186698 research outputs found
Sort by
Rethinking Holistic Ai Development Through Social Diversity, Interdisciplinary Collaboration and Integrative Knowledge Production
No full textThe rapid deployment of AI reveals persistent socio-technical and data-driven biases that reflect profound epistemic limitations in knowledge production. These biases are not accidental, but symptomatic of deeper epistemic limitations in the way AI knowledge is produced — often by homogeneous teams within technocentric paradigms that exclude alternative perspectives. This paper argues that the underrepresentation of diverse social actors in AI development not only perpetuates inequality, but also severely limits the epistemic and ethical robustness of AI systems.
The focus of this paper arises in particular from the preliminary findings obtained in the Horizon Europe project STEP, which highlight the potential of the framework to improve the inclusivity and trustworthiness of AI. The central thesis is that social diversity must be considered as an epistemic condition and not just an ethical or demographic ideal. Drawing on sociology, psychology and educational science, the authors show how integrating plural forms of knowledge, lived experiences and cultural perspectives into the design and development process can lead to AI systems that are more context-sensitive, equitable and trustworthy. Rather than proposing inclusion as an external corrective, this paper discusses a paradigm shift in AI development - a paradigm shift that embeds diversity into the infrastructure of knowledge production itself.
The contribution of this paper is twofold. First, it proposes a theoretical model of integrative knowledge production that identifies mechanisms through which interdisciplinary collaboration can challenge dominant epistemologies and promote systemic reflexivity. Second, a participatory design framework is outlined to operationalise this model through concrete methodological tools, including dialogic co-design workshops, ethnographic participation in data selection and cross-functional team structuring. These practises aim to break through technocratic compartmentalisation by creating space for social critique and situated intelligence within AI development cycles.
Finally, the authors reflect on the transformative potential of this approach and suggest that rethinking who is involved in AI knowledge production will not only change the outcomes of AI systems, but also the normative foundations of the technological future. From this perspective, ethical AI is not just explainable or compliant — it is structurally inclusive, responsive to different lifeworlds and open to critical reinvention
Iatrogenic femoral neck fractures or separation of the proximal femoral epiphysis during closed reduction of irreducible femoral head fracture–dislocations in children: a review of 12 cases
No full textObjectives: This study aimed to analyze the radiologic characteristics of irreducible pediatric femoral head fracture-dislocations to prevent potential iatrogenic femoral neck fractures (FNF) or separation of the proximal femoral epiphysis (SPFE). Methods: This is a retrospective review of patients who were skeletally immature and diagnosed with traumatic hip dislocations combined with femoral head fractures. The collected data included patient demographics, fracture classification, fragment ratio, combined injuries, urgent reductions, treatment strategies, complications, and final outcomes. Results: We treated 12 patients with femoral head fractures and dislocations; 11 out of 12 patients underwent urgent closed reduction (91.7%). Six of the patients failed closed reduction and experienced FNF (n = 2, 33.3%) or SPFE (n = 4, 66.7%). Five patients presented with avascular necrosis of the femoral head after open reduction with internal fixation via a surgical hip dislocation approach, and two patients required further surgical treatment. Analysis of radiographs and computed tomography (CT) scans of irreducible femoral head fracture-dislocations revealed that the fractured femoral head was perched on the sharp angle of the posterior wall of the acetabulum, with a fragment ratio of 18-27%. After recognizing the irreducibility, one case with a fragment ratio of 26% underwent immediate open reduction without further attempts at closed reduction. This procedure preserved the natural hip joint, preventing FNF and SPFE. Conclusions: For pediatric patients with irreducible femoral head fracture-dislocation injuries and a fragment ratio greater than 10%, repeated or forceful closed reduction may result in iatrogenic FNF or SPFE. Examining plain radiographs and CT images carefully before attempting reduction may help determine the safest treatment strategy
I rapporti diplomatici tra la Francia e la Repubblica di Venezia nella seconda metà del XVIII secolo (1749-1797)
No full textIn the second half of the eighteenth century, relations between France and the Republic of Venice unfolded within a European landscape deeply transformed by wars, imperial competition, and the Enlightenment. France, still a major power but weakened by colonial conflicts, financial difficulties, and internal political tensions, pursued ambitious diplomatic aims while confronting the rise of Great Britain and shifting continental balances. Venice, diminished yet still influential in the Mediterranean, adopted after 1718 a principled neutrality meant to preserve its autonomy in the face of Austria and the broader geopolitical dynamics at play in the Italian peninsula.
