Mirko Mazzoleni

In the last decade, kernel-based learning approaches typically employed for classification and regression have shown outstanding performance also in dynamic system identification. The typical way to compute the solution of this learning problem subsumes the inversion of the kernel matrix. However, due to limited machine precision, this might not be possible in many practical applications. In this article, we analyze the aforementioned problem and show that the typical estimate is just one of the possible infinite solutions that can be leveraged, considering both the supervised and the semisupervised settings. We show under which conditions the infinite solutions are equivalent, and if it is not the case, we provide a bound on the mismatch between two generic solutions. Then, we propose two specific solutions that are particularly suited to boost sparsity or performance.

In many engineering applications, continuous-time models are preferred to discrete-time ones, in that they provide good physical insight and can be derived also from non-uniformly sampled data. However, for such models, model selection is a hard task if no prior physical knowledge is given. In this paper, we propose a non-parametric approach to infer a continuous-time linear model from data, by automatically selecting a proper structure of the transfer function and guaranteeing to preserve the system stability properties. By means of benchmark simulation examples, the proposed approach is shown to outperform state-of-the-art continuous-time methods, also in the critical case when short sequences of canonical input signals, like impulses or steps, are used for model learning.

This paper presents a nonparametric Bayesian interpretation of kernel-based function learning with manifold regularization. We show that manifold regularization corresponds to an additional likelihood term derived from noisy observations of the function gradient along the regressors graph. The hyperparameters of the method are estimated by a suitable empirical Bayes approach. The effectiveness of the method in the context of dynamical system identification is evaluated on a simulated linear system and on an experimental switching system setup.

The advent of Industry 5.0 envisages production systems that are more resilient, embrace human–machine collaboration and promote sustainability driven by technological research. The development of supervision solutions for industrial equipment fills in this picture as a basis for more proactive Condition-Based Maintenance strategies. The goal of this paper is to provide a self-contained set of guidelines to design such supervision solutions. With respect to existing literature on the topic, we provide a design process with a strong focus on experimental data collection and failure reproduction activities. Moreover, the connections between the steps of the proposed process are clearly highlighted to guide the user. First, the paper provides a set of tools to select the critical items and the methodological approaches for supervision. Then, these tools are used and referenced in the proposed design process. Finally, the proposed process is exemplified on two industrial case studies to show its effectiveness. Considerations, hints, and a user guidelines are given at the end of most sections.