Neural Network Applications for Robotic Motion Control
Toshio Fukuda and Takanori Shibata
Nagoya University, Dept. of Mechanical Engineering, 1 Furo-cho, Chikusa-ku, Nagoya 464-01, Japan
This paper deals with neural network applications for the robotic motion control. The neural network can be employed for both the long term “learning” of the control process and the short term “adaptation” of the dynamic process. In this paper, we demonstrate some dynamic controls of robotic manipulators using the “Neural Servo Controller” which is applicable to the position and force control of robotic manipulators. The “Neural Servo Controller” is based on the neural network which here consists of two hidden layers and input/output layers. The controller can adjust the neural network output to the robot in the forward manner to cooperate with the feedback loop, depending on unknown characteristics of handling objects. In particular, the proposed neural network has time delay elements in itself, so that the neural network can learn the dynamics of the system. Simulations are carried out for position and force control of a two dimensional robotic manipulator. Moreover, we propose a “Fuzzy Turbo” so that the neural network can learn the dynamic system quickly. The results show the applicability and adaptability of the proposed “Neural Servo Controller” to the nonlinear and dynamic system, and the ability of the proposed “Fuzzy Turbo” on the adaptive process.
This article is published under a Creative Commons Attribution-NoDerivatives 4.0 International License.