The French perception of the Serenissima oscillated between admiration for its stability and criticism of its political immobility. Diplomats and philosophers often denounced the institutional opacity and slow decision-making of the Venetian government, while others noted that the Republic’s longevity rested precisely on strict caution and faithful adherence to its laws. This ambivalence shaped a diplomacy marked by distrust, respect, and continuous interest in Venice’s role in maintaining the Italian equilibrium.
Diplomatic practice constituted the main arena of contact. Ambassadors acted as key mediators, tasked with observing domestic policies, negotiating, and managing incidents. In Paris, Venetian envoys enjoyed broad access to political circles; in Venice, by contrast, French representatives had to comply with an exacting protocol and face the close scrutiny of the State Inquisitors. Consuls in Mediterranean ports (Marseille, Genoa, the Barbary regencies) supplemented this structure, revealing frequent frictions linked to commerce, flags, and navigation rights.
Tensions between the two states stemmed from maritime incidents, disputes over neutrality, violations of diplomatic immunity, or commercial rivalries. Yet no lasting rupture occurred: France, eager to prevent any rapprochement between Venice and Vienna, prioritized mediation and sought to preserve the Republic’s neutrality, which it considered preferable to an uncertain alliance.
French attempts to secure an alliance, especially in the 1750s, failed because Venice feared Austrian retaliation. Neutrality thus became the Republic’s central strategy, despite its limitations. This fragile balance collapsed with the Italian campaign of 1796-1797, when Revolutionary France dismantled a centuries-old diplomatic system. The fall of Venice ultimately revealed the complexity and depth of Franco-Venetian relations throughout the century.Dans la seconde moitié du XVIIIe siècle, les relations entre la France et la République de Venise s’inscrivent dans un paysage européen profondément transformé par les guerres, la compétition impériale et les Lumières. La France, puissance majeure mais fragilisée par les conflits coloniaux, les difficultés financières et les tensions politiques internes, poursuit des ambitions diplomatiques ambitieuses tout en subissant l’ascension de la Grande-Bretagne et la redéfinition des équilibres continentaux. Venise, affaiblie mais encore influente en Méditerranée, adopte après 1718 une neutralité de principe, destinée à préserver son autonomie face à l’Autriche et aux dynamiques géopolitiques qui se jouent dans la péninsule italienne.
La perception française de la Sérénissime oscille entre admiration pour sa stabilité et critiques de son immobilisme. Diplomates et philosophes dénoncent souvent l’opacité institutionnelle et la lenteur décisionnelle du gouvernement vénitien, tandis que d’autres rappellent que la longévité de la République se fonde précisément sur une prudence rigoureuse et sur l’observation stricte de ses lois. Cette ambivalence nourrit une diplomatie faite de méfiance, de respect et d’un intérêt constant pour le rôle de Venise dans l’équilibre italien.
La pratique diplomatique constitue le principal terrain de contact. Les ambassadeurs sont des médiateurs essentiels, chargés d’observer les politiques intérieures, de négocier et de régler les incidents. À Paris, les envoyés vénitiens jouissent d’un large accès aux cercles politiques ; à Venise, en revanche, les représentants français doivent respecter un protocole strict et se confronter au contrôle vigilant des Inquisiteurs d’État. Les consuls dans les ports méditerranéens (Marseille, Gênes, les régences barbaresques) complètent ce dispositif, révélant des conflits fréquents liés au commerce, aux pavillons et aux droits de navigation.
Les tensions entre les deux États proviennent d’incidents maritimes, de contestations de neutralité, de violations d’immunités diplomatiques ou de rivalités commerciales. Cependant, aucune rupture durable n’intervient : la France, soucieuse d’éviter un rapprochement entre Venise et Vienne, privilégie la médiation et cherche à préserver la neutralité de la République, jugée préférable à une alliance incertaine.
Les tentatives françaises d’alliance, surtout dans les années 1750, échouent face à la crainte vénitienne d’une réaction autrichienne. La neutralité devient ainsi la stratégie centrale de Venise, malgré ses limites. Cet équilibre fragile s’effondre avec la campagne d’Italie de 1796-1797, lorsque la France révolutionnaire met fin à un système diplomatique séculaire, révélant a posteriori la complexité et la richesse des rapports franco-vénitiens tout au long du siècle
Design, Emulation, and Control of Edge Computing Systems for the Space Cloud in LEO Satellite Networks
No full textThe advent of mega-constellations in Low Earth Orbit (LEO) has transformed satellite communications, enabling global connectivity and paving the way for integrating computational services in orbit.
This thesis explores the concept of a Space Cloud, in which in-orbit infrastructures provide distributed computing capabilities beyond their traditional role as communication relays.
While Multi-access Edge Computing (MEC) technologies have proven effective in terrestrial networks, their adaptation to dynamic and latency-sensitive space environments raises unresolved challenges in computing resource orchestration, including the Edge Server Activation (ESA) and Edge Server Placement (ESP) problems defined in this work.
