Formalizing a Spectrum of Plan Generalizations Based on Modal Truth Criteria

Plan generalizations are of considerable importance in improving planning performance through reuse. In this paper, we provide a unified framework for analytic (non-inductive) plan generalization based on explanation of plan correctness with respect to modal truth criteria. Within this framework, we explore a large spectrum of generalizations based on the type of constraints on the plan that are being relaxed (ordering, binding, initial state specification), the strength and type of truth criteria used for explaining correctness and the number of explanations used as a basis for generalization. Apart from the straightforward precondition and order generalizations, the spectrum also includes such novel ones as possible correctness generalizations and disjunctive generalizations. In each case, we characterize the cost of producing the generalization, and the storage and usage advantages provided by it during plan reuse. Although there has been previous work on plan generalizations, this is the first time that a complete spectrum of generalizations are characterized within a single unified framework, facilitating a comparative analysis of their costs and benefits. Submitted to AIPS-94 (2nd Intl. Planning Conference) Content Areas: Learning, Explanation-based Generalization Submitted also to Canadian AI Conference, 1994 This research is supported in part by National Science Foundation under grant IRI-9210997, and ARPA/Rome Laboratory planning initiative under grant F30602-93-C-0039.

[1]  S. Kambhampati.  Characterizing multi-contributor causal structures for planning , 1992 .

[2]  Steven Minton,et al.  Quantitative Results Concerning the Utility of Explanation-based Learning , 1988, Artif. Intell..

[3]  Steven Minton,et al.  Total Order vs. Partial Order Planning: Factors Influencing Performance , 1992, KR.

[4]  James A. Hendler,et al.  A Validation-Structure-Based Theory of Plan Modification and Reuse , 1992, Artif. Intell..

[5]  Scott Bennett,et al.  A Domain Independent Explanation-Based Generalizer , 1986, AAAI.

[6]  David Chapman,et al.  Planning for Conjunctive Goals , 1987, Artif. Intell..

[7]  Eugene Fink,et al.  A Spectrum of Plan Justifications , 1993, AAAI 1993.

[8]  Subbarao Kambhampati,et al.  On the Nature and Role of Modal Truth Criteria in Planning , 1996, Artif. Intell..

[9]  M. Veloso,et al.  Nonlinear Planning with Parallel Resource Allocation , 1990 .

[10]  Richard Fikes,et al.  Learning and Executing Generalized Robot Plans , 1993, Artif. Intell..

[11]  Steve Ankuo Chien.  An explanation-based learning approach to incremental planning , 1991 .

[12]  Subbarao Kambhampati,et al.  Relative Utility of EBG based Plan Reuse in Partial Ordering vs. Total Ordering Planning , 1993, AAAI.

[13]  Subbarao Kambhampati,et al.  A Unified Framework for Explanation-Based Generalization of Partially Ordered and Partially Instantiated Plans , 1994, Artif. Intell..

[14]  Subbarao Kambhampati,et al.  Explanation-Based Generalization of Partially Ordered Plans , 1991, AAAI.

[15]  David A. McAllester,et al.  Systematic Nonlinear Planning , 1991, AAAI.

[16]  Raymond J. Mooney,et al.  Generalizing the Order of Operators in Macro-Operators , 1988, ML.