@article{AugustineSchwabKlarmannetal.2024, author = {Augustine, Nitin and Schwab, Maximilian and Klarmann, Steffen and Pfefferer, Christian and Schiendorfer, Alexander}, title = {Impact of Blemish Artefacts on Object Detection Models in Autonomous Driving: A Study on Camera Module Manufacturing Defects}, volume = {2024}, journal = {Procedia Computer Science}, number = {232}, publisher = {Elsevier}, address = {Amsterdam}, issn = {1877-0509}, doi = {https://doi.org/10.1016/j.procs.2024.01.061}, pages = {616 -- 625}, year = {2024}, abstract = {With the rise of autonomous driving technology, concerns over the safety of AI-based systems have increased, particularly regarding their effect on human life and infrastructure. The complexity and unpredictable nature of AI-enabled software functions have disrupted traditional safety evaluation methods. In such a scenario, the quality control methods used at the manufacturing environments of sensors used for deploying AI-based perception functions need to be coupled with the downstream functions. This study focuses on investigating the impact of a specific production-related defect called blemish, which is caused by debris on image sensors of camera modules. The research aims to answer two fundamental questions: 1) How can blemish artifacts be mathematically modeled, and what is the process for developing a suitable dataset that incorporates this defect? 2) How can the influence of blemish artifacts on performance of object detection models be traced back to define quality requirements of the camera modules? This work takes a novel perspective by shifting the focus from improving model resilience to improving hardware quality camera modules by studying the effects of manufacturing defects on downstream functions. This research seeks to establish a framework to derive quality targets for front camera module manufacturing which can be extended to other sensors for similar applications.}, language = {en} } @inproceedings{BaschinBaschinBoeseltetal.2024, author = {Baschin, Anja and Baschin, Michelle and B{\"o}selt, Reinhard and Felix, Rudolf and Fernandez, Cesareo and Gehring, Sven and G{\"o}rtz, Alexander and Harpeng, Lars and Hei, Yuguang and Hildebrandt, Niclas and H{\"u}gging, Thomas and K{\"a}tzel, Pascal and Kolberg, Kristoffer-Robin and Kuoza, Leonid and Luzius, Lukas and Lisca, Gheorghe and Michailov, Lilia and Multhaup, Werner and Outafraout, Karima and Proch, Fabian and Schiendorfer, Alexander and Simmanek, Marcel and Streubel, Tom}, title = {Industrielle K{\"u}nstliche Intelligenz f{\"u}r sichere Gasnetze}, booktitle = {Tagungsband 18. Symposium Energieinnovation 2024}, publisher = {Technische Universit{\"a}t Graz}, address = {Graz}, url = {https://www.tugraz.at/events/eninnov2024/nachlese/download-beitraege/stream-a/\#c590685}, year = {2024}, language = {de} } @inproceedings{GajekSchiendorferReif2023, author = {Gajek, Carola and Schiendorfer, Alexander and Reif, Wolfgang}, title = {A Recommendation System for CAD Assembly Modeling based on Graph Neural Networks}, booktitle = {Machine Learning and Knowledge Discovery in Databases: European Conference, ECML PKDD 2022, Proceedings, Part I}, editor = {Amini, Massih-Reza and Canu, St{\´e}phane and Fischer, Asja and Guns, Tias and Kralj Novak, Petra and Tsoumakas, Grigorios}, publisher = {Springer}, address = {Cham}, isbn = {978-3-031-26387-3}, doi = {https://doi.org/10.1007/978-3-031-26387-3_28}, pages = {457 -- 473}, year = {2023}, language = {en} } @inproceedings{SchoenmannLodesSchiendorfer2023, author = {Sch{\"o}nmann, Alexander and Lodes, Lukas and Schiendorfer, Alexander}, title = {KI-Augmentation der Reifegradbewertung von Technologien: K{\"o}nnen KI-Sprachmodelle Expertenwissen ersetzen?}, booktitle = {Vorausschau und Technologieplanung: 17. Symposium f{\"u}r Vorausschau und Technologieplanung, 14. und 15. September 2023, Berlin}, editor = {Dumitrescu, Roman and H{\"o}lzle, Katharina}, publisher = {Heinz-Nixdorf-Institut}, address = {Paderborn}, isbn = {978-3-947647-32-3}, doi = {https://doi.org/10.17619/UNIPB/1-1821}, pages = {371 -- 387}, year = {2023}, language = {de} } @inproceedings{LodesSchiendorfer2023, author = {Lodes, Lukas and Schiendorfer, Alexander}, title = {A Deep Learning Bootcamp for Engineering \& Management Students}, booktitle = {Proceedings of the Third Teaching Machine Learning and Artificial Intelligence Workshop}, editor = {Kinnaird, Katherine M. and Steinbach, Peter and Guhr, Oliver}, publisher = {PMLR}, address = {[s. l.]