Motion stabilization concept have been a critical interest within various mechanical systems such as Camera gimbal, line of sight stabilization, or optical stabilization system. The basic concept of motion stabilization exploits successful mechanical functionality within the presence of disturbances such as mechanical friction, gravity, heat, or fluidic wave. Among other engineering applications, such concepts, especially, have been important topic in designing military weapon such as a turret on the moving platform due to the precise targeting issue. Turrets placed on the moving platforms such as wheeled vehicles or battleship on the Ocean, require extraordinary accuracy since they are mostly designed to pinpoint a target in a lengthy distance. Thus, maintaining a Line of Sight (LOS) corresponding to the command reference is crucial as even the slightest offset in LOS would result in great failure in aiming long-distance target. In order to possess successful aiming performance, disturbances must be eliminated as much as possible. This paper suggests double loop control algorithm for two Degrees of Freedom (DOF) turret placed on the moving platform, in order to achieve ultimate aiming precision by cancelling disturbances. The double loop control algorithm contains conventional position controller designed based on mathematical system modelling. A type of disturbance observer known as Time Delay Estimation (TDE) placed in the inner loop. The outer controller mainly designed to achieve high position tracking performance while TDE designed to cancel both external and inner disturbances on the system. TDE, however, does have drawback that it is vulnerable to sensor noise. This is because not only does TDE contains second derivative of the position but also the input of the TDE is a time step before. In order to solve such issue, this paper also suggests the sensor fusion algorithm, exploiting only the advantages of two given sensors. The fusion algorithm is known as the modified-track-to-track algorithm and it is based on the Kalman Filter. The suggested control structure is first verified on the simulation using MATLAB/Simulink, and second is tested on the real system.