149544 research outputs found
Sort by
Analyzing Spatial Eye-Tracking Data of Teleoperators to Assess Workload in Automotive Fleet Management
No full textWith the rise of automated vehicles, teleoperation plays a key role in ensuring safe and efficient drives, particularly in partially automated systems where human operators provide high-level commands. This research focuses on understanding how mental states, specifically cognitive workload, impact ocular behavior during teleoperation tasks using a visual interface. Data from a previously-conducted user study is analysed, where task difficulty and frequency were manipulated
in a naturalistic setting. The data-driven approach stresses the use of spatial area-of-interest metrics and its evaluation in providing mental state insights. The results display workload to be a significant factor in influencing the selected AoI metrics, including fixation duration, fixation frequency, time-to-first fixation, visit frequency, dwell time, and stationary entropy. Moreover, the findings partially support that a high workload induces a tunneling effect, although modulated by task-related and interface factors. The influence of difficulty and frequency independently act on the AoI metrics, with the former eliciting a broader effect. The study also demonstrates that workload can be predicted using machine learning models, with binary workload and frequency predictions achieving high recall rates (above 85%), and difficulty prediction reaching a maximum of 75%. A 4-class workload classification has been attempted, too, with the best predictive model reaching a recall of 49%. These outcomes highlight the potential of AoI metrics for real-time workload assessment and detection in teleoperation, paving the way for intelligent interfaces that adapt to operator mental states
Effects of head-down tilt bed rest on the human gut microbiome
No full textA healthy gut microbiome, particularly including probiotic
microorganisms that metabolize complex carbohydrates and
produce short-chain fatty acids (SCFAs), plays a critical role
in maintaining intestinal homeostasis and is generally
associated with improved health outcomes. Stressors, both
physical and psychological, result in the release of stress-related neurochemicals to which gastrointestinal bacteria are
particularly sensitive. This can lead to gastrointestinal
distress and inflammation, and a dysbiotic gut microbiome
166
can lead to anxiety and depression. During spaceflight,
astronauts are exposed to numerous stressors such as
microgravity, radiation, sleep disruption, and isolation, which
can lead to challenges such as muscle wasting, immune
dysfunction, and neuro-ocular syndrome. However, the
impact of space conditions on the human microbiome is not
well understood. Since the availability of samples from
astronauts during spaceflight is limited, bed rest studies on
Earth offer a great opportunity to study the simulated effects
of spaceflight-induced changes in the human body in more
detail and with a larger number of subjects. In particular,
head-down tilt (HDT) bed rest studies provide valuable
insights into the effects of microgravity on the human body.
The DLR Institute of Aerospace Medicine has conducted
several HDT studies, including "AGBRESA", "SANS-CM"
and "SMC", to evaluate the influence of different
countermeasures on physiological changes. Subjects
underwent 30 to 60 days of HDT with countermeasures such
as exercise and centrifugation. To assess microbiome
changes, stool samples were collected from subjects in
these bedrest studies at baseline, during HDT, and after
recovery and analyzed using next-generation sequencing of
the 16S rRNA gene. Results show high stability of gut
microbial communities but interindividual variability in gut
microbiome composition. Improving this understanding of the
interactions between the gut microbiome and individual host
responses to prolonged bed rest, might provide valuable
insights to help mitigate the risks associated with long-duration space missions
Adaptation of the EnMAP Level 2A land product to water
Full text linkEnMAP data are available as top-of-atmosphere Level-1 product in radiance units, and as bottom-of-atmosphere Level-2 product in reflectance units. The conversion from Level-1 to Level-2 is performed using two different atmospheric processors: PACO, which is applied to both land and water pixels, and MIP, which is used exclusively for water pixels. Both correct for adjacency effects, which can be quite large over dark water. The PACO product has been validated over land (de los Reyes et al., 2020) and used extensively for terrestrial applications, and correspondingly the MIP product was validated for water (Soppa et al., 2024) and used frequently for aquatic applications. However, the PACO product over water is rarely used because it is primarily designed for land and does not account for specular reflections; thus, PACO is not validated for water. A novel tool has been developed for the open-source software WASI for adapting land atmospheric correction to water by correcting the reflections at the water surface. It has been applied to PACO products from water, and the results are compared with the MIP product and validated using in-situ measurements
Tele-Tf - Menschzentrierte Gestaltung eines neuen Berufsbilds und Arbeitsplatzes. Remote train operator - Human-centred design of a new job profile and workplace
Full text linkEin Thema der aktuellen Forschung und Entwicklung im Eisenbahnsektor ist die hochautomatisierte Steuerung von Zügen (Automatic Train Operation, ATO). Ziele sind hierbei u.a. die Erhöhung der betrieblichen Stabilität und Kapazität, die Verbesserung der Sicherheit und die Senkung von Personalkosten. Menschliches Eingreifen wird jedoch auch bei ATO weiterhin erforderlich sein, um mit unvorhergesehenen Ereignissen umzugehen. Gemäß der aktuellen Vorstellung wird dafür ein teleoperierender Tf (Tele-Tf) von einer Fernsteuerungszentrale aus verantwortlich sein. Wenn das ATO-System jedoch gut funktioniert, und Tele-Tf das System ansonsten nur passiv überwachen müssten, besteht die Gefahr der Monotonie mit negativen Auswirkungen auf Leistung, Fähigkeiten und Fertigkeiten, Arbeitszufriedenheit und Gesundheit. Daher ist es wichtig, den Arbeitsplatz menschzentriert zu entwickeln. Im Vortrag werden aktuelle Forschungsergebnisse des DLR zur Erarbeitung eines auf den Menschen ausgerichtetes Konzepts für das neue Berufsbild und den Arbeitsplatz vorgestellt und weitere Forschungsbedarfe spezifiziert.
