Adding Preferences to Answer Set Planning

Abstract

Planning—in its classical sense—is the problem of finding a sequence of actions that achieves a predefined goal. As such, much of the research in AI planning has been focused on methodologies and issues related to the development of efficient planners. To date, several efficient planning systems have been developed (e.g., see [3] for a summary of planners that competed in the International Conference on Artificial Intelligent Planning and Scheduling). These developments can be attributed to the discovery of good domain-independent heuristics, the use of domain-specific knowledge, and the development of efficient data structures used in the implementation of the planning algorithms. Logic programming has played a significant role in this line of research, providing a declarative framework for the encoding of different forms of knowledge and its effective use during the planning process [5].