Abstract: Limited by the operation time window and working space, the space teleoperation tasks need to be completed in a finite time while ensuring that the end effector meets the physical constraints. In addition, time delay and external disturbance seriously affect the stability and control performance of uncertain teleoperation system. Therefore, a finitetime control method for robot teleoperation based on time-varying output constraints is proposed. Firstly, the integral barrier Lyapunov function is used to deal with the time-varying constraints of the operation space, and the rapid stability of the system is guaranteed by the practical finite-time Lyapunov stability theorem. Then, the neural network is utilized to estimate the environment force and resolve the impact of model uncertainty, and the robust term is used to compensate for the estimation bias of the neural network and eliminate the influence of unknown external disturbances. Finally, the proposed algorithm is compared with other algorithms in Matlab/Simulink simulation environments and verified on the ground experiment platform. The results of theoretical simulation and experiment show that the proposed method can further improve the error convergence rate and convergence accuracy, and the output of the system never violates the prescribed time-varying boundary.