Anzahl der Treffer: 466
Erstellt: Mon, 24 Jan 2022 14:39:36 +0100 in 0.0687 sec


Döring, Nicola; Mikhailova, Veronika; Brandenburg, Karlheinz; Broll, Wolfgang; Groß, Horst-Michael; Werner, Stephan; Raake, Alexander;
Saying "Hi" to grandma in nine different ways : established and innovative communication media in the grandparent-grandchild relationship. - In: Technology, Mind, and Behavior, ISSN 2689-0208, (2021), insges. 1 S.

https://tmb.apaopen.org/pub/8je5p43m/release/1?readingCollection=b5d405be
Keßler, Jens;
Planungsmethoden für eine sozial akzeptable Navigation von Assistenzrobotern. - Ilmenau : Universitätsbibliothek, 2021. - 1 Online-Ressource (xiii, 195 Seiten)
Technische Universität Ilmenau, Dissertation 2021

In den vergangenen Jahren wurde im Bereich der mobilen Assistenzrobotik auch das Themenfeld der sozial-akzeptablen Navigation weiterentwickelt. Dabei kommt zu den vorhandenen Herausforderungen der Navigation nun auch der Faktor Mensch hinzu. Aus dem Themenbereich der sozial-akzeptablen Navigation widmet sich die vorliegende Arbeit zwei Aspekten, wobei der Fokus immer auf einem mobilen Roboter liegt, welcher sich im häuslichen Umfeld bewegt. Der erste Aspekt der Arbeit behandelt, wie ein Roboter sich in einer häuslichen Umgebung bewegen soll. Dabei ist von Interesse, welche Pfade der mobile Roboter im Beisein eines Bewohners plant. Die wissenschaftliche Leistung der Arbeit ist, vorhandene Planungsansätze auf Raum und Zeit zu erweitern und es so zu ermöglichen, die Bewegung einer Person in die Planungsphase zu integrieren. Dazu werden simple Bewegungsprädiktionsmethoden untersucht und es wird eine mathematische Beschreibung entwickelt, die Bewegungsvorhersagen von Personen in die Planung mit einbezieht. Durch die so weiterentwickelten Planungsverfahren werden zwei Szenarien experimentell näher untersucht: A) frontales Heranfahren an eine Person, B) frühzeitiges Ausweichen einer entgegenkommenden Person. Im zweiten Aspekt der vorliegenden Arbeit wird auf ein bisher nur sehr wenig beachtetes Problem der mobilen Robotik eingegangen: wie findet ein mobiler Roboter für seine jeweilige - häufig nur abstrakt formulierbare - Aufgabe sinnvolle Zielpunkte in seiner Einsatzumgebung, ohne dass diese statisch vorgegeben sein müssen? Als Lösungsansatz wird in dieser Arbeit ein Verfahren vorgeschlagen, welches eine abstrakte Aufgabe auf mehrere Kriterien abbildet und diese nach Extrempunkten untersucht. Diese Extrempunkte ergeben einzeln oder in Kombination mögliche Zielpunkte des Roboters. Die wissenschaftlichen Beiträge sind, für dieses Verfahren zu untersuchen, welche mathematische Formulierung der Einzelkriterien sinnvoll ist, welche Ergebnisse eine Optimierung mit unabhängigen Einzelkriterien liefert (Pareto-Optimalität) bzw. welche Ergebnisse die Kombination der einzelnen Kriterien zu einer Gesamtfunktion erzielt werden können (Superpositionsprinzip). Durch den hier präsentierten Ansatz werden folgende zwei Szenarien für die häusliche Navigation experimentell untersucht: C) Beobachten einer Person, D) Finden einer Parkposition bei potentiellen Engstellen. Für alle Szenarien gilt: es soll insbesondere das Weltwissen des Roboters zur Lösungsfindung genutzt werden. Dies alles setzt eine Akquise von Umweltwissen durch den Roboter voraus. Hierzu werden in der Arbeit praktisch einsetzbare Verfahren vorgestellt, welche das nötige Umweltwissen zur Schätzung der Oberkörperpose, Beobachtbarkeit und Bewegungsprädiktion ermitteln können.



https://doi.org/10.22032/dbt.50287
Balada, Christoph; Eisenbach, Markus; Groß, Horst-Michael;
Evaluation of transfer learning for visual road condition assessment. - In: Artificial neural networks and machine learning - ICANN 2021, (2021), S. 540-551

