First International Workshop on Autonomous System Quality Assurance and Prediction with Digital Twins
Co-located with ETAPS 2025
May 4th, 2025, Hamilton (Canada)
Autonomous systems, such as self-driving cars, robots, and unmanned aerial vehicles, demand rigorous quality assurance to ensure their safety, reliability, and performance in diverse and unpredictable environments.
Achieving robust quality assurance involves rigorous testing and analysis in the entire system life-cycle to identify potential failures and optimize system performance under various conditions. Predictive capabilities are equally necessary, leveraging advanced technologies like machine learning and simulation to anticipate failures and adapt systems accordingly. By continuously monitoring and simulating behaviors in virtual environments, such as those created by digital twin technology, autonomous systems can undergo thorough testing, reducing the risks and costs associated with physical trials.
Digital twins offer a powerful solution to these challenges by providing virtual replicas of physical systems. Engineers can conduct extensive development, testing, and validation activities by continuously obtaining real-time data from the physical entity and simulating the behavior and interactions of autonomous systems in a controlled digital environment. Digital twins enable the emulation of various scenarios, including rare or hazardous conditions that are difficult or inconvenient to replicate in the real world. This capability accelerates the verification process and allows iterative improvements based on simulated data before physical deployment. Additionally, digital twins facilitate continuous monitoring and performance optimization throughout the lifecycle of autonomous systems, ensuring ongoing compliance with safety standards and operational requirements.
Using digital twins in autonomous systems is crucial for ensuring quality assurance and enhancing predictive capabilities. This technology enables real-time monitoring and simulation of system behavior, providing a detailed virtual replica that facilitates advanced testing and the correction of potential faults before physical production. Digital twins also support optimization of design and operational processes, reducing development times and enhancing overall efficiency. Moreover, by analyzing large datasets and simulating complex scenarios, digital twins aid in developing advanced algorithms and predictive resource management, improving system resilience and adaptability.
However, several challenges emerge in realizing a digital twin. This process involves creating detailed models of the system and its environment, which must remain aligned with both the environment's dynamic and open nature and the hardware status of the physical twin. This poses a significant challenge, as both autonomous systems and their digital twins must continually adjust to changes to maintain accuracy and functionality. Runtime techniques are therefore required to verify and adapt both autonomous systems and their digital twins.
Despite the great interest in enhancing autonomous systems' quality assurance and prediction through digital twins, no common methodologies, model-based techniques, or formal aspects have been fully established.
ASQAP 2025 aims to provide a forum for sharing and discussing innovative contributions to both formal and practical approaches in the analysis and development of methodologies, including digital twins, for the quality assurance of autonomous systems.
Topics of interests include, but are not limited to:
ASQAP 2025 welcomes research and experience papers, with papers describing novel research contributions and innovative applications being of particular interest.
Contributions can be:
All papers must:
All submissions (except presentation abstracts) must be unpublished and not be under review elsewhere. Submissions will be reviewed by at least three members of the Program Committee. At least one author of an accepted paper should register for the conference and present the paper.
Accepted regular and short papers will be included in the EPTCS proceedings and appear in the digital libraries.
Registration fees and instructions will be available on the ETAPS 2025 website.
Program will be anounced soon.
Marsha Chechik is a Professor in the Department of Computer Science at the University of Toronto. Prof. Chechik's research interests are in modeling and reasoning about software. She has authored over 200 papers in formal methods, software specification and verification, computer security, and requirements engineering.
Arianna Fedeli is a post-doctoral researcher at the Gran Sasso Science Institute, L’Aquila. She received her Ph.D. from the University of Camerino, Italy, in 2024. Her research interests include model-driven engineering applied to distributed systems such as the Internet of Things and Digital Twins domains.
Gianluca Filippone is a post-doctoral researcher at the Gran Sasso Science Institute, L'Aquila, Italy. He received his Ph.D. from the University of L'Aquila, Italy, in 2023. His research study focuses on the model-driven composition, coordination, and adaptation of distributed systems, multi-robot systems, and microservice architectures.
Federico Formica is a PhD student at McMaster University, Canada. His research mainly focuses on automatic testing and automatic test case generation for Cyber-Physical Systems. Specifically, he works on approaches that leverage the developers' domain knowledge and experience to find failure-revealing test cases, i.e., test cases that cause a requirement violation.
Mirgita Frasheri is an Assistant Professor at the Department of Electrical and Computer Engineering at Aarhus University, Denmark. She has a background in autonomous and multi-agent systems. Her research is mainly focused on Digital Twins for mobile robots. In addition, she has an interest in multi-agent collaboration, local re-planning techniques, as well AI ethics.
Nico Hochgeschwender is a Full Professor of Software Engineering for Cognitive Robots and Systems at the University of Bremen. His research interests lie at the intersection of AI-enabled Robotics and Software Engineering, focusing on assuring dependability, transparency, and explainability of robotics and autonomous systems, benchmarking and performance evaluation, and domain-specific modeling and languages for robotics.
Lina Marsso is currently a post-doctoral researcher in the Department of Computer Science at the University of Toronto working with Marsha Chechik. She received her PhD from INRIA Grenoble, France where she was advised by Radu Mateescu and Ioaniss Parissis. Her recent work is in the combination of safety, social, legal, ethical, empathetic, and cultural, verification and analysis of autonomous systems.