Refine
Year of publication
Document Type
- Conference Proceeding (1253) (remove)
Conference Type
- Konferenzartikel (950)
- Konferenz-Abstract (156)
- Konferenzband (77)
- Sonstiges (42)
- Konferenz-Poster (32)
Language
- English (934)
- German (317)
- Multiple languages (1)
- Russian (1)
Keywords
- Mikroelektronik (62)
- RoboCup (32)
- Gamification (12)
- Machine Learning (12)
- injury (10)
- Biomechanik (9)
- Finite-Elemente-Methode (9)
- Kommunikation (9)
- Assistive Technology (8)
- Produktion (8)
Institute
- Fakultät Elektrotechnik und Informationstechnik (E+I) (bis 03/2019) (453)
- Fakultät Maschinenbau und Verfahrenstechnik (M+V) (286)
- Fakultät Elektrotechnik, Medizintechnik und Informatik (EMI) (ab 04/2019) (213)
- Fakultät Wirtschaft (W) (164)
- Fakultät Medien und Informationswesen (M+I) (bis 21.04.2021) (120)
- ivESK - Institut für verlässliche Embedded Systems und Kommunikationselektronik (113)
- INES - Institut für nachhaltige Energiesysteme (59)
- IMLA - Institute for Machine Learning and Analytics (46)
- ACI - Affective and Cognitive Institute (40)
- Fakultät Medien (M) (ab 22.04.2021) (33)
Open Access
- Open Access (560)
- Closed Access (456)
- Closed (223)
- Bronze (214)
- Diamond (29)
- Grün (13)
- Gold (6)
- Hybrid (6)
This paper presents the development of a capacitive level sensor for robotics applications, which is designed for measurements of liquid levels during a pouring process. The proposed sensor design applies the advantages of guard electrodes in combination with passive shielding to increase resistance against external influences. This is important for reliable operations in rapidly changing measurement environments, as they occur in the field of robotics. The non-contact type sensor for liquid level measurement is the solution for avoiding contaminations and suit food guidelines. The designed sensor can be utilized in gastronomic applications. Two versions of the sensor were simulated, fabricated, and compared. The first version is based on copper electrodes, and the other type is fully 3D printed with electrodes made of conductive polylactic acid (PLA).
Despite the success of convolutional neural networks (CNNs) in many computer vision and image analysis tasks, they remain vulnerable against so-called adversarial attacks: Small, crafted perturbations in the input images can lead to false predictions. A possible defense is to detect adversarial examples. In this work, we show how analysis in the Fourier domain of input images and feature maps can be used to distinguish benign test samples from adversarial images. We propose two novel detection methods: Our first method employs the magnitude spectrum of the input images to detect an adversarial attack. This simple and robust classifier can successfully detect adversarial perturbations of three commonly used attack methods. The second method builds upon the first and additionally extracts the phase of Fourier coefficients of feature-maps at different layers of the network. With this extension, we are able to improve adversarial detection rates compared to state-of-the-art detectors on five different attack methods. The code for the methods proposed in the paper is available at github.com/paulaharder/SpectralAdversarialDefense
Aerosol particles play an important role in the climate system by absorbing and scattering radiation and influencing cloud properties. They are also one of the biggest sources of uncertainty for climate modeling. Many climate models do not include aerosols in sufficient detail. In order to achieve higher accuracy, aerosol microphysical properties and processes have to be accounted for. This is done in the ECHAM-HAM global climate aerosol model using the M7 microphysics model, but increased computational costs make it very expensive to run at higher resolutions or for a longer time. We aim to use machine learning to approximate the microphysics model at sufficient accuracy and reduce the computational cost by being fast at inference time. The original M7 model is used to generate data of input-output pairs to train a neural network on it. By using a special logarithmic transform we are able to learn the variables tendencies achieving an average score of . On a GPU we achieve a speed-up of 120 compared to the original model.
Estimation of Scattering and Transfer Parameters in Stratified Dispersive Tissues of the Human Torso
(2021)
The aim of this study is to understand the effect of the various layers of biological tissues on electromagnetic radiation in a certain frequency range. Understanding these effects could prove crucial in the development of dynamic imaging systems under operating environments during catheter ablation in the heart. As the catheter passes through some arterial paths in the region of interest inside the heart through the aorta, a three-dimensional localization of the catheter is required. In this paper, a study is given on the detection of the catheter by using electromagnetic waves. Therefor, an appropriate model for the layers of the human torso is defined and simulated without and with an inserted electrode.
Android is an operating system which was developed for use in smart mobile phones and is the current leader in this market. A lot of efforts are being spent to make Android available to the embedded world, as well. Many embedded systems do not have a local GUI and are therefore called headless devices. This paper presents the results of an analysis of the general suitability of Anroid in headless embedded systems and ponders the advantages and disadvantages. It focuses on the hardware related issues, i.e. to what extent Android supports hardware peripherals normally used in embedded systems.
Cast aluminum cylinder blocks are frequently used in gasoline and diesel internal combustion engines because of their light-weight advantage. However, the disadvantage of aluminum alloys is their relatively low strength and fatigue resistance which make aluminum blocks prone to fatigue cracking. Engine blocks must withstand a combination of low-cycle fatigue (LCF) thermal loads and high-cycle fatigue (HCF) combustion and dynamic loads. Reliable computational methods are needed that allow for accurate fatigue assessment of cylinder blocks under this combined loading. In several publications, the mechanism-based thermomechanical fatigue (TMF) damage model DTMF describing the growth of short fatigue cracks has been extended to include the effect of both LCF thermal loads and superimposed HCF loadings. This approach is applied to the finite life fatigue assessment of an aluminum cylinder block. The required material properties related to LCF are determined from uniaxial LCF tests. The additional material properties required for the assessment of superimposed HCF are obtained from the literature for similar materials. The predictions of the model agree well with engine dyno test results. Finally, some improvements to the current process are discussed.
The present work ties in with the problem of bicycle road assessment that is currently done using expensive special measuring vehicles. Our alternative approach for road condition assessment is to mount a sensor device on a bicycle which sends accelerometer and gyroscope data via WiFi to a classification server. There, a prediction model determines road type and condition based on the sensor data. For the classification task, we compare different machine learning methods with each other, whereby validation accuracies of 99% can be achieved with deep residual networks such as InceptionTime. The main contribution of this work with respect to comparable work is that we achieve excellent accuracies on a realistic dataset classifying road conditions into nine distinct classes that are highly relevant for practice.
New employees are supposed to quickly understand their tasks, internal processes and familiarize with colleagues. This process is called “onboarding” and is still mainly realized by organizational methods from human resource management, such as introductory events or special employee sessions. Software tools and especially mobile applications are an innovative means to support provide onboarding processes in a modern, even remote, way. In this paper we analyze how the use of gamification can enhance onboarding processes. Firstly, we describe a mobile onboarding application specifically developed for the young, technically literate generations Y and Z, who are just about to start their career. Secondly, we report on a study with 98 students and young employees. We found that participants enjoyed the gamified application. They especially appreciated the feature “Team Bingo” which facilitates social integration and teambuilding. Based on the OCEAN personality model (“Big Five”), the personality traits agreeableness and openness revealed significant correlations with a preference for the gamified onboarding application.
Cardiac resynchronization therapy (CRT) is an established class I level A biventricular pacing therapy in chronic heart failure patients with left bundle branch block and reduced left ventricular ejection fraction, but not all patients improved clinically. Purpose of the study was to evaluate electrical interatrial conduction delay (IACD) to interventricular conduction delay (IVCD) ratio with focused transesophageal left atrial and left ventricular electrocardiography.
Methods: Thirty eight chronic heart failure patients (age 63.4±10.2 years; 3 females, 35 males) with New York Heart Association (NYHA) functional class 3.0±0.2 and 171.71±36.17ms QRS duration were analysed using posterior left atrial and left ventricular transesophageal electrocardiography with hemispherical electrodes before CRT. Electrical IACD was measured between onset of P-wave in the surface ECG and onset of left atrial signal. Electrical IVCD was measured between onset of QRS complex in the surface ECG and onset of left ventricular signal.
Results: Electrical IACD and IVCD could be evaluated by transesophageal left atrial and left ventricular electrocardiography in all heart failure patients with correlation to 1.18±0.92 IACD-IVCD-ratio (r=-0.57, P<0.001; r=0.66, P<0.001). There were 32 CRT responder with reduction of NYHA class from 3.0±0.22 to 1.97±0.31 (P<0.001) during 16.5±18.9 month CRT with 75.19±33.49ms IACD, 78.91±24.73ms IVCD, 1.04±0.66 IACD-IVCD-ratio and correlation between IACD and IACDIVCD- ratio (r=0.84, P<0.001). There were 6 CRT nonresponder with no reduction of NYHA class from 3.0±0.3 to 2.9±0.5 during 14.3±13.7 month biventricular pacing, 50.0±28.26ms IVCD (P=0.014), 1.92±1.65 IACD-IVCD-ratio (P=0,029) and correlation between 67.0±24.9ms IACD and IACD-IVCD-ratio (r=0.85, P=0.031).
Conclusions: Focused transesophageal left atrial and left ventricular electrocardiography can be utilized to analyse electrical IACD and IVCD in heart failure patients. IACDIVDC- ratio may be a useful parameter to evaluate electrical left cardiac desynchronization in heart failure patients.
Cardiac resynchronization therapy is an established therapy for heart failure patients with sinus rhythm, reduced left ventricular ejection fraction and prolongation of QRS duration. The aim of the study was to evaluate ventricular desynchronization with electrical interventricular delay (IVD) to left ventricular delay (LVD) ratio in atrial fibrillation heart failure patients. IVD and LVD were measured by transesophageal posterior left ventricular ECG recording. In atrial fibrillation heart failure patients with prolonged QRS duration, the mean IVD-to-LVD-ratio was 0.84 +/- 0.42 with a range from 0.17 to 2.2 IVD-to-LVD-ratio. IVD-to-LVD-ratio correlated with QRS duration. IVD-to-LVD-ratio may be a useful parameter to evaluate electrical ventricular desynchronization in atrial fibrillation heart failure patients.
Hintergrund: Das elektrische interventrikuläre Delay (IVD) und die Lage der linksventrikulären (LV) Elektrode zum Ort der spätesten LV Erregung sind bei Patienten (P) mit Herzinsuffizienz (HF), reduzierter LV Funktion und breiter QRS Dauer (QRSD) von Bedeutung für den Erfolg der kardialen Resynchronisationstherapie (CRT). Die LV Elektrokardiographie ermöglicht eine Abschätzung des elektrischen IVD. Ziel der Studie besteht in der nicht-invasiven Evaluierung des elektrischen IVD bei Patienten (P) mit Vorhofflimmern (AFib) mit und ohne CRT mit biventrikulärer (BV) Stimulation.
Methoden: Bei 49 HF P mit AFib (Alter 63,9 ± 10,8 Jahre; 43 Männer und 6 Frauen) mit New York Heart Association (NYHA) Klasse 2,9 ± 0,4, LV Ejektionsfraktion 26,03 ± 7,99 % und QRS-Dauer (QRSD) 143,69 ± 35,62 ms wurde das elektrische IVD als Intervall zwischen Beginn des QRS-Komplexes im Oberflächen EKG und Beginn des LV Signals im transösophagealen LV EKG bei 31 HF P mit AFib und bei 18 HF P mit AFib und CRT präoperativ bestimmt. Das fokussierte bipolare LV EKG wurde mittels Osypka TO Sonde mit halbkugelförmigen Elektroden in Höhe des maximalen LV Signals registriert.
