Evolutionary fuzzy extreme learning machine for inverse kinematic modeling of robotic arms

Abstract

Evolutionary fuzzy extreme learning machine (EF-ELM) is one of the neuro-fuzzy system, which combines the learning capabilities of extreme learning machine (ELM) and the explicit knowledge of the fuzzy systems. In EF-ELM, the differential evolutionary technique is used to tune the membership function parameters were as the consequent parameters are tuned by Moore-Penrose generalized inverse techniques. In this paper, inverse kinematic modelings of 2-DOF and 3-DOF robotic arms are proposed. Evolutionary fuzzy extreme learning machine is used to predict the inverse kinematics of robotic arms. Extensive simulations are performed to study the prediction behavior of EF-ELM and comparative analysis is included against ELM and back propagation (BP) based neural networks. It is observed that the EF-ELM technique produces good generalization with minimum root mean square error for predicting the inverse kinematics solution of robotic arms.