FB Technik
Filtern
Dokumenttyp
Volltext vorhanden
- ja (14) (entfernen)
Gehört zur Bibliographie
- nein (14)
Schlagworte
- Elektrode (2)
- Flüssigkristalline Polymere (2)
- Parkinson-Krankheit (2)
- Rauigkeit (2)
- electrode model (2)
- surface roughness (2)
- Algorithmus (1)
- Astrozyt (1)
- BEV (1)
- Bewegung (1)
Institut
- FB Technik (14)
Gait analysis is a systematic study of human movement. Combining wearable foot pressure sensors and machine learning (ML) solutions for a high-fidelity body pose tracking from RGB video frames could reveal more insights into gait abnormalities. However, accurate detection of heel strike (HS) and toe-off (TO) events is crucial to compute interpretable gait parameters. In this work, we present an experimental platform to study the timing of gait events using a new wearable foot pressure sensor (ActiSense System, IEE S.A., Luxembourg), and Google’s open-source ML solution MediaPipe Pose. For this purpose, two StereoPi systems were built to capture stereoscopic videos and images in real time. MediaPipe Pose was applied to the synchronized StereoPi cameras, and two algorithms (ALs) were developed to detect HS and TO events for gait and analysis. Preliminary results from a healthy subject walking on a treadmill show a mean relative deviation across all time spans of less than 4% for the ActiSense device and less than 16% for AL2 (33% for AL1) employing MediaPipe Pose on StereoPi videos. Finally, this work offers a platform for the development of sensor- and video-based ALs to automatically identify the timing of gait events in healthy individuals and those with gait disorders.
Abstract: This paper is about detecting the difference between fully-random and semi-random shuffleing data sets, with the use of unsupervised learning algorithms. Because of the limits of the k-means algorithm alone, a recurrent autoencoder is used for feature extraction to improve the results of k-means. In the next step the autoencoder alone is used for clustering.
Introduction: In the last years, machine learning has been used more and more in different areas and it is also appropriate for for pattern recognition in data. Random data is characterized through the missing of defined patterns. Permutations without repetitions have the highest amount of entropy for a sequence of its length, which is similar to random data according to Andrei Kolmogorov, who states that random data have the highest amount of information and can’t be compressed. Therefore, this paper analyses the difference between random permutations and good shuffled permutations, which have some remaining patterns left. This is done via a recurrent autoencoder.
In this paper, the radio frequency (RF) behavior of mechanically stressed coaxial and for the first time also twisted-pair transmission lines is investigated over their service life. The main goal is to enable predictive maintenance for cables in moving applications and avoid preventive replacement. This also reduces the use of high-cost resources. For this purpose, stranded and solid-core variants of coaxial and twisted-pair type cables are mechanically loaded on the two-pulley apparatus according to EN 50396. Their RF transmission (S21) behavior is measured using a vector network analyzer and presented over bending cycles. For the first time, the phase response of mechanically loaded transmission lines is evaluated with respect to their service life. Two significant causes for the increasing attenuation and altered phase response are identified: breakage in foil screen and increasing surface roughness on the copper conductors. The identified causes are supported with literature evidence. Through measurements and theoretical calculations, it is proven that the phase is much more suitable for an assessment of the remaining service life than the amplitude. The findings can be used to implement a cable monitoring system in industrial environments which monitors the lines in-situ and reminds the user to replace them, whenever a certain wear-level is reached.
Dielectric properties of unidirectional and biaxial flax/epoxy composites at frequencies up to 1 GHz
(2023)
The relative permittivity of flax/epoxy composites in unidirectional and biaxial orientations was mapped in the frequency range of 1 kHz to 200 kHz, and for the first time in the range of 1 MHz to 1 GHz. In addition, permittivity was investigated for the first time in the temperature range between − 20 °C and 50 °C. These composites, produced using the vacuum infusion process, are increasingly used for sustainable and lightweight structural components in the automotive industry. The relative permittivity was determined using a self-developed plate capacitor with an LCR bridge and an impedance analyzer. An examination of the microstructure of the flax/epoxy composites shows that the fibers are disordered in the composite, resulting in local variations in fiber volume fraction. Furthermore, it was shown that the matrix also infiltrates into the fiber itself, resulting in an increase of the matrix fraction. It was found that unidirectional fabrics had a higher relative permittivity than biaxial fabrics, due to a higher fiber volume fraction and lower proportion of epoxy. The results suggest that it is the fiber volume fraction, rather than the manufacturing process and fiber orientation, that primarily determines the relative permittivity. It was also found that the permittivity continues to decrease below room temperature and thus behaves in a manner typical of the material in this temperature range as well.