## Answer Set Programming and Plan Generation (2002)

Venue: | ARTIFICIAL INTELLIGENCE |

Citations: | 144 - 6 self |

### BibTeX

@ARTICLE{Lifschitz02answerset,

author = {Vladimir Lifschitz},

title = {Answer Set Programming and Plan Generation},

journal = {ARTIFICIAL INTELLIGENCE},

year = {2002},

volume = {138},

pages = {2002}

}

### Years of Citing Articles

### OpenURL

### Abstract

The idea of answer set programming is to represent a given computational problem by a logic program whose answer sets correspond to solutions, and then use an answer set solver, such as smodels or dlv, to find an answer set for this program. Applications of this method to planning are related to the line of research on the frame problem that started with the invention of formal nonmonotonic reasoning in 1980.

### Citations

1850 |
STRIPS: A new approach to the application of theorem proving to problem solving
- Fikes, Nilsson
- 1971
(Show Context)
Citation Context ... from the fact that block 2 is now on the table. The ramication problem|the problem of describing indirect eects of actions|is not addressed in the classical action representation formalisms STRIPS [7=-=]-=- and ADL [32]. Finally, the executability of actions is described in this example indirectly as well. As discussed above, the impossibility of moving two blocks b 1 , b 2 onto the same block b is impl... |

1542 | The Stable Model Semantics for Logic Programming
- Gelfond, Lifschitz
- 1988
(Show Context)
Citation Context ...e will see that descriptions of actions in the style of [33, 37] can be used as a basis for planning using answer set solvers. In the next section, we review the concept of an answer set as dened in [=-=9, 10, 25]-=- and its relation to default logic. Then we describe some of the computational possibilities of answer set solvers (Section 3) and illustrate the answer set programming method [27, 30] by applying it ... |

1467 |
A logic for default reasoning
- Reiter
- 1980
(Show Context)
Citation Context ...the perspective of the research on the frame problem and nonmonotonic reasoning done in AI since 1980. Specically, we relate them to the line of work that started with the invention of default logic [=-=33]|the -=-nonmonotonic formalism that turned out to be particularly closely related to logic programming [2, 26, 10]. After the publication of the \Yale Shooting Scenario" [14] it was widely believed that ... |

974 |
Negation as failure
- Clark
- 1978
(Show Context)
Citation Context ...nd only if that query is p or r, that is to say, if and only if the query belongs to the answer set for (5). In this sense, the role of answer sets is similar to the role of the concept of completion =-=[3-=-], which provides an alternative explanation for the behavior of Prolog (p and r are entailed by the program's completion). 2.3 Comparison with Default Logic Let be a program such that the head of ev... |

452 | Planning as satisfiability
- Kautz, Selman
- 1992
(Show Context)
Citation Context ...lications of this method to planning are related to the line of research on the frame problem that started with the invention of formal nonmonotonic reasoning in 1980. 1 Introduction Kautz and Selman =-=[14]-=- proposed to approach the problem of plan generation by reducing it to the problem of finding a satisfying interpretation for a set of propositional formulas. This method, known as satisfiability plan... |

395 | Representing action and change by logic programs
- Gelfond, Lifschitz
- 1993
(Show Context)
Citation Context ...easily. 6 Relation to Action Languages and Satisability Planning Some of the recent work on representing properties of actions is formulated in terms of \high-level" action languages [12], such a=-=s A [11]-=- and C [13]. Descriptions of actions in these languages are more concise than logic programming representations. For example, the counterparts of thesrst two DEFINE rules from Figure 3 in language C a... |

342 |
Logic programs with classical negation
- Gelfond, Lifschitz
- 1990
(Show Context)
Citation Context ...pecically, we relate them to the line of work that started with the invention of default logic [33]|the nonmonotonic formalism that turned out to be particularly closely related to logic programming [=-=2, 26, 10]. Aft-=-er the publication of the \Yale Shooting Scenario" [14] it was widely believed that the solution to the frame problem outlined in [33] was inadequate. Several alternatives have been proposed [18,... |

334 | Logic programs with stable model semantics as a constraint programming paradigm
- Niemelä
- 1999
(Show Context)
Citation Context ... Two best known answer set solvers (systems for computing answer sets) available today are smodels 2 and dlv 3 . The results of computational experiments that use smodels for planning are reported in =-=[4, 3-=-0]. 1 See http://www.research.att.com/kautz/blackbox/ for the latest system of this kind created by the inventors of satisability planning. 2 http://www.tcs.hut.fi/Software/smodels/ . 3 http://www.dba... |

