Autonomous systems have high potential in many application domains. However, most discussions seem to take place with respect to autonomous road vehicles. The automotive industry promised substantial progress in this field but many predictions have not come true. Companies have stepped back and corrected their predictions. However, autonomous behavior is obviously not limited to road vehicles. Various kinds of systems can benefit from autonomous behavior in various domains such as health and pharmaceutics, energy, manufacturing, farming, mining, and so on. We thus take a broader perspective on autonomous systems design and discuss benefits, challenges, and risks in various application domains.

Autonomous cyber-physical systems have enormous potential to make our lives more sustainable, more comfortable, and more economical. Artificial Intelligence and connectivity enable autonomous behavior, but often stand in the way of market launch. Traditional engineering techniques are no longer sufficient to achieve the desired dependability; current legal and normative regulations are inappropriate or insufficient. This paper discusses these issues, proposes advanced systems engineering to overcome these issues, and provides a roadmap by structuring fields of action.

Software and System Product Lines (SSPL) are the state-of-the-art for systematically reusing a common set of core assets in the development of similar products in a product family. A large number of SSPL success stories have been published in the last decade and commercial tool support is also available. SSPLs promise to reduce cost, to shorten time-to-market for new features, and to increase product quality by systematically reusing core assets in the development of three or more systems. However, an open challenge is SSPL engineering for safety-relevant systems such as automotive, avionic, or industrial automation systems. Safety-relevant systems have to be developed, analyzed, and certified according to safety standards such as IEC 61508. These standards demand the application of safety analyses such as Fault Tree Analysis and Failure Mode and Effect Analysis. Starting the safety analysis of each product variant of a SSPL from scratch is complex and very time-consuming. However, there are only few convincing cases, where SSPL approaches have been followed in safety engineering. To pave the way for a broader adoption of SSPL approaches, this paper reports practical experiences with industrial-strength methods and tools along an adaptive cruise control SSPL. The paper demonstrates how commercial tools can be used (i) to analyze safety-related aspects already in the architectural design, (ii) to model the results as component integrated component fault trees (C2FT), and (iii) to systematically reuse C2FT in the safety analysis of a concrete product. The results of the case study show that C2FT (i) can be easily integrated into a feature-oriented development process of SSPL, (ii) facilitate early consideration of safety in domain engineering, and (iii) reduce effort and complexity of safety analyses in application engineering.

Resilience has been successfully realized in automotive systems to increase system reliability at reasonable costs. Using dynamic adaptation, the system adapts to runtime errors - caused by internal system faults or adverse environmental situations like critical driving situations - in order to provide the best possible functionality and to guarantee system safety in any given system and environmental state. This paper introduces an engineering approach for developing resilient systems using dynamic adaptation. The approach is based on component-oriented modeling and on analyses of component compositions. We describe how component-oriented modeling and compositional analyses enable the usage of dynamic adaptation for achieving a trade-off between availability and cost in safety-critical, resilient systems and how it helps to manage the complexity inherent in component composition.

Adaptation is increasingly used in the development of safety-critical embedded systems, in particular to reduce hardware needs and to increase availability. However, composing a system from many reconfigurable components can lead to a huge number of possible system configurations, inducing a complexity that cannot be handled during system design. To overcome this problem, we propose a new component-based modeling and verification method for adaptive embedded systems. The component-based modeling approach facilitates abstracting a composition of components to a hierarchical component. In the hierarchical component, the number of possible configurations of the composition is reduced to a small number of hierarchical configurations. Only these hierarchical configurations have to be considered when the hierarchical component is used in further compositions such that design complexity is reduced at each hierarchical level. In order to ensure well-definedness of components, we provide a model of computation enabling the formal verification of critical requirements of the adaptation behavior.