Applicability Study of the Active Search Algorithm Based on Machine Learning Scheme in the Case of Servo Tuning Problems
Akio Noda, Hikaru Nagano, Tatsuya Nagatani, Yukiyasu Domae, Tetsuaki Nagano, Ken‐ichi Tanaka, Tsukasa Ogasawara · Transactions of the Society of Instrument and Control Engineers · 2017
Optimization task of the control parameters of industrial positioning systems is a daily occurrence. It is necessary to explore unknown response landscape of the system by performing plural sampling of “control parameters and output combination” if the request specification is severe. Skilled operators have been conducting such tasks based on their experience and knowledge. The challenge is “The Optimization of Unknown Objective Function”. In their study, the authors have proposed their original optimization algorithm as a solution. This paper reports applicability study of their algorithm. Experimental results discovered that the algorithm found the optimal control parameter in 100 combination data set for seven times trial with some required specifications.