Offenburg University of Applied Sciences
Hochschulschriftenserver der Hochschule OffenburgNot a member yet
6641 research outputs found
Sort by
Fully 3D-printed gripper jaw with embedded sensitive sensor structures for robotic applications
Full text linkIn this contribution, we present a novel, fully 3D-printed sensitive gripper jaw for high gripping forces up to 40 N. The fabrication process is based on fused layer manufacturing, in which two different materials are sequentially extruded. The gripper jaw is based on a sensor design with multiple-stacked bending beams and four sensor elements, each printed directly into one of the upper and lower bending beams. The research focuses on the sensor design of the gripper jaw, the mathematical description and simulation, the fabrication process, the electrical characterization of the sensor material, and the sensitivity behavior of the gripper jaw in terms of zero-point deviation, characteristic value deviation, linearity behavior, repeatability, and viscoelastic behavior. In addition, the gripper jaw is compared with a similarly manufactured gripper jaw with conventionally attached strain gauges, as well as with other comparable fully 3D-printed sensors to classify the sensor quality. The results demonstrate that the fully 3D-printed gripper jaw is partially suitable for sensitive and fragile components. The gripper jaw is well suited for detecting the gripping state (part gripped / not gripped)
Isothermal and thermomechanical low-cycle fatigue of the hot work steel X38CrMoV5-3 with and without overaging heat treatment
Full text linkIn this work, the martensitic hot work steel X38CrMoV5-3 is investigated by isothermal and thermomechanical fatigue tests conducted between 20 and 650 °C with mechanical strain amplitudes from 0.4 to 1%. The fatigue behavior of the material hardened and tempered to 54 HRC is compared to conditions subjected to different overaging heat treatments, resulting in reduced hardness levels. Time- and temperature-dependent deformation including thermal and cyclic softening as well as stress relaxation and rate-dependent effects are analyzed with pronounced changes above 500 °C depending on heat treatment. Overaging affects fatigue life positively or negatively, depending on temperature and strain amplitude. At lower temperatures, materials with high hardness tend to exhibit partially brittle fracture surfaces, while at high temperatures a creep-enhanced low-deformation ductile fracture and oxidation is observed. In-phase thermomechanical fatigue conditions are more detrimental than out-of-phase TMF conditions. Classical fatigue damage parameters are not capable of adequately capturing the effects of temperature, phase relationship and heat treatment on fatigue life
Assessing Foundation Models for Mold Colony Detection with Limited Training Data
No full textThe process of quantifying mold colonies on Petri dish samples is of critical importance for the assessment of indoor air quality, as high colony counts can indicate potential health risks and deficiencies in ventilation systems. Conventionally the automation of such a labor-intensive process, as well as other tasks in microbiology, relies on the manual annotation of large datasets and the subsequent extensive training of models like YoloV9. To demonstrate that exhaustive annotation is not a prerequisite anymore when tackling a new vision task, we compile a representative dataset of 5000 Petri dish images annotated with bounding boxes, simulating both a traditional data collection approach as well as few-shot and low-shot scenarios with well curated subsets with instance level masks. We benchmark three vision foundation models against traditional baselines on task specific metrics, reflecting realistic real-world requirements. Notably, MaskDINO attains near-parity with an extensively trained YoloV9 model while finetuned only on 150 images, retaining competitive performance with as few as 25 images, still being reliable on 70% of the samples. Our results show that data-efficient foundation models can match traditional approaches with only a fraction of the required data, enabling earlier development and faster iterative improvement of automated microbiological systems with a superior upper-bound performance than traditional models would achieve