Ergebnisse: Bei 31 HF P mit AFib betrugen QRSD 135,48 ± 38,78 ms, IVD 49,55 ± 26,38 ms, QRSD-IVD-Verhältnis 3,12 ± 1,11 und das IVD korrelierte mit der QRSD (r=0,75, P<0,001) und dem QRSD-IVD-Verhältnis (r=-0,67, P<0,001) (Fig.). Bei 18 HF P mit AFib und CRT Defibrillator betrugen QRSD 157,83 ± 24,38 ms, IVD 61,94 ± 26,88 ms, QRSD-IVD-Verhältnis 3,12 ± 1,89 und das IVD korrelierte mit der QRSD (r=0,47, P=0,049) und dem QRSD-IVD-Verhältnis (r=-0,73, P<0,001). Bei 72,2 % CRT Responder (R) (n=13) betrugen QRSD 158,15 ± 22,4 ms, IVD 64,23 ± 24,62 ms, QRSD-IVD-Verhältnis 2,82 ± 1,32 und das IVD korrelierte mit der QRSD (r=0,57, P=0,043) und dem QRSD-IVD-Verhältnis (r=-0,76, P=0,0024). Bei 27,8 % CRT Non-Responder (NR) (n=5) betrugen QRSD 157 ± 31,94 ms, IVD 56 ± 34,52 ms, QRSD-IVD-Verhältnis 3,88 ± 2,98 und das IVD korrelierte nicht mit der QRSD (r=0,33, P=0,591) und dem QRSD-IVD-Verhältnis (r=-0,732, P=0,159). Die CRT R verbesserten sich in der NYHA Klasse von 3 ± 0,2 auf 2,2 ± 0,3 (P<0,001) während 15,3 ± 13,1 Monaten BV Stimulation. Bei 15 CRT NR kam es zu keiner Verbesserung der NYHA Klasse von 3 auf 3,3 ± 0,97 (P=0,529) während 18,8 ± 20,7 Monaten BV Stimulation.
Schlussfolgerungen: Das transösophageale LV EKG ermöglicht bei HF-P mit AFib die nichtinvasive Messung des elektrischen IVD präoperativ vor CRT. IVD und QRSD-IVD-Verhältnis sind möglicherweise einfach anwendbare Parameter zur Vorhersage von CRT R und CRT NR bei P mit AFib.
Background: Cardiac resynchronization therapy (CRT) with biventricular (BV) pacing is an established therapy for heart failure (HF) patients (P) with sinus rhythm, reduced left ventricular (LV) ejection fraction (EF) and electrical ventricular desynchronization. The aim of the study was to evaluate electrical interventricular delay (IVD) and left ventricular delay (LVD) in right ventricular (RV) pacemaker pacing before upgrading to CRT BV pacing.
Methods: HF P (n=11, age 69.0 ± 7.9 years, 1 female, 10 males) with DDD pacemaker (n=10), DDD defibrillator (n=1), RV pacing, New York Heart Association (NYHA) class 3.0 ± 0.2 and 24.5 ± 4.9 % LVEF were measured by surface ECG and transesophageal bipolar LV ECG before upgrading to CRT defibrillator (n=8) and CRT pacemaker (n=3). IVD was measured between onset of QRS in the surface ECG and onset of LV signal in the transesophageal ECG. LVD was measured between onset and offset of LV signal in the transesophageal ECG. CRT atrioventricular (AV) and BV pacing delay were optimized by impedance cardiography.
Results: Interventricular and intraventricular desynchronization in RV pacemaker pacing were 228.2 ± 44.8 ms QRS duration, 86.5 ± 32.8ms IVD, 94.4 ± 23.8ms LVD, 2.6 ± 0.8 QRS-IVD-ratio with correlation between IVD and QRS-IVD-ratio (r=-0.668 P=0.0248) and 2.3 ± 0.7 QRS-LVD-ratio. The LVEF-IVD-ratio was 0.3 ± 0.1 with correlation between IVD and LVEF-IVD-ratio (r=-0.8063 P=0.00272) and with correlation between QRS duration and LVEF-IVD-ratio (r=-0.7251 P=0.01157). Optimal sensing and pacing AV delay were 128.3 ± 24.8 ms AV delay after atrial sensing (n=6) and 173.3 ± 40.4 ms AV delay after atrial pacing (n=3). Optimal BV pacing delay was -4.3 ± 11.3 ms between LV and RV pacing (n=7). During 30.4 ± 29.6 month CRT follow-up, the NYHA class improved from 3.1 ± 0.2 to 2.2 ± 0.3.
Conclusions: Transesophageal electrical IVD and LVD in RV pacemaker pacing may be additional useful ventricular desynchronization parameters to improve P selection for upgrading RV pacemaker pacing to CRT BV pacing.
Cardiac resynchronization therapy with atrioventricular and interventricular pacing delay optimized biventricular pacing is an established therapy for heart failure patients with sinus rhythm and reduced left ventricular ejection fraction. The aim of the study was to evaluate atrioventricular and interventricular pacing delay optimization in cardiac resynchroniza-tion therapy by transthoracic impedance cardiography in biventricular pacing with different left ventricular electrode po-sition. In biventricular pacing heart failure patients with lateral, posterolateral and anterolateral left ventricular electrode position, the mean optimal atrioventricular sening delay was 108.6 ± 20.3 ms and the mean optimal interventricular pac-ing delay -12.3 ± 25.9 ms. Transthoracic impedance cardiography may be a useful technique to optimize atrioventricular and interventricular pacing delay in biventricular pacing with different left ventricular electrode position.
Das Ausmaß der elektrischen ventrikulären Desynchronisation bei reduzierter linksventrikulärer Funktion ist von Bedeutung für den Erfolg der Resynchronisationstherapie der Herzinsuffizienz mit biventrikulärer Stimulation. Das Ziel der Untersuchung besteht in der nichtinvasiven Messung der elektrischen inter-ventrikulären Desynchronisation mit und ohne ischämische Herzerkrankung bei kardialen Resynchronisationstherapie Respondern. Bei Patienten mit 25,3 ± 7,3 % reduzierter linksventrikulärer Ejektionsfraktion und 166,9 ± 38,5 ms QRS-Dauer wurde das transösophageale linksventrikuläre EKG abgeleitet. Die QRS-Dauer korrelierte mit dem interventrikulären und links-ventrikulären Delay bei Resynchronisationstherapie Respondern mit nicht-ischämischer Herzerkrankung.
Termination of atrial flutter (AFL) is not possible in all AFL patients (P) with transesophageal left atrial pacing (TLAP) with undirected electrical pacing field (EPF) and high atrial pacing threshold. Purpose of the study was to evaluate bipo-lar transesophageal left atrial electrocardiography (TLAE) and TLAP with directed EPF for evaluation and termination of AFL with and without simultaneous transesophageal echocardiography (TEE).
Methods: AFL P were analysed using either a TO electrode with one cylindrical (CE) and three or seven hemispherical electrodes (HE) or TEE electrode with four HE (Osypka, Rheinfelden, Germany). Burst TLAP cycle length was between 200msand 50ms.
Results: AFL cycle length was 233±30 ms with mean ventricular cycle length of 540±149 ms. AFL could be terminated by rapid bipolar TLAP with directed EPF using HE-HE and CE-HE with induction of atrial fibrillation (AF), induction of AF and spontaneous conversion to sinus rhythm and direct conversion to sinus rhythm. Directed EPF was simulated with finite element method.
Conclusions: AFL can be evaluated by bipolar TLAE. AFL can be terminated with rapid TLAP with directed EPF with and without simultaneous TEE. Bipolar TLAE with rapid TLAP is a safe, simple and useful method for evaluation and termination of AFL.
Cardiac resynchronisation therapy (CRT) with biventricular pacing (BV) is an established therapy for heart failure (HF) patients with interventricular conduction delay (IVCD). The aim of the study was to evaluate transesophageal IVCD and left ventricular (LV) pacing with directed electrical pacing field (EPF) in HF patients.
Methods: HF patients were analysed with bipolar transesophageal LV electrocardiogram recording and LV pacing with constant voltage stimulus output, 4 ms stimulus duration, distal cylindrical electrode (CE) and seven 6 mm hemispherical electrodes (HE) with 15 mm electrode distance (TO, Dr. Osypka, Rheinfelden, Germany).
Results: LV electrocardiogram recording with HE-HE and CE-HE evaluated a mean IVCD of 79.9 ± 36.7 ms. Directed EPF with CE-HE and HE-HE allowed LV VAT (n=12) and LV D00 pacing (n=5) with a mean effective capture output of 97.35 ± 6.64 V. In 15 responders with IVCD of 87 ± 33 ms arterial pulse pressure (PP) increased from 65 ± 24 mmHg to 79 ± 27 mmHg (p < 0.001). EPF was simulated with finite element method.
Conclusions: Transesophageal LV electrocardiography and directed EPF pacing with CE and HE allowed the evaluation of IVCD and PP to select patients for BV pacing.
Transösophageales interventrikuläres Delay bei Vorhofflimmern und kardialer Resynchronisation
(2013)
Die transösophageale linksventrikuläre Elektrokardiographie ermöglicht die Evaluierung der elektrischen ventrikulären Desynchronisation im Rahmen der kardialen Resynchronisationstherapie der Herzinsuffizienz. Das Ziel der Untersuchung besteht in der präoperativen Abschätzung des transösophagealen interventrikulären Delays bei Vorhofflimmern und kardialer Resynchronisationstherapie. Bei Patienten mit Vorhofflimmern, Herzinsuffizienz New York Heart Association Klasse 3,0 ± 0,2 und QRS-Dauer 159,6 ± 23,9 ms wurde das fokusierte transösophageale linksventrikuläre EKG abgeleitet. Die kardiale Resynchronisationstherapie Responder QRS-Dauer korrelierte mit dem transösophagealen interventrikulären Delay bei Vorhofflimmern.
Introduction: Cardiac resynchronization therapy (CRT) with biventricular pacing (BV) is an established therapy for heart failure (HF) patients (P) with ventricular desynchronisation, but not all patients improved clinically. Aim of this study was to evaluate electrical intra-left ventricular conduction delay (LVCD) and interventricular conduction delay (IVCD), to better select patients for CRT.
Methods: 65 HF patients (age 63.4 ± 10.6 years; 7 females, 58 males) with New York Heart Association (NYHA) class 3 ± 0.2, 24.4 ± 6.7 % left ventricular (LV) ejection fraction and 167.4 ± 35.6 ms QRSD were included. Esophageal TO Osypka focused hemispherical electrodes catheter was perorally applied in position of maximum LV deflection to measure LVCD between onset and offset of LV deflection and IVCD between earliest onset of QRS in the 12-channel surface ECG and onset of LV deflection in the focused bipolar transesophageal LV electrogram.
