In this paper, active vibration control of constrained manipulators used in automated deburring process is addressed. In order to suppress the random vibration generated due to the interaction force between robot end-effector and workpiece, an adaptive critic-based neurofuzzy controller is developed. This controller can adapt itself with different burr size. The control signal is fed to a piezoelectric actuator, which is mounted beside the workpiece. The piezoelectric actuator moves the workpiece to compensate the vibration. According to the intelligent behavior and system independent structure, the proposed design can be used for any industrial manipulator in various automated finishing processes. A two-link planar manipulator with flexible joints is simulated in edge deburring process to illustrate the performance of the proposed scheme.