Real-time deformation of structure using finite element and neural networks in virtual reality applications

Abstract

In this paper, an approach for 'virtual' real-time deformation of complex structures is developed. The approach combines the finite element method and Neural Network calculations to allow the user to perform interactive shape changes and view the resultant deformation changes in a virtual environment. A tennis racket and ball are used to illustrate the capability of the proposed method. Because real time deformation simulation is a time consuming repeated analyses, the neural networks are employed in this investigation as numerical devices for substituting the finite element code needed for the tennis racket and ball deformation. The input data for the artificial neural network are a set of parameters generated randomly (ball impact velocity, impact angle and the zone impact). The output data are the deformation of the ball/strings and the impact force-feedback value. The work contribute toward the development of real virtual simulator which uses a haptic device that "feels" the feedback force generated by the deformation. © 2006 Elsevier B.V. All rights reserved.