A convex optimization approach to online set-membership EIV identification of LTV systems

TL;DR

This paper introduces a convex optimization method using McCormick envelopes for online set-membership identification of LTV systems with noisy measurements, enabling efficient and global solutions to a nonconvex problem.

Contribution

It presents a novel convex relaxation approach for recursive set-membership identification of LTV systems affected by noise, improving computational efficiency and solution optimality.

Findings

Effective identification demonstrated through simulation examples

Convex relaxation enables global solutions via linear programming

Method handles bounded noise in input and output measurements

Abstract

This paper addresses the problem of recursive set-membership identification for linear time varying (LTV) systems when both input and output measurements are affected by bounded additive noise. First we formulate the problem of online computation of the parameter uncertainty intervals (PUIs) in terms of nonconvex polynomial optimization. Then, we propose a convex relaxation approach based on McCormick envelopes to solve the formulated problem to the global optimum by means of linear programming. The effectiveness of the proposed identification scheme is demonstrated by means of two simulation examples.