Through deep learning, major advances have been made in the field of visual road condition assessment in recent years. However, many approaches train from scratch and avoid transfer learning due to the different nature of road surface data and the ImageNet dataset, which is commonly used for pre-training neural networks for visual recognition. We show that, despite the huge differences in the data, transfer learning outperforms training from scratch in terms of generalization. In extensive experiments, we explore the underlying cause by examining various transfer learning effects. For our experiments, we are incorporating seven known architectures. Therefore, this is the first comprehensive study of transfer learning in the field of visual road condition assessment.



Aganian, Dustin; Eisenbach, Markus; Wagner, Joachim; Seichter, Daniel; Groß, Horst-Michael;
Revisiting loss functions for person re-identification. - In: Artificial neural networks and machine learning - ICANN 2021, (2021), S. 30-42

Appearance-based person re-identification is very challenging, i.a. due to changing illumination, image distortion, and differences in viewpoint. Therefore, it is crucial to learn an expressive feature embedding that compensates for changing environmental conditions. There are many loss functions available to achieve this goal. However, it is hard to judge which one is the best. In related work, the experiments are only performed on the same datasets, but the use of different setups and different training techniques compromises the comparability. Therefore, we compare the most widely used and most promising loss functions under identical conditions on three different setups. We provide insights into why some of the loss functions work better than others and what additional benefits they provide. We further propose sequential training as an additional training trick that improves the performance of most loss functions. In our conclusion, we provide guidance for future usage an d research regarding loss functions for appearance-based person re-identification. Source code is available (Source code: https://www.tu-ilmenau.de/neurob/data-sets-code/re-id-loss/).



Fischer, Kai; Simon, Martin; Ölsner, Florian; Milz, Stefan; Groß, Horst-Michael; Mäder, Patrick;
StickyPillars: robust and efficient feature matching on point clouds using graph neural networks. - In: IEEE Xplore digital library, ISSN 2473-2001, (2021), S. 313-323

Robust point cloud registration in real-time is an important prerequisite for many mapping and localization algorithms. Traditional methods like ICP tend to fail without good initialization, insufficient overlap or in the presence of dynamic objects. Modern deep learning based registration approaches present much better results, but suffer from a heavy runtime. We overcome these drawbacks by introducing StickyPillars, a fast, accurate and extremely robust deep middle-end 3D feature matching method on point clouds. It uses graph neural networks and performs context aggregation on sparse 3D key-points with the aid of transformer based multi-head self and cross-attention. The network output is used as the cost for an optimal transport problem whose solution yields the final matching probabilities. The system does not rely on hand crafted feature descriptors or heuristic matching strategies. We present state-of-art art accuracy results on the registration problem demonstrated on the KITTI dataset while being four times faster then leading deep methods. Furthermore, we integrate our matching system into a LiDAR odometry pipeline yielding most accurate results on the KITTI odometry dataset. Finally, we demonstrate robustness on KITTI odometry. Our method remains stable in accuracy where state-of-the-art procedures fail on frame drops and higher speeds.



https://doi.org/10.1109/CVPR46437.2021.00038
Stricker, Ronny; Aganian, Dustin; Sesselmann, Maximilian; Seichter, Daniel; Engelhardt, Marius; Spielhofer, Roland; Hahn, Matthias; Hautz, Astrid; Debes, Klaus; Groß, Horst-Michael;
Road surface segmentation - pixel-perfect distress and object detection for road assessment. - In: IEEE Xplore digital library, ISSN 2473-2001, (2021), S. 1789-1796

Visual road assessment, which is carried out by many countries, involves the evaluation of millions of surface images. This exhaustive task is usually done manually and therefore is costly in terms of time and prone to failure. Different methods for automatic distress detection have been presented in the literature recently. However, most of the approaches are focused on crack detection only. This paper focuses on detecting multiple distress types and object classes on asphalt roads, aiming to fully automate distress detection on road surfaces in Austria, Switzerland, and Germany using image segmentation with neural networks. The paper introduces a distress and object catalog developed by experts of the involved countries that guarantees convertibility into federal distress catalogs. We evaluate the performance gain of different neural network architectures and advanced training techniques by conducting extensive experiments.