Results: There were 50 responders with LVCD of 76.5 ± 20.4 ms, IVCD of 80.5 ± 26.1 ms (P=0.34) and QRSD of 171 ± 37.7 ms. 15 non-responders had longer LVCD of 90 ± 28.5 ms (P = 0.045), shorter IVCD of 50.1 ± 29.1 ms (P < 0.001) and QRSD of 155.3 ± 25 ms (P=0.14). During 21.3 ± 20.3 month BV pacing follow-up, the responder`s NYHA classes improved from 3 ± 0.2 to 2. ± 0.3 (P < 0.001) whereas the non-responders NYHA classes did not improve from 3 ± 0.2 to 2.9 ± 0.3 (P = 0.43) during 15.7 ± 13.9 month BV pacing follow-up (53 Boston, 10 Medtronic and 2 St. Jude CRT devices).
Conclusion: Determination of electrical LVCD and IVCD by focused bipolar transesophageal LV electrogram recording may be an additional useful technique to improve patient selection for CRT.
Cardiac resynchronization therapy (CRT) with biventricular (BV) pacing is an established therapy in approximately two-thirds of symptomatic heart failure (HF) patients (P) with left bundle branch block (LBBB). The aim of this study was to evaluate left atrial (LA) conduction delay (LACD) and left ventricular (LV) conduction delay (LVCD) using pre-implantational transesophageal electrocardiography (ECG) in sinus rhythm (SR) CRT responder (R) and non-responder (NR).
Methods: SR HF P (n=52, age 63.6±10.4 years; 6 females, 46 males) with New York Heart Association (NYHA) class 3.0±0.2, 24.4±7.1 % LV ejection fraction and 171.2±37.6 ms QRS duration (QRSD) were measured by bipolar filtered transesophageal LA and LV ECG recording with hemispherical electrodes (HE) TO catheter (Osypka AG, Rheinfelden, Germany). LACD was measured between onset of P-wave in the surface ECG and onset of LA deflection in the LA ECG. LVCD was measured between onset of QRS in the surface ECG and onset of LV deflection in the LV ECG.
Results: There were 78.8 % SR CRT R (n=41) with 171.2±36.9 ms QRSD, 73.3±25.7 ms LACD, 80.0±24.0 ms LVCD and 2.3±0.5 QRSD-LVCD-ratio. SR CRT R QRSD correlated with LACD (r=0.688, P<0.001) and LVCD (r=0.699, P<0.001). There were 21.2 % SR CRT NR (n=11) with 153.4±22.4 ms QRSD (P=0.133), 69.8±24.8 ms LACD (n=6, P=0.767), 54.2±31.0 ms LVCD (P<0.0046) and 3.9±2.5 QRSD-LVCD-ratio (P<0.001). SR CRT NR QRSD not corre-lated with IACD (r=-0.218, P=0.678) and IVCD (r=0.042, P=0.903). During a 22.8±21.3 month CRT follow-up, the CRT R NYHA class improved from 3.1±0.3 to 1.9±0.3 (P<0.001). In CRT NR, NYHA class not improved (2.9±0.4 to 2.9±0.2, P=1) during 11.2±9.8 months BV pacing.
Conclusions: Transesophageal LA and LV ECG with HE can be utilized to analyse LACD and LVCD in HF P. Pre-implantational LVCD and QRSD-LVCD-ratio may be additional useful parameters to improve P selection for SR CRT.
Capture threshold (CT) for transesophageal left atrial (LA) pacing (TLAP) and transesophageal left ventricular (LV) pacing (TLVP) with conventional cylindrical electrodes (CE) are higher than TLAP feeling threshold (FT). Purpose of the study was to evaluate focused TLAP CT and FT for supraventricular tachycardia (SVT) initiation and focused TLVP CT for cardiac resynchronisation therapy (CRT) simulation.
Methods: SVT initiation in patients (P) with palpitations (n=49, age 47 ± 17 years) was analysed during spontaneous rhythm and during focused bipolar TLAP with atrial constant current stimulus output, distal CE and three or seven 6 mm hemispherical electrodes (HE) (TO, Osypka AG, Rheinfelden, Germany). CRT simulation in heart failure P (n=75, age 62 ± 11 years) was evaluated by focused bipolar TLAP and/or TLVP with ventricular constant voltage stimulus output and different pacing mode.
Results: Focused electrical pacing field between CE and HE (n=28) allowed low threshold TLAP with 8.0 ± 2.6 mA CT at 9.9 ms stimulus duration (SD) which was lower than 9.2 ± 4.5 mA FT at 9.9 ms SD. Focused electrical pacing field between HE and HE (n=21) allowed low threshold TLAP with 8.1 ± 2.2 mA CT at 9.9 ms SD which was lower than 9.8 ± 5.0 mA FT at 9.9 ms SD. SVT initiation by programmed AAI TLAP was possible in 23 P and not possible in 26 P. CRT simulation was evaluated with TLAP and TLVP with VAT, D00 and V00 pacing mode and 95.5 ± 10.9 V TLVP CT at 4.0 ms SD.
Conclusions: Programmed focused AAI TLAP allowed initiation of SVT with very low CT and high FT and focused electrical pacing field between CE-HE and HE-HE.CRT simulation with focused TLAP and/or TLVP with VAT, D00 and V00 pacing mode may be a useful technique to detect responders to CRT.
Introduction: Cardiac resynchronisation therapy (CRT) with atrioventricular (AV) and interventricular (VV) optimized biventricular pacing (BV) is an established therapy for heart failure (HF) patients. The aim of the study was to compare AV and VV delay optimization with cardiac output (CO), cardiac index (CI), contractility index (IC) and acceleration index (ACI) impedance cardiographic (ICG) methods in CRT.
Methods: 15 HF patients (age 66 ± 10 years; 2 females, 13 males) in New York Heart Association (NYHA) class 3.1 ± 0.4, left ventricular (LV) ejection fraction 21.3 ± 7.8 % and QRS duration 176.1 ± 31.7 ms underwent AV and VV delay optimization with CO, CI, IC and ACI (Cardioscreen ®, Medis GmbH, Ilmenau, Germany) at different AV and VV delay BV pacing settings versus right ventricular (RV) pacing one day after implantation of a CRT device.
Results: Optimal AV delay after atrial sensing was 108.6 ± 20.3 ms (n=14) and optimal AV delay after atrial pacing 190 ± 14.1 ms (n=2) with AV delay range from 80 ms to 200 ms. Optimal VV delay was -12.3 ± 25.9 ms left ventricular before RV pacing. RV versus BV pacing mode resulted in improvement of CO from 3.4 ± 1.2 l/min to 4.4 ± 1.4 l/min (p<0.001), CI from 1.8 ± 0.64 l/min/m² to 2.4 ± 0.78 l/min/m² (p<0.001), IC from 0.028 ± 0.011 1/s to 0.036 ± 0.013 1/s (p<0.001) and ACI from 0.667 ± 0.227 1/s² to 0.834 ± 0.282 1/s² (p<0.002). During 34 ± 26 month BV pacing, the NYHA class improved from 3.1 ± 0.4 to 2.1 ± 0.4 (p<0.001).
Conclusion: AV and VV delay optimized BV pacing acutely improve hemodynamic parameters of transthoracic ICG and their NYHA class during long-term follow-up. ICG may be a simple and useful technique to optimize AV and VV delay in CRT.
Introduction: Cardiac resynchronisation therapy (CRT) with atrioventricular (AV) and interventricular (VV) optimized biventricular pacing (BV) is an established therapy for heart failure (HF) patients with electrical interventricular conduction delay (IVCD). The aim of the study was to compare AV and VV delay optimization with cardiac output (CO) and acceleration index (ACI) impedance cardiographic (ICG) methods.
Methods: HF patients with IVCD 86.8 ± 33 ms (n=15, age 66 ± 10 years; 2 females, 13 males), New York Heart Association (NYHA) functional class 3.1 ± 0.4, left ventricular (LV) ejection fraction 21.3 ± 7.8 % and QRS duration 176.1 ± 31.7 ms underwent AV and VV delay optimization with CO and ACI methods (Cardioscreen, Medis GmbH, Ilmenau, Germany). After evaluation of optimal AV delay, we evaluated optimal VV delay during simultaneous LV and right ventricular (RV) pacing (LV=RV), LV before RV pacing (LV-RV) and RV before LV pacing (RV-LV).
Results: Optimal VV delay was -12.3 ± 25.9 ms LV-RV pacing with VV delay range from -80 ms LV-RV pacing to +20 ms RV-LV pacing and RV=LV pacing. Optimal AV delay after atrial sensing was 108.6 ± 20.3 ms (n=14) and optimal AV delay after atrial pacing 190 ± 14.1 ms (n=2) with AV delay range from 80 ms to 200 ms. RV versus BV pacing mode resulted in improvement of CO from 3.4 ± 1.2 l/min to 4.4 ± 1.4 l/min (p<0.001) and ACI from 0.667 ± 0.227 1/s² to 0.834 ± 0.282 1/s² (p<0.002). During 34 ± 26 month BV pacing, the NYHA class improved from 3.1 ± 0.4 to 2.1 ± 0.4 (p<0.001).
Conclusion: AV and VV delay optimized BV pacing acutely improve ICG CO and ACI and their NYHA class during long-term follow-up. ICG may be a simple and useful technique to optimize AV and VV delay in CRT.
Introduction: Cardiac resynchronization therapy (CRT) with left ventricular (LV) pacing is an established therapy for heart failure (HF) patients (P) with ventricular desynchronisation and reduced LV ejection fraction (EF). The aim of this study was to test the utilization of the transesophageal approach to measure arterial pulse pressure (PP) during LV pacing and electrical interventricular conduction delay (IVCD), to better select patients for CRT.
Methods: 32 HF patients (age 64 ± 10 years; 5 females, 27 males) with New York Heart Association (NYHA) class 2.8 ± 0.6, 27 ± 11 % LV EF and 155 ± 35 ms QRS duration were analysed with semi-invasive left cardiac pacing and electrocardiography. Esophageal TO8 Osypka catheter of 10.5 F diameter was perorally applied to the esophagus and placed in the position of maximum left atrial (LA) deflection and maximum LV deflection to measure PP with VAT or D00 pacing modes.
Results: Temporary transesophageal LV pacing was possible with VAT mode (n=16) and D00 mode (n=16) in all patients. In 15 Δ-PP-responders, PP was higher during LV pacing on than LV pacing off (78.3 ± 26.6 versus 65.9 ± 23.7 mmHg, P < 0.001) and NYHA class improved from 3.1 ± 0.35 to 2.1 ± 0.35 (P < 0.001) during 29 ± 26 month biventricular (BV) pacing follow-up (6 Medtronic and 9 Boston BV pacing devices). In 17 Δ-PP-non-responders, PP was not higher during LV pacing on than LV pacing off (61.5 ± 23.9 versus 60.9 ± 23.5 mmHg, P = 0.066). IVCD was significant longer in Δ-PP-responders than in Δ-PP-non-responders (87 ± 33 ms versus 37± 29 ms, P < 0.001).
Conclusion: Semi-invasive transesophageale LA and LV pacing with D00 and VAT mode and LV electrogram recording may be useful techniques to predict CRT improvement.
Hintergrund: Richtung und Stärke des elektrischen Feldes (E-Feld) der biventrikulären (BV) Stimulation und elektrische interventrikuläre Desynchronisation sind bei Patienten mit Herzinsuffizienz und verbreitertem QRS Komplex von Bedeutung für den Erfolg der kardialen Resynchronisationstherapie (CRT). Das 3D Herzrhythmusmodell (HRM) ermöglicht die
Simulation von CRT und Hochfrequenz (HF) Ablation. Das Ziel der Studie besteht in der Integration von Schrittmacher- und Ablationselektroden in das HRM zur E-Feld Simulation der BV Stimulation und thermischen Feld (T-Feld) Simulation der HF Ablation von Vorhofflimmern (AF).
