Repository der Technischen Hochschule Ingolstadt
Not a member yet
4170 research outputs found
Sort by
Sensor Calibration and Data Analysis of the MuFoRa Dataset
Get PDFAutonomous driving sensors face significant challenges under adverse weather conditions such as fog and rain, which can seriously degrade their performance and reliability. Existing datasets often lack the reproducible and measurable data needed to adequately quantify these effects. To address this gap, a new multimodal dataset (MuFoRa) has been collected under controlled adverse weather conditions at the CARISSMA facility, using a stereo camera and two solid-state LiDAR sensors. This dataset is used to quantitatively assess sensor degradation by measuring the entropy for images and the number of inliers for point clouds on a spherical target. These metrics are used to evaluate the impact on performance under varying conditions of fog (5 to 150 m visibility) and rain (20 to 100 mm/h intensity) at different distances (5 to 50 m). Additionally, two calibration target detection approaches - Deep-learning and Hough-based - are evaluated to achieve accurate sensor alignment. The contributions include the introduction of a new dataset focused on fog and rain, the evaluation of sensor degradation, and an improved calibration approach. This dataset is intended to support the development of more robust sensor fusion and object detection algorithms for autonomous driving
AI-Driven Point Cloud Framework for Predicting Solder Joint Reliability using 3D FEA Data
Get PDFCrack propagation in solder joints remains a critical challenge affecting the thermo-mechanical reliability of electronic devices, emphasizing the need for optimized package and solder pad designs. Traditional Finite Element Analysis (FEA) methods for predicting solder joint lifespan rely heavily on manual post-processing, where high-risk regions for plastic strain accumulation are identified. However, these approaches often overlook intricate failure mechanisms, as they primarily average creep strain and correlate it with experimental lifetime data using the Coffin-Manson equation, limiting their predictive accuracy.
To overcome these limitations, this study introduces a novel AI-driven framework that automates 3D FEA post-processing for surface-mounted devices (SMDs) connected to printed circuit boards (PCBs). Unlike traditional methods, this framework leverages deep learning architectures—specifically, 3D Convolutional Neural Networks (CNNs) and PointNet—to extract complex spatial features directly from 3D FEA data, eliminating the need for manual interpretation. These learned features are then mapped to experimentally measured solder joint lifetimes through fully connected neural network layers, allowing the model to capture nonlinear failure behaviours that conventional methods fail to recognize.
The research focuses on crack propagation in ceramic-based high-power LED packages used in automotive lighting systems, incorporating variations in two-pad and three-pad configurations, as well as thin and thick film metallized ceramic substrates with validated FEA models. Comparative analysis shows that PointNet significantly outperforms 3D CNNs, achieving an exceptionally high correlation with experimental data (R² = 99.99%). This AI-driven automated feature extraction and lifetime prediction approach marks a major advancement over traditional FEA-based methods, offering superior accuracy, reliability, and scalability for predicting solder joint reliability in microelectronics
GPUs All Grown-Up: Fully Device-Driven SpMV Using GPU Work Graphs
Get PDFSparse matrix-vector multiplication (SpMV) is a key operation across high-performance computing, graph analytics, and many more applications. In these applications, the matrix characteristics, notably non-zero elements per row, can vary widely and impact which algorithm performs best. Thus, Graphics Processing Unit (GPU) SpMV algorithms often rely on costly preprocessing to determine what per-row algorithm to select to achieve high performance. In this work we combine SpMV preprocessing and the subsequent per-row processing on the GPU by leveraging the novel “Work Graphs” GPU programming model—initially designed for graphics applications—for dynamic on-device self-scheduling. Work Graphs allow for fine-grain dataflow execution of individual workgroups using emerging hardware and firmware support. As soon as preprocessing has generated sufficient work, workgroups of individual processing kernels are self-scheduled and executed, interleaved with those of other kernels. This improves cache locality and eliminates host interaction altogether. Across a suite of 59 sparse matrices, the best of various novel Work Graphs SpMV implementations outperforms state-of-the-art rocSPARSE “LRB” for a single SpMV by up to 7.19 × (mean: 3.35 ×, SD: 1.89). Furthermore, it achieves much more stable performance across various sparsity patterns than the rocSPARSE CSR-General algorithm, and even beats the advanced rocSPARSE CSR-Adaptive algorithm for up to 92 consecutive SpMV calculations. In addition, compared to rocSPARSE LRB, it reduces code complexity by 75%. Its memory footprint for supporting data structures is a fixed ∼ 25 MiB independent of matrix size, compared to rocSPARSE LRB’s data structures that scale with matrix size to hundreds of megabytes. Overall, this work showcases the performance potential of emerging dynamic on-device scheduling techniques for GPU compute applications
The interplay of future making and foresight: Insights from a family business case study
No full textExploring Synthetic Radar Data and Deep Learning for Road User Classification in Autonomous Vehicles
No full text
Evaluating Optimistic Synchronization of Network Graphs in Mobile Networks With MEC Support
No full textSeismic soil-structure interaction of a nuclear building: Comparison of two different methods
No full textConference Report: Mapping Situational Analysis—An International Conference
No full textIn this conference report we provide insights into some of the key topics presented at the first international conference on situational analysis in Germany, held in November 2024 in Magdeburg. Central issues discussed include collaborations and feminist perspectives in practice for power-sensitive knowledge production, nonhumans and how they can be included in situational analysis through all types of mappings as well as methodical and methodological questions regarding mapping processes, interdisciplinary mapping, and the conceptualization of space in and through situational analysis. Adele CLARKE and her work were honored and celebrated at the conference