In this study three types of optimized controllers are developed and tested on two laboratory structures. The two structures represented a progression in complexity and challenge to the controllers. The first structure was simple enough to be accurately modeled so the analytical frequencies and mode shapes agreed with the experimental measurements. The second structure being more complex was more difficult to model so differences between the analytical results and experimental measurements were present. These differences required the application a correction method to the reduced models developed for the second structure. The correction method was shown to work with good results on one reduced model and with poor results on the second reduced model.

Two direct rate feedback control laws and a linear quadratic regulator with state estimation (LQG controller) were designed and implemented on both structures. It was shown that the performance of the LQG controller can be approached with a much simpler direct rate feedback controller with better analytical-experimental agreement. The best analytical-experimental agreement occurred with the simplest controller applied analytically to the corrected reduced model demonstrating the validity of the correction method as well as giving a strong reason to use simpler controller designs.