Methoden: Es wurden fünf multipolare linksventrikuläre (LV) Elektroden, eine epikardiale LV Elektrode, vier bipolare rechtsatriale (RA) Elektroden, zwei rechtsventrikuläre (RV) Elektroden und ein HF Ablationskatheter mit CST (Computer Simulation Technology, Darmstadt) modelliert und das HRM (Schalk et al: Clin Res Cardiol 106, Suppl 1, April 2017, P1812) um den Koronarvenensinus (CS) erweitert (HRM-CS). E-Feld Simulationen bei vorhofsynchroner BV Stimulation und bei RA Stimulation mit RV und LV Ableitung erfolgten mit den Elektroden Select Secure 3830, Capsure VDD-2 5038 und Attain OTW 4194 im HRM+CS (Fig.). F-Feld Simulationen der HF Ablation von AF bei CRT wurden mit integriertem Ablationskatheter AlCath G FullCircle (Biotronik) simuliert.
Ergebnisse: HRM-CS ermöglichte 3D E-Feld Simulationen bei vorhofsynchroner bipolarer BV Stimulation und bei bipolarer RA Stimulation mit bipolarer RV und LV Ableitung. RV und LV Stimulation erfolgten zeitgleich bei einer Amplitude von 3 V an der LV Elektrode und 1 V an der RV Elektrode mit einer Impulsbreite von jeweils 0,5 ms. Die von der BV Stimulationen erzeugten Fernpotentiale konnten von der RA Elektrode wahrgenommen werden. Das Fernpotential an der RA Elektrodenspitze betrug 32,86 mV und in 1 mm Abstand von der RA Elektrodenspitze ergab sich ein Fernpotential von 185,97 mV. HRM-CS ermöglichte 3D T-Feld Simulationen der HF Ablation von AF bei CRT. Das T-Feld bei HF Ablation des AV-Knotens wurde mit einer anliegenden Leistung von 5 W bei 420 kHz an der distalen 8 mm Ablationselektrode simuliert. Die Temperatur an der Katheterspitze betrug nach 5 s Ablationsdauer 88,66 °C, in 1 mm Abstand von der Katheterspitze im Myokard 42,17 °C und in 2 mm Abstand 37,49 °C.
Schlussfolgerungen: HRM-CS und Elektrodenmodelle ermöglichen die 3D Simulationen von E-Feldern bei vorhofsynchroner BV Stimulation, RA Stimulation mit RV und LV Wahrnehmung und von T-Feldern bei HF Ablation. E-Feld Simulationen von RA, RV und LV Stimulation und Sensing können möglicherweise zur Vorhersage von CRT Respondern genutzt werden.
Background: Cardiac resynchronization therapy (CRT) is an established therapy for heart failure (HF) patients (P) with reduced left ventricular (LV) ejection fraction and electrical interventricular desynchronization, but not all P improved clinically. The aim of the study was to evaluate electrical interventricular delay (IVD) to LV delay (LVD) ratio in atrial fibrillation (AF) CRT responder (R) and non-responder (NR).
Methods: AF P (n = 18, age 60.6 ± 11.4 years, 1 female, 17 males) with HF New York Heart Association (NYHA) class 3.0 ± 0.2, 25.3 ± 5.9 % LV ejection fraction and 157.8 ± 24.4 ms QRS duration (QRSD) were measured by surface ECG and focused transesophageal bipolar LV ECG before implantation of CRT pacemaker (n = 2) or CRT defibrillator (n = 16). IVD was measured between onset of QRS in the surface ECG and onset of LV signal in the LV ECG. LVD was measured between onset and offset of LV signal in the LV ECG.
Results: Electrical ventricular desynchronization in AF CRT P were 61.9 ± 26.9ms IVD, 80.6 ± 24.3ms LVD, 0.85 ± 0.41 IVD-LVD-ratio (Figure), 3.12 ± 1.89 QRSD-IVD-ratio and 2.07 ± 0.47 QRSD-LVD-ratio. There were 72.2 % AF CRT R (n = 13) with 64.2 ± 24.6ms IVD and 77.8 ± 21.6ms LVD with Pearson correlation to 0.89 ± 0.39 IVD-LVD-ratio (r = 0.87, P < 0.01; r = -0.69, P < 0.01), 2.82 ± 1.32 QRSD-IVD-ratio (r = -0.76, P < 0.01; r = 0.67, P = 0.011) and 2.13 ± 0.46 QRSD-LVD-ratio (r = 0.57, P = 0.041; r = -0.85, P < 0.01). There were 27.8% AF CRT NR (n = 5) with 56.0 ± 34.5ms IVD and 87.8 ± 31.9ms LVD without correlation to 0.74 ± 0.48 IVD-LVD-ratio, 3.88 ± 2.98 QRSD-IVD-ratio and 1.90 ± 0.48 QRSD-LVD-ratio. During 15.3 ± 13.1 month CRT follow-up, the AF CRT R NYHA class improved from 3.0 ± 0.2 to 2.2 ± 0.3 (P < 0.001). During 18.8 ± 20.7 month CRT follow-up, the AF CRT NR NYHA class not improved from 3 to 3.3 ± 0.97.
Hintergrund: Das elektrische interventrikuläre Delay (IVD) ist bei Patienten (P) mit Herzinsuffizienz (HF), reduzierter linksventrikulärer (LV) Funktion und verbreitertem QRS Komplex von Bedeutung für den Erfolg der kardialen Resynchronisationstherapie (CRT). Die transösophageale LV Elektrokardiographie (EKG) ermöglicht die Bestimmung des elektrischen IVD und linksventrikulären Delays (LVD). Das Ziel der Studie besteht in der Untersuchung des transösophagealen elektrischen IVD, LVD und deren Verhältnis zur QRS Dauer bei rechtsventrikulärer (RV) Stimulation vor Aufrüstung auf eine biventrikuläre (BV) Stimulation.
Methoden: Bei 11 HF P (Alter 69,0 ± 7,9 Jahre; 10 Männer und 1 Frau) mit DDD Schrittmacher (n=10), DDD Defibrillator (n=1) und RV Stimulation, New York Heart Association (NYHA) Klasse 3,0 ± 0,2, LV Ejektionsfraktion 24,5 ± 4,9 % und QRS-Dauer 228,2 ± 44,8 ms wurden das elektrische IVD als Intervall zwischen Beginn des QRS-Komplexes im Oberflächen EKG und Beginn des LV Signals im transösophagealen LV EKG und das elektrische LVD als Intervall zwischen Beginn und Ende des LV Signals im transösophagealen LV EKG präoperativ vor Aufrüstung auf CRT Defibrillator (n=8) und CRT Schrittmacher (n=3) bestimmt. Der Anstieg des arteriellen Pulse Pressure (PP) wurde zwischen RV Stimulation und transösophagealer LV Stimulation mit unterschiedlichem AV-Delay (n=5) vor Aufrüstung von RV auf BV Stimulation getestet.
Ergebnisse: Bei RV Stimulation betrugen IVD 86,54 ± 32,80 ms, LVD 94,45 ± 23,80 ms, QRS-IVD-Verhältnis 2,63 ± 0,81 mit negativer Korrelation zwischen IVD und QRS-IVD-Verhältnis (r=-0,668 P=0,0248) (Fig.) und QRS-LVD-Verhältnis 2,33 ± 0,73. Vorhofsynchrone ventrikuläre Stimulation führte zu 63,6 ± 27,7 mmHg PP bei RV Stimulation und 80,6 ± 38,5 mmHg PP bei LV Stimulation und der PP erhöhte sich bei LV Stimulation mit optimalem AV Delay um 17 ± 11,2 mmHg gegenüber RV Stimulation (P<0,001). Nach Aufrüstung von RV Stimulation auf BV Stimulation verbesserten sich die NYHA Klasse von 3,1 ± 0,2 auf 2,2 ± 0,3 während 30,4 ± 29,6 Monaten CRT.
Schlussfolgerungen: Das transösophageale LV EKG ermöglicht die Bestimmung des elektrischen IVD und LVD bei RV Stimulation zur Evaluierung der interventrikulären und linksventrikulären elektrischen Desynchronisation. IVD, LVD und deren Verhältnis zur QRS Dauer können möglicherweise zur Vorhersage einer CRT Response vor Aufrüstung von RV auf BV Stimulation genutzt werden.
Cardiac resynchronization therapy (CRT) is an established biventricular pacing therapy in heart failure patients with left bundle branch block and reduced left ventricular ejection fraction, but not all patients improved clinically as CRT responder. Purpose of the study was to evaluate electrical left atrial conduction delay (LACD) with focused transesophageal electrocardiography in CRT responder and CRT non-responder.
Methods: Twenty heart failure patients (age 66.6±8.2 years; 2 females, 18 males) with New York Heart Association functional class 3.0±0.3 and 174.2±40.2ms QRS duration were analysed using posterior left atrial transesophageal electrocardiography with hemispherical electrodes. Electrical LACD was measured between onset and offset of transesophageal left atrial signal before implantation of CRT devices.
Results: Electrical LACD could be evaluated by bipolar transesophageal left atrial electrocardiography using TO Osypka electrode in all heart failure patients with negative correlation between 54.7±18.1ms LACD and 24.9±6.4% left ventricular ejection fraction (r=-0.65, P=0.002). There were 16 CRT responders with reduction of New York Heart Association functional class from 3.0±0.29 to 2.1±0.2 (r=0.522, P=0.038) during 9.41±10.96 month biventricular pacing and negative correlation between 49.6±14.2ms LACD and 26.0±6.2% left ventricular ejection fraction (r=-0.533, P=0.034). There were 4 CRT non-responders with no reduction of New York Heart Association functional class from 3.0±0.4 to 2.8±0.5 (r=0.816, P=0.184) during with 13.88±16.39 month biventricular pacing and no correlation between 75.25±19.17ms LACD and 20.75±6.4% left ventricular ejection fraction (r=-0.831, P=0.169).
Conclusions: Focused transesophageal left atrial electrocardiography can be utilized to analyse electrical LACD in heart failure patients. LACD correlated negative with left ventricular ejection fraction in CRT responders. LACD may be a useful parameter to evaluate electrical left atrial desynchronization in heart failure patients.
Targeting complex fractionated atrial electrocardiograms by automated algorithms during ablation of persistent atrial fibrillation has produced conflicting outcomes in previous electrophysiological studies and catheter ablation of atrial fibrillation and ventricular tachycardia. The aim of the investigation was to evaluate atrial and ventricular high frequency fractionated electrical signals with signal averaging technique.
Methods: Signal averaging electrocardigraphy allows high resolution ECG technique to eliminate interference noise signals in the recorded ECG. The algorithm use automatic ECG trigger function for signal averaged transthoracic, transesophageal and intra-cardiac ECG signals with novel LabVIEW software.
