Archivio della ricerca della Scuola Superiore Sant'Anna
Not a member yet
26957 research outputs found
Sort by
Machine Learning for Predicting User Satisfaction in Human–Robot Interaction (HRI) Teleoperation Tasks
No full textPredicting user satisfaction in Human-Robot Interaction (HRI) tasks is essential to enhancing system adaptability and user experience. This study addresses this challenge in the context of teleoperated Autonomous Ground Vehicles (AGVs) by introducing the Robot Motion Dataset (RMD). This multimodal dataset integrates motion, control, and haptic feedback signals with user satisfaction scores. Data were collected from 30 participants performing navigation tasks with and without a haptic-actuated glove interface. A set of statistical features was extracted from inertial measurements, control commands, and haptic feedback signals, and the most informative features were selected through a sequential forward selection process. Several machine learning algorithms were trained to classify satisfaction levels, with evaluation performed on both internal and participant-independent external test sets. The best performance was achieved by a Weighted k-Nearest Neighbors classifier, reaching an accuracy above 80% in both experimental conditions. The results demonstrate the feasibility of predicting user satisfaction from multimodal sensor data in real time, highlighting the potential of the proposed framework for adaptive HRI systems
Methods for optimizing energy distribution in the integrated steel making industry|Metodologie per l’ottimizzazione della distribuzione di energia nell’industria siderurgica a ciclo integrale
No full textIntegrated steelworks are energy-intensive facilities that also have a significant environmental impact. Their internal energy demand is characterized by a large diversification of energy sources, and this demand can vary widely during a single production day. However, during steel production, large amounts of process gases are recovered which, besides being used for the internal heat production, are valorized for producing electricity and steam in the power plant. Optimizing their distribution requires accurate forecasting of energy flows and solving complex real-time optimization problems. In this context, this paper presents some methodologies used for developing a decision support system based on machine learning techniques, suitable for the prediction of energy consumption and production, and linear mixed integer optimization techniques. These methodologies have been applied for the optimization of energy distribution in a real steel plant with promising results
RTilience: Fault-Tolerant Time-Critical Kubernetes
No full textThis paper tackles the problem of optimal configuration and deployment of fault-tolerant time-critical service chains with arbitrary DAG-alike topologies. We propose RTilience, designed according to a scalable cloud microservice paradigm, and prototyped on top of the well-known Kubernetes cloud orchestrator. It features real-time reservation scheduling of containers to guarantee temporal isolation of time-critical tasks, leading to fine-grained control of compute latencies, while allowing for sharing physical CPUs among containers. A distributed routing library, ReqRoute, is configured with a timeout and primary and secondary routes, enabling autonomous and decentralized handling of failing requests. The routes are configured by a centralized controller that performs admission control, resource management of microservice instances, task placement, and fault detection and recovery, extending the features available in Kubernetes. Admission control is based on a theoretical framework enclosing a worst-case performance model for the experienced end-to-end response-time under various fault handling options, and an optimization framework that computes the optimum resource allocation for admitted services. Extensive experimentation of the proposed solution has been performed with synthetic examples, and an autonomous transport robot use-case, verifying that end-to-end deadlines are effectively respected, even in presence of high fault rates of individual microservice instances, according to the theoretical expectations. RTilience is made available as open-source software, released under a MIT license
Trend in the Diagnosis of Pyoderma Gangrenosum in Italy: A Multicenter Study
No full textIntroduction: Pyoderma gangrenosum (PG) is a rare neutrophilic dermatosis characterized by painful, rapidly progressive ulcerations, often associated with systemic inflammatory diseases. Despite advancements in diagnostic criteria, PG remains a diagnostic challenge, and recent increases in reported cases may reflect improved recognition rather than true incidence. This study evaluates epidemiological trends in PG diagnosis in Italy (2013-2024) across four dermatology centers (Bologna, Milano, Pisa, Torino) to determine whether increased number of diagnoses are due to refined diagnostic methodologies or external immunological factors. Methods: A cross-sectional analysis was conducted using anonymized data from four major dermatology centers in Italy. Data analysis deployed one-way ANOVA and Poisson regression models to assess temporal trends in PG diagnoses. Significance was set at α = 0.05. Results: Between 2013 and 2024, 213 PG cases were diagnosed across four Italian referral centers, with a female predominance (126 vs. 87). Diagnostic rates showed marked annual variability, including a 150% rise in 2015, a 