Skip to Main content Skip to Navigation

Traduction logique et résolution de problèmes : application à la planification

Maël Valais 1, 2
1 IRIT-ADRIA - Argumentation, Décision, Raisonnement, Incertitude et Apprentissage
IRIT - Institut de recherche en informatique de Toulouse
2 IRIT-LILaC - Logique, Interaction, Langue et Calcul
IRIT - Institut de recherche en informatique de Toulouse
Abstract : This thesis deals with logical translation and solving of problem using solvers. In particular, we are interested in solving planning problems in Artificial Intelligence. We present the automatic translator TouIST that we developed and that allows us to use a simple language to generate logical formulas from a problem description. Our tool allows us to model many static or dynamic combinatorial problems as Sudoku, Takuzu or Nim game, and to benefit from the regular improvements to SAT, QBF or SMT solvers to solve them efficiently. We then present reference encodings to solve classical planning problems with SAT and QBF, or temporal planning problems with SMT. In each case, we introduce new encodings in plan-spaces based on open condition modeling to represent causal links. Finally, we show, thanks to an experimental study, that our encodings are more efficient than the existing ones on the reference problems of international planning competitions (IPC)
Document type :
Complete list of metadata

Cited literature [103 references]  Display  Hide  Download
Contributor : Abes Star :  Contact
Submitted on : Wednesday, July 8, 2020 - 3:10:52 PM
Last modification on : Wednesday, November 3, 2021 - 7:16:36 AM
Long-term archiving on: : Monday, November 30, 2020 - 3:19:35 PM


Version validated by the jury (STAR)


  • HAL Id : tel-02893759, version 1


Maël Valais. Traduction logique et résolution de problèmes : application à la planification. Performance et fiabilité [cs.PF]. Université Paul Sabatier - Toulouse III, 2019. Français. ⟨NNT : 2019TOU30079⟩. ⟨tel-02893759⟩



Les métriques sont temporairement indisponibles