There exists an urgent need to develop new and critical control methodologies that will explicitly address the multiple performance requirements of controlled mechanical systems, while deemphasizing absolute reliance on very accurate process models. The problem can be stated as follows; given a process model with mixed uncertainty, a set of performance specifications, and a high frequency stability margin, develop a controller which will provide output performance within the given tolerances and simultaneously maintain closed loop stability in the presence of the uncertainties.This research uses the frequency response method of quantitative feedback theory to investigate (1) the design of control systems for complex mechanical systems whose models display low frequency parametric uncertainty as well as high frequency flexibility effects, (2) the development of control algorithms for high performance trackiing accuracy of such systems, and (3) the experimental validation of the algorithms on an electrohydraulic rig.