20005 research outputs found
Sort by
Zusammenfassung und Ausblick
No full textIm Jahr 2024 konnte insgesamt eine Zunahme derNutzung des erneuerbaren Energieangebots innerhalb des Energiesystems in Deutschland festgestellt werden. Jedoch zeigen sich nach wie vor deutliche Unterschiede zwischen den Bereichen Strom, Wärme und Mobilität sowohl in Bezug auf die absolute Nutzung als auch hinsichtlich der aktuellen Entwicklungen
Deep denoising of volumetric OCT images for in vivo motion detection
No full textOptical Coherence Tomography (OCT) is an imaging modality with high temporal and spatial resolution. There are different applications, e.g. in ophthalmology, dermatology, cardiology and recently OCT image based motion compensation has been considered. However, OCT suffers from speckle noise which affects image quality and subsequent interpretation. We propose denoising based on unsupervised 3D convolutional Autoencoder (AE) and systematically evaluate the model on in vivo and post mortem datasets. Moreover, we study motion detection as a subsequent task and to investigate the effects of AE based denoising. Our results demonstrate that speckle noise in the OCT images can lead to substantial outliers. Denoising based on AEs is effective in reducing the outliers and results in improved motion detection. Furthermore, the proposed AE processes OCT volumes in 0.5 ms, making it suitable for real-time applications. Finally, the results illustrate that the AE can effectively improve motion detection performance on in vivo data, despite being trained on different data. In conclusion, our AE model presents a simple and unsupervised deep learning approach to obtain fast denoising adapted to OCT imaging. For motion detection, denoising can be crucial to avoid artifacts due to outliers.Clinical relevance: This work provides an efficient denoising Autoencoder for fast data processing that can be applied in various clinical scenarios to improve noisy 3D Optical Coherence Tomography images
A cascaded strategy with embodied artificial intelligence: forward kinematics solutions for CCRobot-S
No full textThis paper presents a novel cable-climbing mechanism: the Collaborative Climbing Robot Squad (CCRobot-S), a variant of Reconfigurable Cable-Driven Parallel Robots (R-CDPR), specifically designed for the inspection and maintenance of stay cables. The forward kinematics of the CCRobot-S robotic system, however, is inherently mathematically intractable. This research proposes a novel cascaded strategy with Embodied Artificial Intelligence (EAI) to effectively tackle the forward kinematics problem. In this proposed strategy, a lightweight deep learning-based model integrated with numerical method optimization supplants traditional methods, providing feedback on the poses of the flying platform to the control loop of the CCRobot-S robotic system. It provides an approximate solution as initial values through a deep neural network by learning from physical or simulated interactive experiences of CCRobot-S, and then transfers the suitable initial values with kinematic constraints or physical constraints that are near the real solution to the numerical method. This process achieves a stable and robust solution for the forward kinematics of CCRobot-S. This article includes the foundational kinematic analysis of CCRobot-S, the formulation of the CCRobot-S model, a comprehensive introduction and analysis of the cascaded strategy, including the dataset preparation, the training configuration, the solution inference, and the numerical method optimization. Comprehensive evaluations and experiments were undertaken to examine the proposed strategy. The results reveal and confirm that the deep-learning neural network implemented in the CCRobot-S robotic system is effective. Additionally, the proposed cascaded strategy achieves higher prediction accuracy than the standalone neural network approach under the condition of real-time execution (position error reduced from (Formula presented.) mm to (Formula presented.) mm in the X direction, from (Formula presented.) mm to (Formula presented.) mm in the Y direction, and from (Formula presented.) to (Formula presented.) in the (Formula presented.) orientation). The cascaded strategy also guarantees convergence in 100 (Formula presented.) of test cases (50/50) and demonstrates enhanced stability and robustness (1:1 mapping from the joint space to the task space)relative to the conventional Newton-Raphson algorithm's numerical method. These attributes are crucial and necessary for the CCRobot-S system to be effectively deployed in real-world applications
Torsion of α-connections on the density manifold
No full textWe study the torsion of the α-connections defined on the density manifold in terms of a regular Riemannian metric. In the case of the Fisher-Rao metric our results confirm the fact that all α-connections are torsion free. For the α-connections ∇(O,α) obtained by the Otto metric, we show that, except for α=−1, they are not torsion free and that ∇(O,0) is compatible with the Otto metric, but not its Levi-Civita connection. In fact, we derive an explicit formula for this torsion and show that the ∇(O,0)-geodesics differ from those of the Otto metric
Gestaltung KI-basierter Geschäftsmodelle in der Produktion: eine Fallstudienanalyse KI-basierte Geschäftsmodelle in produzierenden KMU
No full textDer Beitrag untersucht auf Basis von fünf Fallstudien die Gestaltung KI-basierter Geschäftsmodelle in produzierenden KMU. Mithilfe des 4V-Modells werden strukturelle Implikationen entlang von Value Proposition, Value Delivery, Value Creation und Value Capture analysiert. Die Ergebnisse zeigen gemeinsame Muster der datenbasierten Wertschöpfung sowie differenzierende Faktoren im Hinblick auf Branchen kontext und organisationale Voraussetzungen.Based on five case studies, this article exami nes the design of AI-based business models in manufacturing SMEs. Using the 4V model, structural implications are analy zed along the dimensions of value proposition, value delivery, value creation, and value capture. The findings reveal recur ring patterns of data-driven value creation and distinguishable factors shaped by industry context and organizational capabilities
The TUHH process imaging system
No full textChemical engineering is a crucial field in the production of many essential goods, with its processes undergoing major changes in the transition to a less fossil-dependent product chain. Yet, understanding these processes is often based on modeling and measurements with integral or local invasive sensors. Tomographic techniques such as magnetic resonance imaging (MRI) can overcome these limitations and provide essential information. However, MRI systems are typically not constructed with the needs of (bio)chemical reactors in mind. Most vertical MRI systems feature probe diameters of below 8 cm and maximum sample heights of below 1 meter. On the other end, clinical MRI systems are usually built horizontally and feature bore sizes of bigger than 30 cm and maximum sample lengths of a few meters. In these system, processes based on gravity have to conform to the bore size and are therefore limited.