One topic of current research and development in the railway sector is the highly automated control of trains (Automatic Train Operation, ATO). The aims here include increasing operational stability and capacity, improving safety and reducing labour costs. However, human intervention will still be necessary even with ATO in order to deal with unforeseen events. According to current conceptions, this will be the responsibility of a remote train operator (RO) in a remote control centre. However, if the ATO system works well and RO would otherwise only have to passively monitor the system, there is a risk of monotony with negative effects on performance, skills and abilities, job satisfaction and health. It is therefore important to develop the workplace in a human-centred way. The talk presents current DLR research results on the development of a human-centred concept for the new job profile and workplace and specifies further research requirements
Presenting BACARDI's Sub-Catalogue of drifting Objects close to the Geostationary Ring
Full text linkSince the beginning of its operational phase in 2017, observations with the Small Aperture Robotic Telescope Network (SMARTnet) have primarily focused on observations of Resident Space Objects in the geostationary ring, including geosynchronous objects and Resident
Space Objects in so-called graveyard orbits. Those latter Resident Space Objects are uncontrolled and are potential candidates for fragmentation events and pose a threat for the geostationary ring as a whole.
After two network extensions with telescope stations in Australia (2019) and Chile (2024), almost the entire geostationary ring is now within the observational reach of the SMARTnet stations. Drifters may be observable throughout their drifting period, i. e., the time after which a Resident Space Object is again above the same longitude. We therefore launched a study to investigate the number of drifters in the Backbone Catalouge of Relational Debris Information (BACARDI) database.
In terms of orbital regions as defined by the European Space Agency, we focused on Resident Space Objects that spend some time in the geostationary ring, deliberately excluding Resident Space Objects of the Geostationary Orbit themselves. One particular focus was put on orbits with a perigee below the geostationary ring, an apogee above it, and an inclination lower than 15°. These Resident Space Objects are crossing the geostationary ring regularly and pose a threat to active satellites. Tracklets were selected that represent Resident Space Objects in these regions. The filter process happened against the publicly available Two Line Element set catalogue.
Additionally, we present a number of tracklets from the BACARDI database that are associated with Resident Space Objects in those regions but do not have a counterpart in the Two Line Elements catalogue. These Resident Space Objects shall be observed from every available telescope station in order to determine orbit updates.
It is imperative not to lose track of them again, for they still pose a threat to the space environment
Online 3D-Objekterkennung mit Tiefenschätzung und Unsicherheitsbewusstsein durch Stereo Vision beim hochautomatisierten Fahren
Full text linkDie Wahrnehmung der dreidimensionalen Umwelt bildet die Grundlage für den Betrieb autonomer Fahrzeuge. Besonders entscheidend ist dabei die 3D-Objekterkennung, um Verkehrssituationen genau einschätzen zu können. Für die 3D-Objekterkennung gibt es verschiedene Methoden wie z. B. das laserbasierte Light detection and ranging (Lidar) oder die kamerabasierte Stereo Vision. Stereo Vision hat den Vorteil, dass weniger Daten als bei Lidar anfallen, was weniger Datenverarbeitung notwendig macht und zudem kostengünstiger ist. Bei der Stereo Vision wird das räumliche Sehen des Menschen, mit zwei Augen, technisch umgesetzt. Dabei wird aus minimalen Unterschieden
in dem linkem beziehungsweise rechtem Bild, die Tiefe berechnet. In Kombination mit einer
panoptischen Segmentierung, bei der alle Pixel in einem Bild klassifiziert und dabei verschiedene
Objekte der gleichen Klasse berücksichtigt werden, können 3D-Objekte berechnet werden. Mit
der Verwendung des neuronalen Netzes Evidential Panoptic Segmentation Network (EvPSNet)
wird versucht die Unsicherheit, die bei der Segmentierung auftritt, zu berücksichtigen. Im Zuge
der Umsetzung wird ein Konzept für eine online Computer Vision (CV)-Pipeline dargelegt, in
Verbindung mit der Simulationssoftware für autonomes Fahren Car Learning to Act (CARLA), und
dem Robot Operating System (ROS) 2 umgesetzt werden soll. Im Rahmen der Evaluation wurde die
Tiefenschätzung und die Segmentierung im Vergleich zu der Ground Truth aus CARLA verglichen.
Dabei sind für beide Aspekte gute Ergebnisse ausgewertet worden, die die Zuverlässigkeit des
Ansatzes in den evaluierten Szenarien zeig
Agile and optimized methodology for VENQS An automated, cross-platform, modules versioning managment system
Full text linkEntwicklung und Exploration einer Prozessautomation für ein 10-kW-Brennstoffzellentestsystem
No full textIn dieser Masterarbeit wird eine Automation in eine bestehende Steuerungsumgebung einer Polymer-Elektrolyt-Membran-Brennstoffzelle (PEMFC) entwickelt, implementiert und
analysiert. Hierfür werden Deterministische Endliche Automaten (DEA) untersucht, um ein
besseres Verständnis für Automatisierungsprozesse zu erlangen. Die Automatisierung gliedert sich in zwei Bereiche: die externe und die interne Automation. Die externe Automation
beschäftigt sich mit der Automatisierungsumgebung, insbesondere mit der Struktur und
dem Aufbau der Daten, die das System von außen empfängt. Die interne Automation verarbeitet diese Daten in LabVIEW und wandelt sie in eine Trajektorie um, die anschließend
von der Steuerungsumgebung angesteuert wird. Im Rahmen der Arbeit wurde die Totzeit
des Systems analysiert und reduziert. Abschließend wurde eine automatisierte Untersuchung
einer Rekonditionierungsprozedur durchgeführt, um Aussagen über die Reproduzierbarkeit
der Messungen sowie des Systemverhaltens treffen zu können