Kinematic Control of Redundant Robot Arms Using Neural Networks

7Neural Network Based Stewart Platform Control

7.1 Introduction

Kinematically redundant manipulators are referred to as those which have more degrees of freedom (DOFs) than required to determine the position and orientation. The redundancy of parallel manipulators can be utilized to overcome the obstacles, singularities [90], increasing workspace, improving dexterity and to optimize the performance and to achieve a smooth end‐effector motion task. As redundant robots have more DOFs than required, there usually exist multiple solutions for kinematic control, which motivates us to consider exploiting the extra DOFs to improve the control performance.

The inverse kinematics problem is one of the fundamental tasks in understanding the operability of parallel manipulators, i.e. to find the actuator inputs, providing the desired end‐effector trajectories. Conventional design of the parallel mechanisms often encounter singularity problems. The design of redundancy in the parallel mechanism often provides an effective remedy. In [91], the authors proposed a new 3‐DOF symmetric spherical 3‐UPS/S parallel mechanism with three prismatic actuators, and studied the kinematics, statics, and workspace of the mechanism. In [91], a 2(SP+SPR+SPU) serial‐parallel manipulator was considered. Based on the analysis, they designed three new types of kinematically redundant parallel mechanisms, including a new redundant 7‐DOF Stewart platform. In [92], the damped least square method was utilized ...