In a social networked industry, the focus is on collaboration between humans and technology. Communication is the basic prerequisite for synergetic collaboration between all players. It includes non-verbal as well as verbal interactions. To enable non-verbal interaction, machines must be able to detect and understand human movements. This article presents the ongoing fundamental research on the analysis of human movements using sensor-based activity recognition and identifies potential for a transfer to industrial applications. The focus is on the practical feasibility of activity recognition by adding further data streams such as the position data of logistical objects and tools, meaning the context in which a certain activity is carried out. -- In der Social Networked Industry steht die Zusammenarbeit von Mensch und Technik im Vordergrund. Grundvoraussetzung für eine synergetische Zusammenarbeit aller Akteure ist die Kommunikation, welche neben verbalen auch nonverbale Interaktionen umfasst. Um eine nonverbale Interaktion zu ermöglichen, müssen Maschinen in der Lage sein, menschliche Bewegungen zu erfassen und zu verstehen. Dieser Beitrag stellt die laufende Grundlagenforschung z
Nichtverbale Signale sind ein elementarer Bestandteil der menschlichen Kommunikation. Sie erfüllen eine Vielzahl von Funktionen bei der Klärung von Mehrdeutigkeiten, der subtilen Aushandlung von Rollen oder dem Ausdruck dessen, was im Inneren der Gesprächspartner vorgeht. Viele Studien mit sozial-interaktiven Robotern zeigen, dass vom Menschen inspirierte Bewegungsmuster ähnlich interpretiert werden wie die von realen Personen. Dieses Kapitel erläutert daher die wichtigsten Funktionen, welche die jeweiligen Bewegungsmuster in der Kommunikation erfüllen, und gibt einen Überblick darüber, wie sie auf Roboter übertragen werden können. -- Non-verbal signals are a fundamental part of human communication. They serve a variety of functions in clarifying ambiguities, subtly negotiating roles, or expressing what is going on inside the interlocutors. Many studies with socially-interactive robots show that human-inspired movement patterns are interpreted similarly to those of real people. This chapter therefore explains the most important functions that the respective movement patterns fulfill in communication and gives an overview of how they can be transferred to robots.
This paper proposes an unsupervised workflow to pseudo-label extracellular spikes from human brain slice MEA recordings into two putative cell types: pyramidal cells and interneurons. Here, the raw data from the data acquisition system is used and processed. The pipeline for pre-processing includes bandpass filtering, threshold--based spike detection, frame alignment and normalization. In the ML workflow, dimensionality reduction (PCA, t-SNE, UMAP), clustering (GMM, k-means). To achieve an online system, template matching and OSort under varying curation strictness is also considered. All pipelines are evaluated by different cluster quality with within-cluster Pearson correlation, Silhouette score, and Calinski-Harabasz index. Applying stricter curation improves separation at some cost to inclusivity.
Model-based process design and operation involves here-and-now and wait-and-see decisions. Here-and-now decisions include design variables like the size of heat exchangers or the height of distillation columns, whereas wait-and-see decisions are directed towards operational variables like reflux and split ratios. In this contribution, we describe how to deal with these different types of decisions in a multicriteria framework, offering an adjustability for the wait-and-see variables while at the same time respecting optimality guarantees on process KPIs.
Autonomous driving needs good roads, but 85% of Brazilian roads have damages that deep learning models may not regard as most semantic segmentation datasets for autonomous driving are high-resolution images of well-maintained urban roads. A representative dataset for emerging countries consists of low-resolution images of poorly maintained roads and includes labels of damage classes; in this scenario, three challenges arise: objects with few pixels, objects with undefined shapes, and highly underrepresented classes. To tackle these challenges, this work proposes the Performance Increment Strategy for Semantic Segmentation (PISSS) as a methodology of 14 training experiments to boost performance. With PISSS, we reached state-of-the-art results of 79.8 and 68.8 mIoU on the Road Traversing Knowledge (RTK) and Technik Autonomer Systeme 500 (TAS500) test sets, respectively. Furthermore, we also offer an analysis of DeepLabV3+ pitfalls for small object segmentation.
We consider the shape optimization of flow fields for electrochemical cells. Our goal is to improve the cell by modifying the shape of its flow field. To do so, we introduce simulation models of the flow field with and without the porous transport layer. The latter is less detailed and used for shape optimization, whereas the former is used to validate our obtained results. We propose three objective functions based on the uniformity of the flow and residence time as well as the wall shear stress. After considering the respective optimization problems separately, we use techniques from multi-criteria optimization to treat the conflicting objective functions systematically. Our results highlight the potential of our approach for generating novel flow field designs for electrochemical cells.
Diazo compounds are gathering interest for their potential in promoting greener synthesis routes. We investigate, at a lab-scale, the continuous synthesis of diazo acetonitrile (DAN) using a micro-structured flow reactor and a flow reaction calorimeter. Data concerning DAN formation in the former, and relative to reaction heat and gas flow rate in the latter, are collected. We present both a physical and a grey-box simulation model, both of which are calibrated to our measurements. Both models provide valuable insights into the DAN synthesis. The grey-box approach is useful to incorporate the complex chemical reaction pathways for DAN synthesis and decomposition that are currently hard to address with the physical model.