310 |
ADL: Exploring the middle ground between STRIPS and the situation calculus
- Pednault
- 1989
(Show Context)
Citation Context ...ct that block 2 is now on the table. The ramication problem|the problem of describing indirect eects of actions|is not addressed in the classical action representation formalisms STRIPS [7] and ADL [3=-=2]-=-. Finally, the executability of actions is described in this example indirectly as well. As discussed above, the impossibility of moving two blocks b 1 , b 2 onto the same block b is implicit in our d... |

259 |
Nonmonotonic logic and temporal projection
- Hanks, McDermott
- 1987
(Show Context)
Citation Context ...e invention of default logic [33]|the nonmonotonic formalism that turned out to be particularly closely related to logic programming [2, 26, 10]. After the publication of the \Yale Shooting Scenario&q=-=uot; [14]-=- it was widely believed that the solution to the frame problem outlined in [33] was inadequate. Several alternatives have been proposed [18, 20, 34, 15, 21, 29, 8]. It turned out, however, that the ap... |

256 | Stable models and an alternative logic programming paradigm
- Marek, Truszczyński
- 1999
(Show Context)
Citation Context ...as defined in [8, 9, 17] and its relation to default logic. Then we describe some of the computational possibilities of answer set solvers (Section 3) and illustrate the answer set programming method =-=[19, 21]-=- by applying it to a graphtheoretic search problem (Section 4). In Section 5 we turn to the use of answer set solvers for plan generation. Section 6 describes the relation of this work to other resear... |

193 | Action languages
- Gelfond, Lifschitz
- 1998
(Show Context)
Citation Context ...criptions quite easily. 6 Relation to Action Languages and Satisability Planning Some of the recent work on representing properties of actions is formulated in terms of \high-level" action langua=-=ges [12]-=-, such as A [11] and C [13]. Descriptions of actions in these languages are more concise than logic programming representations. For example, the counterparts of thesrst two DEFINE rules from Figure 3... |

183 | Causal theories of action and change
- McCain, Turner
- 1997
(Show Context)
Citation Context ... counterparts of thesrst two DEFINE rules from Figure 3 in language C are move(b; l) causes on(b; l) and inertial on(b; l): The design of language C is based on the system of causal logic proposed in =-=[28-=-]. For a large class of action descriptions in C, an equivalent translation into logic programming notation is dened in [24]. The possibility of such a translation further illustrates the expressive p... |

155 | Answer set planning
- Lifschitz
- 1999
(Show Context)
Citation Context ...ystem of this kind created by the inventors of satisability planning. 2 http://www.tcs.hut.fi/Software/smodels/ . 3 http://www.dbai.tuwien.ac.at/proj/dlv/ . 1 In this paper, based on earlier reports [=-=22, 23-=-], applications of answer set programming to planning are discussed from the perspective of the research on the frame problem and nonmonotonic reasoning done in AI since 1980. Specically, we relate th... |

147 | Consistency of Clark’s completion and existence of stable models
- Fages
- 1993
(Show Context)
Citation Context ...ection between the two methods. If a program without classical negation is \positive-order-consistent," or \tight," then its answer sets can be characterized by a collection of propositional=-= formulas [6]-=-|the formulas obtained by applying the completion process [3] to the program. The translations from language C described in [24] happen to produce tight programs. Describing a planning problem by a pr... |

138 | An Action Language Based on Causal Explanation: Preliminary Report
- Giunchiglia, Lifschitz
- 1998
(Show Context)
Citation Context ...elation to Action Languages and Satisability Planning Some of the recent work on representing properties of actions is formulated in terms of \high-level" action languages [12], such as A [11] an=-=d C [13]-=-. Descriptions of actions in these languages are more concise than logic programming representations. For example, the counterparts of thesrst two DEFINE rules from Figure 3 in language C are move(b; ... |

118 | Encoding planning problems in nonmonotonic logic programs
- Dimopoulos, Nebel, et al.
- 1997
(Show Context)
Citation Context ... Two best known answer set solvers (systems for computing answer sets) available today are smodels 2 and dlv 3 . The results of computational experiments that use smodels for planning are reported in =-=[4, 3-=-0]. 1 See http://www.research.att.com/kautz/blackbox/ for the latest system of this kind created by the inventors of satisability planning. 2 http://www.tcs.hut.fi/Software/smodels/ . 3 http://www.dba... |

102 | Answer sets in general nonmonotonic reasoning (preliminary report
- Lifschitz, Woo
- 1992
(Show Context)
Citation Context ...e will see that descriptions of actions in the style of [33, 37] can be used as a basis for planning using answer set solvers. In the next section, we review the concept of an answer set as dened in [=-=9, 10, 25]-=- and its relation to default logic. Then we describe some of the computational possibilities of answer set solvers (Section 3) and illustrate the answer set programming method [27, 30] by applying it ... |

