TAILIEUCHUNG - Using active learning in motor control and matlab simulation
Designing any intelligent control system is possible by modeling. In this paper, solid modeling, and the four axes of motion of a robot arm, the starting point positioning with servo motors of the simulation of neural networks with active learning strategies which types of learning are presented. | Journal of Automation and Control Engineering Vol. 3, No. 4, August 2015 Using Active Learning in Motor Control and Matlab Simulation Ercan Nurcan Yilmaz Department of Electrical & Electronic Engineering, Faculty of Technology, Gazi University, Teknikokullar, Ankara, Turkey E-mail: enyilmaz@ Onur Battal Haci Bektas Veli Vocational School, Nevşehir University, Hacibektas, Nevsehir, Turkey E-mail: onurbattal@ Abstract—People's desire to produce systems capable of learning and decision making, has led to the concept of artificial intelligence. One of the many ways that is used in the design of intelligent systems is artificial neural networks. Artificial neural networks are computational networks which attempt to simulate the networks of nerve cell (neurons) like central nervous system of the living. Designing any intelligent control system is possible by modeling. In this paper, solid modeling, and the four axes of motion of a robot arm, the starting point positioning with servo motors of the simulation of neural networks with active learning strategies which types of learning are presented. Index Terms—active learning, artificial neural networks, servo motor I. INTRODUCTION In this study, we would like to bring a novel active machine learning simulation result in order to discuss for which problems the autonomous learning loop can be closed using learning, and to identify the machine learning methods that can be used to close it. Training phase of learner systems that will be trained with supervised learning is based on input and output data set that produced by this system. Sometimes, training data sets cannot be easily achieved in complex systems by using the system modeling as a result of modeling and control of these systems difficult by using artificial neural networks. In addition, the approach function quality used for the training of the neural network and longer duration of training depending on the size of the training data
đang nạp các trang xem trước
