Daniel Bamal hat seine Masterarbeit an der TU Clausthal erfolgreich abgeschlossen. Das Thema seiner Arbeit lautet: “Machine Learning-based intelligent Fault Detection and Classification for Automotive Software Systems based on real-time Fault Injection and Hardware-in-the-Loop Simulation“.
Hardware-in-the-Loop (HIL) Simulation has been recommended by ISO 26262 as an essential test bench for Automotive Software Systems (ASS) testing. Due to the complexity and huge amount of data recorded by the HIL platform during the testing process, the conventional data analysis methods used for detecting and classifying faults based on the human expert are not feasible. Therefore, the development of an effective means based on the historical dataset is required to analyze the records of the testing process in an efficient manner. Although data-driven fault diagnosis is superior to other approaches, selecting an appropriate technique from the wide range of machine learning techniques is challenging. Moreover, the training data contains the automotive faults are very rare and considered highly confidential by the automotive industry. This study aims to develop and implement an intelligent fault detection and classification model using hybrid machine learning techniques. To this end, a HIL-based fault injection framework is developed and implemented to generate faulty data without altering the original system model. Besides, a combination of Convolutional Neural Network (CNN) and Long Short-Term Memory (LSTM) is used to build the model. In this study, different types of faults are considered to cover the most common potential faults in automotive software signals, and the collected data are used to train and evaluate the developed model. As a case study, a gasoline engine system is used to demonstrate the capabilities and advantages of the proposed method. The results prove that the proposed method shows better diagnosis performance than other individual methods.