Publications Lists     [ all BibTeX ]  

Data-driven Anomaly Detection approaches have received increasing attention in  many other application areas in the past fewyears as a tool for monitoring of complex systems in addition to classical univariatecontrol charts. Tree-based approaches have proven to be particularly effective when dealing with high-dimensional Anomaly Detection problems and with underlying non-gaussian data distributions. The most popular approach in this family is the Isolation Forest, which is currently one of the mostpopular choices for scientists and practitioners when dealing with Anomaly Detection tasks. The Isolation Forest represents a seminal algorithm upon which many extended approaches have been presented in the past years aiming at improving the original method or at dealing with peculiar applicationscenarios. In this work, we revise some of the most popular and powerful Tree-based approaches to Anomaly Detection (extensions of the Isolation Forest and other approaches), considering both batch and streaming datascenarios. This work will review several relevant aspects of the methods, like computational costs and interpretability traits. To help practitioners, we also report available relevant libraries and open implementations and we review real-world industrial applications of the considered approaches.

The monitoring of rotating machinery is an essential task in today's production processes. Currently, several machine learning and deep learning-based modules have achieved excellent results in fault detection and diagnosis. Nevertheless, to further increase user adoption and diffusion of such technologies, users and human experts must be provided with explanations and insights by the modules. Another issue is related, in most cases, with the unavailability of labeled historical data that makes the use of supervised models unfeasible. Therefore, a new approach for fault detection and diagnosis in rotating machinery is here proposed. The methodology consists of three parts: feature extraction, fault detection and fault diagnosis. In the first part, the vibration features in the time and frequency domains are extracted. Secondly, in the fault detection, the presence of fault is verified in an unsupervised manner based on anomaly detection algorithms. The modularity of the methodology allows different algorithms to be implemented. Finally, in fault diagnosis, Shapley Additive Explanations (SHAP), a technique to interpret black-box models, is used. Through the feature importance ranking obtained by the model explainability, the fault diagnosis is performed. Two tools for diagnosis are proposed, namely: unsupervised classification and root cause analysis. The effectiveness of the proposed approach is shown on three datasets containing different mechanical faults in rotating machinery. The study also presents a comparison between models used in machine learning explainability: SHAP and Local Depth-based Feature Importance for the Isolation Forest (Local- DIFFI). Lastly, an analysis of several state-of-art anomaly detection algorithms in rotating machinery is included.

Learning useful representations of complex data has been the subject of extensive research for many years. With the diffusion of Deep Neural Networks, Variational Autoencoders have gained lots of attention since they provide an explicit model of the data distribution based on an encoder/decoder architecture which is able to both generate images and encode them in a low-dimensional subspace. However, the latent space is not easily interpretable and the generation capabilities show some limitations since images typically look blurry and lack details. In this paper, we propose the Introspective Variational Classifier (IntroVAC), a model that learns interpretable latent subspaces by exploiting information from an additional label and provides improved image quality thanks to an adversarial training strategy.We show that IntroVAC is able to learn meaningful directions in the latent space enabling fine-grained manipulation of image attributes. We validate our approach on the CelebA dataset.

LEarning TO Rank (LETOR) algorithms are usually trained on annotated corpora where a single relevance label is assigned to each available document-topic pair. Within the Cranfield framework, relevance labels result from merging either multiple expertly curated or crowdsourced human assessments. In this paper, we explore how to train LETOR models with relevance judgments distributions (either real or synthetically generated) assigned to document-topic pairs instead of single-valued relevance labels. We propose five new probabilistic loss functions to deal with the higher expressive power provided by relevance judgments distributions and show how they can be applied both to neural and gradient boosting machine (GBM) architectures. Moreover, we show how training a LETOR model on a sampled version of the relevance judgments from certain probability distributions can improve its performance when relying either on traditional or probabilistic loss functions. Finally, we validate our hypothesis on real-world crowdsourced relevance judgments distributions. Overall, we observe that relying on relevance judgments distributions to train different LETOR models can boost their performance and even outperform strong baselines such as LambdaMART on several test collections.