Results: The analysis in the time domain evaluated fractionated atrial signals at the end of the signal averaged P-wave and fractionated ventricular signals at the end of the QRS complex. We evaluated atrial flutter in the time domain with two-to-one atrioventricular conduction, 212.0 ± 4.1 ms atrial cycle length, 426.0 ± 8.2 ms ventricular cycle length, 58.2 ± 1.8 ms P-wave duration, 119.6 ± 6.4 ms PQ duration, 103.0 ± 2.4 ms QRS duration and 296.4 ± 6.8 ms QT duration. The analysis in the frequency domain evaluated high frequency fractionated atrial signals during the P-wave and high frequency fractionated ventricular signals during QRS complex.
Conclusions: Spectral analysis of signal averaging electrocardiography with novel LabVIEW software can be utilized to evaluate atrial and ventricular conduction delays in patients with atrial fibrillation and ventricular tachycardia. Complex fractionated atrial and ventricular electrocardiograms may be useful parameters to evaluate electrical cardiac bradycardia and tachycardia signals in atrial fibrillation and ventricular tachycardia ablation.
Robotic Process Automation (RPA) is a technology for automating business processes and connecting systems by means of software robots in organizations that is gaining traction and growing out of its infancy. Thus, it is no longer just a question of what is technologically feasible, but rather how this technology can be used most profitably. However, business models for RPA remain underinvestigated in literature. Existing work is highly heterogenous, lacking structure and applicability in practice. To close this gap, we present an approach to sustainably establish RPA as a driver of digitization and automation within a company based on an iterative, holistic view of business models with the Business Model Canvas as analysis tool.
Die in dieser Arbeit vorgestellte Vorgehensweise erlaubt die Ortung von Schienenfahrzeugen in topologischen Karten allein mit Hilfe eines Wirbelstromsensorsystems (WSS). Zur Ortung primär erforderlich ist die Identifizierung des befahrenen Gleises selbst, wofür unterschiedliche in einer Karte gespeicherte Merkmale herangezogen werden sowie der zurückgelegte Weg, der durch Zählen der passierten Schwellen ermittelt wird. Diese Merkmale werden mittels eigens definierter, virtueller Sensoren aus dem Signal des WSS gewonnen und mittels einem Bayes’schen Formalismus mit den Referenzdaten aus der vorliegenden topologischen Karte abgeglichen. Diese auf virtuellen Sensoren basierende Vorgehensweise erlaubt eine Parallelisierung der Sensorsignalverarbeitung und eine flexible Einbindung von Sensoren in das Ortungssystem. Die Möglichkeit, Weichen mit einer Trefferquote von 99% zu detektieren, erlaubt die Verfolgung der Fahrzeugposition über die gesamte Fahrstrecke hinweg, unter alleiniger Verwendung der vom WSS gelieferten Messdaten.
Für die genaue Positionsbestimmung in Innenräumen, beispielsweise in Bahnhöfen oder Einkaufszentren, soll in dem beschriebenen Projekt untersucht werden, inwiefern lokale Magnetfelder genutzt werden können, um Genauigkeit und Robustheit zu erhöhen. Hierzu wird untersucht, ob und wie kostengünstige Magnetfeldsensoren und mobile Roboterplattformen genutzt werden können, um Karten zu erstellen, die eine spätere Navigation, beispielsweise mit Smartphones oder mit anderen mobilen Geräten.
Solar irradiance prediction is vital for the power management and the cost reduction when integrating solar energy. The study is towards a ground image based solar irradiance prediction which is highly dependent on the cloud coverage. The sky images are collected by using ground based sky imager (fisheye lens). In this work, different algorithms for cloud detection being a preparation step for their segmentation are compared.
This study focuses on the autonomous navigation and mapping of indoor environments using a drone equipped only with a monocular camera and height measurement sensors. A visual SLAM algorithm was employed to generate a preliminary map of the environment and to determine the drone's position within the map. A deep neural network was utilized to generate a depth image from the monocular camera's input, which was subsequently transformed into a point cloud to be projected into the map. By aligning the depth point cloud with the map, 3D occupancy grid maps were constructed by using ray tracing techniques to get a precise depiction of obstacles and the surroundings. Due to the absence of IMU data from the low-cost drone for the SLAM algorithm, the created maps are inherently unscaled. However, preliminary tests with relative navigation in unscaled maps have revealed potential accuracy issues, which can only be overcome by incorporating additional information from the given sensors for scale estimation.
Modern industrial production is heavily dependent on efficient workflow processes and automation. The steady flow of raw materials as well as the separation of vital parts and semi-finished products are at the core of these automated procedures. Commonly used systems for this work are bowl feeders, which separate the parts and material by a combination of mechanical vibration and friction. The production of these tools, especially the design of the ramping spiral, is delicate and time-consuming work, as the shape, slope, and material must be carefully adjusted for the corresponding parts. In this work, we propose an automated approach, making use of optimization procedures from artificial intelligence, to design the spiral ramps of the bowl feeders. Therefore, the whole system and considered parts are physically simulated and the optimized geometry is subsequently exported into a CAD system for the actual building, respectively printing. The employment of evolutionary optimization gives the need to develop a mathematical model for the whole setup and find an efficient representation of integral features.
The need to measure basic aerosol parameters has increased dramatically in the last decade. This is due mainly to their harmful effect on the environment and on public health. Legislation requires that particle emissions and ambient levels, workplace particle concentrations and exposure to them are measured to confirm that the defined limits are met and the public is not exposed to harmful concentrations of aerosols.
The precise positioning of mobile systems is a prerequisite for any autonomous behavior, in an industrial environment as well as for field robotics. The paper describes the set up for an experimental platform and its use for the evaluation of simultaneous localization and mapping (SLAM) algorithms. Two approaches are compared. First, a local method based on point cloud matching and integration of inertial measurement units is evaluated. Subsequent matching makes it possible to create a three-dimensional point cloud that can be used as a map in subsequent runs. The second approach is a full SLAM algorithm, based on graph relaxation models, incorporating the full sensor suite of odometry, inertial sensors, and 3D laser scan data.
Die Positionierung mobiler Systeme mit hoher Genauigkeit ist eine Voraussetzung für intelligentes autonomes Verhalten, sowohl in der Feldrobotik als auch in industriellen Umgebungen. Dieser Beitrag beschreibt den Aufbau einer Roboterplattform und ihre Verwendung für den Test und die Bewertung von Kalman-Filter-Konfigurationen. Der Aufbau wurde mit einem mobilen Roboter Husky A200 und einem LiDAR-Sensor (Light Detection and Ranging) realisiert. Zur Verifizierung des vorgeschlagenen Aufbaus wurden fünf verschiedene Szenarien ausgearbeitet. Mit denen wurden die Filter auf ihre Leistungsfähigkeit hinsichtlich der Genauigkeit der Positionsbestimmung getestet.
Evaluierung von Kalman Filter Konfigurationen zur Roboterlokaliserung mittels Sensordatenfusion
(2023)
In dieser Arbeit werden drei verschiedene Konfigurationen der von Tom Moore, für das Robot Operating System, entwickelte Kalman-Filter vorgestellt. Diese bilden die Grundlage für eine Lokalisierung mittels Sensorfusion in dem verwendeten ROS-Framework. Ziel dieser Arbeit ist der Aufbau und die Verifikation einer Lokalisierung für ein mobiles Robotersystem Husky A200 der Firma Clearpath Robotics. Hierzu wurden die Möglichkeiten des bestehenden Systems untersucht und mehrere Versionen von Lokalisierungsfiltern konfiguriert. Am an Ende, wird eine Verifikation der Ergebnisse in verschiedenen Szenarien gegeneinandergestellt. Hierzu werden die Ergebnisse einer Variante des Extended Kalman-Filters in 2D (EKF2D), eine Variante des Unscented Kalman-Filter in 2D (UKF2D) und eine Variante des Extended Kalman-Filters in 3D (EKF3D) verifiziert und verglichen. Die Untersuchungen ergaben das der EKF2D die besten und robustesten Ergebnisse für eine Lokalisierung erbringt, trotz, im Vergleich zu der UKF2D Variante, 17,3 % höhere Endpositionsabweichung aufweist. Die in diesem Projekt gewählte EKF3D Konfigurationsvariante eignet sich, wegen seinen starken Ungenauigkeiten in der Höhenbestimmung nicht für eine aussagekräftige Positionsbestimmung.
Mit der Implementierung sowie einer anschließenden aussagekräftigen Evaluierung, soll das, visuelle-inertiale Kartierungs- und Lokalisierungssystem maplab analysiert werden. Hierbei basiert die Kartierung bzw. Lokalisierung auf der Detektion von Umgebungsmerkmalen. Neben der Möglichkeit der Kartenerstellung besteht ferner die Option, mehrere Karten zu fusionieren und somit weitreichende Gebiete zu kartieren sowie für weitere Datenauswertungen zu nutzen. Aufgrund der Durchführung und Bewertung der Ergebnisse in unterschiedlichen Anwendungsszenarien zeigt sich, dass maplab besonders zur Kartierung von Räumen bzw. kleinen Gebäudekomplexen geeignet ist. Die Möglichkeit der Kartenfusionierung bietet weiterhin die Option, den Informationsgehalt von Karten zu erhöhen, welches die Effektivität für eine anschließende Lokalisierung steigert. Bei wachsender Kartierungsgröße hingegen zeigt sich jedoch eine Vergrößerung geometrischer Inkonsistenzen.
A novel approach for synchronization and calibration of a camera and an inertial measurement unit (IMU) in the research-oriented visual-inertial mapping-and localization-framework maplab is presented. Mapping and localization are based on detecting different features in the environment. In addition to the possibility of creating single-case maps, the included algorithms allow merging maps to increase mapping accuracy and obtain large-scale maps. Furthermore, the algorithms can be used to optimize the collected data. The preliminary results show that after appropriate calibration and synchronization maplab can be used efficiently for mapping, especially in rooms and small building environments.
The visual-inertial mapping and localization system maplab is analyzed by its implementation and subsequent evaluation. The mapping or localization is based on environmental feature detection. In addition to creating maps, there is also the option of fusion of several maps and thus mapping extensive areas and using them for further analysis of data. In this way, various software tools can be used to optimize the existing data sets.
Two sensor components are needed: an inertial measuring unit (IMU) and a monochrome camera, which are combined by a hardware rig and put into operation for the analysis of the visual-inertial system. System calibration is crucial for precision and system functioning and is based on nonlinear dynamic state estimation. This ensures the best possible estimate of the position of the environmental feature and the map. Maplab is particularly suitable for mapping rooms or small building complexes as the implementation and evaluation of the results in different application scenarios show. Special emphasis is laid on the evaluation of larger scenarios, in which is shown, that the system is struggling to keep up geometric consistencies and thus provide an accurate map.
Object Detection and Mapping with Unmanned Aerial Vehicles Using Convolutional Neural Networks
(2021)
Significant progress has been made in the field of deep learning through intensive research over the last decade. So-called convolutional neural networks are an essential component of this research. In this type of neural network, the mathematical convolution operator is used to extract characteristics or anomalies. The purpose of this work is to investigate the extent to which it is possible in certain initial settings to input aerial recordings and flight data of Unmanned Aerial Vehicles (UAVs) in the architecture of a neural network and to detect and map an object. Using the calculated contours or dimensions of the so-called bounding boxes, the position of the objects can be determined relative to the current UAV location.
