Abstarct: In this thesis, the goal of arms control expert Robotics, Robot "Puma 560" is. This robot has six joints that hinge joint of each robot 6 by a stimulus ( motor DC) to the move to position the robot end points to reach our desired location. This robot has many applications in industry. Between methods of control, adaptive fuzzy control method for indirect control of the robot we have chosen, for the design of fuzzy systems in this control method, the parameters of these systems with a gradient descending method we adjust to the tracking error between the desired position with the robot joints is minimum. In doing this project, before the controller design for robot Puma 560, this type of controller for a DC motor design we have followed its result, the controller voltage signal to control each of the six motor to the robot joints motion we use brings. The control on each joint is done separately, which is independent of the detailed control strategy. In this case the coupling between the robot joints have been ignored, but because the control system, Adaptive coupling between joints is that the unrest and uncertainty in the system are considered. It should be noted that our original system, no relationship of robot "Puma 560" and not related to the equations of the robot is unknown to us. Simulation system for robot control in the toolbox Puma 560 Robotic content use software. Finally, this control method with a classical control methods that classical PD control method and to compare this result we will find that the method of indirect adaptive fuzzy control, steady state error, but have less control compared to classic PD, later control system should be matched.