}, url = {https://proceedings.mlr.press/v207/lodes23a.html}, pages = {32 -- 36}, year = {2023}, language = {en} } @inproceedings{DachtlerOrtnerFerrietal.2023, author = {Dachtler, Kristina and Ortner, Michael and Ferri, Massimo and Eberst, Christof and Schiendorfer, Alexander}, title = {Data-centric and Goal-oriented AI for Robotic Repair Tasks}, booktitle = {ISR Europe 2023, 56th International Symposium on Robotics, in cooperation with Fraunhofer IPA September 26 - 27, 2023 in Stuttgart}, publisher = {VDE Verlag}, address = {Berlin}, isbn = {978-3-8007-6141-8}, year = {2023}, language = {en} } @inproceedings{LodesSchiendorfer2022, author = {Lodes, Lukas and Schiendorfer, Alexander}, title = {Certainty Groups: A Practical Approach to Distinguish Confidence Levels in Neural Networks}, booktitle = {Proceedings of the European Conference of the PHM Society 2022}, editor = {Do, Phuc and Michau, Gabriel and Ezhilarasu, Cordelia}, publisher = {PHM Society}, address = {State College}, isbn = {978-1-936263-36-3}, doi = {https://doi.org/10.36001/phme.2022.v7i1.3331}, pages = {294 -- 305}, year = {2022}, abstract = {Machine Learning (ML), in particular classification with deep neural nets, can be applied to a variety of industrial tasks. It can augment established methods for controlling manufacturing processes such as statistical process control (SPC) to detect non-obvious patterns in high-dimensional input data. However, due to the widespread issue of model miscalibration in neural networks, there is a need for estimating the predictive uncertainty of these models. Many established approaches for uncertainty estimation output scores that are difficult to put into actionable insight. We therefore introduce the concept of certainty groups which distinguish the predictions of a neural network into the normal group and the certainty group. The certainty group contains only predictions with a very high accuracy that can be set up to 100\%. We present an approach to compute these certainty groups and demonstrate our approach on two datasets from a PHM setting.}, language = {en} } @article{StieberSchroeterFausteretal.2022, author = {Stieber, Simon and Schr{\"o}ter, Niklas and Fauster, Ewald and Bender, Marcel and Schiendorfer, Alexander and Reif, Wolfgang}, title = {Inferring material properties from FRP processes via sim-to-real learning}, volume = {128}, journal = {The International Journal of Advanced Manufacturing Technology}, number = {3-4}, publisher = {Springer}, address = {London}, issn = {1433-3015}, doi = {https://doi.org/10.1007/s00170-023-11509-8}, pages = {1517 -- 1533}, year = {2022}, abstract = {Fiber reinforced polymers (FRP) provide favorable properties such as weight-specific strength and stiffness that are central for certain industries, such as aerospace or automotive manufacturing. Liquid composite molding (LCM) is a family of often employed, inexpensive, out-of-autoclave manufacturing techniques. Among them, resin transfer molding (RTM), offers a high degree of automation. Herein, textile preforms are saturated by a fluid polymer matrix in a closed mold.Both impregnation quality and level of fiber volume content are of crucial importance for the final part quality. We propose to simultaneously learn three major textile properties (fiber volume content and permeability in X and Y direction) presented as a three-dimensional map based on a sequence of camera images acquired in flow experiments and compare CNNs, ConvLSTMs, and Transformers. Moreover, we show how simulation-to-real transfer learning can improve a digital twin in FRP manufacturing, compared to simulation-only models and models based on sparse real data. The overall best metrics are: IOU 0.5031 and Accuracy 95.929 \%, obtained by pretrained transformer models.}, language = {en} } @inproceedings{BhavnaniSchiendorfer2022, author = {Bhavnani, Sidhant and Schiendorfer, Alexander}, title = {Towards copeland optimization in combinatorial problems}, booktitle = {Integration of Constraint Programming, Artificial Intelligence, and Operations Research; 19th International Conference, CPAIOR 2022, Los Angeles, CA, USA, June 20-23, 2022; Proceedings}, editor = {Schaus, Pierre}, publisher = {Springer}, address = {Cham}, isbn = {978-3-031-08010-4}, doi = {https://doi.org/10.1007/978-3-031-08011-1_4}, pages = {34 -- 43}, year = {2022}, language = {en} } @inproceedings{StieberSchroeterFausteretal.2021, author = {Stieber, Simon and Schr{\"o}ter, Niklas and Fauster, Ewald and Schiendorfer, Alexander and Reif, Wolfgang}, title = {PermeabilityNets: Comparing Neural Network Architectures on a Sequence-to-Instance Task in CFRP Manufacturing}, booktitle = {Proceedings: 20th IEEE International Conference on Machine Learning and Applications (ICMLA 2021)}, editor = {Wani, M. Arif and Sethi, Ishwar and Shi, Weisong and Qu, Guangzhi and Raicu, Daniela Stan and Jin, Ruoming}, publisher = {IEEE}, address = {Piscataway}, isbn = {978-1-6654-4337-1}, doi = {https://doi.org/10.1109/ICMLA52953.2021.00116}, pages = {694 -- 697}, year = {2021}, language = {en} }