Aiming at a class of nonlinear systems with unknown time-delay and dead-zone inputs,an adaptive asymptotic tracking control algorithm based on novel dynamic surface is proposed.Firstly,a nonlinear filter with a positive time-varying integral function is constructed,which is different from the traditional low-order linear filter.It can not only avoid the problem of “differential explosion” and reduce the computational burden,but also effectively compensate for the boundary layer error.Secondly,the radial basis neural network and a new inequality are utilized to deal with the unknown nonlinear time-delay terms.Finally, the neural network error and the unknown parameters of the dead zone are estimated online.Theoretical analysis shows that the semi-global uniform boundedness of all signals in the closed-loop system is achieved and that the tracking error converges to zero.The simulation results have verified the effectiveness of the proposed control algorithm.