eshgi ( 4] and Missiaen ([9]) The following logic program is proposed by Missiaen in order to describe what properties hold at a given time: holds at(P; T ) happens(E) initiates(E; P ) 1) succeeds(E) E T; persists(E; P; T ) persists(E; P; T ) notclipped(E; P; T ) 2) clipped(E; P; T ) happens(C) terminates(C; ....

Lode Missiaen. Localized Abductive Planning with the Event Calculus. PhD thesis, Univ. Leuven, 1991.

....P for all literals L in the abducible list Ab two rules of the form: 1. L not:L 2. L notL 3 Abductive Planning with Event Calculus In order to represent and reason about actions and events it was used the event calculus with some changes proposed by Shanahan ( 13] Eshgi ( 3] and Missiaen ([8]) In this framework events are time points with no duration and are represented by the time at which they occur. Time points are ordered by the precedence relationship . Events are action instances and they are related to the actions by the predicate act 2. There are also other prede ned ....

Lode Missiaen. Localized Abductive Planning with the Event Calculus. PhD thesis, Univ. Leuven, 1991.

....definition of the Phi operator) The well founded model of the program P is the F least partial stable model of P . The well founded semantics of P is determined by the set of all partial stable models of P: 2. 1 Events As time formalism we use a variation of the Event Calculus (Quaresma, 1997;Missiaen, 1991) that allows events to have an identification and a duration. As a consequence events may occur simultaneously. The predicate holds at(P; T ) defines the properties P that are true at a specific time T ; the predicate happens(E; T i ; T f ) means that the event E occurred between T i and T f ; ....

Missiaen, L. (1991). Localized Abductive Planning with the Event Calculus. Ph.D. Dissertation, Univ. Leuven.

....set of predicates (see next section) This process is done through the use of this framework and the Event Calculus. 3 Abductive Planning with Event Calculus In order to represent and reason about actions and events a modi ed version of the event calculus proposed by Eshgi ( 4] and Missiaen ([12]) was used. In this formalism, events are action instances. They can be represented by the time at which they ocurred. Time is linear and time points are ordered by the precedence relationship . Events are referents which are related to actions by the predicate act(E; A) happens(E) states that ....

Lode Missiaen. Localized Abductive Planning with the Event Calculus. PhD thesis, Univ. Leuven, 1991.

....we will allow events to be abduced. This process is done through the use of this framework and the Event Calculus. 3. ABDUCTIVE PLANNING WITH EVENT CALCULUS In order to represent and reason about actions and events we use the event calculus with some changes proposed by Eshgi [5] and Missiaen [12]. In this formalism, events are action instances and have no duration. They can be represented by the time point at which they occurred, and time points are ordered by the precedence relationship . Events are related to its actions by the predicate act(E; A) happens(E) states that the event E ....

Lode Missiaen. Localized Abductive Planning with the Event Calculus. PhD thesis, Univ. Leuven, 1991.

....is very important that agents have the capability to reason about their mental state at a given time point. They should also be able to change their mental state as a consequence of some external or internal events. As a time formalism we propose a variation of the Event Calculus ( Sha89; Esh88; Mis91 ] that allows events to have an identification and a duration. As a consequence events may occur simultaneously. The predicate holds at defining the properties that are true at a specific time is: holds at(P; T ) happens(E; T i ; T f ) 1) initiates(E; TP ; P ) TP T ; persists(T P ; P; ....

Lode Missiaen. Localized Abductive Planning with the Event Calculus. PhD thesis, Univ. Leuven, 1991.

....the capability to represent time and events and they should be able to reason about their mental state at a given time point. They should also be able to change their mental state as a consequence of some external or internal events. For a time formalism we use a variation of the Event Calculus [7] that allows events to have an identification and a duration and allows events to occur simultaneously. In this approach time is linear and discrete. The predicate holds at defining the properties that are true at a specific time is: holds at(P; T ) happens(E; T i ; T f ) 1) initiates(E; TP ; ....

Lode Missiaen. Localized Abductive Planning with the Event Calculus. PhD thesis, Univ. Leuven, 1991.

....and Koomen write the expansion of an abstract action as a necessary condition, Eshghi writes it initially as a sufficient condition and then turns the implication around by an abductive inference. This abductive approach to planning has subsequently been followed up by others [ Shanahan, 1989; Missiaen, 1991; Denecker et al. 1992 ] Although our ultimate goal is to reason about actions with the complexity apparent in legal domains (see [ McCarty, 1989; Schlobohm and McCarty, 1989 ] our objectives in the present paper are more modest. We will study hypothetical implications in which the ....

