Abstract: The robust adaptive output-feedback tracking problems for a class of Marino-Tomei model disturbed by uncertain noises are studied. A filtered transformation and a stochastic control Lyapunov design method are applied to Marino-Tomei systems driven by Wiener noises of unknown covariance. A parameter adaptive law and a control law are obtained to ensure that the closed-up systems are noise-to-state stable and the tracking error can converge to a small residual set around the origin in the sense of mean quartic value in average.