Based on the radial basis function neural network(RBFNN), an adaptive formation control method is proposed for the wheeled mobile robot with dynamic uncertainties by using the concept of virtual leader. Through the introduction of virtual leader, the high-dimensional formation control problem is transformed into following problem of a single mobile robot. Then, the RBFNN with the adaptive weights is introduced to estimate the upper bound of the nonlinear uncertain part of the followers, so that the design process of the overall control scheme is simplified significantly. Asymptotic stability of the close-loop system is ensured by using an appropriate Lyapunov function. Numerical results show the effectiveness of the proposed control strategy.