Neuro-fuzzy based modeling of vehicle suspension system

An alternative approach to identify, suspension system models using a neuro-fuzzy technique is presented. A structure and a design scheme based on this approach are briefly discussed, which includes an adaptive network employed as a building block, and the backpropagation gradient method as well as least square estimator as a hybrid learning rule. The objective is to represent the suspension dynamics by a set of fuzzy rules representation. By using this approach, the nonlinear characteristics of the suspension system can also be accommodated. Experimental evaluation of the proposed technique has been conducted using an input-output data collected from a running test vehicle. Observations by comparing the model responses with the actual output measurements revealed that satisfactory model matching were obtained which means that the models have captured the real basic features of the vehicle suspension dynamic characteristics.