A control of hexapod based on the neural networks solution for forward kinematics

Abstract

This research is a part of the work implemented by BSTU "Voenmeh" under the financial support of the Ministry of Education and Science of the Russian Federation for design and development of a precision mechanism with the parallel kinematics called "Hexapod". In the paper the inverse kinematics problem of Hexapod is described. The solution of the forward kinematics problem is solved by the iterative Newton-Raphson method. A solution of the forward kinematics problem and approximation of the Jacobi matrix of the hexapod based on the apparatus of artificial neural networks is proposed. The results of synthesis of neural networks are presented. An algorithm for the application of neural network kinematics in the hexapode control system is described. A model of a digital control system based on the neural network algorithm under investigation is implemented. The estimation of control accuracy in the positional mode is obtained.