An asymptotically stable robust controller formulation for a class of MIMO nonlinear systems with uncertain dynamics

Abstract

In this work, we present a novel continuous robust controller for a class of multi-input/multi-output nonlinear systems that contains unstructured uncertainties in their drift vectors and input matrices. The proposed controller compensates uncertainties in the system dynamics and achieves asymptotic tracking while requiring only the knowledge of the sign of the leading principal minors of the input gain matrix. A Lyapunov-based argument backed up with an integral inequality is applied to prove the asymptotic stability of the closed-loop system. Simulation results are presented to illustrate the viability of the proposed method.Keywords: nonlinear uncertain dynamical systems; multi-input/multi-output systems; robust control; Lyapunov methods Nomenclature x(t) System states x (i) (t) ith-order time derivative of x(t) H(·), h(·) Functional containing uncertain components G(·), g(·)Real-valued matrix with nonzero leading principal minors τ (t), τ 1 (t), τ 2 (t) Control inputs X(t) Combined state vector S(X) Symmetric positive definite matrix D Diagonal matrix with entries ±1 U(X) Unity upper triangular matriẋ τ (t) Time derivative of control input ϕ (X, x (n) ) Auxiliary function m Positive bounding constant m (X) Positive non-decreasing bounding function x r (t) Reference trajectory e 1 (t) Output tracking error e i (t) Auxiliary error signals for i = 2, . . ., n r(t) Filtered error signal a i, j Positive-valued coefficients generated via a Fibonacci number series I m m × m identity matrix 0 m × 1

A robust tracking controller for dynamically positioned surface vessels with added mass

“…Proof: The proof is similar to that of the one given in [24], it can also be found in Appendix B of [23].…”

“…Proof: Due to page limitations, only a highlight of the proof is provided. The reader is referred to [23] for a similar proof. The non-negative function V 1 (z) ∈ R is defined as…”

Robust control design for positioning of an unactuated surface vessel

“…The most general case of the proof is given in Appendix D of [17]. The proof for the Theorem 2 can be obtained by using 2, 3, λ 1 , λ 2 , λ 3 and Γ instead of n, m, λ 2 , λ 3 , λ 4 and α given in the mentioned study, respectively.…”

Lyapunov-based output feedback learning control of robot manipulators

“…Provided the entries of the control gain matrix k 1 are chosen to be greater than the upper bound of the auxiliary term N d (t), the proof in [16] can be traced to demonstrate the non-negativeness of P (t).…”

“…The proof of (44) can be found in [15] or in [16]. At this stage, to prove the overall stability of the closedloop system and the asymptotic convergence of the error signals, we define the following non-negative function, denoted by V (t) ∈ R,…”