Refine
Year of publication
Document Type
- conference proceeding (article) (495) (remove)
Language
- English (495) (remove)
Is part of the Bibliography
- no (495)
Keywords
- field emission (23)
- Current measurement (14)
- silicon (12)
- Iron (10)
- CATHODES (8)
- Software (8)
- Production (7)
- Produktionsplanung (7)
- field emitter array (7)
- field emitter arrays (7)
Institute
- Fakultät Informatik und Mathematik (199)
- Fakultät Elektro- und Informationstechnik (131)
- Fakultät Maschinenbau (65)
- Fakultät Angewandte Natur- und Kulturwissenschaften (61)
- Labor Mikrosensorik (52)
- Laboratory for Safe and Secure Systems (LAS3) (48)
- Fakultät Bauingenieurwesen (43)
- Labor für Digitalisierung (LFD) (27)
- Regensburg Strategic IT Management (ReSITM) (23)
- Labor Finite-Elemente-Methode (FEM) (19)
Begutachtungsstatus
- peer-reviewed (495) (remove)
Preliminary considerations on the form-finding of a tensegrity joint to be used in dynamic orthoses
(2024)
Control and Automation of services of the urban infrastructure offered to citizens and tourists are elementary parts of a smart city. But both rely on a stable supply of data from sensors spread across the whole city, e. g., the fill level sensors of waste bins needed for a waste management tool which we developed in a collaboration with the Regensburg city council for the on-demand collection of waste bins. Europe has a lot of historic cities like Regensburg with narrow streets and huge building walls, some made from granite and fieldstones, which often represents an insurmountable obstacle to wireless data transmission. The reduction of the road traffic volume poses an additional challenge for city planners. By means of networked planning and simulation software, the situation, state and efficiency of citywide logistic services can be monitored and optimized. In the course of such optimizations, we propose the combination of digital and logistic services. As an example, we show that monitoring state information, such as the waste bin fill levels, can be accomplished using the same vehicles and the same planning software, that is used for luggage transportation. Moreover, we describe how we adapted a solver for a variant of the TSP, namely the prize-collecting traveling salesman, to optimize the route planning dynamically.
Training data for Neural Networks is often scarce in the medical domain, which often results in models that struggle to generalize and consequently showpoor performance on unseen datasets. Generally, adding augmentation methods to the training pipeline considerably enhances a model’s performance. Using the dataset of the Foot Ulcer Segmentation Challenge, we analyze two additional augmentation methods in the domain of chronic foot wounds - local warping of wound edges along with projection and blurring of shapes inside wounds. Our experiments show that improvements in the Dice similarity coefficient and Normalized Surface Distance metrics depend on a sensible selection of those augmentation methods.
Due to the complexity and the number of factors involved in factory layout planning, computers were identified as an efficient tool to support the process. However, so far no method for computer-aided layout planning has gained wide acceptance in practical application. One reason for this is that in present approaches either the user or the computer designs the layout, neglecting either the qualitative or the quantitative goals. To bridge this gap, this article introduces a concept for human-computer-integration based on evaluative feedback and inverse reinforcement learning. A key element of the concept is the interactive planning process in which user and computer alternately design and improve the layout until a satisfactory layout is found. The user evaluates the layouts according to qualitative criteria, adjusts them intuitively and specifies objectives and restrictions in an explorative way. The computer on the other hand – in form of a reinforcement algorithm – generates possible layouts and incorporates the user’s feedback into its policy. This synergy is expected to generate better results than an expert or an algorithm alone could. Furthermore, in the context of learning factories, it encourages critical thinking and allows students to develop a deeper understanding of the factors that contribute to efficient manufacturing processes. Both an architecture for the implementation is proposed and the requirements for the user interface are specified.
In this paper, we present a new approach to determine the estimated time of arrival (ETA) for bus routes using (Deep) Graph Convolutional Networks (DGCNs). In addition we use the same DGCN to detect detours within a route. In our application, a classification of routes and their underlying graph structure is performed using Graph Learning. Our model leads to a fast prediction and avoids solving the vehicle routing problem (VRP) through expensive computations. Moreover, we describe how to predict travel time for all routes using the same DGCN Model. This method makes it possible not to use a more computationally intensive approximation algorithm when determining long travel times with many intermediate stops, but to use our network for an early estimate of the quality of a route. Long travel times, in our case result from the use of a call-bus system, which must distribute many passengers among several vehicles and can take them to places without a regular stop. For a case study, the rural town of Roding in Bavaria is used. Our training data for this area results from an approximation algorithm that we implemented to optimize routes, and to generate an archive of routes of varying quality simultaneously.