The fisheye camera has been widely studied in the field of ground based sky imagery and robot vision since it can capture a wide view of the scene at one time. However, serious image distortion is a major drawback hindering its wider use. To remedy this, this paperproposes a lens calibration and distortion correction method for detecting clouds and forecasting solar radiation. Finally, the radial distortion of the fisheye image can be corrected by incorporating the estimated calibration parameters. Experimental results validate the effectiveness of the proposed method.
This paper deals with the detection and segmentation of clouds on high-dynamic-range (HDR) images of the sky as well as the calculation of the position of the sun at any time of the year. In order to predict the movement of clouds and the radiation of the sun for a short period of time, the clouds thickness and position have to be known as precisely as possible. Consequently, the segmentation algorithm has to provide satisfactory results regardless of different weather, illumination and climatic conditions. The principle of the segmentation is based on the classification of each pixel as a cloud or as a sky. This classification is usually based on threshold methods, since these are relatively fast to implement and show a low computational burden. In order to predict if and when the sun will be covered by clouds, the position of the sun on the images has to be determined. For this purpose, the zenith and azimuth angles of the sun are determined and converted into XY coordinates.
The paper describes a systematic approach for a precise short-time cloud coverage prediction based on an optical system. We present a distinct pre-processing stage that uses a model based clear sky simulation to enhance the cloud segmentation in the images. The images are based on a sky imager system with fish-eye lens optic to cover a maximum area. After a calibration step, the image is rectified to enable linear prediction of cloud movement. In a subsequent step, the clear sky model is estimated on actual high dynamic range images and combined with a threshold based approach to segment clouds from sky. In the final stage, a multi hypothesis linear tracking framework estimates cloud movement, velocity and possible coverage of a given photovoltaic power station. We employ a Kalman filter framework that efficiently operates on the rectified images. The evaluation on real world data suggests high coverage prediction accuracy above 75%.
The applicability of characteristics of local magnetic fields for more precise determination of localization of subjects and/or objects in indoor environments, such as railway stations, airports, exhibition halls, showrooms, or shopping centers, is considered. An investigation has been carried out to find out whether and how low-cost magnetic field sensors and mobile robot platforms can be used to create maps that improve the accuracy and robustness of later navigation with smartphones or other devices.
The aim of this work is the application and evaluation of a method to visually detect markers at a distance of up to five meters and determine their real-world position. Combinations of cameras and lenses with different parameters were studied to determine the optimal configuration. Based on this configuration, camera images were taken after proper calibration. These images are then transformed into a bird's eye view using a homography matrix. The homography matrix is calculated with four-point pairs as well as with coordinate transformations. The obtained images show the ground plane un distorted, making it possible to convert a pixel position into a real-world position with a conversion factor. The proposed approach helps to effectively create data sets for training neural networks for navigation purposes.
Bei dem vorgestellten Ansatz soll der Auftreffpunkt des Pfeils durch die Kreuzkorrelation von Audio-Signalen bestimmt werden. Das Auftreffen des Pfeils erzeugt ein charakteristisches Geräusch, welches von mehreren Mikrofonen in bestimmter Anordnung um die Dartscheibe herum in elektrische Signale umgewandelt wird. Mithilfe der Schallgeschwindigkeit und den Zeitdifferenzen, welche die Schallwelle zu den einzelnen Mikrofonen benötigt soll dann der Auftreffpunkt berechnet werden.
This paper presents an approach for implementing an automated hit detection and score calculation system for a steel dartboard using a standard webcam. First, the rectilinear field separations of the dartboard are described mathematically by means of line slopes and are than stored. These slopes serve as a basis for later score calculation. In addition, thrown darts have to be detected and the pixel at which the dart cuts the dartboard has to be determined. When this information is known, a comparison is made using the line slopes, allowing the field number of the hit to be detected. The decision for single, double or triple hit is made by evaluating the defined colors on the dartboard. All these functions are then packaged in a Matlab GUI.
Sweaty has already participated several times in RoboCup soccer competitions (Adult Size). Now the work is focused coordinating the play of two robots. Moreover, we are working on stabilizing the gait by adding additional sensor information. An ongoing work is the optimization of the control strategy by balancing between impedance and position control. By minimizing the jerk, gait and overall gameplay should improve significantly.
Rising societies’ demands require more sustainable products and technologies. Although numerous methods and tools have been developed in the last decades to support environmental-friendly product and process development, an interdisciplinary knowledge base of eco-innovative examples linked to the eco-innovative problems and solution principles is lacking. The paper proposes an ontology of examples for eco-friendly products and technologies assigned to the Inventive Principles (IPs) of the TRIZ methodology in accordance with the German TRIZ Standard VDI 4521. The examples of sustainable technologies and products build a database for sharing and reusing eco-innovation knowledge. The ontology acts as a tool for systematic solving of specific environmental problems in typical life cycle phases, for different environmental impact categories and engineering domains. Finally, the paper defines a future research agenda in the field of the TRIZ-based systematic eco-innovation.
Moderne strombetriebene Wärmepumpenanlagen sind im Hinblick auf die Umweltbilanz und die Jahresenergiekosten selbst gegenüber den effizientesten gasbefeuerten und ölbefeuerten Heizanlagen deutlich im Vorteil. Um diesen Vorteil auch in der Praxis zu erreichen, muss der Auswahl und Dimensionierung der Wärmepumpe einschliesslich der Wärmequelle besondere Beachtung geschenkt werden. Hinsichtlich des Temperaturniveaus bietet sich als Wärmequelle besonders das Erdreich an, beispielsweise durch Erdsonden erschlossen. Abhängig von den hydrogeologischen Rahmenbedingungen ist insbesondere auf die längenbezogene Wärmeentzugsleistung der Erdsonden zu achten. Hierfür hat das Ministerium für Umwelt und Verkehr des Landes Baden-Württemberg einheitliche Grundlagen für die Beurteilung und Bearbeitung geschaffen. Ausgehend von einer Wärmepumpenheizanlage, von der bereits Messungen vorliegen, wurde diese Anlage mit dem Simulationsprogramm TRNSYS nachgebildet und die wesentlichen Temperaturverhältnisse in verschiedenen Erdreichtiefen und Erdsondenabständen ermittelt. Mit diesen Erdreichtemperaturen werden mit Hilfe eines vereinfachten Energiebilanzverfahrens die Jahresarbeitszahlen ermittelt. Dabei zeigt sich, dass neben der Hydraulik der Heizanlage der Art der Trinkwassererwärmung besondere Bedeutung zukommt, da dies sich entscheidend auf die Jahresarbeitszahl und damit auch auf die mögliche Reduzierung des CO2-Ausstosses auswirkt.
Bei einem neuen Kühldeckensystem wird die Luft im Deckenhohlraum oberhalb der abgehängten Decke durch Konvektoren gekühlt, wobei die Wärmeübertragung ausschließlich durch freie Konvektion und Strahlung erfolgt. Die abgehängte Decke nimmt durch Strahlung und freie Konvektion Wärme aus dem darunterliegenden Büroraum auf und gibt sie über freie Konvektion an die Luft im Deckenhohlraum sowie durch Strahlung an die Begrenzungsflächen des Deckenhohlraumes ab. Die vom Kühlwasser durchströmten Rippenrohre nehmen ihrerseits Wärme aus der Luft im Hohlraum auf, die infolge der freien Konvektion durch die Rippenrohre strömt. Zur Aufnahme des bei starker Belastung und Dauerbetrieb anfallenden Kondensats ist unterhalb des Konvektors eine Kondensatauffangwanne angebracht. Mit dem von Rutsch erstellten Rechenprogramm ist es möglich, für verschiedenste Randbedingungen die Rippenrohre auf maximale Kühlleistung zu optimieren. Voraussetzung für die direkte Anwendung dieses Berechnungsverfahrens sind jedoch ungestörte An- und Abströmbedingungen der Luft am Konvektor. Neben der Optimierung der Rippenrohre auf Kühlleistung ist für den Aufbau der eigentlichen Decke die Luftströmung im Deckenhohlraum und die Austrittstemperatur aus den Konvektoren von entscheidender Bedeutung. Die maximale Kühlleistung der Rippenrohre kann nur erreicht werden, wenn alle Einzelrippenrohre nahezu gleichmäßig durchströmt werden und günstige Abströmverhältnisse vorhanden sind. Die Strömungsverhältnisse im Deckenhohlraum und innerhalb der Konvektoren wurden durch Rauch sichtbar gemacht und auf Video aufgezeichnet. Entscheidenden Einfluß auf die Temperaturverteilung an der Kühldecke haben die Breite des Konvektors, die Konstruktion der Kondensatauffangwanne, der Abstand der Konvektoren voneinander und der Konvektortyp. Direkt unterhalb der Kondensatauffangwanne beträgt die Temperaturdifferenz ca. 3 K. Der Temperaturgradient im Abströmbereich liegt bei ca. 0,2 bis 0,3 K/m.
In jüngster Zeit werden vermehrt größere thermische Solaranlagen zur Warmwassererzeugung und zur Heizungsunterstützung in größeren Mehrfamilienhäusern und Neubausiedlungen mit Einfamilien- bzw. Doppelhäusern eingesetzt. Während die hydraulische Einbindung der Solarenergie zur Heizungsunterstützung meist problemlos realisiert wird, entstehen bei der Warmwasserbereitung immer wieder Probleme hinsichtlich der Legionellenproblematik. Die Anforderungen im DVGW Arbeitsblatt W551, den gesamten Wasserinhalt der Vorwärmstufen einmal am Tag auf 60 Grad C zu erwärmen und damit zur Verminderung des Legionellenwachstums beizutragen, führen immer wieder dazu, daß durch fehlerhafte Konzepte, insbesondere im Bereich der hydraulischen Schaltungen in Verbindung mit den Regelkonzepten, der solare Deckungsgrad bei der Trinkwassererwärmung sehr gering bleibt. Anhand ausgewählter Ausführungsbeispiele wird die Problematik ausführlich behandelt und Lösungsmöglichkeiten aufgezeigt. Abschließend werden funktionsgerechte Anlagentechniken verschiedener Ausführungsvarianten besprochen.
Durch den verstärkten Einsatz von EDV-Geräten an den Arbeitsplätzen in Büroräumen stiegen die flächenbezogenen inneren Kühllasten deutlich an. Für die Arbeitsplätze in Fensternähe müssen EDV-gerechte Lichtverhältnisse vorhanden sein. Es werden Versuchsergebnisse eines neu entwickelten Kühlkonvektors im Fensterbereich im Zusammenhang mit Innenjalousien vorgestellt. Dieser Kühlkonvektor eignet sich auch für den kombinierten Einsatz mit Kühldecken und raumlufttechnischen (RLT)-Anlagen, wie weitere Versuche bestätigen. Der Kühlkonvektor kann auch beim Ersatz von Induktionsgeräten angewandt werden, die z.B. nach dem 4-Leiter-Prinzip angeschlossen sind. Die Untersuchungen haben gezeigt, daß mit der Kombination von Kühldecke, Kühlkonvektor und RLT-Anlage große flächenbezogene Kühllasten abgeführt werden können, ohne daß dadurch die thermische Behaglichkeit negativ beeinflußt wird. Der Luftstrom läßt sich auf das hygienisch bzw. auf das zur Entfeuchtung notwendige Maß reduzieren, wodurch sich die Betriebskosten senken. Der Anteil der drei Komponenten an der gesamten Kühllast beträgt im Mittel: 15 % RLT-Anlage, 15 % Kühlkonvektor und 70 % Kühldecke.
