Higher Institute on Territorial Systems for Innovation
PORTO@iris (Publications Open Repository TOrino - Politecnico di Torino)Not a member yet
146173 research outputs found
Sort by
Exploiting CO2‐Derived Carbon in Lithium‐Ion Batteries
Get PDFThe extraction of valuable carbon-based products from CO2 reduction represents a promising route to reduce greenhouse gas emissions, promote circular economy practices, and facilitate the integration of energy storage sources such as Li-ion batteries (LIBs). Using such extremely valuable sustainable products obtained from the CO2 capture process can solve not only the global warming problems but also the high demand of the battery industry to provide graphite and highly conductive additives. Herein, a carbon nanomaterial (CNM) extracted from CO2 reduction treatment is used as the conductive additive in LIB graphite anode and lithium iron phosphate cathode. The electrodes are produced by casting, using both the conventional poly(vinylidene fluoride) binder, dissolved in N-methyl-2-pyrrolidone, and sodium alginate as a green, water-soluble, alternative binder. The electrochemical performance of the CNM-based electrodes is here compared to that of LIB cathodes and anodes produced with a commercial carbon additive. The electrodes featuring CNM offer electrochemical performance close to those of conventional electrodes in which commercial conductive additives are utilized
Toward Believable Emotions: Evaluating FACS Coding for Virtual Human Expressions
Get PDFIn interactive computer graphics, the Facial Action Coding System (FACS) is widely used to enhance the emotional expressiveness of Virtual Humans (VHs). By linking specific Action Units (AUs) with facial blendshapes, animators can theoretically reproduce a wide range of human emotions. However, achieving realistic and believable emotional expressions remains challenging, as the same AU intensities do not work equally well across all VHs. This paper explores whether optimal sets of AU intensities can be defined for specific subgroups of VHs, such as those differentiated by gender and visual fidelity, rather than pursuing a one-size-fits-all approach. Through a focused analysis of happiness, sadness, and disgust, we demonstrate that visual fidelity plays a critical role in emotional clarity, while certain emotions require gender-specific adjustments. The findings emphasize the limitations of uniformly maximizing AU intensities across all VHs and offer practical insights for animators, providing a nuanced framework for creating believable emotional expressions in various types of VHs and enhancing realism in interactive applications
Linear inelastic kinetic equations modelling the spread of fake news and its interplay with personal awareness
Get PDFIn this paper, we introduce a kinetic model which describes a learning process leading individuals to build personal awareness about fake news. Next, we embed the results of this model into another kinetic model, which describes the popularity gained by news on social media conditioned to the reliability of the disseminated information. Both models are formulated in terms of linear inelastic Boltzmann-type equations, of which we investigate the main analytical properties - existence and uniqueness of solutions, trend
to equilibrium, identification of the equilibrium distributions - by employing extensively Fourier methods for kinetic equations. We also provide evidence of the analytical results by means of Monte Carlo numerical simulations
Crowdsensing-based automated operational modal analysis for indirect bridge structural health monitoring
Get PDFIn this study, an automated identification procedure for crowdsensing-based indirect Bridge Structural Health Monitoring (iBSHM) is presented. The scope is to estimate the modal parameters of a cycle-pedestrian bridge using only acceleration data collected by smartphones installed on board. The proposed method introduces several innovations. First, natural frequencies are identified using the Stochastic Subspace Identification (SSI) algorithm. Second, the method enables the estimation of damping ratios, which are typically neglected in existing crowdsensing applications. Third, it uses the Singular Value Decomposition (SVD) step within the SSI framework to extract singular vectors corresponding to dominant frequencies, thereby isolating the modal components of the signal and enabling the estimation of mode shapes. The proposed identification procedure is experimentally tested and validated with data from a real footbridge in Bologna (Italy). The field test was carried out with multiple passages of a commercial bicycle, using a single smartphone installed on board. The obtained results are compared with those from a previous test conducted with the same experimental setup and case study, but using a different analysis methodology. Satisfactory comparability and repeatability of the results were achieved
POLYDIM: A C++ library for POLYtopal DIscretization Methods
Get PDFof polytopal discretization methods for partial differential equations. The library provides robust and modular
tools to support advanced numerical techniques, with a focus on the Virtual Element Method in both 2D and 3D
settings. PolyDiM is designed to address a wide range of challenging problems, including those involving nonconvex geometries, domain decomposition and mixed-dimensional coupling applications. It is integrated with the
geometry library GeDiM, and offers interfaces for MATLAB and Python to enhance accessibility. Distinguishing
features include support for multiple polynomial bases, advanced stabilization strategies, and efficient local-toglobal assembly procedures. PolyDiM aims to serve both as a research tool and a foundation for scalable scientific
computing in complex geometrical setting
Characterizing the performance of classification models through conformal correlation matrices
