Supporting Flexible Plan Reuse

Publisher Summary This chapter discusses a framework called PRIAR that effectively supports retrieval and modification of plans for reuse. PRIAR enables a hierarchical nonlinear planner to improve its planning by reusing previously generated plans. PRIAR uses the causal structures of plans to judge the appropriateness of reusing them in new problem situations and to modify them minimally to make them solve new problems. PRIAR provides an integrated framework for supporting reuse within hierarchical nonlinear planning. Hierarchical nonlinear planning is a dominant method of abstraction and least commitment in domain-independent planning. In this approach, plans are represented as partially ordered network of tasks at various levels of abstraction. PRIAR facilitates the modification of a hierarchical plan by justifying the individual decisions underlying its development in terms of its causal dependency structure. PRIAR also provides a domain-independent similarity metric based on the plan validation structure to measure the utility of modifying a given plan to solve a new problem. The key insight used in developing the similarity metric is that the modification cost depends on the number and types of inconsistencies that are caused by the specifications changes in the plan validation structure.

[1]  Kristian J. Hammond,et al.  Explaining and Repairing Plans that Fail , 1987, IJCAI.

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

[3]  Steven Minton Issues in the Design of Operator Composition Systems , 1990, ML.

[4]  James A. Hendler,et al.  Control of Refitting during Plan Reuse , 1989, IJCAI.

[5]  Austin Tate,et al.  Generating Project Networks , 1977, IJCAI.

[6]  Subbarao Kambhampati,et al.  Approach toward incremental and interactive planning for concurrent product and process design , 1990 .

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

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

[9]  Patrick J. Hayes,et al.  A Representation For Robot Plans , 1975, IJCAI.

[10]  Qiang Yang,et al.  The Downward Refinement Property , 1991, IJCAI.

[11]  David E. Wilkins,et al.  Domain-Independent Planning: Representation and Plan Generation , 1984, Artif. Intell..

[12]  Subbarao Kambhampati,et al.  Combining Specialized Reasoners and General Purpose Planners: A Case Study , 1991, AAAI.

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

[14]  Earl D. Sacerdoti,et al.  The Nonlinear Nature of Plans , 1975, IJCAI.

[15]  Richard E. Korf,et al.  Planning as Search: A Quantitative Approach , 1987, Artif. Intell..

[16]  Jaime G. Carbonell,et al.  Derivational Analogy and Its Role in Problem Solving , 1983, AAAI.

[17]  Subbarao Kambhampati,et al.  Mapping and Retrieval During Plan Reuse: A Validation Structure Based Approach , 1990, AAAI.

[18]  David E. Wilkins,et al.  Recovering from execution errors in SIPE , 1985, Comput. Intell..

[19]  Janet L. Kolodner,et al.  Reconstructive Memory: A Computer Model , 1983, Cogn. Sci..