Harshini Vishwanatha Reddy hat ihre Masterarbeit an der TU Clausthal erfolgreich abgeschlossen. Das Thema ihrer Arbeit lautet: “Machine Learning-based Simultaneous Fault Detection and Classification for Recordings Analysis of Hardware-in-the-Loop Tests“.
Abstrakt:
Functional validation of complex Automotive Software Systems (ASSs) is important to avoid any upcoming dangerous consequences. Hardware-in-the-Loop (HIL) is recommended by ISO 26262 as a test bench to ensure the safety aspects of such complex systems. Failure is common occurrence in any system which are hard to analyze and more often is treated manually. Most of the failures occur due to associated faults. Faults can be either single or simultaneous fault. Simultaneous fault is the scenario where two or more faults raise during the same time but in different location. In this study, simultaneous fault with two different kinds of the faults is considered for detection and classification. Data driven method is associated with many machines learning techniques which enable to take decision based on data generated from system rather than based in intuition. An intelligent novel method for Simultaneous Fault Detection and Classification (SFDC) is proposed based on hybrid machine learning algorithms. SFDC model is designed such a way that it can also deal with noise in the data. The proposed model represents a significant advancement in addressing the challenges of simultaneous fault analysis, showcasing its potential to contribute to safer and more robust automotive systems.