Massiv- Heiz-/Kühldecken bestehen im wesentlichen aus in der tragenden Deckenkonstruktion eingegossenen Kunststoffrohren, die je nach Jahreszeit von Heiz- bzw. Kühlwasser durchströmt verden. Das instationäre Verhalten wurde mit dem Anlagen und Simulationsprogramm TRNSYS untersucht. Anhand ausgesuchter Ergebnisse können die Anforderungen an die notwendigen Regelkonzepte abgeleitet werden. Es eignen sich für derartige Systeme, insbesondere wegen ihrer großen Speicherfähigkeit und damit Trägheit, modifizierende selbstadaptierende Regler. Aufgrund der großen Trägheit der Massivdecke ist eine direkte außentemperaturgeführte Vorlauftemperaturregelung des Wasserstromes nicht sinnvoll. Es sollte ein selbstadaptierendes Regelsystem verwendet werden, das anhand der vorangegangenen Lastverläufe die Gebäudekonstante und damit das Speicherverhalten der Räume ermittelt. Damit der Regelaufwand reduziert werden kann, sollen mehrere Räume als Gruppe zusammengefaßt werden. Als weitere Größe für das Regelkonzept ist die Speichertemperatur am Ende der Ladezeit zu messen und in das Programm mit einzubinden.
Die Untersuchungen wurden im Raumlufttechnischen Labor der Fachhochschule Offenburg zur Beurteilung der Effizienz der Schadstoffabfuhr unter Einhaltung der thermischen Behaglichkeit beim Einsatz verschiedener Luftführungssysteme- Quellauslaß, Deckendrallauslaß und Fußbodenauslaß- in Verbindung mit einer Kühldecke durchgeführt. Die Ergebnisse zeigen, daß keine signifikanten Vorteile für ein bestimmtes RLT-Konzept bestehen. Je nach Wahl der Randbedingungen der Untersuchungen liegen die erreichbaren Kühlleistungen innerhalb einer Systemkombination weiter auseinander als im Verhältnis zu den Vergleichssystemen. Insgesamt ist hinsichtlich der von beiden Systemen zusammen abgebbaren, maximalen Kühlleistung die Kombination Kühldecke mit Deckendralluftauslaß vorteilhaft.
Bei modernen Lokomotiven mit Drehstrom-Asynchronmotoren und mit bis zu 7 Megawatt Leistung, neigt das Antriebssystem bei nicht ausreichendem Kraftschluß zwischen Treibradsatz und Schiene zum 'Durchdrehen'. Bei diesem Vorgang wird nahezu die gesamte Energie zur Beschleunigung des Radsatzes eingesetzt, was zu mechanischen Schäden an den Rädern und Schienen führen kann. Beim Bremsen ist dies ähnlich, die Räder gleiten auf den Schienen, wenn zuviel Bremskraft gefordert wird. Die übertragbaren Zug- und Bremskräfte werden primär durch die Radsatzlast und den Kraftschlußbeiwert bestimmt, wobei der Verlauf der Kraftschlußkennlinie als Funktion des Schlupfs oder der Schlupfgeschwindigkeit im wesentlichen durch den Schienenzustand (naß oder trocken) bestimmt wird. Eine hohe Kraftschlußnutzung wird dann erreicht, wenn man laufend denjenigen Schlupfwert einstellt, der zum jeweiligen Kraftschlußmaximum führt. Hierzu werden in der Praxis verschiedene Konzepte und Methoden eingesetzt, es ist bis heute jedoch keine Methode bekannt, den Verlauf der Kraftschlußkennlinie meßtechnisch laufend zu erfassen oder rechentechnisch zu bestimmen. Bei der hier vorgestellten Vorgehensweise wurde der mechanische Antrieb zusammen mit dem Rad-Schiene-Kontakt als Zustandsraumodell beschrieben. Die Betrachtungen beruhen dann auf einer Frequenzganguntersuchung des eingeführten linearen Zustandsraummodells. Aufgrund der Linearisierung gelten die Ergebnisse der Frequenzgangsberechnung nur für den jeweiligen Betriebspunkt der Kraftschlußkennlinie, also für eine bestimmte Steigung. Variiert man nun die Steigung, so läßt sich der Einfluß der nichtlinearen Kraftschlußkennlinien ermitteln. Zur Messung von Frequenzgängen eignen sich insbesondere Verfahren der Orthogonalen Korrelation. Die technische Realisierung wird skizziert. Die Meßinformation ist dann die Basis für eine Regelung, die ein permanentes optimales Fahren im Kraftschlußmaximum zuläßt und zwar beim Beschleunigen und beim Bremsen. Das beschriebene Meß- und Regelungsverfahren ist derzeit in der Schweiz in der Betriebserprobung.
In der Wertanalyse ist die Methodik TRIZ (Theorie der Lösung erfinderischer Problemstellungen) seit vielen Jahren als Werkzeug zur Kostensenkung oder zur Steigerung der Funktionalität von Produkten bekannt. Seit ihrem ersten Bekanntwerden in Westeuropa hat sich auch TRIZ weiterentwickelt. So wurden Methoden zur Modellierung von Systemen inzwischen erweitert und um Werkzeuge zur schnellen Lösungsfindung, zur Fehlervoraussage und zur Produktplanung neu entwickelt. Durch den weltweiten wissenschaftlichen Fortschritt, die Verwendung unterschiedlicher Sprachen und neue Literatur ist andererseits auch die verwendete Terminologie angewachsen und nicht mehr eindeutig. Die neue VDI-Richtlinie 4521, von deren erstem Teil nun der Gründruck vorliegt, zielt deswegen auf eine Standardisierung der Terminologie und eine vereinheitlichte Beschreibung der Methoden ab. Mit ihrer Hilfe sollen das Studium der Methodik erleichtert, die Benutzung von Literatur vereinfacht und Inhalte der TRIZ klarer darstellbar werden.
The process of establishing an industry standard for TRIZ has been initiated: VDI Guideline 4521 will cover TRIZ. Work is going on on the first part of the standard which will define and explain basic TRIZ vocabulary and notions. A first draft of a list of terms has been compiled by V. Souchkov and is currently being discussed at MATRIZ. The standardization committee consists of TRIZ specialists of various degrees together with TRIZ users from industry. It is working in close connection with MATRIZ. In parallel, translations for the elements of TRIZ terminology into several languages are being sought. According to schedule, work on the first part of the standard may be finished by July 2014 and may go into print by the end of the year.
VDI Standard 4521: Status
(2016)
VDI Guideline 4521 Part 1: “Inventive problem solving with TRIZ: Part 1 – Fundamentals and definitions” has been published on 2015-04-01. The standard will sharpen the image of TRIZ, facilitate cooperation, and support studying and teaching. It is not a textbook but concisely summarizes basic assumptions of TRIZ and its terminology. It gives an overview on specific methods and tools which will be described in the following parts.
Separation Estimation with Thermal Cameras for Separation Monitoring in Human-Robot Collaboration
(2022)
Human-Robot Collaborative applications have the drawback of being less efficient than their non-collaborative counterparts. One of the main reasons is, that the robot has to slow down when a human being is within the operating space of the robot. There are different approaches on dynamic speed and separation monitoring in human-robot collaborative applications. One approach additionally differentiates between human and non-human objects to increase efficiency in speed and separation monitoring. This paper proposes to estimate the separation distance by measuring the temperature of the approaching object. Measurements show that the measured temperature of a human being decreases with 1 deg C per meter distance from the sensor. This allows an estimation of separation between a robotic system and a human being.
Differentiation between human and non-human objects can increase efficiency of human-robot collaborative applications. This paper proposes to use convolutional neural networks for classifying objects in robotic applications. The body temperature of human beings is used to classify humans and to estimate the distance to the sensor. Using image classification with convolutional neural networks it is possible to detect humans in the surroundings of a robot up to five meters distance with low-cost and low-weight thermal cameras. Using transfer learning technique we trained the GoogLeNet and MobilenetV2. Results show accuracies of 99.48 % and 99.06 % respectively.
Efficient collaborative robotic applications need a combination of speed and separation monitoring, and power and force limiting operations. While most collaborative robots have built-in sensors for power and force limiting operations, there are none with built-in sensor systems for speed and separation monitoring. This paper proposes a system for speed and separation monitoring directly from the gripper of the robot. It can monitor separation distances of up to three meters. We used single-pixel Time-of-Flight sensors to measure the separation distance between the gripper and the next obstacle perpendicular to it. This is the first system capable of measuring separation distances of up to three meters directly from the robot's gripper.
Avoiding collisions between a robot arm and any obstacle in its path is essential to human-robot collaboration. Multiple systems are available that can detect obstacles in the robot's way prior and subsequent to a collision. The systems work well in different areas surrounding the robot. One area that is difficult to handle is the area that is hidden by the robot arm. This paper focuses on pick and place maneuvers, especially on obstacle detection in between the robot arm and the table that robot is located on. It introduces the use of single pixel time-of-flight sensors to detect obstacles directly from the robot arm. The proposed approach reduces the complexity of the problem by locking axes of the robot that are not needed for the pick and place movement. The comparison of simulated results and laboratory measurements show concordance.
Human-robot collaboration plays a strong role in industrial production processes. The ISO/TS 15066 defines four different methods of collaboration between humans and robots. So far, there was no robotic system available that incorporates all four collaboration methods at once. Especially for the speed and separation monitoring, there was no sensor system available that can easily be attached directly to an off-the-shelf industrial robot arm and that is capable of detecting obstacles in distances from a few millimeters up to five meters. This paper presented first results of using a 3D time-of-flight camera directly on an industrial robot arm for obstacle detection in human-robot collaboration. We attached a Visionary-T camera from SICK to the flange of a KUKA LBR iiwa 7 R800. With Matlab, we evaluated the pictures and found that it works very well for detecting obstacles in a distance range starting from 0.5 m and up to 5 m.
Correlation Clustering, also called the minimum cost Multicut problem, is the process of grouping data by pairwise similarities. It has proven to be effective on clustering problems, where the number of classes is unknown. However, not only is the Multicut problem NP-hard, an undirected graph G with n vertices representing single images has at most edges, thus making it challenging to implement correlation clustering for large datasets. In this work, we propose Multi-Stage Multicuts (MSM) as a scalable approach for image clustering. Specifically, we solve minimum cost Multicut problems across multiple distributed compute units. Our approach not only allows to solve problem instances which are too large to fit into the shared memory of a single compute node, but it also achieves significant speedups while preserving the clustering accuracy at the same time. We evaluate our proposed method on the CIFAR10 …
Multiple Object Tracking (MOT) is a long-standing task in computer vision. Current approaches based on the tracking by detection paradigm either require some sort of domain knowledge or supervision to associate data correctly into tracks. In this work, we present a self-supervised multiple object tracking approach based on visual features and minimum cost lifted multicuts. Our method is based on straight-forward spatio-temporal cues that can be extracted from neighboring frames in an image sequences without supervision. Clustering based on these cues enables us to learn the required appearance invariances for the tracking task at hand and train an AutoEncoder to generate suitable latent representations. Thus, the resulting latent representations can serve as robust appearance cues for tracking even over large temporal distances where no reliable spatio-temporal features can be extracted. We show that, despite being trained without using the provided annotations, our model provides competitive results on the challenging MOT Benchmark for pedestrian tracking.