We address modeling and control of a gate access automation system. A model of the mechatronic system is derived and identified. Then, an approximate explicit feedback linearization scheme is proposed, which ensures almost linear response between the electronic driver duty cycle input and the delivered torque. A nonlinear optimization problem is solved offline to generate a feasible trajectory associated with a feedforward action, and a low-level feedback controller is designed to track it. The feedback gains can be conveniently tuned by solving a set of convex linear matrix inequalities, performing a multiobjective tradeoff between disturbance attenuation and transient response. The proposed control strategy is tested on an industrial device. The experiments show that it can effectively meet the requirements in terms of robustness, load disturbance rejection, and tracking performance.

Nowadays, aerial formations are frequently employed in outdoor scenarios to cooperatively explore and monitor wide areas of interest. In these applications, the vehicles are often exposed to relevant security vulnerabilities, as, for instance, the alteration of navigation signals from an attacker with map counterfeiting (if not even hijacking) purposes. In this work, we focus on an Unmanned Aerial Vehicle (UAV) formation that monitors an area, wherein navigation spoofing attacks may occur. Letting the UAVs cooperate and exploiting the redundancy in the available sensing information, a distributed procedure is proposed to i) detect spoofing attacks, and ii) support the navigation in adverse conditions. The validity of the designed approach is confirmed by numerical results. Aerial vehicles for outdoor operation are generally endowed with inertial measurements, relative ranging, and GNSS sensing capability. In this work, two cascaded estimation algorithms for concurrent GNSS spoofing detection and localization in a multi-UAV scenario is proposed, to attain robust navigation in areas subject to GNSS spoofing attacks. The attack detection leverages on information theoretic tools to provide a practical threshold test by checking the multimodal measurement consistency. The localization procedures exploit a decision logic relying on measurement reliability to combine information sources that are different in nature, for UAV self-localization in both safe and under-attack conditions.

Visual sensor networks (VSNs) constitute a fundamental class of distributed sensing systems, with unique complexity and appealing performance features, which correspondingly bring in quite active lines of research. An important research direction consists in the identification and estimation of the VSN sensing features: these are practically useful when scaling with the number of cameras or with the observed scene complexity. With this context in mind, this paper introduces for the first time the idea of Stimulation Model (SM), as a mathematical relation between the set of detectable events and the corresponding stimulated cameras observing those events. The formulation of the related SM identification problem is proposed, along with a proper network observations model, and a solution approach based on deep embedded features and soft clustering. In detail: first, the Gaussian Mixture Modeling is employed to provide a suitable description for data distribution, while an autoencoder is used to reduce undesired effects due to the so-called curse of dimensionality emerging in case of large scale networks. Then, it is shown that a SM can be learnt by solving Maximum A-Posteriori estimation on the encoded features belonging to a space with lower dimensionality. Numerical results on synthetic scenarios are reported to validate the devised estimation algorithm.

The recent trend toward the development of IoT architectures has entailed the transformation of the standard camera networks into smart multi-device systems, capable of acquiring, elaborating, exchanging data and, often, dynamically adapting to the environment. Along this line, this work proposes a novel distributed solution that guarantees the real-time monitoring of 3D indoor structured areas and also the tracking of multiple targets, by employing an heterogeneous visual sensor network composed of both fixed and Pan-Tilt-Zoom (PTZ) cameras. Specifically, the fulfilment of the twofold mentioned goal is ensured through the implementation of a suitable optimization procedure regarding the PTZ devices controllable parameters, inspired by game theory. Numerical simulations in realistic scenarios confirm the capability of the outlined strategy of securing the simultaneous tracking of several targets, maintaining the total coverage of the surveilled area. In particular, the proposed solution results to be effective in dealing with conflicting goals like achieving a good tracking precision while obtaining high resolution frames of the tracked subjects.