The RBF neural network compensation based output feedback dynamic surface controller is proposed for N link manipulators with actuator saturation. An observer is designed to estimate unknown velocity states. The RBF neural network is designed to overcome the saturation nonlinearity. Based on the Lyapunov stability analysis, it is proved that the control strategy can guarantee the stability of the closed-loop system, and high tracking performance can be achieved by adjusting the controller parameters. Simulation results show that the proposed control system can compensate for the actuator saturation effectively, reduce the tracking error dramatically and improve tracking performance, and the control system shows robustness to external disturbances.