DaRUS (University of Stuttgart)
Not a member yet
2037 research outputs found
Sort by
Supplementary Material for: Safe-by-Design Approximate Nonlinear Model Predictive Control with Realtime Feasibility
No full textThis repository contains supplementary material for the paper "Safe-by-Design Approximate Nonlinear Model Predictive Control with Realtime Feasibility" submitted to IEEE Transactions on Automatic Control. It contains additional information regarding the implementation and the MATLAB source code to generate the numerical results. See the README file for more information on how to use the source code and the provided data.Important information:
The TAC-Inf-MPC.zip is a copy of the GitHub repository TAC25-Inf-MPC.
The usage of the .zip version is similar to the github version, with the difference that the CaSoS submodule is already available as a copy. This version was used to generate the results in the paper, together with the data provided in this dataverse.
For the latest version, e.g., from CaSoS or to get updates, we recommended to use the GitHub version
Replication Data for: Deep learning based computational imaging and optical metrology
No full textA new study explores using deep learning to speed up optical scatterometry, a key quality control technique in computer chip manufacturing. This method could replace older, slower, and data-heavy processes.
Manufacturing computer chips involves creating tiny, complex patterns on silicon wafers. The accuracy of these microscopic features is vital for the chips' performance and reliability. Optical scatterometry, which uses light scattering to measure these features, is essential for this quality control. However, traditional scatterometry methods have drawbacks: iterative fitting methods are slow, and library search methods need massive amounts of data.
To overcome these issues, this project investigated the use of deep learning, a powerful form of artificial intelligence. Specifically, a neural network called ResNet was trained with simulated optical measurements that mimic how light scatters off a chip's microscopic features. By analyzing this scattered light, the ResNet learned to predict key parameters of the features' shape and size, such as their width, height, and sidewall angles.
The research compared different architectures and prediction strategies:
UniNet: A single network that predicts all parameters at once.
MonoNet: Separate networks for each parameter, predicting them one by one.
ExpertNet: A combination of networks that predicts groups of related parameters.
The results showed that MonoNet performed the best. This approach successfully decoupled the parameters and achieved high accuracy even with smaller datasets.
Since real-world measurements are always affected by noise and uncertainties, the study also examined how noise impacts the deep learning models. By adding simulated noise to the training data, the researchers assessed the robustness of the different network architectures. These findings offer valuable insights for developing more reliable and accurate measurement techniques, even in the presence of real-world imperfections
Numerical Data of the Asymmetric Inner Film Flow during Oblique Droplet Impact onto a Wall Film
No full textSupplementary material to a submitted manuscript [full citation link follows when it is published], referred to as related publication in the following.
Direct Numerical Simulation of an oblique droplet impact onto a thin wall film (Isopropanol, air) performed with the ITLR in-house program package Free Surface 3D (FS3D). The code solves the incompressible Navier-Stokes equations in a one-field formulation with a Volume of Fluid (VoF) approach, see related publication. The simulation was performed in three dimensions (1024*512*512), cropped to 512*103*293.
The y- and z-axes are interchanged compared to the related publication.
The data is saved in hdf5 files. For the here published timestep, there is one file storing the VoF-variable, which is a scalar field with values between 0 and 1 indicating the phases being ambient gas (0) or drop liquid (1), (funs0007.hdf). Additionally, there is another file for storing the velocity vector field (velv0007.hdf). In total, there are 7 different cases (varying impact conditions) published, each for the evaluated timestep of tau=1.5 of the related publication.
The reference impact conditions are, details can be found in the related publication: D=1.5mm, U=2.868m/s, h_f0=0.3mm, alpha=70°, delta=0.2, We=450, Oh=0.014.
1: Reference Case
2: alpha = 60°
3: alpha = 50°
4: We = 0.5*We_ref
5: Oh = 0.3*Oh_ref
6: delta = 2*delta_ref
7: free-slip condition
The simulation was performed as part of the GRK 2160 within the subproject SP-C2
Replication Data for 'Measurement-free quantum error correction optimized for biased noise'
No full textThis is the data used for the publication 'Measurement-free quantum error correction optimized for biased noise'. The collection includes the scripts and data to reproduce the figures. The directories 'simulation_results_measurement_based' and 'simulation_results_measurement_free' contain the data, which are needed to reproduce the figures of the paper. The Jupiter notebook 'plots_for_paper.ipynb' was used for generating the figures. The README-file contains information on the required python-packages. The data were produced using Qiskit (v1.2.0) and the Qiskit QRyd Provider (v1.0.6)
Replication Data for: A Modular Approach for Mesh-Particle Coupling
No full textThis dataset contains software as well as setup and result files to reproduce the numerical experiments in Chapter 7 of my dissertation titled "Flexible and Efficient Data Mapping for Simulation of Coupled Problems". For further instructions on how to run the experiments see the README.md of the dataset
Atmospyre v1.0.0
No full textRelease v1.0.0 of AtmosPyre.
AtmosPyre is a flexible library for atmospheric measurements. Monitor CO₂, Radon, and more with an intuitive Python API
Experimental Data for Fault Diagnosis in the Adaptive High-Rise D1244
No full text
General information:
This dataset is meant to serve as a benchmark problem for fault detection and isolation in dynamic systems. It contains preprocessed sensor data from the adaptive high-rise demonstrator building D1244, built in the scope of the CRC1244. Parts of the measurements have been artificially corrupted and labeled accordingly. Please note that although the measurements are stored in Matlab's .mat-format (Version 7.0), they can easily be processed using free software such as the SciPy library in Python.
