Loading...
| Friend's email | |
| Your name | |
| Your email | |
| enter code | |
This page was sent successfuly
- Type of Document: M.Sc. Thesis
- Language: Farsi
- Document No: 44901 (19)
- University: Sharif University of Technology
- Department: Computer Engineering
- Advisor(s): Ghassem-Sani, Gholamreza
- Abstract:
- Planning, as one of the major area of the artificial intelligence, has substantial applications in different science and technology fields. In the classical planning, usually some restricted assumptions are made to simplify the planning procedure. For example, for the world representation, states and actions are defined precisely; however, in the real world such information is mostly inacurate. Therefore, in order to address this inacuracy, flexible planning has been developed. In this technique, robust restrictions such as complete satisfaction of preconditions are interpreted as preferences or priority. In other words, the flexible planning aims at fulfilling planning goals with maximum satisfaction degrees, instead of completely satisfying them. In this field, up to date, some methods such as Flexible Graph Plan and Fuzzy CSP have been presented, which employ the fuzzy concept to build an inacurate model. Besides, in the classical planning area, heuristic planning methods such as FF have demonstrated proper outcomes.
This project aims at introducing a novel flexible planning technique incorporating heuristic methods in classical planning such as FF and some other state of the art planners. A comparison between results and performance of the suggested methods and other alternative techniques is presented to validate the functionality of these new planning techniques with a proper satisfaction degree - Keywords:
- Classical Planning ; Heuristic Search ; Flexible Planning ; Artificial Intelligence ; Heuristic Planning
Digital Object List
-
محتواي کتاب
- view
Bookmark