....whether an abductive explanation that satisfies the integrity constraints, i.e. a countermodel, is impossible. This is precisely the point of our inductive proof procedure in Section 3 and our decision procedure in Section 4. Moreover, most abductive theories of planning [ Eshghi, 1988; Missiaen, 1991; Denecker et al. 1992 ] have adopted a different logical framework from the one presented here, a framework that formalizes the effects of primitive actions on the state of the world. Abductive reasoning is then used within this framework to implement a nonlinear goal directed regression style ....

L. Missiaen. Localized Abductive Planning with the Event Calculus. PhD thesis, Katholieke Universiteit Leuven, 1991.

....into a simplified version ( Sadri 87] Kowalski 92] which employed time points instead of time intervals. This simplified event calculus has re29 placed the event calculus in practice and has been applied, in addition to the original intended applications, to planning (e.g. Eshghi 88] Missiaen 91] explanation (e.g. Shanahan 89] and air traffic management (e.g. Sripada 94] In the simplified event calculus there is a real line of time points. Properties may hold or not at some of these time points. The meaning of the formula holds(P; T ) is that property P holds at time T . An ....

Missiaen L. 1991. Localized abductive planning with the event calculus, PhD Thesis, Department of Computer Science, K.U. Leuven.

....interest has been clearly established on several realistic domains. Mots cl es : planification, hi erarchie d abstraction. Keywords : planning, abstraction hierarchy. y Les travaux ayant conduit a cet article remontent au s ejour post doctoral de Fr ed erick Garcia au LAAS, entre avril et aout 1994, financ e par le PRC IA. z Philippe Laborie travaille actuellement a l IRISA, Rennes. 1 Introduction L utilisation de techniques de hi erarchisation se r ev ele en g en eral tr es fructueuse lorsqu il s agit de r esoudre des probl emes de planification de grande taille. De mani ere ....

....le plan donn e en entr ee les contient d ej a toutes, la hi erarchie ne peut etre g en er ee automatiquement. Dans notre test, elle est donn ee en entr ee par l utilisateur. ffl la planification des travaux de finition d une pi ece (installation eau, electricit e, papier peint) inspir ee de [19] Les principaux r esultats sont r esum es sur les tableaux 2 et 3. Le tableau 2 donne respectivement pour chaque domaine: le facteur de gain (d efini comme le rapport entre le temps CPU sans utiliser d abstraction et celui en utilisant notre approche de hi erarchie dynamique) la profondeur du ....

[Article contains additional citation context not shown here]

L. Missiaen. Localized Abductive Planning with the Event Calculus. PhD thesis, Dept of Computer Science, K.U.Leuven, 1991.

....of actions and the initial situation of the planning domain, and Q describes the desired goal state. The abductive procedure then generates a plan which describes the events and their order, necessary to obtain the goal state. In event calculus, the frame axiom can be represented as follows [33, 32]: holds at(P; T ) happens(E) E T; initiates(E; P ) succeeds(E) clipped(E; P; T ) clipped(E; P; T ) happens(C) terminates(C;P ) succeeds(C) E C; C T The role of the negation as failure call :clipped(E; P; T ) is to assume that the property P is not clipped between time points ....

....Thus, the procedure flounders on these non ground abductive goals: it terminates without a complete computation. In the past, special abductive procedures which do not suffer from this limitation, have been presented for temporal reasoning with abductive event calculus [13] 41] Recently, [33, 32] described an implementation of such a planner based on a special purpose abductive procedure. However, the procedure in [13] is for definite programs with integrity constraints; no formalisation is given and soundness and completeness results are lacking. These results are also lacking for the ....

[Article contains additional citation context not shown here]

L.R. Missiaen. Localized abductive planning with the event calculus. PhD thesis, Department of Computer Science, K.U.Leuven, 1991.

....of a definition. resulting logic program P LTTS 6 contains definitions for all predicates. Note that the OLPFOL theory consisting of P LTTS augmented with the theory of linear order is inconsistent: P LTTS violates the linearity constraint since it entails :e1 e2 :e2 e1 :e1 = e2. In [21], examples were given where this approach failed. Also in this case the approach fails: P LTTS entails holds at(alive; e4) and Prolog can be used to prove it. In more detail, one easily verifies that P LTTS entails holds at(loaded; e1) and holds at(loaded; e2) Therefore, both e1 and e2 unload ....

L.R. Missiaen. Localized abductive planning with the event calculus. PhD thesis, Department of Computer Science, K.U.Leuven, 1991.

