TY - GEN
T1 - Combining Specialized Reasoners and General Purpose Planners
T2 - 9th National Conference on Artificial Intelligence, AAAI 1991
AU - Kambhampati, Subbarao
AU - Cutkosky, Mark
AU - Tenenbaum, Marty
AU - Lee, Soo Hong
N1 - Funding Information: Many realistic planning problems require significant amounts of deep domain-specific reasoning. As an example, process planning for machining involves extensive reasoning about geometry, kinematics and cutting and clamping forces. The classical planning framework, in which the planner is modeled as an isolated module with all knowledge relevant to plan generation at its disposal, is inadequate for addressing such problems because it is impractical to encode deep models of specialized considerations in the constrained-action representations used by classical planners. While extending the action representation sufficiently to encode these considerations is possible, the cost of planning becomes prohibitive as the expressiveness of the domain models increases [ 141. Most previous approaches for planning in such situations have dealt with these issues either through very domain specific planning algorithms (e.g. [5]), or by restricting themselves to shallow models of the specialized considerations (e.g. [3, 161). *We acknowledge the support of Office of Naval Research under contract N00014-88-K-0620T.h e authors’ ePncse’l addresses are [email protected], marty@c&s.stanford.edu, [email protected] and [email protected],edu. Publisher Copyright: Copyright © 1991, AAAI (www.aaai.org). All rights reserved.
PY - 1991
Y1 - 1991
N2 - Many real-world planning problems involve substantial amounts of domain-specific reasoning that is either awkward or inefficient to encode in a general purpose planner. Previous approaches for planning in such domains have either been largely domain specific or have employed shallow models of the domain-specific considerations. In this paper we investigate a hybrid planning model that utilizes a set of specialists to complement both the overall expressiveness and the reasoning power of a traditional hierarchical planner. Such a model retains the flexibility and generality of classical planning framework while allowing deeper and more efficient domain-specific reasoning through specialists. We describe a preliminary implementation of a planning architecture based on this model in a manufacturing planning domain, and use it to explore issues regarding the effect of the specialists on the planning, and the interactions and interfaces between them and the planner.
AB - Many real-world planning problems involve substantial amounts of domain-specific reasoning that is either awkward or inefficient to encode in a general purpose planner. Previous approaches for planning in such domains have either been largely domain specific or have employed shallow models of the domain-specific considerations. In this paper we investigate a hybrid planning model that utilizes a set of specialists to complement both the overall expressiveness and the reasoning power of a traditional hierarchical planner. Such a model retains the flexibility and generality of classical planning framework while allowing deeper and more efficient domain-specific reasoning through specialists. We describe a preliminary implementation of a planning architecture based on this model in a manufacturing planning domain, and use it to explore issues regarding the effect of the specialists on the planning, and the interactions and interfaces between them and the planner.
UR - https://www.scopus.com/pages/publications/84891474141
UR - https://www.scopus.com/pages/publications/84891474141#tab=citedBy
M3 - Conference contribution
T3 - Proceedings of the 9th National Conference on Artificial Intelligence, AAAI 1991
SP - 199
EP - 205
BT - Proceedings of the 9th National Conference on Artificial Intelligence, AAAI 1991
PB - AAAI press
Y2 - 14 July 1991 through 19 July 1991
ER -