Adaptive Neuro Fuzzy Inference System (ANFIS) for Generation of Joint Angle Trajectory

Abstract

In this paper, Adaptive Neuro-Fuzzy Inference System is utilized to learn from training data and create ANFIS with limited mathematical representation of the system. The proposed system consists of three phases i.e. Generation of training data, Execution of ANFIS, Generation of joint angle trajectory. The schematic of the proposed system is shown in Figure 4. The predicted joint angle configurations are further to be used to determine the trajectory for the task execution of the robot. The simulation studies conducted on a 5-DOF SCORBOT ER-IV robot manipulator shows the effectiveness of the approach over conventional techniques