A direct adaptive fuzzy cerebellar model articulation controller (CMAC) is developed for a class of uncertain nonlinear systems. Only the system output measurement is available for feedback. The adjustable weighting factors of the direct adaptive fuzzy CMAC are updated online by an adaptive law. Strictly positive real Lyapunov theory is used to verify the stability of the closed-loop system, and the proposed control scheme guarantees that all signals involved are bounded and the system output asymptotically tracks the desired output. Finally, simulation results demonstrate the effectiveness of the proposed control scheme.