Participating in public trafﬁc always involves accepting a trade-off between mobility and safety. This applies to both, human-driven and automated vehicles. Thus, there is an inherent risk to operate SAE level 3+  automated vehicles in public trafﬁc.
Different types of uncertainty can be distinguished to describe the root causes of this risk. Machine perception is the enabling technology for the automation of road vehicles. However, different sensor principles are always subject to physical limitations. Fusing data from different sensor types can overcome some of these deﬁciencies, but not all. Moreover, the interpretation of the environment data is still an active ﬁeld of research. Even assuming perfect environment perception, occluded areas introduce uncertainty to the task of situation assessment.
The prediction of other trafﬁc participants’ behavior is thus based on inherently limited information about their states and – in general – based on unknown intentions. This even applies to human drivers, who basically assume that other trafﬁc participants will behave mostly compliant to trafﬁc rules. Still, it is an open question whether this assumption should be adopted to automated vehicles. Even more challenging, potential reckless behavior, e. g., of children, must also be considered.
The generation of appropriate vehicle behavior and its execution are another source of uncertainty. On the one hand, it must be assured that automated vehicles are able to make safe driving decisions for any perceived situation. On the other hand, the execution of these decisions in terms of trajectory planing and tracking relies on vehicle dynamics models that are always an abstraction of the reality. Hence, the effects of unmodeled vehicle dynamics and external disturbances must also be considered.
The execution of perception and decision making software moreover requires data, software and hardware of unprecedented complexity – requiring new approaches to functional safety  and safety of the intended functionality , for instance cf. . This already starts with the speciﬁcation of an automated driving functionality as foundation for the development. Basically, the speciﬁcation is required to deﬁne the vehicle behavior for each possible scenario encountered during the automated operation. However, it appears impossible to consider all relevant scenarios during system design since automated vehicles operate in an open environment. Consequently, the vehicle’s correct behavior is not guaranteed in scenarios which go beyond the speciﬁcation. In contrast, human drivers are able to adapt to unforeseen scenarios. Hence, approaches are needed to explicitly address this deﬁnitional gap. Industry has recently put forward some of these approaches [7, 8].
Last but not least, conventional validation approaches do not sufﬁce for validating SAE level 3+ automation systems as has been pointed out by Wachenfeld and Winner . Also, new approaches such as scenario based validation in simulation are subject to research and development. Validated sensor models have to be established and rare corner cases must be incorporated in test suites. Moreover, the impact of discretization of a continuous environment and its implications for the completeness of validation have to be understood. Besides the need for novel validation approaches, surrogate safety metrics for the operational assessment of the technology are neither well defined nor harmonized across the industry and the regulatory frameworks being discussed to date. The necessary safety level for automated vehicles accepted by society is not determined yet, as well.
Approaches to reduce all these uncertainties to an accepted level (which is yet to be defined) and methods to argue why the remaining risks are reasonable are key for deploying automated vehicles into public traffic. To achieve a conclusive argumentation of reasonable residual risk, a variety of stakeholder values need to be considered in a systematic and traceable manner . In summary, the many unresolved challenges in ensuring safety of automated vehicles necessitate a holistic consideration of safety and its validation throughout the development of automated vehicles. Thus, we see an increased interest on this field derived from the urgent need of solutions from industry and regulators.
The proposed workshop aims at encouraging contributions and promoting scientiﬁc exchange among researchers and practitioners from academia, industry, and regulatory bodies. Although safety and its validation are more and more focused in the ITS community, the numerous participants of the successful workshops on “Ensuring and Validating Safety for Automated Vehicles” (IEEE IV & ITSC) as well as on “Automated Vehicle Safety: Veriﬁcation, Validation, and Transparency” (IEEE ITSC) commonly agreed that research with respect to ensuring and validating safety is still underrepresented in the ITS community. This joint workshop will continue the fruitful discussions of the previous workshops at the same time with increasing the cadence of the dialogue to accelerate consensus among practitioners. Held in the neighborhood to several companies developing automated vehicles, we expect a high number of contributions as well as participants for the workshop. This workshop proposal is supported by the IEEE ITSS Technical Committee on Self Driving Automobiles.