When does a course count as 'inverted'? When first faced with a new teaching method, deliberate lectures may want to test it without abandoning the advantages of the established method. Aftes the experiment they have to evaluate additional benefits of the new system for teacher and student. Considering the high workload of lecturers, the help of a third party may be useful to obtain an objective assessment of the results during such a transition. At the University of Applied Sciences 'Hochschule Karlsruhe - Technik und Wirtschaft' in Germany, the SKATING project, funded by the German Federal Ministry of Education and Research (BMBF), offers among other things individual technical and didactic advice. This includes practical help with the development, implementation and testing of innovative lecture formats like the Inverted Classroom Method (also known as flipped class). When Professors at the University stated their interest in this approach, varying teaching projects on its implementation and evaluation were initiated. The main focus was the adaption of the method to the individual needs and wishes of professors, course and students. Several examples of a gradual lecture transfor
New solution method for the systems of linear equations in commutative integral domains is proposed. Its complexity is the same that the complexity of the matrix multiplication.
An overview of the cosmological models with expansion, shear and rotation is presented. Problems of the rotating models are discussed, their general kinematic properties and dynamical realizations are described. A particular attention is paid to the discussion of possible observational effects which could be used to estimate or to put limits on the cosmic rotation.
We present a reformulation of quantum mechanics in terms of probability measures and functions on a general classical sample space and in particular in terms of probability densities and functions on phase space. The basis of our proceeding is the existence of so-called statistically complete observables and the duality between the state spaces and the spaces of the observables, the latter holding in the quantum as well as in the classical case. In the phase-space context, we further discuss joint position-momentum observables, Hilbert spaces of infinitely differentiable functions on phase space, and dequantizations. Finally, the relation of quantum dynamics to the classical Liouville dynamics is investigated.
Synchronisation of wireless inertial measurement units in human movement analysis is often achieved using event-based synchronisation techniques. However, these techniques lack precise event generation and accuracy. An inaccurate synchronisation could lead to large errors in motion estimation and reconstruction and therefore wrong analysis outputs. We propose a novel event-based synchronisation technique based on a magnetic field, which allows sub-sample accuracy. A setup featuring Shimmer3 inertial measurement units is designed to test the approach. The proposed technique shows to be able to synchronise with a maximum offset of below 2.6~ms with sensors measuring at 100~Hz. Also, the results indicate a reliable event generation and detection. The investigated parameters suggest a required synchronisation time of eight seconds. Further research should investigate the temperature changes that the sensors are exposed to during human motion analysis and their influence on the internal time measurement of the sensors. In addition, the approach should be tested using inertial measurement units from different manufacturers to investigate an identified constant offset in the accuracy meas
Physical head phantoms allow assessing source reconstruction procedures in electroencephalography and electrical stimulation profiles during transcranial electric stimulation. Volume conduction in the head is strongly influenced by the skull representing the main conductivity barrier. Realistic modeling of its characteristics is thus important for phantom development. In the present study, we proposed plastic clay as a material for modeling the skull in phantoms. We analyzed five clay types varying in granularity and fractions of fireclay, each with firing temperatures from 550 °C to 950 °C. We investigated the conductivity of standardized clay samples when immersed in a 0.9% sodium chloride solution with time-resolved four-point impedance measurements. To test the reusability of the clay model, these measurements were repeated after cleaning the samples by rinsing in deionized water for 5 h. We found time-dependent impedance changes for approximately 5 min after immersion in the solution. Thereafter, the conductivities stabilized between 0.0716 S/m and 0.0224 S/m depending on clay type and firing temperatures. The reproducibility of the measurement results proved the effectiveness
The S-DALINAC is a 130 MeV superconducting recirculating electron accelerator serving several nuclear and radiation physics experiments as well as driving an infrared free-electron laser. A system of normal conducting rf resonators for noninvasive beam position and current measurement was established. For the measurement of gamma-radiation inside the accelerator cave a system of Compton diodes has been developed and tested. Detailed investigations of the transverse phasespace were carried out with a tomographical reconstruction method of optical transition radiation spots. The method can be applied also to non-Gaussian phasespace distributions. The results are in good accordance with simulations. To improve the quality factor of the superconducting 3 GHz cavities, an external 2K testcryostat was commissioned. The influence of electro-chemical polishing and magnetic shielding is currently under investigation. A digital rf-feedback-system for the accelerator cavities is being developed in order to improve the energy spread of the beam of the S-DALINAC. * Supported by the BMBF under contract no. 06 DA 820, the DFG under contract no. Ri 242/12-1 and -2 and the DFG Graduiertenkolleg 'Ph
PrivacyScore ist ein öffentliches Web-Portal, mit dem automatisiert überprüft werden kann, ob Webseiten gängige Mechanismen zum Schutz von Sicherheit und Privatheit korrekt implementieren. Im Gegensatz zu existierenden Diensten ermöglicht PrivacyScore, mehrere Webseiten in Benchmarks miteinander zu vergleichen, die Ergebnisse differenziert und im Zeitverlauf zu analysieren sowie nutzerdefinierte Kriterien für die Auswertung zu definieren. PrivacyScore verbessert dadurch nicht nur die Transparenz für Endanwender, sondern erleichtert auch die Arbeit der Datenschutz-Aufsichtsbehörden. In diesem Beitrag stellen wir das Konzept des Dienstes vor und wir erörtern, unter welchen Umständen das automatische Scannen und öffentliche "Anprangern" von Schwächen aus rechtlicher Sicht zulässig ist. -- This German article describes the technical and legal considerations surrounding PrivacyScore, a public web portal that allows automatic scans of websites for privacy and security problems. For an English article discussing the same system in more technical detail, but lacking the legal interpretation, see arXiv:1705.05139.