Identifying and Exploiting Features for Effective Plan Retrieval in Case-Based Planning

Case-Based planning can fruitfully exploit knowledge
gained by solving a large number of problems, storing
the corresponding solutions in a plan library and reusing
them for solving similar planning problems in the future.
Case-based planning is extremely effective when
similar reuse candidates can be efficiently chosen.
In this paper, we study an innovative technique based
on planning problem features for efficiently retrieving
solved planning problems (and relative plans) from
large plan libraries. A problem feature is a characteristic
of the instance that can be automatically derived from
the problem specification, domain and search space
analyses, and different problem encodings.
Since the use of existing planning features are not always
able to effectively distinguish between problems
within the same planning domain, we introduce a new
class of features.
An experimental analysis in this paper shows that our
features-based retrieval approach can significantly improve
the performance of a state-of-the-art case-based
planning system.