Adaptive Nonlinear Observer Design Using Feedforward Neural Networks

This paper concerns the design of a neu ral state observer for nonli near dyna mic systems withnoisy measurement channels and in the presence of small model errors. the proposed observerconsists of three feedforward neural parts, two of which are mlp universal approximators, whichare being trained off-line and the last one being a linearly parameterized neural network (lpnn),which is being updated on-line. the off-line trained parts are able to generate state estimationsinstantly and almost accurately, if there are not catastrophic errors in the mathematical modelused. the contribution of the on-line adapting part is to compensate the remainder estimationerror due to uncertain parameters and/or unmodeled dynamics. a time delay term is also addedto compensate the arising differential effects in the observer. the proposed observer can learnthe noise cancellation property by using noise corrupted data sets in the mlp's off-line training.simulation results in two case studies show the high effectiveness of the proposed state observingmethod.