68.8% drop in 2020, and a 360% rebound in 2021. Diagnoses more than doubled after 2018 (68 vs. 145; p < 0.05). Trends varied by center: Bologna (+0.25/year, p = 0.003) and Milano (+0.09/year, p = 0.03) showed significant increases; Pisa and Torino did not. Overall, the rise in diagnoses post-2018 aligns with the broader adoption of standardized diagnostic criteria. Conclusion: The increase in PG diagnoses recorded since 2018 is more plausibly explained by improved clinical recognition and widespread adoption of structured diagnostic frameworks than by a true rise in incidence. Regional variability and limitations of retrospective data caution against firm epidemiological conclusions. Prospective multicenter studies and standardized registries are needed to validate diagnostic tools, reduce misclassification, and clarify the true burden of PG
Corporate Venture Clienting: A Governance Perspective Through the Lens of Stewardship Theory
No full text
Rails of Progress? Exploring the nexus between railroad access and innovation in Italy (19th-20th centuries)
No full textThis paper provides new evidence on the nexus between railroads and inventive activities in Italy in the period 1861–1936. We develop two new georeferenced datasets on railroad stations and patents covering about 8,000 municipalities. By adopting a staggered difference in differences identification strategy, we show that the impact of railroad construction on innovation is noticeable for the first wave of construction of the period of the Destra storica (1861–1878), when the network was expanded following a state building strategy. However, these effects became noticeable only after two decades and concern mostly independent inventors and low-quality patents. Finally, we show that railroad access fostered innovation, particularly in locations with more advanced pre-existing capabilities
Enhancing Motor Synchrony in Rhythmic Dyadic Tasks Through Portable Elbow Exoskeletons
No full textSynchrony is a cornerstone for the successful physical interaction between humans while cooperating or competing towards a goal and is achieved by correct and smooth information exchange between subjects. Recently, Human-Robot-Human (HRH) interaction arose as an emerging paradigm for improving motor control in collaborative and dyadic motor tasks. Among the robotic solutions explored for agent coupling, exoskeletons are powerful tools for exerting torque and force feedback at the joint level. In this work, two identical torque-controlled elbow exoskeletons were used in dyadic interaction, to provide haptic feedback and improve synchrony between two individuals performing a tapping task. Each exoskeleton is lightweight and compact, weighing 0.8 kg on the arm. Bench tests to verify the performance of closed-loop torque control showed a residual torque below 0.2 Nm when the reference torque was set to zero, and a bandwidth higher than 6 Hz, thus achieving adequate performance for applications in HRH scenarios. In human subjects' experiments, the root-mean-squared error between the two users' joint trajectories was 50% lower when users received haptic feedback compared to the condition without feedback; the relative phase error was lower than 60%. The results of this study suggest that exoskeletons can enhance synchrony in HRH interactions, being potentially useful in rehabilitation training, collaborative industrial tasks or sport and music learning
Left ventricular geometry, brain architecture, and cognition: an observational study
No full textBackground and aims: Cardiovascular (CV) diseases and dementia share common risk factors and often coexist in older adults. Understanding the CV-brain interaction is essential for tackling their interconnected burden. This study investigated the link between CV phenotypes, brain architecture, and cognition. Methods: Overall, 15 519 UK Biobank participants without neurodegenerative diseases or stroke (median age 64 years, 49% female) were analysed. Confirmatory factor analysis aggregated 18 CV magnetic resonance imaging (MRI) biomarkers into latent variables for left ventricular systolic function (gSyst), diastolic (gDiast) function, and geometry (gGeom); arterial compliance was measured by aortic distensibility (AoD). Multivariable linear regression models evaluated associations with brain MRI phenotypes, including grey matter volume, white matter hyperintensities, MRI diffusion white matter microstructure, and hippocampal volume. Models were adjusted for age, body mass index, height, mean blood pressure, and CV risk factors. Exploratory mediation models evaluated whether hippocampal volume accounted for associations between CV phenotypes and cognition. Results: gGeom, reflecting greater myocardial mass, wall thickness, and ventricular volumes, showed the strongest association with hippocampal volume [β = 0.082 (0.048-0.117) in females; β = 0.039 (0.006-0.071) in males] and was the only CV phenotype associated with better cognition, including higher fluid intelligence and faster reaction time. Hippocampal volume significantly accounted for the positive relationship between gGeom and cognition across sexes. In contrast, gSyst, gDiast, and AoD demonstrated weaker and less consistent associations with brain structure and were unrelated to cognition. Conclusions: Ventricular geometry emerged as the CV phenotype most strongly associated with brain architecture and cognition. Hippocampal volume may help explain this association, but further studies are needed to investigate causality