The enhancement of the passengers comfort and their safety are part of the constant concerns for car manufacturers. As a solution, the semi-active damping control systems have emerged to adapt the suspension features, where the road profile is one of the most important factors that determine the automotive vehicle performance. Because direct measurements of the road condition represent expensive solutions and, are susceptible to be contaminated, this paper proposes a novel road profile estimator that offers the essential information (road roughness and its frequency) for the adjustment of the vehicle dynamics by using conventional sensors of cars. Based on the Q-parametrization approach, an adaptive observer estimates the dynamic road signal, posteriorly, a Fourier analysis is used to compute online the road roughness condition and perform an ISO 8608 classification. Experimental results on the rear-left corner of a 1:5 scale vehicle, equipped with Electro-Rheological (ER) dampers, have been used to validate the proposed road profile estimation method. Different ISO road classes evaluate online the performance of the road identification algorithm, whose results show that any road can be identified successfully at least 70% with a false alarm rate lower than 5%; the general accuracy of the road classifier is 95%. A second test with variable vehicle velocity shows the importance of the online frequency estimation to adapt the road estimation algorithm to any driving velocity, in this test the road is correctly estimated 868 of 1,042 m (error of 16.7%). Finally, the adaptability of the parametric road estimator to the semi-activeness property of the ER damper is tested at different damping coefficients.