119092 research outputs found
Sort by
Designing Conversational AI for Social Robots in Corporate Contexts: A Case Study on Customizing LLMs through Action Research
No full text
Probing the selectivity of acidic sites in the oxidative catalysis of ethanol: AlₓFe₂₋ₓ(MoO₄)₃
No full textWe report the catalytic reactions of ethanol under oxidizing conditions using ball-milled AlₓFe₂₋ₓ(MoO₄)₃ (0.25⩽x_Al⩽1.0) samples of similar particle sizes and surface areas. Characteristic of this series is the increased Mo:(Al+Fe) ratio of 1.7:1.0 in surface near volumes of the bulk. Acetaldehyde (AcH) results from the oxidative dehydrogenation of ethanol as the main product. The selectivity of AcH decreases by about 20% around 250°C in favor of diethyl ether (DEE), regardless of the Al³⁺ content. This suggests that the Lewis acidic Al³⁺ cations act on the general bulk structure and, in particular, influence the lattice oxygen activity of Mo–O–Fe redox centers. At high conversion rates at elevated temperatures, the overoxidation of ethanol to COₓ is reduced compared to Fe₂(MoO₄)₃ in favor of the formation of ethene (dehydration of ethanol). Beyond reporting on the catalytic reactions of these compounds, we provide insights into defect sites in subsurface volumes through methods for characterizing solid-state materials such as X-ray powder diffraction, SEM-EDX, ⁵⁷Fe Mössbauer spectroscopy, and thermal analysis
Influence of donor and acceptor doping on conductivity in potassium niobate
No full textLead-free ferroelectrics based on the (K, Na)NbO 3 solid solution have emerged as strong candidates to replace lead-containing ceramics due to their excellent piezoelectric and thermal properties. However, their
practical application is hindered by high leakage currents and low ceramic density. Doping provides a strategy for enhancing material performance, highlighting the need for a deeper understanding of defect chemistry. For
this purpose, we investigated the equilibrium of intrinsic defects and the doping behavior of Ca and Fe in the boundary phase KNbO3 . We have found all elements to be prone to vacancy formation. Moreover, Ca acts as a donor and leads to a large concentration of charge carriers and therefore larger electronic conductivity. On the other hand, Fe behaves as an acceptor and shows oxidation states ranging from 2+ to 5+, whose concentrations depend on the total Fe concentration. As a consequence, the number of charge carriers and the electronic conductivity are reduced
Evaluation of generalised soil water retention models and their role in hydro–mechanical modelling of unsaturated soils
No full textFoundation models in autonomous driving: a survey on scenario generation and scenario analysis
No full textEnsuring the safety of autonomous vehicles in real-world environments requires handling a wide spectrum of diverse and rare driving scenarios. Scenario-based testing addresses this need by offering a scalable and controlled approach to develop and validate autonomous driving systems. However, traditional scenario generation methods relying on rule-based logic, knowledge-driven models, or data-driven synthesis often yield limited diversity and unrealistic cases. With the emergence of foundation models, which represent a new generation of pre-trained, general-purpose Artificial Intelligence (AI) models, developers can process heterogeneous inputs (e.g., natural language, sensor data, maps, and control actions), enabling the synthesis, interpretation, analysis of complex driving scenarios. In this paper, we review the use of foundation models for scenario generation and scenario analysis in autonomous driving. Our survey presents a unified taxonomy that includes large language models, vision language models, multimodal large language models, diffusion models, and world models for the generation and analysis of autonomous driving scenarios, outlining their fundamental principles, applications, and corresponding evaluation metrics. In addition, we review the methodologies, open-source datasets, simulation platforms, and benchmark challenges. Finally, the survey concludes by highlighting the open challenges, research questions and promising future directions in applying foundation models to scenario generation and analysis in autonomous driving. All reviewed papers are listed in a continuously maintained repository, which is publicly available and updated with new research: GitHub.com/TUM-AVS/FM-for-Scenario-Generation-Analysis
Group closeness effects on co-owned information sharing: A multilevel perspective
No full textCo-owned information contains personal details about multiple individuals, often nested within a social group. It
is important to study the sharing of such information because its careless disclosure can violate the privacy of all
co-owners. What makes such sharing decisions unique is that they are often conducted within a tight social
context, the attributes of which can systematically affect the decisions of all individuals nested within the group.
