Hochschule Konstanz University of Applied Sciences
Hochschulschriftenserver der HTWG KonstanzNot a member yet
2319 research outputs found
Sort by
Development of Cutaneous Feedback Haptic Glove for VR Industrial Training
No full textTraditional industrial training methods often require extensive time, cost, and resources for effective preparation and execution. This paper presents a virtual reality (VR)-based solution for industrial training, enhanced by a cutaneous feedback haptic glove that improves the immersive learning experience. The developed haptic glove integrates a high-resolution tactile pin array that provides realistic feedback through the fingertips, allowing users to sense interactions with virtual objects. The VR training environment, created using Unity, simulates a crane sorting system that replicates realworld industrial operations. This setup enables users to manipulate a virtual crane through hand motions, with the haptic glove delivering tactile feedback that enhances the sense of realism. This approach not only advances skill acquisition but also opens opportunities for safe and cost-effective training in complex industrial tasks. By leveraging VR and cutaneous feedback, this study contributes to the modernization of training methods, reducing risks associated with real-world training scenarios and offering a pathway toward more immersive, effective industrial training
Paths and Ambient Spaces in Neural Loss Landscapes
No full textUnderstanding the structure of neural network loss surfaces, particularly the emergence of low-loss tunnels, is critical for advancing neural network theory and practice. In this paper, we propose a novel approach to directly embed loss tunnels into the loss landscape of neural networks. Exploring the properties of these loss tunnels offers new insights into their length and structure and sheds light on some common misconceptions. We then apply our approach to Bayesian neural networks, where we improve subspace inference by identifying pitfalls and proposing a more natural prior that better guides the sampling procedure
Optimising Sleep Stage Detection Using a Minimal Non-EEG Physiological Signal Set and Deep Learning
No full textAutomatic sleep stage classification is essential for enabling non-invasive, at-home monitoring. However, current methods often rely on electroencephalogram (EEG) signals and ad-hoc development approaches that limit reproducibility. We present a reproducible engineering framework for a deep learning model based on the U-Net architecture that classifies sleep into five stages (Wake, N1, N2, N3 and REM) or four (Wake, Light Sleep, Deep Sleep and REM) using only three easily acquired physiological signals: oxygen saturation (SpO), heart rate (HR) and abdominal respiratory effort (AbdRes). In contrast to most previous studies, our model provides sleep stage predictions on a per-second basis, thus overcoming the limitations associated with fixed 30-s epochs. The model was trained on the Sleep Heart Health Study—Visit 2 (SHHS2) dataset and externally validated on the Multi-Ethnic Study of Atherosclerosis (MESA). Optimisation of the model was achieved via Keras Tuner with the Hyperband algorithm. The study achieved weighted F1-scores of 68% (five-stage) and 71% (four-stage) with Cohen's Kappa of 0.61 and 0.67 on SHHS2, with consistent performance on MESA. These results demonstrate strong generalisation and suggest that this lightweight, EEG-free approach offers a practical path towards scalable, clinically relevant sleep monitoring
A System Architecture for AI-Driven Market Entry Strategy Generation Using Large Language Models
No full textBusinesses looking to expand into international markets need high-quality market research to make strategic decisions and develop effective market entry strategies. However, conducting extensive secondary market research that contains all the necessary data is very time-consuming and resource-intensive. In recent years, Artificial Intelligence (AI) has been widely used for marketing applications. Conducting market research and generating market entry strategies are functions that AI, especially Large Language Models (LLMs), could support businesses with. In this paper we propose a theoretical model of how an LLM could be trained to ensure the quality of the output by conducting secondary market research and, based on it, to generate realistic market entry strategy for decision making about international expansion. One of the key elements of the proposed pipeline is an adjustable scoring system which ensures input data reliability and transparency and, as a result, helps to provide a reliable output which could be used for strategic decision-making. Compared to traditional market research approaches, the proposed methodology offers significant improvements in speed, resource efficiency, and accessibility and is aimed to support the businesses in their expansion into new international markets
FX sentiment analysis with large language models
No full textWe enhance sentiment analysis in the foreign exchange (FX) market by fine-tuning large language models (LLMs) to better understand and interpret the complex language specific to FX markets. We build on existing methods by using state-of-the-art open source LLMs, fine-tuning them with labelled FX news articles and then comparing their performance against traditional approaches and alternative models. Furthermore, we test these fine-tuned LLMs by creating investment strategies based on the sentiment they detect in FX analysis articles with the goal of demonstrating how well these strategies perform in real-world trading scenarios. Our findings indicate that the fine-tuned LLMs outperform the existing methods in terms of both the classification accuracy and trading performance, highlighting their potential for improving FX market sentiment analysis and investment decision-making
Jahresbericht 2025
No full textEin Rückblick auf das akademische Jahr
Berichtszeitraum: 1.9.2024 - 31.8.202
Forschung und Transfer Jahresbericht 2024
Full text linkÜbersicht über Drittmitteleinnahmen im Bereich Forschung und Transfer sowie wissenschaftliche Publikationen inkl. abgeschlossener Promotionen
Power grid operation in distribution grids with convolutional neural networks
No full textThe efficient and reliable operation of power grids is of great importance for ensuring a stable and uninterrupted supply of electricity. Traditional grid operation techniques have faced challenges due to the increasing integration of renewable energy sources and fluctuating demand patterns caused by the electrification of the heat and mobility sector. This paper presents a novel application of convolutional neural networks in grid operation, utilising their capabilities to recognise fault patterns and finding solutions. Different input data arrangements were investigated to reflect the relationships between neighbouring nodes as imposed by the grid topology. As disturbances we consider voltage deviations exceeding 3% of the nominal voltage or transformer and line overloads. To counteract, we use tab position changes of the transformer stations as well as remote controllable switches installed in the grid. The algorithms are trained and tested on a virtual grid based on real measurement data. Our models show excellent results with test accuracy of up to 99.06% in detecting disturbances in the grid and suggest a suitable solution without performing time-consuming load flow calculations. The proposed approach holds significant potential to address the challenges associated with modern grid operation, paving the way for more efficient and sustainable energy systems
Start-ups und EU-Recht
No full textIm ersten Teil geht es zunächst darum, einen Überblick darüber zu geben, was EU-Recht überhaupt ist, welche Arten es gibt und welche Akteure auf europäischer Ebene für die Rechtsetzung und die Rechtsprechung zuständig sind.
Im zweiten Teil geht es dann um die zur Schaffung und Aufrechterhaltung eines Binnenmarktes wichtigen „Grundfreiheiten“ sowie um die grundsätzliche Sicherung eines freien Wettbewerbs durch die Regelungen des europäischen Wettbewerbsrechts.
Im dritten Teil werden die für Start-ups relevanten und durch EU-Recht harmonisierten Rechtsbereiche vorgestellt. Dabei geht es um den Datenschutz, das Preisrecht und - ganz aktuell – um die neue KI-Verordnung der EU