@article{SiefkeSommerWudkaetal., author = {Siefke, Lennart and Sommer, Volker and Wudka, Bj{\"o}rn and Thomas, Carsten}, title = {Robotic Systems of Systems Based on a Decentralized Service-Oriented Architecture}, series = {Robotics}, volume = {9}, journal = {Robotics}, number = {4}, publisher = {MDPI}, issn = {2218-6581}, doi = {10.3390/robotics9040078}, url = {http://nbn-resolving.de/urn:nbn:de:kobv:523-14461}, pages = {10}, language = {en} } @article{HiebertHofstetterThomasetal., author = {Hiebert, Tina and Hofstetter, Florian and Thomas, Carsten and Mushtaq, Savera and Kaan, Eero and Parameswaran, Biranavan}, title = {Labels4Rails: A Railway Image Annotation Tool and Associated Reference Dataset}, series = {Data}, volume = {10}, journal = {Data}, number = {12}, editor = {Torres-Sospedra, Joaqu{\´i}n}, publisher = {MDPI}, issn = {2306-5729}, doi = {10.3390/data10120210}, url = {http://nbn-resolving.de/urn:nbn:de:kobv:523-22070}, abstract = {The development of autonomous train systems relies heavily on machine learning (ML) models, which in turn depend on large, high-quality annotated datasets for training and evaluation. The railway domain lacks adequate public datasets and efficient annotation tools. To address this gap, we present Labels4Rails, a tool designed specifically for the annotation of railway scenes. It captures track topology, switch states including switch directions, and informational tags regarding the images' content and leverages consistent camera perspectives and the fixed track geometries inherent to railways for annotation efficiency. We used Labels4Rails to create the L4R_NLB reference dataset from Norwegian railway footage. The dataset contains 10,253 annotated images across four seasons, including 1415 switch annotations. Both the tool and dataset are publicly available.}, subject = {Artificial intelligence}, language = {en} } @article{JassThomas, author = {Jaß, Philipp and Thomas, Carsten}, title = {Using N-Version Architectures for Railway Segmentation with Deep Neural Networks}, series = {Machine Learning and Knowledge Extraction}, volume = {7}, journal = {Machine Learning and Knowledge Extraction}, number = {2}, editor = {Javaheri, Danial}, publisher = {MDPI}, issn = {2504-4990}, doi = {10.3390/make7020049}, url = {http://nbn-resolving.de/urn:nbn:de:kobv:523-20867}, abstract = {Autonomous trains require reliable and accurate environmental perception to take over safety-critical tasks from the driver. This paper investigates the application of N-version architectures to rail track detection using Deep Neural Networks (DNNs) as a means to improve the safety of machine learning (ML)-enabled perception systems. We combine three different neural network architectures (WCID, VGG16-UNet, MobileNet-SegNet) in a 3M1I configuration. In this configuration, we apply two fusion methods to increase accuracy and to enable error detection: Maximum Confidence Voting (MCV), combining the DNN predictions at the image level, and Pixel Majority Voting (PMV), a novel approach for combining the predictions at the pixel level. In addition, we implement a new method for evaluating and combining prediction confidence values in the N-version architecture during runtime. We adjust the overall prediction confidence according to the conformity of all individual predictions, which is not possible with an individual network. Our results show that the N-version architecture not only enables a detection of erroneous predictions by utilizing those adjusted confidence values, but it can also partially improve the predictions by using the PMV combination algorithm. This work emphasizes the importance of model diversity and appropriate thresholds for an accurate assessment of prediction safety. These approaches can significantly improve the practical applicability of ML-based systems in safety-critical domains such as rail transportation.}, subject = {Machine learning}, language = {en} }