Publikationen an der Fakultät für Informatik und Automatisierung ab 2015

Anzahl der Treffer: 1907
Erstellt: Thu, 28 Mar 2024 23:11:04 +0100 in 0.0515 sec


Willkomm, Julian; Wulff, Kai; Reger, Johann
Quantitative robustness analysis of model following control for nonlinear systems subject to model uncertainties. - In: IFAC-PapersOnLine, ISSN 2405-8963, Bd. 54 (2021), 14, S. 167-172

We investigate a model following control (MFC) design for nonlinear minimumphase systems subject to model uncertainties. The model following control architecture is a two degrees-of-freedom structure consisting of two control loops. The model control loop (MCL) includes a nominal model of the process. The design of the process control loop (PCL) is based on the error system resulting from the nominal design and the actual process. Both control loops are designed using (partial) feedback linearisation. We analyse the robustness in view of the norm of the uncertainty and the region of attraction compared to a single-loop (partial) feedback linearisation control. It turns out that the proposed approach is able to stabilize significantly larger uncertainties, shows better tracking performance, and exhibits a larger region of attraction (based on a quadratic Lyapunov function).



https://doi.org/10.1016/j.ifacol.2021.10.347
Steinmetz, Nadine; Sattler, Kai-Uwe
What is in the KGQA benchmark datasets? Survey on challenges in datasets for question answering on knowledge graphs. - In: Journal on data semantics, ISSN 1861-2040, Bd. 10 (2021), 3/4, S. 241-265

Question Answering based on Knowledge Graphs (KGQA) still faces difficult challenges when transforming natural language (NL) to SPARQL queries. Simple questions only referring to one triple are answerable by most QA systems, but more complex questions requiring complex queries containing subqueries or several functions are still a tough challenge within this field of research. Evaluation results of QA systems therefore also might depend on the benchmark dataset the system has been tested on. For the purpose to give an overview and reveal specific characteristics, we examined currently available KGQA datasets regarding several challenging aspects. This paper presents a detailed look into the datasets and compares them in terms of challenges a KGQA system is facing.



https://doi.org/10.1007/s13740-021-00128-9
Ravi Kumar, Varun; Yogamani, Senthil; Milz, Stefan; Mäder, Patrick
FisheyeDistanceNet++: self-supervised fisheye distance estimation with self-attention, robust loss function and camera view generalization. - In: Electronic imaging, ISSN 2470-1173, Bd. 33 (2021), 17, art00011, S. 181-1-181-10

FisheyeDistanceNet [1] proposed a self-supervised monocular depth estimation method for fisheye cameras with a large field of view (> 180˚). To achieve scale-invariant depth estimation, FisheyeDistanceNet supervises depth map predictions over multiple scales during training. To overcome this bottleneck, we incorporate self-attention layers and robust loss function [2] to FisheyeDistanceNet. A general adaptive robust loss function helps obtain sharp depth maps without a need to train over multiple scales and allows us to learn hyperparameters in loss function to aid in better optimization in terms of convergence speed and accuracy. We also ablate the importance of Instance Normalization over Batch Normalization in the network architecture. Finally, we generalize the network to be invariant to camera views by training multiple perspectives using front, rear, and side cameras. Proposed algorithm improvements, FisheyeDistanceNet++, result in 30% relative improvement in RMSE while reducing the training time by 25% on the WoodScape dataset. We also obtain state-of-the-art results on the KITTI dataset, in comparison to other self-supervised monocular methods.



https://doi.org/10.2352/ISSN.2470-1173.2021.17.AVM-181
Krauß, Benedikt; Link, Dietmar; Stodtmeister, Richard; Nagel, Edgar; Vilser, Walthard; Klee, Sascha
Modulation of human intraocular pressure using a pneumatic system. - In: Translational Vision Science & Technology, ISSN 2164-2591, Bd. 10 (2021), 14, 4, S. 1-9

https://doi.org/10.1167/tvst.10.14.4
Ebner, Christian; Gorelik, Kirill; Zimmermann, Armin
Model-based design space exploration for fail-operational mechatronic systems. - In: 2021 IEEE International Symposium on Systems Engineering (ISSE), (2021), insges. 8 S.

The increasing level of automation in safety-critical applications such as automated driving leads to additional requirements for functional availability of mechatronic systems. It is thus important to consider fail-operability and functional safety in order to optimize the performance and cost of the overall system in early design phases. The evaluation of feasible designs requires also the analysis of possible malfunctions and robustness of appropriate safety mechanisms in erroneous system states. This includes the examination of not only single faults but also of the sequence and impact of possible fault combinations in dynamic systems. The work presented in this paper proposes a methodology and a modeling approach suitable for the design exploration of mechatronic systems under consideration of functional safety. It enables to automatically generate feasible design variants by varying the functional system architecture at different abstraction levels and by mapping the functions to a set of hardware components. Furthermore, it combines logical and behavioral modeling with the goal to automatically evaluate the impact of component failures for various design variants under consideration of system dynamics and possible reconfigurations. A stochastic process of each design variant is set up automatically to estimate relevant safety metrics. The applicability and benefits of the proposed model-based design exploration of fail-operational mechatronic systems are demonstrated on exemplary drivetrain variants by investigating multiple safety mechanisms at different abstraction levels.



https://doi.org/10.1109/ISSE51541.2021.9582505
Blum, Maren-Christina; Klee, Sascha
Effects of ocular direct current stimulation on oscillatory potentials. - In: Der Ophthalmologe, ISSN 1433-0423, Volume 118 (2021), Suppl. 3, NM19-06, Seite 234

https://doi.org/10.1007/s00347-021-01465-7
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/).



Gast, Richard; Knösche, Thomas R.; Schmidt, Helmut
Mean-field approximations of networks of spiking neurons with short-term synaptic plasticity. - In: Physical review, ISSN 2470-0053, Bd. 104 (2021), 4, 044310, insges. 15 S.

Low-dimensional descriptions of spiking neural network dynamics are an effective tool for bridging different scales of organization of brain structure and function. Recent advances in deriving mean-field descriptions for networks of coupled oscillators have sparked the development of a new generation of neural mass models. Of notable interest are mean-field descriptions of all-to-all coupled quadratic integrate-and-fire (QIF) neurons, which have already seen numerous extensions and applications. These extensions include different forms of short-term adaptation considered to play an important role in generating and sustaining dynamic regimes of interest in the brain. It is an open question, however, whether the incorporation of presynaptic forms of synaptic plasticity driven by single neuron activity would still permit the derivation of mean-field equations using the same method. Here we discuss this problem using an established model of short-term synaptic plasticity at the single neuron level, for which we present two different approaches for the derivation of the mean-field equations. We compare these models with a recently proposed mean-field approximation that assumes stochastic spike timings. In general, the latter fails to accurately reproduce the macroscopic activity in networks of deterministic QIF neurons with distributed parameters. We show that the mean-field models we propose provide a more accurate description of the network dynamics, although they are mathematically more involved. Using bifurcation analysis, we find that QIF networks with presynaptic short-term plasticity can express regimes of periodic bursting activity as well as bistable regimes. Together, we provide novel insight into the macroscopic effects of short-term synaptic plasticity in spiking neural networks, as well as two different mean-field descriptions for future investigations of such networks.



https://doi.org/10.1103/PhysRevE.104.044310
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: 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition, (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