....well founded model of the program P is the F least partial stable model of P . The well founded semantics of P is determined by the set of all partial stable models of P . 102 P. Quaresma and J. G. Lopes 2. 1 Events As a time formalism, we use a variation of the Event Calculus (Quaresma, 1997; Missiaen, 1991) that allows events to have an identification and a duration. As a consequence, events may occur simultaneously. The predicate holds at(P; T ) defines the properties P that are true at a specific time T ; the predicate happens(E; T i ; T f ) means that the event E occurred between T i and T f ; ....

Missiaen, L. (1991). Localized Abductive Planning With the Event Calculus. Ph.D. Dissertation, University of Leuven, Belgium.

....as abduction using Horn rules, based on the Event Calculus of Kowalski and Sergot [66] Eshghi writes Horn rules which formalize the effects of actions, and then uses abduction to implement a regression style planner. This approach has subsequently been simplified and followed up by others [115, 92, 30]. We refer the reader to [86] for further discussion on the relation between the results of the present chapter and AI planning problems. 188 Chapter 7 Conclusion While we have stated a large number of results, there are many obvious questions we have left unanswered. In this concluding chapter ....

L. Missiaen. Localized Abductive Planning with the Event Calculus. PhD thesis, Katholieke Universiteit Leuven, 1991.

....(see [14] is a well known formalism for temporal representation and reasoning. The basic concepts of the formalism are events and properties: events initiate and terminate periods of time during which properties hold. The Event Calculus has been modified in several ways, for example in [20] 9] [17] and [13] mainly to simplify the ontology and to eliminate problems occurring because of bidirectional persistence of properties (forward as well as backward in time) In the context of the Event Calculus, 8] 19] and [17] have introduced abduction to solve planning problems and [7] showed how ....

.... Calculus has been modified in several ways, for example in [20] 9] 17] and [13] mainly to simplify the ontology and to eliminate problems occurring because of bidirectional persistence of properties (forward as well as backward in time) In the context of the Event Calculus, 8] 19] and [17] have introduced abduction to solve planning problems and [7] showed how abduction can be used to solve general temporal postdiction problems in the presence of incomplete information. Though abduction is clearly a useful computational paradigm, it was not recognized earlier that the formalism of ....

[Article contains additional citation context not shown here]

L. Missiaen. Localized abductive planning with the event calculus. PhD thesis, Department of Computer Science, K.U.Leuven, 1991.

....become ground. A procedure for planning should be able to cope with such situations. This condition is not satisfied by the abductive procedure defined in [7] In the past, special abductive procedures for temporal reasoning with abductive event calculus have been presented ( 3] 12] Recently, [9, 10] described an implementation of such a planner based on a special purpose abductive procedure. However, for the procedures in [3] and [12] no formalisation of the procedure is given and soundness and completeness results are lacking. In [9, 10] the abductive procedure is designed for a specific ....

....calculus have been presented ( 3] 12] Recently, 9, 10] described an implementation of such a planner based on a special purpose abductive procedure. However, for the procedures in [3] and [12] no formalisation of the procedure is given and soundness and completeness results are lacking. In [9, 10], the abductive procedure is designed for a specific class of planning problems, but is unsound in the general case. Moreover, as the three authors argue, the treatment of non ground abductive failure goals can be very inefficient. As a first contribution, we present in section 2 an improved ....

[Article contains additional citation context not shown here]

L.R. Missiaen. Localized abductive planning with the event calculus. PhD thesis, Department of Computer Science, K.U.Leuven, 1991.

....and the reasoning about, time [Kowalski and Sergot 86] It was originally used as a mechanism for updating databases and understanding natural language. Since then, event calculus has been presented in a variety of other ways [Sadri 87, Shanahan 89, Shanahan 90, Borillo and Gaume 90, Sripada 91, Missiaen 91b] The event calculus is based on events which initiate and terminate properties. For example, the event taking a shower would make the property one s hair is wet true. Events are assumed to be atomic; no other events can happen during the execution of an event. The occurrence of events allows ....

....their truth. The goals in Deltac are also called constraints. Checking every constraint in Deltac would be computationally expensive. The actual implementation uses a method based on locality to determine the constraints in Deltac selectively that could be affected by adding a new event [Missiaen 91b] Techniques for checking integrity constraints in deductive databases could be used for this same purpose [Sadri and Kowalski 88, Bry Manthey and M. 90] 4.2.4 Maintenance mechanism If a goal in Deltac can no longer be proven via the persistence rules, then the abductive refutation would ....

[Article contains additional citation context not shown here]

Lode R. Missiaen. Localized Abductive Planning with the Event Calculus. PhD thesis, KU Leuven, Department of Computer Science, Celestijnenlaan 200A, B-3001 Leuven, Belgium, September 1991. REFERENCES 27

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