Overview Structure, Solutions and Filtering Search and Search Space Reduction Experimental Results Conclusion and Future Work Filtering, Decomposition and Search Space Reduction for Optimal Sequential Planning (FDP System) St´ ephane Grandcolas Cyril Pain-Barre InCA Team Laboratoire des Sciences de l’Information et des Syst` emes (LSIS) UMR CNRS 6168 Marseille, France St´ ephane Grandcolas, Cyril Pain-Barre FDP System. AAAI’07, 24 July 2007, Vancouver 1/36
Overview Structure, Solutions and Filtering Search and Search Space Reduction Experimental Results Conclusion and Future Work Plan of the Talk Overview 1 Framework Quick Presentation Structure, Solutions and Filtering 2 Planning-Structure Solution Consistency and Filtering Search and Search Space Reduction 3 Search and Problem Decomposition Search Space Reduction Experimental Results 4 Conclusion and Future Work 5 St´ ephane Grandcolas, Cyril Pain-Barre FDP System. AAAI’07, 24 July 2007, Vancouver 2/36
Overview Structure, Solutions and Filtering Framework Search and Search Space Reduction Quick Presentation Experimental Results Conclusion and Future Work Overview 1 Framework Quick Presentation Structure, Solutions and Filtering 2 Planning-Structure Solution Consistency and Filtering Search and Search Space Reduction 3 Search and Problem Decomposition Search Space Reduction Experimental Results 4 Conclusion and Future Work 5 St´ ephane Grandcolas, Cyril Pain-Barre FDP System. AAAI’07, 24 July 2007, Vancouver 3/36
Overview Structure, Solutions and Filtering Framework Search and Search Space Reduction Quick Presentation Experimental Results Conclusion and Future Work A Hybrid Planner FDP (Filtering and Decomposition for Planning) optimal sequential planner classical planning assumptions several techniques: from planning : consistency rules goal reachability estimation (future) heuristic search from the CSP field : filtering decomposition no-good recording St´ ephane Grandcolas, Cyril Pain-Barre FDP System. AAAI’07, 24 July 2007, Vancouver 4/36
Overview Structure, Solutions and Filtering Framework Search and Search Space Reduction Quick Presentation Experimental Results Conclusion and Future Work FDP in a Nutshell planning-structure kind of planning graph incrementally extended with a given maximal size (no termination criterion on unsolvable problems) filtering of inconsistent actions and literals search for a solution in a planning-structure currently, a kind of depth-first search (complete procedure) several decomposition strategies reduction of the search space : no-good recording goal-reachability estimation pruning of some redundant action sequences St´ ephane Grandcolas, Cyril Pain-Barre FDP System. AAAI’07, 24 July 2007, Vancouver 5/36
Overview Structure, Solutions and Filtering Framework Search and Search Space Reduction Quick Presentation Experimental Results Conclusion and Future Work FDP in a Nutshell planning-structure kind of planning graph incrementally extended with a given maximal size (no termination criterion on unsolvable problems) filtering of inconsistent actions and literals search for a solution in a planning-structure currently, a kind of depth-first search (complete procedure) several decomposition strategies reduction of the search space : no-good recording goal-reachability estimation pruning of some redundant action sequences St´ ephane Grandcolas, Cyril Pain-Barre FDP System. AAAI’07, 24 July 2007, Vancouver 5/36
Overview Structure, Solutions and Filtering Planning-Structure Search and Search Space Reduction Solution Experimental Results Consistency and Filtering Conclusion and Future Work Overview 1 Framework Quick Presentation Structure, Solutions and Filtering 2 Planning-Structure Solution Consistency and Filtering Search and Search Space Reduction 3 Search and Problem Decomposition Search Space Reduction Experimental Results 4 Conclusion and Future Work 5 St´ ephane Grandcolas, Cyril Pain-Barre FDP System. AAAI’07, 24 July 2007, Vancouver 6/36
Overview Structure, Solutions and Filtering Planning-Structure Search and Search Space Reduction Solution Experimental Results Consistency and Filtering Conclusion and Future Work Planning-Structure : Definition Given the problem P = ( A , I , G , L ), � k , V a , V l , d � is a planning-structure for P where: k : size of the planning-structure, V a = { y 0 , . . . , y k − 1 } : action variables , V l = { x i , l } 0 ≤ i ≤ k , l ∈ L : literal variables , d : domain function ∀ y i ∈ V a �− → domain d ( y i ) ⊆ A , denoted A i , ∀ x i , l ∈ V l �− → domain d ( x i , l ) ⊆ {⊤ , ⊥} , denoted D i , l . St´ ephane Grandcolas, Cyril Pain-Barre FDP System. AAAI’07, 24 July 2007, Vancouver 7/36
