Autonomous systems have enormous potential and they are bound to be a major driver in future economical and societal transformations. In contexts where safety, or other critical properties, need to be guaranteed it is, however, presently not possible to exploit autonomous systems to their full potential. Unknowns and uncertainties are induced due to the high complexity of the autonomous behaviors, the utilized technology and the volatile and highly complex system contexts. These characteristics render the base assumptions of established assurance methodologies (and standards) void, hence new approaches need to be investigated.

One general approach for making autonomous systems dependable is to make them aware of risks and empower them to assess and control those risks. Implementing such a Dynamic Risk Management (DRM) approach comes along with many challenges concerning the necessary self- and context awareness. On the one hand, powerful and thus complex self- and context awareness is necessary to minimize risks, to resolve conflicting objectives and to make acceptable trade-off decisions. On the other hand, the complexity of DRM is in conflict with assurance and high confidence in adequate risk management. DRM has the potential to not only outright enable certain types of systems or applications, but also to significantly increase the performance of already existing ones. This is due to the fact that by resolving unknowns and dealing with uncertainties at runtime it will be possible to get rid of worst-case assumptions that are typically detrimental to the systems performance properties.

The DREAMS workshop intends to explore concepts, techniques and technology for realizing DRM.

It invites experts, researchers, and practitioners for presentations and in-depth discussions about the current status of DRM in practice, its relevance for specific use cases, its relation to exiting assurance frameworks for autonomous systems and standardization activities. DREAMS aims at bringing together communities from diverse disciplines, such as safety engineering, runtime adaptation, predictive modelling, control theory, and from different application domains such as automotive, healthcare, manufacturing, agriculture and critical infrastructures.

Topics of interest include but are not limited to

definition and modelling of the Operational Design Domain
approaches for monitoring AI-based components
(runtime) estimation and handling of uncertainties
approaches to formalize risk and risk assessment
ethics of risk / machine ethics
(dynamic) assurance cases for autonomous systems and AI
collecting data from operation for continuous risk management / MLOps
Case studies