The TUHH system combines the advantages of both systems. It is a vertical bore magnet with a 40 cm bore diameter. Sitting on legs at 4 m height, samples of up to 3 meters can be measured. This is especially relevant for reactors based on gravity such as fluidized beds or bubble columns. The magnet itself is a cryogen-free magnet with a field strength of 3 T.
The system is freshly commissioned in 2024 and is planned to be used in various projects. In particular, the work of Alexander Penn in ultrafast imaging will be continued and extended in scope[1,2]. A main focus here are fluidized beds and liquid-gas reactors, which can be investigated with temporal resolutions below 30 ms. Furthermore, techniques such as thermometry and chemically resolved imaging will be employed to reveal the processes inside important reactor systems.
This work has been funded by the Deutsche Forschungsgemeinschaft: instrumentation proposal 422037415, research proposals 471615686 and 544956881 and the collaborative research center ”SMART Reactors” (503850735).Deutsche Forschungsgemeinschaft (DFG
Entwicklungen und Perspektiven in Forschung und Lehre
No full textIn diesem Kapitel werden ausgewählte Aktivitäten des PKT aus den Jahren 2021 bis 2025 vorgestellt. Neben einem Rückblick auf die rege Zusammenarbeit mit zahlreichen Forschungs- und Industriepartnern und den wissenschaftlichen Diskurs unter anderem in Fachgesellschaften und auf Fachtagungen, werden Einblicke in die Lehre gegeben. Weiter werden die Promotionen der vergangen Jahre vorgestellt
Die Auswirkungen von Priorisierungsansätzen von Schiffen mithilfe von Vorabanmeldung auf die Hafenanlaufkoordination
No full textThis simulation study investigates the impacts of pre-arrival vessel prioritization strategies on port call coordination using a simulation model that integrates discrete-event and agent-based techniques. Motivated by the environmental and operational inefficiencies of the traditional first-come-first-served (FCFS) port policies, the simulation assesses how early and structured communication of arrival intentions, as first-announced-first-served (FAFS), can enhance berth allocation, improve resource utilization in the port, and reduce ship emissions. The model replicates a real port environment using empirical data and Python-based libraries evaluating multiple prioritization strategies under varying timing rules for port call announcements. Results demonstrate that structured pre-arrival announcements improve turnaround times and berth occupancy, particularly under strategies setting upper timing limits. However, results vary by terminal type and installed capacity. The findings underscore the need for improved digital infrastructure and cooperative governance to enable Just-in-Time (JIT) arrivals, highlighting the potential for simulation to support decision-making in port operations modernization
Modeling of Geographic Greedy Routing in Sparse LDACS Air-to-Air Networks Using Absorbing Markov Chains
No full textL-band Digital Aeronautical Communications System (LDACS) is the selected Air-to-Ground (A2G) technology for future aeronautical communications and a proposed candidate for Air-to-Air (A2A) links. Geographic greedy routing in sparse LDACS A2A networks, typical during gradual system deployment, frequently encounters local minima, necessitating backup mechanisms that are inefficient. Previous research has primarily focused on refining backup mechanisms, neglecting the root cause of geographic greedy routing failures. This paper investigates why geographic greedy routing performance deteriorates in sparse network scenarios, where failures occur more frequently than in dense deployments. We introduce a novel metric to quantify the quality of hop-by-hop forwarding decisions in geographic greedy routing. Furthermore, we develop a second-order absorbing Markov chain model to predict the success ratio and hop stretch factor. The model is validated through Monte-Carlo simulations over the French airspace with varying LDACS equipage fractions, achieving an average difference from simulation results of less than 3.4% for the success ratio and 1.5% for the hop stretch factor. The proposed model demonstrates high accuracy and can be generalized to evaluate other geographic routing protocols. Consequently, the outcomes provide valuable insights toward designing optimized geographic greedy routing protocols