Smart Systems are not automatically safe

Research advances in cyber-physical systems augmented by machine learning and artificial intelligence promise to transform our world. However, systems and software that operate in dangerous or unpredictable environments, provide large-scale distributed coordination, augment human capabilities, or enhance societal wellbeing are not without critical challenges. QONFEST 2020 offers state-of-the-art research on all the concerns of critical systems, cyber-physical systems, software systems driven by interoperable blockchain technology, as well as smart grids, health-care systems and logistics or industrial production processes.

Cyber-physical Systems

The autonomous systems are being used to explore the deepest oceans or space, zero-net energy buildings, robotic surgeries, smart manufacturing and scheduling, self-driving cars avoiding collisions autonomously, or contact tracing apps. Among others, they all have one thing in common: they are cyber-physical systems that bridge the gap between the cyber world and the physical world in which we live.

For the first time, between August 31 and September 5, 2020, Vienna is hosting QONFEST, a joint congress of the most important international scientific conferences and workshops on the topic of concurrency theory and its applications, timed systems, semantics, logics, and verification techniques. The organizers of CONCUR (31st International Conference on Concurrency Theory), QEST (17th International Conference on Quantitative Evaluation of SysTems), FORMATS (18th International Conference on Formal Modeling and Analysis of Timed Systems) and FMICS (25th International Conference on Formal Methods for Industrial Critical Systems) have managed to bring together researchers and practitioners from industry and academia who all seek to solve the primary demand of the users of technology: placing a high level of trust in the operation of the systems – whereas trust is a combination of many characteristics, mainly reliability, safety, security, privacy, and usability.

Trust the Machine

New computer science methods such as machine learning and artificial intelligence tools help realize the modern grand vision for societal-level services that transcend space and time at scales never possible before. „At the same time, they represent one of the critical challenges to ensure certification and thus trustworthiness of cyber-physical systems,” explains Ezio Bartocci, head of the Trustworthy Cyber-Physical Systems (TrustCPS) Group at TU Wien Informatics and the general chair of the QONFEST 2020. „For example, a CPS with machine learning components such as a self-driving car cannot be designed just using a model (as in the past), because their operational behavior is now data-driven. Then, how can we make sure (verify) that the system will behave correctly in every situation? Can we verify the autonomous system at runtime so that in the case of an unforeseen event, it can still trigger a safety mechanism?”

Verification and Model Checking in Austria

These are also some of the research questions which also connect the Austrian organizers of QONFEST, who in 2010 created the national Research Network “RiSE” and the Austrian Society for Rigorous Systems Engineering (ARiSE), funded by the FWF (Austrian Science Fund). Since then, the Austrian community has been occupied with a critical re-examination of existing hardware design and programming abstractions for computing. Together with research institutes such as Austrian Institute of Technology (AIT), they have introduced new scientific and engineering principles, algorithms, models, and theories for the analysis and design of cyber-physical systems, which has also received numerous ERC grants. “ERC Proof of Concept grant is allowing us to bring our ideas closer to the market since we want the big companies to apply our method of automated error detection,” explains Laura Kovacs (TU Wien Informatics).

Among the internationally acclaimed founding members of RiSE are also two of the QONFEt’s invited speakers. Thomas Henzinger (IST Austria) will present some of the game-theoretical aspects in his talk “A Survey of Bidding Games on Graphs. “ Roderick Bloem (TU Graz) will address a case of reinforcement learning (RL) as a machine learning technique that has been deployed together with formal verification on a case study involving service robots in his invited talk “Safe Reinforcement Learning using Probabilistic Shields. “

Privacy and Security

Smart systems that include digital cyber technologies, software, and physical components and intelligent interaction with other systems across information and physical interfaces lead to collecting a vast amount of human-centric data at various scales. Although cyber-physical systems greatly enrich our life qualities and experiences, they also bring about privacy and security concerns, warns Bartocci. Addressing those concerns is the point of the QONFEST´s plenary speaker Annabelle McIver (Macquarie University). She will address the trade-off between individuals’ data privacy and the necessary information flow between other systems and people in her talk “On Privacy and Accuracy in Data Releases.”

Date, Time and Place

Due to the current COVID-19 situation, the QONFEST takes place online, from August 31 until September 5, 2020. QONFEST is organized by the research units Formal Methods in System Engineering and Cyber-physical Systems at TU Wien Informatics. It is supported by the Interchain Foundation, the Vienna Center for Logic and Algorithms at the TU Wien Informatics as well as by Springer.

Information on access, schedules and contributors

Curious about our other news? Subscribe to our news feed, calendar, or newsletter, or follow us on social media.