A Semi-Supervised Learning Approach for Identification of Piecewise Affine Systems

Abstract

Piecewise affine (PWA) models are attractive frameworks that can represent various hybrid systems with local affine submodels and polyhedral regions due to their universal approximation properties. The PWA identification problem amounts to estimating both the submodel parameters and the polyhedral partitions from data. In this paper, we propose a novel approach to address the identification problem of PWA systems such that the number of submodels, parameters of submodels, and the polyhedral partitions are obtained. In particular, a cluster-based algorithm is designed to acquire the number of submodels, the initial labeled data set, and initial parameters corresponding to each submodel. Additionally, we develop a modified self-training support vector machine algorithm to simultaneously identify the hyperplanes and parameter of each submodel with the ouputs of the cluster-based algorithm. The proposed algorithm is computationally efficient for region estimation and able to accomplish this task with only a small quantity of classified regression vectors. The effectiveness of the proposed identification approach is illustrated via simulation results.

An Improved Double-Branch Network for Estimation of Crater Ages Based on Semisupervised Learning and Multi-Source Lunar Data

“…The labeled data augmentation method, based on a semisupervised learning strategy, involves incorporating pseudolabeled data during training [28], [29], [30], [31], [32]. Craters sharing similar morphology and geology are typically indicative of the same period.…”

Learning Linear Complementarity Systems

“…Many alternate approaches that enable PWA system identification from data exist such as (Ferrari-Trecate et al, 2003;Nakada et al, 2005;Bemporad et al, 2005;Hartmann et al, 2015;Du et al, 2020). Researchers also suggested recursive PWA identification algorithms (Bako et al, 2011), (Breschi et al, 2016).…”

Determinação da Estrutura de Modelos PWARX com Base no Erro Dinâmico de Predição

“…Nosúltimos anos, tem-se observado um empenho crescente em técnicas de identificação de sistemas com modelos afim chaveados (do inglês -switched affine models) (Du et al, 2018;Zwart, 2019;Hojjatinia et al, 2019;Fey et al, 2020) e modelos afim por partes (do inglês -piecewise affine models) (Barbosa et al, 2018;Schirrer et al, 2018;Lassoued e Abderrahim, 2019;Kersting e Buss, 2019;Du et al, 2020). Isto se deve ao fato de que identificar um modelo global que consiga cobrir diversas situações pode torná-lo muito complexo.…”

“…Trabalhos presentes na literatura com a identificação dessa classe de modelo ainda carecem de atenção quantoà escolha dos regressores que irão compor o vetor de regressores, como exemplos, (Nakada et al, 2005;Sun et al, 2018;Lassoued e Abderrahim, 2019;Du et al, 2020). O presente trabalho tem como foco a escolha de diferentes regressores nesse tipo de representação.…”