SOM Network with Weighted Least Norm Matrix Based Redundancy Resolution for a 5-DOF Spatial Robotic Manipulator

Abstract

This paper proposes an intelligent control technique for a spatial robotic manipulator to track the desired 3D trajectory while resolving redundancy (joint limit avoidance). The proposed control strategy is designed based on Self Organizing Map (SOM) and the Weighted Least Norm (WLN) concept, which eliminates the Jacobian pseudo inverse computation of the spatial robotic manipulator. A thorough simulation study is performed on 5 degrees of freedom (DOF) spatial robotic manipulator for 3D trajectories (line, circular, and elliptical) tracking objectives. To show the efficacy of our proposed approach, we have compared the performance of our developed algorithm with the existing standard algorithms, such as task decomposition and WLN. From the simulation and experimental results, it is strongly observed that the combination of SOM network and WLN matrix methodology can generate a smooth minimum joint angle space motion.