We address these challenges through models, software tools, and algorithms that couple orbital propagation with real-time network emulation to evaluate control and placement strategies under realistic conditions.
Central to our contributions is MeteorNet, an open-source, continuous-time platform that integrates orbital dynamics, network virtualization, and containerized execution to reproduce operational network behavior using Docker, Mininet, and Software-Defined Networking (SDN) controllers.
MeteorNet dynamically adapts link states and propagation delays to satellite motion and line-of-sight conditions, generating high-fidelity synthetic datasets for training and validating learning-based algorithms in space systems, where real data are scarce.
For ESA, we develop two adaptive controllers—Fuzzy Logic and Reinforcement Learning—that implement decentralized policies based on each node’s current state and historical performance data to minimize activation time while maintaining low task failure rates, outperforming Access and Bernoulli baselines across traffic regimes.
For ESP, we introduce the Farthest Point Sampling (FPS) method, a topology-aware deployment strategy for space systems.
We compare FPS with alternative space, terrestrial, and hybrid deployments, showing that it can reduce mean round-trip time by 15–20\% and that, at sufficient infrastructure budgets, space-based configurations can surpass terrestrial clouds.
Experiments conducted with MeteorNet, together with the results obtained for the MEC resource orchestration problems, confirm that intelligent orchestration and informed placement substantially outperform reactive or naive baselines in complex, dynamic constellations.
Overall, this thesis lays the groundwork for scalable, autonomous, and latency-aware edge computing in satellite constellations by unifying mixed-integer formulations, learning-based ESA controllers, topology-aware ESP strategies, and a real-time emulation framework to advance the Space Cloud paradigm
Pre-feasibility analysis of a cold ironing system for cruise ships. An engineering, economic and environmental analysis
No full textThe present paper presents a pre-feasibility study for the development of a cold ironing system for cruise ships in the Italian context. The aim of the study is to understand if the development of a cold ironing system is more sustainable from the energy, environmental, and economic point of view with respect to onboard generation. To this end, a simple model is developed for the estimation of energy consumption, pollutant, and carbon emissions of a cruise ship while hotelling in port. The model allows to assess the level of consumption and emissions for both onboard generation and cold ironing. Since the parameters of the model are affected by uncertainty, a Monte Carlo analysis is also developed to understand the overall range of variability of the results. Then, a financial model is developed to estimate an appropriate tariff for the cold ironing system and to compare with the fuel cost. Results show that savings of 3ktoe/year of energy, 200 t/day of CO2, 9 M2021 in operating cost, 12 M2021 in externality costs can be achieved if a cold ironing system is used. Monte Carlo analysis shows that there is a probability of 95% to save between 10.3 and 8.1 M in operating costs.Graphical AbstractSchematic of a cold ironing system (Abu Bakar, Guerrero, et al. 2022
Buffer-free class-incremental learning with out-of-distribution detection
No full textClass-incremental learning (CIL) poses significant challenges in open-world scenarios, where models must learn new classes over time without forgetting previous ones and handle inputs from unknown classes that a closed-set model would misclassify. In this paper, we present an in-depth analysis of post-hoc OOD detection methods and investigate their potential to eliminate the need for a memory buffer. When post hoc OOD detection is applied at inference time, we discover that it can effectively replace buffer-based strategies. We examine the performance of these methods in terms of classification accuracy of seen samples and rejection rates of unseen samples. We show that our approach achieves competitive performance compared to recent multi-head and single-head methods that rely on memory buffers and other buffer-free approaches. The results show that the proposed approach outperforms them in a closed-world setting and detects unseen samples while being significantly resource-efficient. Experimental results on CIFAR-10, CIFAR-100, and Tiny ImageNet support our findings and offer new insights into the design of efficient and privacy-preserving CIL systems for open-world settings
Design and Integration of Multi-Modal Sensor Systems for Enhanced Robotic Manipulation: Applications in Grasping & Control Across Adaptive Grippers and Robotic Platforms
No full textThis thesis presents the design, development, and application of three new variants of multi-modal sensor systems, each created to address key challenges in robotic manipulation, perception, and human-robot interaction. Modern robots are increasingly expected to operate in unstructured, sensitive, and shared environments. To perform effectively in such contexts, they require compact, high-resolution, and versatile sensing platforms that can combine tactile, proximity, and orientation or posture-based data. This research responds to that need through a step-by-step development process, where each sensor builds on the previous design to progressively expand capability and application.
