In this paper, the design of a position tracking control scheme for an industrial robot manipulator is considered. The design is based on the normalized left coprime factors robust stabilization approach. By coupling this H/sub /spl infin// synthesis to a sequential quadratic programming optimization procedure, a systematic design method is developed optimizing the performance of the controller. Further, the role of the nominal model for the design is explored. The control scheme is successfully implemented and tested on a real industrial robot manipulator. The computed controller worked immediately in the first experiment without any further tuning and showed excellent robust performance.