Structure of the dataset:
train contains training data (only nominal)
validation contains validation data (nominal and faulty). Faulty samples were obtained by manipulating a single signal in a random nominal sample from the validation data.
test contains test data (nominal and faulty). Faulty samples were obtained by manipulating a single signal in a random nominal sample from the test data.
meta contains textual labels for all signals as well as additional information on the considered fault classes
File contents:
Each file contains the following data from 1200 timesteps (60 seconds sampled at 20 Hz):
t: time in seconds
u: actuator forces (obtained from pressure measurements) in newtons
y: relative elongations as well as bending curvatures of structural elements obtained from strain gauge measurements, and actuator displacements measured by position encoders
label: categorical label of the present fault class, where 0 denotes the nominal class and faults in the different signals are encoded according to their index in the list of fault types in meta/labels.mat
Faulty samples additionally include the corresponding nominal values for reference
u_true: actuator forces without faults
y_true: measured outputs without faults
Textual labels for all in- and output signals as well as all faults are given in the struct labels. Each sample's textual fault label is additionally contained in its filename (between the first and second underscore).
</p
Replication Data for "Cavity enhancement of V2 centers in 4H-SiC with a fiber-based Fabry-Pérot microcavity"
No full textCSV data sets to reproduce all plots in the paper.
The datasets are either 1D plots with the x-values in the first column and all plotted y-curves in the following columns or 2D plots. In the latter case, the data consists of a vector containing the values of the x- and y-axis (first row and column, respectively) as well as the 2D data.
Information on how the data was acquired and what it actually shows can be found in the related manuscript
NGS data related to Sogl et al. "Systematic analysis of specificities and flanking sequence preferences of bacterial DNA-(cytosine C5)-methyltransferases reveals mechanisms of enzyme- and sequence-specific DNA readout"
No full textAnalysis of flanking sequence preference with a randomized substrate and bioinformatic data
For analysis of the flanking sequence preference, substrate with different target sites sites in a 9 or 10 bp randomized sequence context were prepared as described (Dukatz, et al. 2020; Dukatz, et al. 2022). Substrate methylation reactions were performed with different enzyme concentrations in methylation buffers as indicated in the main manuscript . Methylation reactions were incubated at 37 °C for 30-60 min. The reactions were stopped by freezing in liquid N2, followed by 2 h digestion with proteinase K (NEB) at 42 °C and purification with NucleoSpin® Gel and PCR Clean-up Kit (Macherey-Nagel). Bisulfite conversion was performed as described in the standard protocol EZ DNA Methylation-Lightning™ Kit (Zymo Research). Samples were eluted with RNase free H2O. Library preparation was performed with two PCRs using variable primer pairs to introduce sample specific barcodes and indices for sample distinction and the sequencing reactions. Bioinformatic analysis of the NGS data was conducted as described (Dukatz, et al. 2020; Dukatz, et al. 2022). For determination of the methylation rates of all 256 NNCGNN sequences by one enzyme, methylation reactions of individual substrates were assumed to be independent and reaction velocities are of first order with respect to the substrate concentrations. The results of the individual reactions with different enzyme concentrations and incubation times were fitted to monoexponential reaction progress curves using variable virtual time values. Fitting was conducted with MatLab as described except that convergence was validated by serial fitting (Adam, et al. 2022).
References
Adam S, Bräcker J, Klingel V, Osteresch B, Radde NE, Brockmeyer J, Bashtrykov P, Jeltsch A. Flanking sequences influence the activity of TET1 and TET2 methylcytosine dioxygenases and affect genomic 5hmC patterns. Communications Biology 5, 92 (2022)
Dukatz M, Dittrich M, Stahl E, Adam S, de Mendoza A, Bashtrykov P, Jeltsch A. DNA methyltransferase DNMT3A forms interaction networks with the CpG site and flanking sequence elements for efficient methylation. J. Biol. Chem. 298(10), 102462 (2022)
Dukatz M, Adam S, Biswal M, Song J, Bashtrykov P, Jeltsch A. Complex DNA sequence readout mechanisms of the DNMT3B DNA methyltransferase. Nucleic Acids Res 48, 11495-11509 (2020)<br
Replication Data for: Simple Estimation Algorithm for Laser Tracker Localization in Industrial Robot Calibration
No full textThis paper presents a straightforward algorithm for locating laser trackers in the calibration tasks of industrial robots. Laser trackers are important for enhancing the positioning precision of industrial robots, which generally experience positioning inaccuracies. The algorithm meets the requirement for accurate kinematic models to guarantee the visibility of the laser target and facilitate precise calibration.
The approach employs singular value decomposition (SVD) along with two identification paths to assess the transformation from the laser tracker to the robot's reference frame. Experimental testing on a KUKA KR210-2 robot shows the method's precision, resulting in an average positioning error of 2.002 mm and a highest error of 5.040 mm in novel test positions. The method reduces dependence on stationary calibration fixtures, streamlines configuration, and readily adjusts to modifications in the laser tracker’s location.
This method serves as an economical and effective substitute for robot calibration and activities that need ongoing laser tracker measurements, including path tracking and quality assurance