Structural optimization of porous CPC scaffolds and the effect of eliminating the outer wall on mechanical properties for bone regeneration
Full text linkAdditive manufacturing was utilized to fabricate rotationally symmetrical scaffolds from CPC, which exhibit sufficient mechanical stability to function as bone replacement and possess sufficient accessible surface area for subsequent release of active ingredients. An existing geometry was further developed for this purpose. The experimental protocol entailed an initial phase of solidification in an atmosphere saturated with water, followed by a post-solidification phase in Phosphate Buffered Saline (PBS). Furthermore, a pause was inserted after every five layers during three-dimensional plotting, and the green bodies were sprayed with water. The study also investigated the influence of water content on mechanical strength. A comprehensive examination of the test specimens was conducted under macroscopic, microscopic, and mechanical scrutiny. The scaffolds demonstrated an adequate capacity to withstand a load of 2,000 N (N). Subsequent to consolidation in Phosphate Buffered Saline (PBS), there was no observed increase in the maximum tolerated force. At this breaking load, the majority of test series exhibited an average deformation of 5%. The resultant stiffness was measured at 1,100 MPa. Consequently, the samples exhibited a strength level that was lower than that of spongy bone. The investigation revealed that the novel geometry, featuring an open outer ring, exhibited adequate mechanical stability while concomitantly augmenting the surface area accessible from the exterior for subsequent drug release. The advent of mass production with the new geometry is now a possibility
Öffentlichkeitsarbeit als integrierter Teil von Schulentwicklung
No full textSchulentwicklung ist uns als mehr oder weniger kontinuierlicher Prozess, bei dem Personal-, Organisations- und Unterrichtsentwicklung ineinandergreifen, bekannt. Angesichts der vielen Veränderungen im Umfeld von Schulen ist es jedoch ein sehr kurzzyklischer Prozess mit mittel- bis langfristigem Planungshorizont. Kommunikation ist das Bindeglied zwischen den drei Teilbereichen und eine zentrale Aufgabe, um proaktiv oder im Krisenfall Prozess und Einbindung zentraler Stakeholder nachzujustieren
Der Einsatz Künstlicher Intelligenz in der Unternehmensfinanzierung und -steuerung
No full textDer Einsatz Künstlicher Intelligenz gewinnt in der Unternehmensfinanzierung und -steuerung zunehmend an Bedeutung. Fortschritte in den Bereichen Machine Learning, Predictive Analytics und automatisierte Datenverarbeitung eröffnen neue Möglichkeiten zur Effizienzsteigerung, Prognoseverbesserung und Entscheidungsunterstützung in finanzwirtschaftlichen Steuerungsprozessen. Gleichzeitig stehen Unternehmen vor erheblichen organisatorischen, technologischen und regulatorischen Herausforderungen, die den erfolgreichen Einsatz von KI begrenzen können.
Ziel der vorliegenden Arbeit ist es, den Einsatz von KI in der Unternehmensfinanzierung und -steuerung systematisch zu analysieren und kritisch zu bewerten. Im Mittelpunkt steht die Untersuchung zentraler Einsatzfelder, darunter Finanzplanung und -analyse, Budgetierung und Forecasting, Rechnungswesen sowie Risikomanagement sowie deren Einordnung entlang vier Bewertungsdimensionen: Nutzenpotenzial, technologischer und organisatorischer Reifegrad, Risiken und Herausforderungen sowie Steuerungs- und Governancefähigkeiten. Auf dieser Basis wird ein analytischer Bewertungsrahmen entwickelt undauf die ausgewählten Einsatzfelder angewendet.
