Abstract: Adaptive an wavelet backstepping control(WARBC) systems is proposed for electrohydraulic servo systems. Two wavelet neural networks are used to approach the arbitrary model uncertainties of electro-hydraulic servo system, and all their weights are trained online. Novel update and control laws are proposed to guarantee that all the signals in the closedloop control system are uniformly ultimately bounded(UUB) in a Lyapunov sense. As a result, the aim of electrohydraulic servo systems tracking control can be achieved. In addition, a robust compensator is designed to improve the tracking performance. Finally, a computer simulation example is presented to illustrate the effectiveness and applicability of the suggested proposal.