Active control of vehicle suspension systems typically relies on linear, time-invariant, lumped-parameter dynamic models. While these models are convenient, nominally accurate, and tractable due to the abundance of linear control techniques, they neglect potentially significant nonlinearities and time-varying dynamics present in real suspension systems. One approach to improving the effectiveness of such linear control applications is to introduce time and spatially-dependent coefficients, making the model adaptable to parameter variations and unmodeled dynamics. In this paper, the authors demonstrate an intelligent parameter estimation approach, using structured artificial neural networks, to continually adapt the lumped parameters of a linear, quarter-car suspension model. Results are presented for simulated and experimental quarter-vehicle suspension system data, and clearly demonstrate the viability of this approach.