Modelling of the adaptive neuro-fuzzy inference system based controlof 5-DOF robotic manipulator “Intelbot”

Author(s):  Thu Rain, Kursk State University, Kursk, Russia, thurein.48@gmail.com

V.M. Dovgal, Dr., Kursk State University, Kursk, Russia

Yan Naing Soe, South-West State University, Kursk, Russia

Issue:  Volume 45, № 3

Rubric:  Computer simulation history

Annotation:  This article lays attention on the usage of a hybrid artificial intelligent technique called ‘ANFIS’ in kinematics control of robotic manipulator, “Intelbot”. In this article, by using the Adaptive Networkbased Fuzzy Inference System (ANFIS) to learn from training data, the ANFIS for kinematic control of Intelbot was created with mathematical representation of the system. In the proposed method, forward kinematics relations of robot manipulator “Intelbot”are used to obtain the orientation and position of the end-effector of the Intelbot. To reduce the number of inputs, the processing time and memory space that need to train the ANFIS, the received orientation values are converted from the form of rotation matrices to the form of quaternions. And then the training data sets were created using the quaternion values of orientations, the coordinates of positions and joint-angle values of Intelbot. For simulation the 3-D model of Intelbot was created using SolidWorks. Simulation is carried out with the help of MATLAB/NeuroFuzzyDesigner /Robotics Toolbox/GUIDE to show the effectiveness of the proposed method. The experiments were carried out on the actual physical model of 5-DOF robotic manipulator “Intelbot”

Keywords:  ANFIS, adaptive network-based fuzzy inference system, Intelbot, artificail intelligent, control, kinematics, Matlab, modelling, robotic manipulator

Full text (PDF):  Download

Downloads count:  405