Efficient Method for Inverse Dynamics of Robot Manipulators by using Adaptive-network-based Fuzzy Inference System

Abstract

The inverse dynamics calculation of a robot manipulator is essential in a robot manipulator control system. The commonly used approaches to formulate dynamic models of robot manipulators are Newton-Euler and Lagrange-Euler methods. Since both methods are numerical recursive methods, they are computationally expensive and not suitable to be used directly in real time applications. In this paper, the adaptive-network-based fuzzy inference system (ANFIS) is used to construct the input-output mapping of inverse dynamics model of a robot manipulator. To create the training data sets for the ANFIS systems, the dynamic model of a 5 degree-of-freedom robot manipulator is computed using Newton-Euler dynamic formulation. The proposed method is tested in formulating the required torques of a point-to-point trajectory. The results show that the proposed method possesses shorter operational time and its performance is comparable to numerical method. The proposed method can be used for the rigid-body robot manipulators whose dynamical characteristics are known.