Planning and Problem Solving
Abstract
Problem solving is the process of developing a sequence of actions to achieve a goal. This broad definition admits all goal-directed artificial intelligence programs to the ranks of problem solvers; for example, MYCIN solves the problem of determining a bacteremia infection. HARPY solves the problem of understanding speech signals, and solves the problem of filling in slots in its representations of concepts. It follows that this chapter is not about problem solvers-the entire Handbook is about problem solvers. This, is about problem- solving techniques. In particular, it is about planning. In everyday terms, planning means deciding on a course of action before acting. This definition accurately describes the planning systems of this chapter, so we will adopt it. A plan is, thus, a representation of a course of action. It can be an unordered list of goals, such as a grocery list, but usually a plan has an implicit ordering of its goals; for example, most people plan to get dressed to go to the theater, not the other way around. Many plans include steps that are vague and require further specification. These serve as placeholders in a plan; for example, a daily plan includes the goal eat-lunch, although the details-where to eat, what to eat, when to leave- are not specified. Reprints.
Document Details
- Document Type
- Technical Report
- Publication Date
- Oct 01, 1982
- Accession Number
- ADA222298
Entities
People
- Paul R. Cohen
Organizations
- Stanford University