The 10th International Conference on Fuzzy Systems and Data Mining (FSDM 2024)

Abstract: Deep Neural Networks [DNNs] are being integrated into Automated Driving Systems [ADS] to perform complex perception and control problems. However, DNNs are generally challenging or impossible to interpret for the purpose of functional safety [FuSa] or Safety of the intended functionality [SOTIF] assessment. In contrast, physical models of the driving task are generally much easier to explain and assess than the abstract statistical models encoded in a DNN. In this paper, we present a statistical modelling and evaluation workflow that can be easily explained to FuSa and SOTIF assessors. Our workflow uses Bayesian networks [BN] refining fault trees and a physical model of an ADS in a given scenario. The Dominant Factors [DF] that impact the ADS risk can then be identified based on simulations of the physical model and simulations sampled from the BN. The workflow can evaluate under which conditions a tolerable risk target [TRT] can be achieved. We evaluate our proposed workflow in an example high-frequency traffic scenario, a highway cut-in scenario. We compare two methods to identify and confirm the DF for meeting the TRT. The DF found show that a static operating design domain [ODD] definition is insufficient. In the example, if the sense-plan-act control architecture is extended by a dynamic traffic monitoring protection layer, the TRT can be achieved.

Biography: Mr. Faller has over forty years of experience and extensive knowledge in the safety and reliability field. He is widely known as the world’s top safety and reliability expert. After graduating from Munich University, Mr. Faller began working for MAN Technology where he co-developed a self-steering bus for public transportation. In 1985, he joined TÜV Bayern as chief engineer for safety of computer systems. Several years later, Mr. Faller became technical manager of IQSE, the Institute for Quality and Safety in Electronics. Also, during this time, Mr. Faller was named Chairman of the German DIN/DKE committee on “Computers in Safety–Related Applications”. In 1998, he became Business Unit Manager of “Automation, Software, and Electronics” at TUV Product Service. This position provided him with world-wide responsibility. His signature appeared on most of the world’s equipment safety certifications. Since 2000, Mr. Faller is the co-founder and Principal Partner for exida.com, LLC, and was CEO of exida.com, GmbH. Mr. Faller was a member of many different international standardization committees and was member of the Nuclear Reactor Safety Commission for Electrotechnical equipment of the German government. He holds various patents on functional safety of programmable electronic systems and semiconductors and is frequently sought after for speaking engagements and publications based on his unparalleled expertise.