This necessitates multi-level theorizing and testing. Doing so, we theorize the impact of group closeness (a group-level attribute) on co-owned information sharing by the group members (individual-level reflections and behaviors). We tested our ideas through a deceptive procedure: ninety participants in 40 groups were asked to
voluntarily share a co-owned photo of 2–3 group members, for algorithm training purposes (cover story). Hierarchical Linear Modeling revealed (1) the retained relevance of self-centered private information sharing
motivators and deterrents in group contexts, and (2) a cross-level effect of group closeness: it weakened the
negative effect of privacy concerns on actual co-owned information sharing. The findings underscore the role of
social context in determining the potency of privacy concerns to drive the privacy behaviors of individuals nested
within this context
Stable crack propagation in dislocation-engineered oxide visualized by double cleavage drilled compression test
No full textUnderstanding crack tip – dislocation interaction is critical for improving the fracture resistance of semi-brittle materials like room-temperature plastically deformable ceramics. Here, we use a modified double cleavage
drilled compression (DCDC) specimen geometry, which facilitates stable crack propagation, to achieve in situ observation of crack tip – dislocation interaction inside a scanning electron microscope. Single-crystal magne-
sium oxide specimens, furnished with dislocation-rich barriers across the intended crack path, were employed to study how localized dislocation structures influence crack dynamics. While individual dislocation behavior could not be resolved due to mechanical drift limitations, crack progression was clearly observed to decelerate within dislocation-rich regions, slowing to 15 % of its velocity as compared to the pristine crystal. Upon exiting these regions, cracks reaccelerated until reaching the next dislocation-rich barrier. Complementary phase field modeling coupled with crystal plasticity has been employed to replicate the experimental observations and
provide mechanistic insight into crack tip – dislocation interactions, confirming that dislocation-rich zones serve as effective barriers to crack propagation. The aligned experiment and simulation results underscore the
robustness of the technique and its potential to inform the design of more fracture-resistant ceramic materials
A robust finite strain isogeometric chemo-mechanics solid-beam element
No full textAn efficient isogeometric solid-beam element for finite strain chemo-mechanical analysis is presented in this work. The formulation is particularly well-suited for simulating diffusion-induced large deformations and buckling of slender micro-lattice structures, such as micro-architected Li-ion battery electrode materials. This multi-field solid-beam element extends our previous work by fully coupling chemical diffusion, finite volumetric chemical expansion, and large elastic deformations. The mechanics driven drifting effect on the flux of the chemical species induces higher-order issue. Consequently, a mixed formulation with displacements, concentration and chemical potential as degrees of freedom is employed. The two-way coupled variational problem is discretized using NURBS basis functions. As with solid finite elements based on Lagrange polynomials, NURBS-based formulations are also affected by the non-physical phenomena of locking. To mitigate such effects in the solid-beam context, the assumed natural strain (ANS) method is employed to alleviate both membrane and transverse shear locking, while quadratic or higher-degree NURBS along the beam’s cross-section naturally alleviate Poisson thickness locking. Additionally, a mixed integration point (MIP) strategy is adopted to enhance computational efficiency and robustness. The proposed element is evaluated through a series of single-patch and multi-patch benchmark problems, and validated against a coupled 3D chemo-mechanical solid element. Results demonstrate that the developed solid-beam element inherits the high accuracy typical of solid elements, while also offering the computational efficiency and locking-alleviation performance characteristic of slender beam analysis. This element offers a promising tool for robust simulations of eigenstrain-induced large deformations and buckling of slender structures within a multi-physics framework like chemo-mechanics and thermo-mechanics
