Robot Kinematics and Neural Networks: A Summary

Abstract

The robot control problem can be divided into two main areas: kinematics control (coordinating the kinematics chain links to generate the robot’s desired motion) and dynamics control (driving the mechanism's actuator to obey the programmed positions and speeds). Further, the kinematics analysis is performed in two ways in robotics, with forward kinematics (FK) and inverse kinematics (IK). FK is defined as the mapping from joint space to Cartesian space (also called task space), whereas IK is defined as the mapping from Cartesian space to joint space, that is, computation needs to find the joint angles for a given Cartesian position and orientation of the end effectors. This calculation is important for controlling robot arms, but it is very difficult to calculate a robot manipulator's inverse kinematic solution
Keywords: ANN, degree of freedom, forward kinematics, FFN, kinematics, neural network, reverse kinematics, robot control