Overview Structure, Solutions and Filtering Planning-Structure Search and Search Space Reduction Solution Experimental Results Consistency and Filtering Conclusion and Future Work Planning-Structure : Graphical Representation planning-structure of size k action steps A A 1 A k−1 0 . . . L L L L L 0 1 2 k−1 k literal steps St´ ephane Grandcolas, Cyril Pain-Barre FDP System. AAAI’07, 24 July 2007, Vancouver 8/36
Overview Structure, Solutions and Filtering Planning-Structure Search and Search Space Reduction Solution Experimental Results Consistency and Filtering Conclusion and Future Work Planning-Structure : Detailed Representation Step i of the planning-structure A i the literal p p variable x i,p has a 1 domain { , } q q a 2 r r a 3 the literal s s variable has x i,s domain { } L L i+1 i action variable y i has domain { , } a 1 a 3 Initially, all the variables are undefined literal variables have {⊤ , ⊥} as domain action variables have A as domain St´ ephane Grandcolas, Cyril Pain-Barre FDP System. AAAI’07, 24 July 2007, Vancouver 9/36
Overview Structure, Solutions and Filtering Planning-Structure Search and Search Space Reduction Solution Experimental Results Consistency and Filtering Conclusion and Future Work Valid Plan in a Planning-Structure p p q a 1 a 1 a 1 q r a 2 a 2 a 2 r a 3 a 3 a 3 s s t t goals A valid plan is an assignment of the variables of V l ∪ V a that : respects the domains valids the initial state literals (closed-world assumption) valids the goals in the last state respects the preconditions and effects of the chosen actions respects the persistence of the literals not affected by actions St´ ephane Grandcolas, Cyril Pain-Barre FDP System. AAAI’07, 24 July 2007, Vancouver 10/36
Overview Structure, Solutions and Filtering Planning-Structure Search and Search Space Reduction Solution Experimental Results Consistency and Filtering Conclusion and Future Work Preliminary Removals in the Structure Before searching, some values for literals are removed : p p a 1 a 1 a 1 q q a 2 a 2 a 2 r r a 3 a 3 a 3 s s in L 0 according to I in L k according to G ∀ l ∈ I , ⊥ is removed ∀ l ∈ G , ⊥ is removed from D 0 , l from D k , l ∀ p (proposition) �∈ I , ⊤ is removed from D 0 , p Consequences some actions and some literals become then inconsistent St´ ephane Grandcolas, Cyril Pain-Barre FDP System. AAAI’07, 24 July 2007, Vancouver 11/36
Overview Structure, Solutions and Filtering Planning-Structure Search and Search Space Reduction Solution Experimental Results Consistency and Filtering Conclusion and Future Work Inconsistent Values for Literal and Action Variables 5 possibilities for a literal value to become inconsistent logical consistency forward persistence backward persistence all actions delete opposite always required 3 possibilities for an action value to become inconsistent falsified precondition falsified effect effect required St´ ephane Grandcolas, Cyril Pain-Barre FDP System. AAAI’07, 24 July 2007, Vancouver 12/36
Overview Structure, Solutions and Filtering Planning-Structure Search and Search Space Reduction Solution Experimental Results Consistency and Filtering Conclusion and Future Work Consistency Rules and Filtering : Example Initial situation A i p p p p a 1 q q a 2 r r a 3 s s a 4 t t L L i+1 i St´ ephane Grandcolas, Cyril Pain-Barre FDP System. AAAI’07, 24 July 2007, Vancouver 13/36
Overview Structure, Solutions and Filtering Planning-Structure Search and Search Space Reduction Solution Experimental Results Consistency and Filtering Conclusion and Future Work Consistency Rules and Filtering : Example Logical consistency A i logical consistency p p p p a 1 q q a 2 r r a 3 s s a 4 t t L L i+1 i St´ ephane Grandcolas, Cyril Pain-Barre FDP System. AAAI’07, 24 July 2007, Vancouver 13/36
Overview Structure, Solutions and Filtering Planning-Structure Search and Search Space Reduction Solution Experimental Results Consistency and Filtering Conclusion and Future Work Consistency Rules and Filtering : Example Forward persistence A i p p p p a 1 q q a 2 r r a 3 s s a 4 t t forward persistence L L i+1 i St´ ephane Grandcolas, Cyril Pain-Barre FDP System. AAAI’07, 24 July 2007, Vancouver 13/36
Recommend
More recommend