The first prototype integrates eight optoelectronic components, a time-of-flight (ToF) sensor for depth perception, and a 6-axis Inertial Measurement Unit (IMU). It provided the foundation for multimodal sensing, enabling the collection of proximity, motion, and acceleration data. This early platform served as a testbed for calibration, signal integration, and basic interaction tasks, establishing the principles for sensor fusion in robotic systems.
The second variant extended these capabilities through a 4×3 matrix array of optoelectronic elements having enhanced spatial resolution combined with a 6-axis IMU and a modular interface. Its adaptable design allows mounting on a wide range of robotic grippers, including both commercial devices and custom in-lab prototypes. This version was validated in three distinct domains:
1. Delicate object handling, such as fresh food, where sensitive force control was required.
2. Human-Robot Interaction (HRI), where the system enabled safer, more intuitive collaboration between humans and robots.
3. In satellite robotics, Only IMU for feedback to enhance attitude control.
These applications confirmed the sensor’s robustness, real-time performance, and flexibility.
The third and most advanced sensor is a fingertip-integrated module developed for the SEEDS anthropomorphic robotic gripper, very similar to the human fingertip. It incorporates PWM-controlled optoelectronic elements, a 9-axis Bosch IMU with onboard processing, and a high-resolution ToF sensor, all embedded directly within the fingertip. This miniaturized design delivers high-frequency, low-latency data, supporting precise force estimation, object shape recognition, and compliant grasping. The result is a sensing solution that enables dexterous, human-like manipulation.
Alongside these multi-modal sensor developments, this thesis also addresses vision-based perception challenges through the integration of a novel stereo vision approach for detecting thin wire features. Standard 3D cameras are often bulky and costly, making them unsuitable for fine robotic manipulation. To overcome this, a stereo vision system using two low-cost, off-the-shelf endoscopic cameras was developed, allowing integration directly into the robot’s end-effector. By applying state-of-the-art vision algorithms, this system achieved a diameter estimation error below 10% and a location error below 3% for wires of only a few millimeters or less. This contribution demonstrates how compact, affordable vision systems can support fine-grained robotic perception tasks, expanding the scope of multimodal sensing in robotics.
The contributions of this thesis can be summarized as:
1. A progressive methodology for the scalable and modular design of robotic sensor systems.
2. The integration of optoelectronic, inertial, and depth sensing technologies into compact, customizable formats.
3. Development of low-cost stereo vision system for robotic manipulation.
4. Experimental validation of the sensors in diverse application domains including space robotics, food handling, and HRI.
5. Establishing a foundation for future work in soft robotics, teleoperation, and assistive technologies.
Overall, this thesis advances the field of tactile and perceptual sensing in robotics by introducing adaptable, high-performance, and application-specific sensory systems. In addition to tactile and proprioceptive sensing, it establishes a pathway toward vision-augmented multimodal perception, enabling richer environmental awareness and more intelligent robot behavior. These contributions open new opportunities for sensor-driven control strategies, machine learning-based perception, and enhanced robot cognition, while also fostering safer human-robot collaboration and paving the way for next-generation intelligent machine applications
Process integration of enzymatic and microbial PBAT conversion with a Pseudomonas taiwanensis mixed culture
No full textEnzymatic depolymerization of polyesters has been established as green recycling strategy to reduce plastic pollution. However, applied on industrially relevant mixed plastic waste, it generates a wide mixture of various oligomers and monomers, complicating downstream processing and monomer recycling. Several Pseudomonas taiwanensis strains have been engineered to grow on plastic monomers while producing valuable aromatics. This enables metabolic funneling of diverse monomers, supporting efficient bio-upcycling. Integrating hydrolysis and monomer conversion into one intensified process would increase the competitiveness of biotechnological upcycling. Therefore, a one-pot process was developed in which hydrolysis of poly(butylene adipate-co-terephthalate) (PBAT) via a cutinase from Humicola insolens (HiC) was coupled with cultivation of P. taiwanensis strains metabolizing the resulting monomers, adipic acid (AA), terephthalic acid (TA) and 1,4-butanediol (BDO). For this purpose, the buffer strength and stirring rate for PBAT-hydrolysis were adjusted for compatibility with cultivation of P. taiwanensis. An impact of various process settings on enzymatic hydrolysis was found with the temperature as main parameter, where enzymatic and microbial conversion conflict. Hence, two consecutive steps were carried out within one reactor—a 24-h hydrolysis at 70 °C, followed by inoculation with Pseudomonas after changing the conditions to 30 °C. Growth on PBAT was established this way, but the TA metabolism was strongly inhibited by the hydrolysate compared to pure TA. This is probably due to an inhibitory effect of AA and TA-containing oligomers on TA uptake or metabolism. After 10 days, all PBAT monomers were completely consumed, setting the path for a novel, industrially promising plastic upcycling concept
