Neuroadaptive control of information-poor servomechanisms with smooth and nonsmooth uncertainties

Abstract For motion control of uncertain servomechanisms, nonlinear dynamics including smooth and nonsmooth types, external disturbances, signal measurement noises, asymmetric input saturation, so on seriously hinder the further development high-performance closed-loop algorithms. However, already existing strategies cannot address above-mentioned issues at same time. It greatly increases difficulty controller design especially when some states are not measurable. Inspired by above motivations, this paper exploits neural networks to deal with discontinuous combines extended state observers estimate disturbances unmeasurable for servomechanisms. Meanwhile, desired-command-based model compensation approach is integrated into design. worth noting that network weights updated combination estimation error tracking acquire better approximation accuracy. According technologies, a novel extended-state-observer-based adaptive algorithm will be synthesized. The bounded stability whole system proved strictly. In addition, comparisons application results an electro-hydraulic servo verify availability superiority developed algorithm.