91 | The logic of persistence
- Kautz
- 1986
(Show Context)
Citation Context ... 10]. After the publication of the \Yale Shooting Scenario" [14] it was widely believed that the solution to the frame problem outlined in [33] was inadequate. Several alternatives have been prop=-=osed [18, 20, 34, 15, 21, 29, 8]-=-. It turned out, however, that the approach of [33] is completely satisfactory if the rest of the default theory is set up correctly [37]. It is, in fact, very general, as discussed in Section 5.2 bel... |

86 | Representing actions in logic programs and default theories: A situation calculus approach
- Turner
- 1997
(Show Context)
Citation Context ...Several alternatives have been proposed [18, 20, 34, 15, 21, 29, 8]. It turned out, however, that the approach of [33] is completely satisfactory if the rest of the default theory is set up correctly =-=[37]-=-. It is, in fact, very general, as discussed in Section 5.2 below. We will see that descriptions of actions in the style of [33, 37] can be used as a basis for planning using answer set solvers. In th... |

55 | Action languages, answer sets, and planning
- Lifschitz
- 1999
(Show Context)
Citation Context ...ystem of this kind created by the inventors of satisability planning. 2 http://www.tcs.hut.fi/Software/smodels/ . 3 http://www.dbai.tuwien.ac.at/proj/dlv/ . 1 In this paper, based on earlier reports [=-=22, 23-=-], applications of answer set programming to planning are discussed from the perspective of the research on the frame problem and nonmonotonic reasoning done in AI since 1980. Specically, we relate th... |

52 | Using Logic Programs with Stable Model Semantics to Solve Deadlock and Reachability Problems for 1-Safe Petri Nets
- Heljanko
- 1999
(Show Context)
Citation Context ...e sophisticated search strategies somewhat similar to those used in ecient satisability solvers. Answer set programming has found applications to several practically important computational problems [=-=30, 35, 16]-=-. One of these problems is planning. 5 Planning 5.1 Example The code in Figures 3-5 allows us to use smodels to solve planning problems in the blocks world. We imagine that blocks are moved by a robot... |

50 | Representing transition systems by logic programs
- Lifschitz, Turner
- 1999
(Show Context)
Citation Context ...The design of language C is based on the system of causal logic proposed in [28]. For a large class of action descriptions in C, an equivalent translation into logic programming notation is dened in [=-=24]-=-. The possibility of such a translation further illustrates the expressive power of the action representation method used in this paper. As noted in the introduction, the answer set programming approa... |

47 | Relating stable models and ai planning domains
- Subrahmanian, Zaniolo
- 1995
(Show Context)
Citation Context ...retation for a set of propositional formulas. This method, known as satisability planning, is used now in several planners. 1 In this paper we discuss a related idea, due to Subrahmanian and Zaniolo [=-=36]: reducing-=- a planning problem to the problem ofsnding an answer set (\stable model") for a logic program. The advantage of this \answer set programming" approach to planning is that the representation... |

42 |
Minimalism subsumes default logic and circumscription
- Bidoit, Froidevaux
- 1987
(Show Context)
Citation Context ...pecically, we relate them to the line of work that started with the invention of default logic [33]|the nonmonotonic formalism that turned out to be particularly closely related to logic programming [=-=2, 26, 10]. Aft-=-er the publication of the \Yale Shooting Scenario" [14] it was widely believed that the solution to the frame problem outlined in [33] was inadequate. Several alternatives have been proposed [18,... |

35 | Simple Causal Minimizations for Temporal Persistence and Projection
- Haugh
- 1987
(Show Context)
Citation Context ... 10]. After the publication of the \Yale Shooting Scenario" [14] it was widely believed that the solution to the frame problem outlined in [33] was inadequate. Several alternatives have been prop=-=osed [18, 20, 34, 15, 21, 29, 8]-=-. It turned out, however, that the approach of [33] is completely satisfactory if the rest of the default theory is set up correctly [37]. It is, in fact, very general, as discussed in Section 5.2 bel... |

33 | Pointwise circumscription: Preliminary report
- Lifschitz
- 1986
(Show Context)
Citation Context ... 10]. After the publication of the \Yale Shooting Scenario" [14] it was widely believed that the solution to the frame problem outlined in [33] was inadequate. Several alternatives have been prop=-=osed [18, 20, 34, 15, 21, 29, 8]-=-. It turned out, however, that the approach of [33] is completely satisfactory if the rest of the default theory is set up correctly [37]. It is, in fact, very general, as discussed in Section 5.2 bel... |

32 |
Autoepistemic logic and formalization of commonsense reasoning
- Gelfond
- 1989
(Show Context)
Citation Context |

