ニューロ・ファジィ技術の最適制御問題への応用

概要

論文の詳細を見る
The optimum control problem is one of the most interesting problems to apply AI technology, because of its theoretical diffieulty and its human-related characteristics. The minimum time mass movement problem is one of the classical optimum control problems and many studies have been carried out. In this paper as the first step to complex problems, applicability of neuro-fuzzy technology for the simple problems with and without friction force is clarified in comparison with theoretically known solutions. As a conclusion, normalization of state variables and neural net-work capabilities in tuning of membership-functions of fuzzy rules and in representaion of normalized functions are found to play very important roles in solving these preblems.
津山工業高等専門学校の論文
1995-11-01