<p>This article focuses on an adaptive neural network tracking control problem for a robotic manipulator with input dead zone and function constraints on states. A new adaptive neural controller is created, taking into account the dynamics of the manipulator and motor, time-varying asymmetric barrier Lyapunov functions, and input dead zone. Unlike previous studies, the constraint boundaries of this paper are related to both state and time. To mitigate the adverse influence of input dead zone on robotic manipulator, the dead zone function is processed in two parts. Neural network is introduced to estimate the unknown part. This control strategy makes sure that all states are kept within the function constraints. At last, the feasibility of the proposed control method is verified by simulation results.</p>