Using recurrent neuro-fuzzy techniques for the identification and simulation of dynamic systems

The identification and simulation of dynamic systems is still a challenging problem. In this article some basic aspects of neuro-fuzzy techniques for the identification and simulation of time-dependent physical systems are pr esented. In particular, a neuro-fuzzy model that can be used for the identi fication and the (real-time) simulation of viscoelastic models, is describe d. The presented model is motivated by a cooperative neuro-fuzzy approach b ased on a vectorized recurrent neural network architecture. The physical mo tivation of this model is illustrated and specific propagation procedures a nd a learning algorithm are presented. Moreover, the usability in practice is demonstrated by an application of the model in the area of surgical simu lation. (C) 2001 Elsevier Science B.V. All rights reserved.