Identification of robotic systems with hysteresis using Nonlinear AutoRegressive eXogenous input models

Abstract

Identification of robotic systems with hysteresis is the main focus of this article. Nonlinear AutoRegressive eXogenous input models are proposed to describe the systems with hysteresis, with no limitation on the nonlinear characteristics. The article introduces an efficient approach to select model terms. This selection process is achieved using an orthogonal forward regression based on the leave-one-out cross-validation. A sampling rate reduction procedure is proposed to be incorporated into the term selection process. Two simulation examples corresponding to two typical hysteresis phenomena and one experimental example are finally presented to illustrate the applicability and effectiveness of the proposed approach.

Sparse, Interpretable and Transparent Predictive Model Identification for Healthcare Data Analysis

“…The noise terms are included to accommodate the effects of measurement noise, modelling errors, and/or unmeasured disturbances. Now define a group of new variables (i.e., lagged versions of the original input and output variables) as In practice, many types of functions are available to approximate the unknown function f[•] in (1), including power-form polynomial models and rational models [28], radial basis function (RBF) [8,31,32], and wavelet expansions [33]. In this study, power-form polynomial basis is considered.…”

“…System identification and predictive modelling are two important classes of data-driven modelling techniques, the former concerns the development of mathematical models using data observed from dynamical systems, whilst the latter concerns the revealing of relationships of features of interest from any collected data. The past decades have witnessed tremendous developments and applications of system identification and predictive modelling techniques [1][2][3][4], which have been applied in diverse areas including space weather [6][7][8][9][10][11][12][13], climate and geophysics [14][15][16][17][18], medicine and healthcare [19][20][21], environments [23][24][25][26], societal wellbeing studies [27], and engineering [28][29]. In concept, there are some subtle differences in system identification and predictive modelling.…”

An Efficient Hierarchical Identification Method With Kernel-Based SVM for Equivalent Systems of Aircrafts

“…The decomposition ignores the coupling relationship between the roll and sideslip responses. In some situations, the simplified models (15) and (16) can be used as the LOES models, however, when there is a strong coupling between two responses under some control input of the pilot, the roll response due to the rudder input need to be considered. Then the fourth-order model is required.…”

“…Least squares algorithm remains the most widely used identification method in diverse areas [14]- [16]. The measurements are processed using sensor data, which always have noise associates with them.…”

Nonlinear predictive model selection and model averaging using information criteria

“…The NARMAX structure can be identified by an OFR algorithm (Chen, Billings, & Luo, 1989), which can be used to select significant model terms according to an error reduction ratio index (ERR), and estimate model parameters simultaneously Wei, Billings, & Liu, 2004). The NARMAX model and the OFR algorithm have been successfully applied to solve a wide range of real-world problems in various fields including engineering (Zhang, Zhu, & Gu, 2017), ecological (Marshall et al, 2016), environmental (Bigg et al, 2014), geophysical (Balikhin et al, 2011;Boynton, Balikhin, Billings, Wei, & Ganushkina, 2011), medical (Billings, Wei, Thomas, Linnane, & Hope-Gill, 2013), and neurophysiological (Li, Wei, Billings, & Sarrigiannis, 2016) sciences.…”