This article investigates the finite-time command-filtered backstepping control problem for dual- motor servo systems. The advantages of the finite-time controller include fast convergence and high robustness, which can improve the dynamic and steady control performance of the system. However, the controller often contains nonlinear and exponential terms increasing the computational complexity. Consequently, for the convenience of practical applications, the command-filtered technique is utilized to decrease the computational burden causing by repeated derivatives in the backstepping process. Besides, LuGre friction model is used to describe the system friction and a fuzzy logic system is designed to handle it. Only one adaptive parameter is estimated, which makes it suitable for industrial applications. Finally, the stability of the closed-loop system is proved and two experiments are carried out to show the advantages of the proposed scheme.