This paper proposes a hybrid synchronization scheme for chaotic systems with input time delay and uncertainty. In the proposed framework, radial basis function (RBF) neural networks (NNs) are constructed to approximate the unknown smooth nonlinear functions of the synchronization error system. The time delay part is dealt with an adaptive controller and the effect of approximate errors, uncertainties and disturbances are reduced to a H∞ norm constraint. By Lyapunov stability theorem, the closed-loop of the controlled synchronization error system is proved to be stable around zero. Thus the synchronization of chaotic systems is obtained. A simulation example with Hindmarsh-Rose neuronal systems is presented to demonstrate the validity of the proposed control method.