Short‐Time Linear Quadratic Form Technique for Estimating Fast‐Varying Parameters in Feedback Loops

The precision of a closed-loop controller system designed for an uncertain plant depends strongly upon the maximum extent to which it is possible to track the trend of time-varying parameters of the plant. The aim of this study is to describe a new parameter estimation algorithm that is able to follow fast-varying parameters in closed-loop systems. The short-time linear quadratic form (STLQF) estimation algorithm introduced in this paper is a technique for tracking time-varying parameters based on short-time analysis of the regressing variables in order to minimize locally a linear quadratic form cost function. The established cost function produces a linear combination of errors with several delays. To meet this objective, mathematical development of the STLQF estimation algorithm is described. To implement the STLQF algorithm, the algorithm is applied to a planar mobile robot with fast-varying parameters of inertia and viscous and coulomb frictions. Next, performance of the proposed algorithm is assessed against noise effects and variation in the type of parameters.