H2 Guaranteed Cost Control in Track-Following Servos

This chapter presents two new control synthesis approaches for dual-stage track-following servo systems. Both approaches are based on H-2 guaranteed cost analysis, in which an upper bound on the worst-case H-2 performance of a discrete-time system with gain-bounded unstructured causal LTI uncertainty is determined by solving either a semi-definite program (SDP) or several Riccati equations. We review the results of a paper on H-2 guaranteed cost analysis and a paper on optimal full information H-2 guaranteed cost control and then use these results to develop two output feedback control synthesis approaches. The first approach is based entirely on the solution of SDPs whereas the second approach exploits Riccati equation structure to reduce the number and complexity of the SDPs that need to be solved. Throughout the paper, we apply the analysis and control techniques to a hard disk drive model with a PZT-actuated suspension and demonstrate that the approaches that exploit Riccati equation structure are faster and at least as accurate as their SDP counterparts.