Abstract
How can digital twins (DTs) scale from representing individual assets to capturing the complexity of entire systems? Building a monolithic twin for such systems is often costly or impractical, which has led to the emergence of composite digital twins (CDTs). A CDT is a combination of DTs that together represent a more complex real-world entity, capturing both their individual characteristics and their interactions. Although this approach improves modularity and reusability, it raises significant challenges in terms of scalability and interoperability. In this talk, we discuss how OpenTwins, an open-source platform for the development of CDTs, has evolved to address these challenges. Starting from the limitations of its first version, built on Eclipse Ditto and supporting a hierarchical form of twin composition, we present OpenTwins v2. This new version builds on our formalization of DT composition to combine open standards such as the W3C Web of Things with knowledge graphs within an event-driven, actor-based architecture. In this way, OpenTwins lays the foundations for moving beyond isolated replicas towards interconnected ecosystems of digital twins.
About the speaker
Cristian Martín is an Associate Professor (Profesor Titular de Universidad) at the University of Málaga (UMA), where he is a member of the ERTIS research group and the ITIS Software Institute. He obtained his Ph.D. in Computer Science from UMA in 2018, with an International Mention and an Extraordinary Ph.D. Thesis Award. He currently serves as Secretary of the Department of Languages and Computer Science at UMA. His research focuses on distributed computing and software architectures, with particular emphasis on the Internet of Things (IoT), machine learning over data streams, and open digital twin platforms. His current research addresses real-time streaming machine learning, TinyML, soft sensors, and the development and composition of distributed digital twins, including approaches based on causality, knowledge graphs, and real-time synchronization. His research has evolved from the IoT and now sits at the intersection of AI and digital twins, two complementary fields. This research follows an open-source code approach, with numerous software solutions publicly available on GitHub and widely used by the research and engineering communities.
Julia Robles is a predoctoral researcher at the University of Málaga (UMA) and a member of the ERTIS research group and the ITIS Software Institute. She is currently in the final stage of her Ph.D. in Computer Science, having previously received her B.Sc. in Software Engineering in 2021 and her M.Sc. in Software Engineering and Artificial Intelligence in 2023, both from UMA. Her research centers on digital twins, with a particular focus on how they can be composed and scaled to represent complex environments. In this context, she also explores software architectures, modeling approaches, the Internet of Things, knowledge graphs, and multi-agent systems to build more interoperable, semantically rich, and intelligent digital twins. She is the lead developer of OpenTwins, an open-source platform for the development of composite digital twins, publicly available on GitHub and applied in projects across diverse domains.