A new neural network learning of inverse kinematics of robot manipulator

In this paper we present a new method of solving the inverse kinematics of robot manipulators. We propose a learning method of a neural network such that the network represents the relations of both the positions and velocities from the task space coordinate to the joint space coordinate simultaneously. The adjoint neural networks for the original neural networks are introduced in order to derive the efficient learning algorithm. It is shown that the proposed method makes it possible to solve the inverse kinematics problem of robot manipulators more accurately