Die Ergebnisse zeigen, dass KI nahezu allen betrachten Einsatzfeldern ein hohes bis sehr hohes Nutzenpotenzial aufweist, insbesondere in datenintensiven, prognoseorientierten und standardisierten Prozessen. Demgegenüber befindet sich der technologische und organisatorische Reifegrad vieler Anwendungen noch im Aufbau. Risiken im Zusammenhang mit Datenqualität, Modelltransparenz, Akzeptanz und regulatorischen Anforderungen sind weiterhin relevant und erfordern gezielte Steuerungs- und Governancefähigkeiten. Insgesamt verdeutlicht die Arbeit, dass der nachhaltige Nutzen von KI im Finanzbereich maßgeblich von klaren organisatorischen Strukturen, einer geeigneten Datenbasis sowie einer verantwortungsvollen Integration in bestehende Steuerungsprozesse abhängt. Aber es ist davon auszugehen, dass KI in den betrachteten Einsatzfeldern eine weiter zunehmende Rolle spielen wird
Aufbau eines Digitalen Zwilling des Schluckspechts S5
No full textDer Schluckspecht S5 ist ein Niedrigenergiefahrzeug, welches im Rahmen des studentischen Projekts „Schluckspecht“ an der Fachhochschule Offenburg entwickelt wird. In dieser Arbeit wird ein digitaler Zwilling in Form eines physikalischen Modells, mit der Modellierungssoftware OpenModelica entwickelt. Der Datenaustausch zwischen dem Modell und dem physischen Objekt erfolgt ausschließlich manuell über die Modellierungssoftware. Als Grundlage dient ein eindimensionales Fahrwiderstandsmodell, in dem Luftwiderstandskraft, Rollwiderstandskraft, Steigungswiderstandskraft, Trägheitskraft und Antriebskraft bilanziert werden. Dieses Fahrwiderstandsmodell wird um Seitenführungskräfte ergänzt und in ein Einspurmodell integriert. Das Einspurmodell verfügt über eine Lenkwinkelregelung, um Fahrten entlang einer beliebigen Strecke im zweidimensionalen Raum zu simulieren. Zum Abschluss werden einige Untersuchungen mit dem Schluckspecht S5 Modell durchgeführt, der Fokus ist dabei der Einfluss der Seitenführungskräfte auf die Energiebilanz des Fahrzeugs
A full-lifetime model for the biodegradation of molybdenum under physiological conditions
No full textMolybdenum (Mo) is an emerging material for resorbable metallic implants due to its favorable biocompatibility and mechanical properties. Understanding its complete degradation behavior under physiological conditions is critical for predicting in vivo performance. Here, we present a macroscopic, kinetic model that quantitatively describes the degradation pathway of Mo, including sequential formation of oxide phases (MoO2 and MoO3) and the subsequent dissolution of the final soluble species. The model is based on coupled first-order rate equations, with parameters that can be directly identified from experimental data. Fits to corrosion measurements and in vitro dissolution experiments reveal excellent agreement with the model, capturing both the oxide formation kinetics and the solubility-driven release of molybdenum into the medium. This approach enables the quantitative separation of multi-stage oxidation and dissolution processes, providing a predictive framework for the complete service life of molybdenum-based implants
Additively Manufactured Inductive Sensor for Translational Motion in Robotic Applications
No full textThe following contribution presents a fully additively manufactured inductive displacement sensor applied to a model of a translationally movable robot flange. In addition to demonstrating the feasibility of 3D-printed coils, the focus is particularly on their inductive properties and the variation of inductance as a function of translational displacement. For this purpose, two coils are entirely additively manufactured and integrated into the flange of a 3D-printed robot model. Through translational movement of the flange, the overlap between the two opposing coils changes, enabling position detection by measuring the series inductance. The advantage of such additively manufactured approaches lies in their adaptability to diverse application requirements and their capability to realize complex geometries
Integrating IPHC and SCHC with DTLS: A Comparative Study of Simulation and Testbed-Based IoT Networks
No full textIntegrating secure communication protocols in constrained IoT networks is challenging due to limitations in processing power, memory, and energy. This study evaluates the performance of a layered protocol stack combining IPv6 Header Compression (IPHC), Static Context Header Compression (SCHC), and Datagram Transport Layer Security (DTLS), under both simulation (using COOJA) and real-world testbed (using CC2538-based hardware) environments. Three core performance metrics were analyzed: energy consumption, packet delivery ratio (PDR), and DTLS handshake latency. The simulation results, reflecting idealized conditions with no radio interference or timing variability, reported consistently high performance across all metrics. In contrast, real-world deployments experienced up to 30% higher energy usage, 20% lower PDR, and 60% increased handshake latency, due to radio collisions, asynchronous scheduling, and processing constraints. Rather than attributing these differences solely to simulation limitations, this work advocates for enhancing simulation fidelity, especially at the MAC and physical layers to better reflect real-world behavior. These findings highlight the importance of validating protocol stacks in both environments, and reinforce the value of improving simulation models for accurate, scalable, and secure IoT protocol design