In this work, we evaluate two different image clustering objectives, k-means clustering and correlation clustering, in the context of Triplet Loss induced feature space embeddings. Specifically, we train a convolutional neural network to learn discriminative features by optimizing two popular versions of the Triplet Loss in order to study their clustering properties under the assumption of noisy labels. Additionally, we propose a new, simple Triplet Loss formulation, which shows desirable properties with respect to formal clustering objectives and outperforms the existing methods. We evaluate all three Triplet loss formulations for K-means and correlation clustering on the CIFAR-10 image classification dataset.
Estimating the Robustness of Classification Models by the Structure of the Learned Feature-Space
(2022)
Over the last decade, the development of deep image classification networks has mostly been driven by the search for the best performance in terms of classification accuracy on standardized benchmarks like ImageNet. More recently, this focus has been expanded by the notion of model robustness, \ie the generalization abilities of models towards previously unseen changes in the data distribution. While new benchmarks, like ImageNet-C, have been introduced to measure robustness properties, we argue that fixed testsets are only able to capture a small portion of possible data variations and are thus limited and prone to generate new overfitted solutions. To overcome these drawbacks, we suggest to estimate the robustness of a model directly from the structure of its learned feature-space. We introduce robustness indicators which are obtained via unsupervised clustering of latent representations from a trained classifier and show very high correlations to the model performance on corrupted test data.
In this work, we evaluate two different image clustering objectives, k-means clustering and correlation clustering, in the context of Triplet Loss induced feature space embeddings. Specifically, we train a convolutional neural network to learn discriminative features by optimizing two popular versions of the Triplet Loss in order to study their clustering properties under the assumption of noisy labels. Additionally, we propose a new, simple Triplet Loss formulation, which shows desirable properties with respect to formal clustering objectives and outperforms the existing methods. We evaluate all three Triplet loss formulations for K-means and correlation clustering on the CIFAR-10 image classification dataset.
Engineering, construction and operation of complex machines involves a wide range of complicated, simultaneous tasks, which potentially could be automated. In this work, we focus on perception tasks in such systems, investigating deep learning approaches for multi-task transfer learning with limited training data. We show an approach that takes advantage of a technical systems’ focus on selected objects and their properties. We create focused representations and simultaneously solve joint objectives in a system through multi-task learning with convolutional autoencoders. The focused representations are used as a starting point for the data-saving solution of the additional tasks. The efficiency of this approach is demonstrated using images and tasks of an autonomous circular crane with a grapple.
Autonomous humanoid robots need high torque actuators to be able to walk and run. One problem in this context is the heat generated. In this paper we propose to use water evaporation to improve cooling of the motors. Simulations based on thermodynamic calculations as well as measurements on real actuators show that, under the assumption of the load of a soccer game, cooling can be considerably improved with relatively small amounts of water.
Die Verwendung von Kameras als Messmittel für medizinische Anwendungen setzt deren präzise Kalibrierung voraus. Gängige Verfahren modellieren die Abbildungseigenschaften einer Kamera mittels perspektivischer Projektion und parametrisierter Funktionen zur Beschreibung von Linsenverzerrung. In den Randbereichen des Kamerabildes sind diese Modelle oft unzureichend. Außerdem bedingt die Verwendung starrer Kalibriermuster eine in der Regel kleine Anzahl an nicht gleichmäßig verteilten Punktkorrespondenzen zur Bestimmung der Modellparameter. In der vorliegenden Arbeit wird ein vollkommen neues und nicht auf Modellen basierendes Kalibrierverfahren vorgestellt, bei dem jedes Kamerapixel unabhängig von jedem anderen kalibriert wird.
WirelessHART protocol was specifically designed for real-time communication in the wireless sensor networks domain for industrial process automation requirements. Whereas the major purpose of WirelessHART is the read-out of sensors with moderate real-time requirements, an increasing demand for integration of actuator applications can be observed. Therefore, it must be verified that the WirelessHART protocol gives sufficient support to real-time industry requirements. As a result, the delay of especially burst and command messages from actuator and sensor nodes to the gateway and vice versa must be analyzed. In this paper, we implemented a WirelessHART network scenario in WirelessHART simulator in NS-2 [8], simulated and analyzed its time characteristics under ideal and noisy conditions. We evaluated the performance of the implementation in order to verify whether the requirements of industrial process and control can be met. This implementation offers an early alternative to expensive test beds for WirelessHART in real-time actuator applications.
The nonlinear behavior of inverters is largely impacted by the interlocking and switching times. A method for online identifying the switching times of semiconductors in inverters is presented in the following work. By being able to identify these times, it is possible to compensate for the nonlinear behavior, reduce interlocking time, and use the information for diagnostic purposes. The method is first theoretically derived by examining different inverter switching cases and determining potential identification possibilities. It is then modified to consider the entire module for more robust identification. The methodology, including limitations and boundary conditions, is investigated and a comparison of two methods of measurement acquisition is provided. Subsequently the developed hardware is described and the implementation in an FPGA is carried out. Finally, the results are presented, discussed, and potential challenges are encountered.
Current Harmonics Control Algorithm for inverter-fed Nonlinear Synchronous Electrical Machines
(2023)
Current harmonics are a well known challenge of electrical machines. They can be undesirable as they can cause instabilities in the control, generate additional losses and lead to torque ripples with noise. However, they can also be specifically generated in new methods in order to improve the machine behavior. In this paper, an algorithm for controlling current harmonics is proposed. It can be described as a combination of different PI controllers for defined angles of the machine with repetitive control characteristics for whole revolutions. The controller design is explained and important points where linearization is necessary are shown. Furthermore, the limits are analyzed and, for validation, measurement results with a permanently excited synchronous machine on the test bench are considered.
The nonlinear behavior of inverters is mainly influenced by the interlocking and switching times of the semiconductors. In the following work, a method is presented that enables the possibility of an online identification of the switching times of the semiconductors. This information allows a compensation of the non-linear behavior, a reduction of the locking time and can be used for diagnostic purposes. First, a theoretical derivation of the method is made by considering different cases when switching of the inverter and deriving identification possibilities. The method is then extended so that the entire module is taken into account. Furthermore, a possible theoretical implementation is shown. After the methodology has been investigated with possible limitations, boundary conditions and with respect to real hardware, an implementation in the FPGA is performed. Finally, the results are presented, discussed
and further improvements are presented in an outlook.
Decrease of non-responder rate is the main chal-lenge in cardiac resynchronization therapy. The problem could be solved, partly, in the follow-up by consequent indi-vidualization of hemodynamic pacing parameters. The eso-phageal electrogram feature of the Biotronik ICS 3000 programmer was used in the follow-up of 20 heart failure patients carrying implants for cardiac resynchronization therapy. Adverse hemodynamic programming of the sensed and paced AV delay could be easily observed and replaced by the individual optimal duration in 3 patients (15%) VDD and DDD operation.This result proves the value of esophageal electrogram recording CRT follow-up.
Distribution of esophageal interventricular conduction delays in CRT patients and healthy subjects
(2015)
The uncertain and time-variant nature of renewable energy results in the need to deal with peaks in the production of energy. One approach is to achieve a load shift and thereby help balancing the grid by using thermally Activated Building Systems (TABS). Control systems currently in place do not exploit the full potential of TABS. This paper reviews how Model Predictive Control can possibly reduce the fluctuations of the demand and supply of (renewable) energy as it enables the TABS to react to the dynamics of weather and its impact on the grid at any time.
Konstrukteure im Maschinenbau stehen häufig vor der Problemstellung, hochfest vorgespannte Schraubenverbindungen und einen durchgehenden Korrosionsschutz zu vereinen. Die einschlägigen Normen und Richtlinien bieten hierzu Stand heute keine ausreichende Hilfestellung. In diesem Beitrag werden an Versuchsblechen ermittelte Setzbeträge von maschinenbautypischen organischen Beschichtungssystemen unter Variation der Belastungshöhe und der Umgebungstemperatur präsentiert und mit in Bauteilversuchen gemessenen Vorspannkraftverlusten vergleichend bewertet.
Die Untersuchungen der Hochschule Offenburg zeigen, dass es durch organische Korrosionsschutzschichten im Kraftfluss von Schraubenverbindungen nicht zu einem verfrühten Abschalten der streckgrenzengesteuerten Schraubmontage kommt. Die fünf untersuchten Lacksysteme zeigten ein sehr unterschiedliches Reibverhalten, der Anzugsvorgang wurde jedoch zuverlässig bei Erreichen der Schraubenstreckgrenze beendet. Durch den ermittelten Drehmoment/Drehwinkelverlauf lässt sich das streckgrenzengesteuerte Anzugsverfahren als Analystetool einsetzen, wodurch für den jeweiligen Schraubfall auch Rückschlüsse auf anderen Anzugsverfahren getroffen werden können. Des Weiteren zeigte sich, dass Pulverlacksysteme widerstandsfähiger gegen die bei der Montage wirkenden Belastungen sind und eine Montage direkt auf Lack ermöglichen können.
Höchste Korrosionsschutzanforderungen sind für bestimmte technische Produkte insbesondere im Offshore Anwendungsbereich, nach ISO 20340, zwingend zu erfüllen, um deren Funktion und Betriebssicherheit dauerhaft gewährleisten zu können. Bis heute werden viele dieser Produkte am Ende ihrer Wertschöpfungskette nass überlackiert, mit einer kompletten Kunststoffhaut, der Korrosionsschutz-Lackschicht, überzogen. Diese Lackierung ist unter anderem deshalb erforderlich, weil es im klassischen Maschinenbau, insbesondere in der Antriebstechnik, viele mechanische Schnittstellen gibt, die vor der endgültigen Produktmontage quasi metallisch blank bleiben müssen, um den erforderlichen und definierten geometrischen Oberflächenzustand nach Form und Lage als Pass- und Fügefläche zu gewährleisten. Eine dieser mechanischen Schnittstellen sind Schraubenverbindungen. Mit dem derzeit gültigen Regelwerk ist die Berechnung einer Schraubenverbindung mit Lackschichten in den Trennfugen oder auf der Kopf- und Mutternauflagefläche nicht möglich, da lackierte Bauteile in der derzeit geltenden VDI-Richtlinie 2230 nicht berücksichtigt sind. Nach einem Praxisbericht anhand von Stellantrieben für Industriearmaturen über deren Umstellung von Nasslackierung des Gesamtproduktes auf Pulverbeschichtung von Einzelteilen wird die experimentelle Validierung der Betriebs- und Funktionssicherheit von Schraubenverbindungen mit lackierten Bauteilen vorgestellt. Daraus resultierend wurde im März 2014 an der Hochschule Offenburg ein Forschungsprojekt gestartet, dessen Ziel es ist für die oben genannte Problemstellung einen systematischen Lösungsansatz zu erarbeiten. Künftig soll es Entwicklungsingenieuren und Konstrukteuren bereits in der Phase von Entwicklung und Konstruktion möglich sein Schraubenverbindungen mit lackierten Bauteilen zuverlässig zu berechnen und auszulegen oder diese in der Prototypenphase zuverlässig zu testen. Die letzten beiden Abschnitte geben den Lösungsansatz und den aktuellen Stand der Forschung wider.