32 |
Planning as satis
- Kautz, Selman
(Show Context)
Citation Context ...lications of this method to planning are related to the line of research on the frame problem that started with the invention of formal nonmonotonic reasoning in 1980. 1 Introduction Kautz and Selman =-=[19-=-] proposed to approach the problem of plan generation by reducing it to the problem ofsnding a satisfying interpretation for a set of propositional formulas. This method, known as satisability plannin... |

32 |
Formal theories of action (preliminary report
- Lifschitz
- 1987
(Show Context)
Citation Context |

31 | Fages’ theorem and answer set programming
- Babovich, Erdem, et al.
- 2000
(Show Context)
Citation Context ...m into propositional logic, and,snally, invoking a satisability solver tosnd a plan is a form of satisability planning that can be viewed also as \answer set programming without answer set solvers&quo=-=t; [1]-=-. This is essentially how planning is performed by the Causal Calculator. 4 7 Conclusion In answer set programming, solutions to a combinatorial search problem are represented by answer sets. Plan gen... |

28 |
The anomalous extension problem in default reasoning
- Morris
- 1988
(Show Context)
Citation Context |

23 | Chronological ignorance: Time, nonmonotonicity, necessity and causal theories
- Shoham
- 1986
(Show Context)
Citation Context |

16 | An Application of Action Theory to the Space Shuttle
- Watson
- 1999
(Show Context)
Citation Context ...ing actions is adopted, specifying action preconditions explicitly is unnecessary. The usefulness of indirect descriptions of action domains for applications of AI was demonstrated in a recent report =-=[38]-=- on modelling the Reaction Control System (RCS) of the Space Shuttle. The system consists of several fuel tanks, oxidizer tanks, helium tanks, maneuvering jets, pipes, valves, and other components. Ho... |

14 | Transformations of logic programs related to causality and planning
- Erdem, Lifschitz
- 1999
(Show Context)
Citation Context ...for several atoms A. A program that consists of n rules of this form has 2 n answer sets. For instance, the program p; :p ; q; :q has 4 answer sets: fp; qg; fp; :qg; f:p; qg; f:p; :qg: As observed in =-=[5]-=-, rule (11) can be equivalently replaced in any program by two nondisjunctive rules Asnot :A ; :Asnot A : In the notation of default logic, these rules can be written as : A A ; : :A :A : Alternativel... |

8 |
and Miroslaw Truszczyński. Stable semantics for logic programs and default theories
- Marek
- 1989
(Show Context)
Citation Context ...ecifically, we relate it to the line of work that started with the invention of default logic [23]---the nonmonotonic formalism that turned out to be particularly closely related to logic programming =-=[2, 18, 9]. After th-=-e publication of the "Yale Shooting Scenario" [13] it was widely believed that the solution to the frame problem proposed in [23] was inadequate. It turned out, however, that that solution i... |

4 |
Control knowledge in planning: Bene and tradeos
- Huang, Selman, et al.
- 1999
(Show Context)
Citation Context ...sed by an answer set solver is crucial for progress in this area, just as the possibility of using control knowledge by propositional solvers is crucial for further progress in satisability planning [=-=17-=-]. This is a topic for future work. Acknowledgements Useful comments on preliminary versions of this paper have been provided by Maurice Bruynooghe, Marc Denecker, Esra Erdem, Selim Erdogan, Paolo Fer... |

4 |
Miros law Truszczynski. Stable models and an alternative logic programming paradigm
- Marek
- 1999
(Show Context)
Citation Context ...as dened in [9, 10, 25] and its relation to default logic. Then we describe some of the computational possibilities of answer set solvers (Section 3) and illustrate the answer set programming method [=-=27, 30]-=- by applying it to a graphtheoretic search problem (Section 4). In Section 5 we turn to the use of answer set solvers for plan generation. Section 6 describes the relation of this work to other resear... |

2 |
Timo Soininen. Extending the stable model semantics. This journal
- Niemela, Simons
(Show Context)
Citation Context ...d from these 6 atoms. Symbol j at the beginning of the rule restricts the answer sets to those whose cardinality is at least j. This is an instance of the \cardinality" construct available in smo=-=dels [31]-=-. It allows the user to bound, from below and from above, the number of atoms of a certain form that are included in the answer set. (A lower bound is placed to the left of the expression in braces, a... |

1 |
Miros law Truszczynski. Stable semantics for logic programs and default theories
- Marek
- 1989
(Show Context)
Citation Context ...pecically, we relate them to the line of work that started with the invention of default logic [33]|the nonmonotonic formalism that turned out to be particularly closely related to logic programming [=-=2, 26, 10]. Aft-=-er the publication of the \Yale Shooting Scenario" [14] it was widely believed that the solution to the frame problem outlined in [33] was inadequate. Several alternatives have been proposed [18,... |