Counterexample-Guided Cartesian Abstraction Refinement

Counterexample-guided abstraction refinement (CEGAR) is a method for incrementally computing abstractions of transition systems. We propose a CEGAR algorithm for computing abstraction heuristics for optimal classical planning. Starting from a coarse abstraction of the planning task, we iteratively compute an optimal abstract solution, check if and why it fails for the concrete planning task and refine the abstraction so that the same failure cannot occur in future iterations. A key ingredient of our approach is a novel class of abstractions for classical planning tasks that admits efficient and very fine-grained refinement. Our implementation performs tens of thousands of refinement steps in a few minutes and produces heuristics that are often significantly more accurate than pattern database heuristics of the same size.

[1]  Drew McDermott,et al.  The 1998 AI Planning Systems Competition , 2000, AI Mag..

[2]  Jörg Hoffmann,et al.  Relaxation Refinement: A New Method to Generate Heuristic Functions , 2009, MoChArt.

[3]  Andreas Podelski,et al.  Boolean and Cartesian abstraction for model checking C programs , 2001, International Journal on Software Tools for Technology Transfer.

[4]  Malte Helmert,et al.  Efficient Implementation of Pattern Database Heuristics for Classical Planning , 2021, SOCS.

[5]  Jonathan Schaeffer,et al.  Pattern Databases , 1998, Comput. Intell..

[6]  Edmund M. Clarke,et al.  Counterexample-guided abstraction refinement , 2003, 10th International Symposium on Temporal Representation and Reasoning, 2003 and Fourth International Conference on Temporal Logic. Proceedings..

[7]  Bernhard Nebel,et al.  COMPLEXITY RESULTS FOR SAS+ PLANNING , 1995, Comput. Intell..

[8]  S. Edelkamp.  Planning with Pattern Databases , 2014 .

[9]  Patrik Haslum.  Incremental Lower Bounds for Additive Cost Planning Problems , 2012, ICAPS.

[10]  Robert C. Holte,et al.  Experiments with Automatically Created Memory-Based Heuristics , 2000, SARA.

[11]  Jörg Hoffmann,et al.  Computing Perfect Heuristics in Polynomial Time: On Bisimulation and Merge-and-Shrink Abstraction in Optimal Planning , 2011, IJCAI.

[12]  Malte Helmert,et al.  Concise finite-domain representations for PDDL planning tasks , 2009, Artif. Intell..

[13]  Patrik Haslum,et al.  Semi-Relaxed Plan Heuristics , 2012, ICAPS.

[14]  Patrik Haslum,et al.  Flexible Abstraction Heuristics for Optimal Sequential Planning , 2007, ICAPS.

[15]  Krishnendu Chatterjee,et al.  Counterexample-guided Planning , 2005, UAI.

[16]  Jendrik Seipp.  Counterexample-guided Abstraction Refinement for Classical Planning Master's Thesis Erklärung (declaration) Zusammenfassung , 2013 .