ABSTRACT
JAGAN: “dk3189_c003” — 2006/3/13 — 11:46 — page 139 — #1
In Chapter 2, standard adaptive controller development in discrete-time, normally referred as self tuning regulator (STR), was covered in detail. In contrast with the available controllers, the STR discussed in the previous chapter guarantees analytically the performance of the controller without persistency of excitation (PE) condition and certainty equivalence (CE) principle. The suite of nonlinear design tools includes adaptive controllers. However, most commercially available systems use proportional, integral, and derivative (PID) control algorithms. PID control allows accuracy acceptable for many applications at a set of via points specified by a human user but it does not allow accurate dynamic trajectory following between via points. As performance requirements on speed and accuracy of motion increase in today’s micro-and nanoscale manufacturing environments, PID controllers lag further behind in providing adequate system performance. Since most commercial controllers do not use any sort of adaptation or learning capabilities, control accuracy is lost when certain nonlinearities change. In this chapter we show how to use biologically inspired control techniques to remedy these problems while further relaxing linear in the unknown parameter (LIP) assumption, which is required in STRs.