https://doi.org/10.1109/CASE49439.2021.9551591
Seichter, Daniel; Köhler, Mona; Lewandowski, Benjamin; Wengefeld, Tim; Groß, Horst-Michael;
Efficient RGB-D semantic segmentation for indoor scene analysis. - In: IEEE Xplore digital library, ISSN 2473-2001, (2021), S. 13525-13531

Analyzing scenes thoroughly is crucial for mobile robots acting in different environments. Semantic segmentation can enhance various subsequent tasks, such as (semantically assisted) person perception, (semantic) free space detection, (semantic) mapping, and (semantic) navigation. In this paper, we propose an efficient and robust RGB-D segmentation approach that can be optimized to a high degree using NVIDIA TensorRT and, thus, is well suited as a common initial processing step in a complex system for scene analysis on mobile robots. We show that RGB-D segmentation is superior to processing RGB images solely and that it can still be performed in real time if the network architecture is carefully designed. We evaluate our proposed Efficient Scene Analysis Network (ESANet) on the common indoor datasets NYUv2 and SUNRGB-D and show that we reach state-of-the-art performance while enabling faster inference. Furthermore, our evaluation on the outdoor dataset Cityscapes shows that our approach is suitable for other areas of application as well. Finally, instead of presenting benchmark results only, we also show qualitative results in one of our indoor application scenarios.



https://doi.org/10.1109/ICRA48506.2021.9561675
Müller, Steffen; Stephan, Benedict; Groß, Horst-Michael;
MDP-based motion planning for grasping in dynamic scenarios. - In: IEEE Xplore digital library, ISSN 2473-2001, (2021), insges. 8 S.

Path planning for robotic manipulation is a well understood topic as long as the execution of the plan takes place in a static scene. Unfortunately, for applications involving human interaction partners a dynamic obstacle configuration has to be considered. Furthermore, if it comes to grasping objects from a human hand, there is not a single goal position and the optimal grasping configuration may change during the execution of the grasp movement. This makes a continuous re-planning in a loop necessary. Besides efficiency and security concerns, such periodic planning raises the additional requirement of consistency, which is hard to achieve with traditional sampling based planners. We present an online capable planner for continuous control of a robotic grasp task. The planner additionally is able to resolve multiple possible grasp poses and additional goal functions by applying an MDP-like optimization of future rewards. Furthermore, we present a heuristic for setting edges in a probabilistic roadmap graph that improves the connectivity and keeps edge count low.



https://doi.org/10.1109/ECMR50962.2021.9568813
Scheidig, Andrea; Schütz, Benjamin; Trinh, Thanh Quang; Vorndran, Alexander; Mayfarth, Anke; Sternitzke, Christian; Röhner, Eric; Groß, Horst-Michael;
Robot-assisted gait self-training: assessing the level achieved. - In: Sensors, ISSN 1424-8220, Bd. 21 (2021), 18, 6213, insges. 15 S.

This paper presents the technological status of robot-assisted gait self-training under real clinical environment conditions. A successful rehabilitation after surgery in hip endoprosthetics comprises self-training of the lessons taught by physiotherapists. While doing this, immediate feedback to the patient about deviations from the expected physiological gait pattern during training is important. Hence, the Socially Assistive Robot (SAR) developed for this type of training employs task-specific, user-centered navigation and autonomous, real-time gait feature classification techniques to enrich the self-training through companionship and timely corrective feedback. The evaluation of the system took place during user tests in a hospital from the point of view of technical benchmarking, considering the therapists' and patients' point of view with regard to training motivation and from the point of view of initial findings on medical efficacy as a prerequisite from an economic perspective. In this paper, the following research questions were primarily considered: Does the level of technology achieved enable autonomous use in everyday clinical practice? Has the gait pattern of patients who used additional robot-assisted gait self-training for several days been changed or improved compared to patients without this training? How does the use of a SAR-based self-training robot affect the motivation of the patients?



https://doi.org/10.3390/s21186213
Zhang, Yan; Müller, Steffen; Stephan, Benedict; Groß, Horst-Michael; Notni, Gunther;
Point cloud hand-object segmentation using multimodal imaging with thermal and color data for safe robotic object handover. - In: Sensors, ISSN 1424-8220, Bd. 21 (2021), 16, 5676, insges. 16 S.

https://doi.org/10.3390/s21165676