Get PDFIn classification tasks, it is critical to accurately distinguish between specific classes, as misclassifications can undermine system reliability and user trust. In this paper, we study how client selection in both centralized and federated learning environments affects the performance of classification models trained on heterogeneous data. When training datasets across clients are statistically diverse, careful client selection becomes crucial to improve the ability of the model to discriminate between classes, while preserving privacy. In particular, we introduce a novel metric based on conformal prediction outcomes – the conformal correlation matrix – which captures the likelihood of class pairs co-occurring within conformal prediction sets. Unlike the traditional confusion matrix, which quantifies actual misclassifications, our metric characterizes potential ambiguities between classes, thus offering a complementary perspective on model performance and uncertainty. Through a series of examples, we demonstrate how our proposed metric can guide informed client selection and enhance model performance in both centralized and federated training settings. Our results highlight the potential of conformal-based metrics to improve classification reliability while safeguarding sensitive information about individual client data
A Novel Model Reference Adaptive Control Approach for Three-Phase Inverter Applications
Get PDFThis paper presents an innovative control strategy for three-phase inverters. The proposed model reference adaptive control utilizes the Torelli Control Box (TCB) methodology, previously applied successfully to DC-DC converters, to regulate a three-phase inverter without load current sensors. This direct control approach overcomes the limitations of using traditional dual-loop control methodologies based on linear control systems and offers a robust solution against non-linearities and system disturbances. This paper formulates the overall framework for using the TCB method to control three-phase inverters. The performance of the TCB method is compared with linear and non-linear control strategies proposed in the literature, demonstrating the relative advantages in managing three-phase inverter complexities. The results have been verified in the laboratory on an experimental three-phase inverter installation that supplies a resistive load, with the proposed controller implemented in a microcontroller
Silicon Elsewhere: Nairobi, Global China, and the Promise of Techno-Capital
Get PDFHeralded as Africa’s “Silicon Savannah”—a cradle of innovation—Nairobi has become a technology and innovation capital for Kenya and for the continent at large. With a national strategy that has prioritized digital technology for the last two decades, many Chinese digital champions, smaller start-ups, and investors have since chosen Nairobi as their African landing pad. Mapping the interface between Nairobi’s innovation scene and China’s digital presence there, Silicon Elsewhere tells a unique story of ingenuity and adaptation, failure and speculation, and hopefulness and pragmatism. Andrea Pollio’s ethnography draws on interviews with cautious venture capitalists, renegade entrepreneurs, dedicated bureaucrats, and ambitious data scientists to explore the competing meanings of contemporary techno-capital. Moving between leafy coworking spaces and the temperature-controlled rooms of brand-new data centers, Pollio locates Nairobi among the experimental capitals, not peripheries, of technological change in the early twenty-first century
Between Light and Shadow: Exploring Spatial and Multi-Dimensional Justice in the EU’s Green Transition
No full textThe Green Transition is the European growth strategy - supported by the European Green Deal (EGD, 2019) - that aims to overcome climate change and environmental degradation, foster climate neutrality by 2050, boost the economy through green technology, create sustainable industry and transport, and cut pollution. EU Member States are requested to implement actions that support the Green Transition, contributing to achieving the goals of the European Green Deal. At the same time, the Just Transition Mechanism (JTM) is the EU’s instrument to guarantee that the shift to a climate-neutral economy proceeds fairly and leaves no one behind. Turning climate and environmental challenges into opportunities implies that the transition must be just and inclusive for all. Even though within the EU policy framework, Green Transition commonly comes together with the justice dimension, the spatial interdependence between these concepts requires more clarification. Therefore, this contribution questions the EU's Just Transition Strategy and addresses its important limitations: Overemphasising the distributive dimension of justice while undermining the procedural one, paying inadequate attention to the spatial dimension of justice, and considering justice as a secondary element of Green Transition. By analysing these constraints, the chapter provides a critical standpoint of the EU's Just Transition Strategy, highlighting significant gaps in its approach to justice. In addition, practical recommendations for policymakers are made to integrate spatial justice into the Green Transition. Some of these recommendations include assessing how environmental policies related to the Green Transition impact different geographies on a local scale; ensuring equitable geographical access to new infrastructures, such as urban green spaces, renewable energy sources, and green jobs across neighbourhood; promoting green job creation in disadvantaged areas; providing fair access to training programs; empowering the least advantaged groups in decision-making; and, finally, promoting bottom-up decision-making processes that are inclusive, transparent, and focused on addressing structural inequities, rather than solely on immediate economic and environmental impacts
Neuro-Symbolic AI in Computer Vision: Toward More Interpretable, Efficient, Generalized, and Logical Visual Understanding Systems
No full textComputer vision has evolved dramatically from traditional handcrafted image processing methods to advanced deep learning models. However, despite achieving notable results, these purely statistical methods often suffer from limitations in interpretability, data efficiency, generalization, and reasoning capabilities. Neuro-Symbolic (NeSy) AI has emerged as a promising paradigm that integrates the powerful pattern recognition of neural networks with the structured, logical reasoning of symbolic systems. This paper provides a comprehensive introduction to NeSy applications in computer vision, covering tasks such as image classification, object detection, scene understanding, and action recognition. We explore key NeSy frameworks, including Logic Tensor Networks (LTNs), highlighting their ability to improve interpretability, robustness, and reasoning. Finally, we discuss the challenges and future directions this promising hybrid approach poses toward explainable and trustworthy computer vision solutions