REFERENCES  T. Woopen, B. Lampe, T. Böddeker, L. Eckstein, A. Kampmann, B. Alrifaee, S. Kowalewski, D. Moormann, T. Stolte, I. Jatzkowski, M. Maurer, M. Möstl, R. Ernst, S. Ackermann, C. Amersbach, S. Leinen, H. Winner, D. Püllen, S. Katzenbeisser, M. Becker, C. Stiller, K. Furmans, K. Bengler, F. Diermeyer, M. Lienkamp, D. Keilhoff, H.-C. Reuss, M. Buchholz, K. Dietmayer, H. Lategahn, N. Siepenkötter, M. Elbs, E. von Hinüber, M. Dupuis, and C. Hecker, “UNICARagil - Disruptive Modular Architectures for Agile, Automated Vehicle Concepts,” in Aachener Kolloquium, Aachen, Germany, 2018.  T. Stolte, A. Reschka, G. Bagschik, and M. Maurer, “Towards Automated Driving: Unmanned Protective Vehicle for Highway Hard Shoulder Road Works,” in 2015 IEEE 18th International Conference on Intelligent Transportation Systems, Las Palmas, Spain, 2015, pp. 672–677.  SAE, “Taxonomy and Deﬁnitions for Terms Related to Driving Automation Systems for On-Road Motor Vehicles,” Society of Automotive Engineers, Standard J3016, 2016.  ISO, “ISO 26262: Road vehicles - Functional Safety,” International Organization for Standardization, Geneva, Switzerland, International Standard ISO 26262:2018, 2018.  ——, “ISO/PAS 21448: Road vehicles – Safety of the intended functionality,” International Organization for Standardization, International Standard ISO/PAS 21448, 2019.  M. Törngren, X. Zhang, N. Mohan, M. Becker, L. Svensson, X. Tao, D.-J. Chen, and J. Westman, “Architecting Safety Supervisors for High Levels of Automated Driving,” in 21st IEEE Intelligent Transportation Systems Conference, Maui, HI, USA, 2018.  S. Shalev-Shwartz, S. Shammah, and A. Shashua, “On a Formal Model of Safe and Scalable Self-driving Cars,” arXiv preprint arXiv:1708.06374, 2017.  M. Wood, P. Robbel, M. Maass, R. Duintjer Tebbens, M. Meijs, M. Harb, J. Reach, K. Robinson, D. Wittmann, T. Srivastava, M. E. Bouzouraa, S. Liu, Y. Wang, C. Knobel, D. Boymanns, M. Löhning, B. Dehlink, D. Kaule, R. Krüger, J. Frtunikj, F. Raisch, M. Gruber, J. Steck, J. Mejia-Hernandez, S. Syguda, P. Blüher, K. Klonecki, P. Schnarz, T. Wiltschko, S. Pukallus, K. Sedlaczek, N. Garbacik, D. Smerza, D. Li, A. Timmons, M. Bellotti, M. O‘Brien, M. Schöllhorn, U. Dannebaum, J. Weast, A. Tatourian, B. Dornieden, P. Schnetter, P. Themann, T. Weidner, and P. Schlicht, “Safety First for Automated Driving,” Aptiv, Audi, Baidu, BMW, Continental, Daimler, Fiat Chrysler Automobiles, Here, Inﬁneon, Intel, Volkswagen, White Paper, 2019.  W. Wachenfeld and H. Winner, “The Release of Autonomous Vehicles,” in Autonomous Driving: Technical, Legal and Social Aspects, M. Maurer, J. C. Gerdes, B. Lenz, and H. Winner, Eds., Berlin, Heidelberg, Germany: Springer Berlin Heidelberg, 2016, pp. 425–449.  “IEEE 7000 – IEEE Standard Model Process for Addressing Ethical Concerns during System Design,” Standard IEEE 7000:2021.