INVERSION OF MULTILAYER NEURAL-NETWORK WITH MODELING ERROR COMPENSATION

The main contribution of this paper is to develop a method, using the Newton-Raphson method, to search for the unknown part of the inputs of a multilayer neural network with given outputs and known inputs. To u se the Newton-Raphson method, a method of expressing a jacobian by neu ral network parameters is developed first. A locally linearized relati on between inputs and outputs of neural network is then derived. With this, iterative Newton-Raphson searches are performed until satisfacto ry results are obtained. The method shows rapid convergence, compared with previous approaches. While deriving the inverse of the neural net work, some types of optimality, which are problem dependent, are resol ved. Although the method shows fast convergence, this type of solution yields some inversion error due to the neural network modelling error . The second contribution of this paper is to propose a novel structur e which can eliminate the inversion error caused by the neural network modelling error. The proposed method has a simple structure, but show s good performance as it has a feedforward structure and other benefic ial features. Through computer experiments, the proposed methods show good performances in solving inverse kinematics of redundant robots an d controlling nonlinear plant.