In this paper, we present a new Hybrid Genetic Search (HGS) algorithm for solving the Capacitated Vehicle Routing Problem for Pickup and Delivery (CVRPPD) as it is required for public transport in rural areas. One of the biggest peculiarities here is that a large area has to be covered with as few vehicles as possible. The basic idea of this algorithm is based on a more general version of HGS, which we adopted to solve the CVRPPD in rural areas. It also implements improvements that lead to the acceleration of the algorithm and, thereby, to a faster generation of a fastest route. We tested the algorithm on real road data from Roding, a rural district in Bavaria, Germany. Moreover, we designed an API for converting data from the Openrouteservice, so that our algorithm can be applied on real world examples as well.
Quantum Machine Learning: Foundation, New Techniques, and Opportunities for Database Research
(2023)
In the last few years, the field of quantum computing has experienced remarkable progress. The prototypes of quantum computers already exist and have been made available to users through cloud services (e.g., IBM Q experience, Google quantum AI, or Xanadu quantum cloud). While fault-tolerant and large-scale quantum computers are not available yet (and may not be for a long time, if ever), the potential of this new technology is undeniable. Quantum algorithms havethe proven ability to either outperform classical approaches for several tasks, or are impossible to be efficiently simulated by classical means under reasonable complexity-theoretic assumptions. Even imperfect current-day technology is speculated to exhibit computational advantages over classical systems. Recent research is using quantum computers to solve machine learning tasks. Meanwhile, the database community already successfully applied various machine learning algorithms for data management tasks, so combining the fields seems to be a promising endeavour. However, quantum machine learning is a new research field for most database researchers. In this tutorial, we provide a fundamental introduction to quantum computing and quantum machine learning and show the potential benefits and applications for database research. In addition, we demonstrate how to apply quantum machine learning to the optimization of join order problem for databases.
Internal transport systems are an essential part of intralogistics in production and distribution facilities. These are characterized by a variety of technologies as well as a multitude of interactions with other processes, such as warehouse, picking, and production processes. Therefore, resource planning and control of these systems is complex, especially for discontinuous conveyors. In this task, users can be supported by Digital Twins for decision-making, as they are suitable for investigating both future system states and possible actions. However, relevant use cases that are generally applicable across sectors as well as a generic system architecture for Digital Twins for resource planning and process control of in-plant transport systems have not yet been sufficiently investigated. In this paper, use cases are presented, relevant functions defined, and, finally, a generic functional and a logical reference architecture described. This is conducted with the design science in information systems research method together with a Systems Engineering approach. The use cases are determined at industrial partners of the research project TwInTraSys, which explores Digital Twins for the planning and control of internal transport systems. They are generalized and, thus, also applicable to other production and distribution facilities in different sectors. Further, the reference architecture can provide a basis for the successful implementation of the Digital Twin.
The glass industry is facing increased challenges regarding climate protection targets and rising energy costs. The integration of renewable energy including conversion and storage is a key for both challenges in this energy-intensive industrial sector, which has been mainly relying on fossil gas so far. The options considered to this point for reducing CO2 emissions and switching to a renewable energy supply involve far-reaching changes of the established melting processes. This entails significant risks in terms of influences on glass quality and stable production volumes. The presented approach for the integration of a Power-to-Methane (PtM) system into the glass industry is a completely new concept and has not been considered in detail before. It allows the use of established oxyfuel melting processes, the integration of fluctuating renewable energy sources and a simultaneous reduction of CO2 emissions by more than 78%. At the same time, natural gas purchases become obsolete. A techno-economic evaluation of the complete PtM process shows, that 1,76 €/m3 or 1,26 €/kg synthetic natural gas are possible with renewable energy supply. Using electricity from the energy grid would require electricity prices < 0,126 €/kWh to allow cost competitive PtM processes in the glass industry. Such electricity prices could be achieved by electricity market-based optimization and operation of the PtM system. This operation strategy would require AI-based algorithms predicting availabilities and prices on future-based markets.
In our experiments we grew electron emitting carbon nanostructures on tungsten tips. Subsequently, we transferred the growth process to pre-structured phosphorus-doped n-type silicon and obtained emitting carbon nanostructures directly grown on silicon. After growth of the nanostructures, the silicon field emitters showed increased emission currents of 76 nA at 1.1 kV (compared to 6 nA under the same conditions before growth).