Abstarct: Robot manipulators are nonlinear multivariable systems with high couplings and various uncertainties. Although, adaptive and robust control methods are suggested to overcome the uncertainties including parametric uncertainty, un-modeled dynamics, external disturbances and discretization error, they face many challenges because of the complexity in robot dynamics. A fuzzy system can be used as a universal approximator for any nonlinear system. This feature has been efficiently used to design the adaptive fuzzy controllers. Adaptive fuzzy control systems are designed baxsed on guaranteeing stability. Since practical implementation of the control law is carried out using digital processors, designing a discrete-time adaptive fuzzy controller for robot manipulators baxsed on the voltage control strategy and proposed control systems stability analysis is suggested in this thesis. In this thesis, a novel method is used for gradient descent algorithm. In most references, the fuzzy estimator is designed such that fuzzy output tracks the system output. However, in this thesis, the system fuzzy is design for estimation of the lumped uncertainty, and this is the most important distinction between proposed method and previous approaches. Another novelty this thesis is presenting a new approach for compensating the approximation error of fuzzy system. In this thesis, a new method is developed for compensating the approximation error of the fuzzy system which does not needed integration of tracking error. Moreover, the proposed discrete-time adaptive fuzzy with position feedback control law requires feedbacks of joint positions only. Simulation studies are performed on a SCARA and articulated robot. Stability analysis and simulation results show its efficiency in the control.