FDL 2019

Abstract:

As engineered systems become more and more complex, challenges and issues, the current system engineering practices face with, get higher. Characterizing and designing such systems and their interactions is not an easy task and it must involve interdisciplinary methods and disciplines that need to ensure that all system aspects are covered and integrated.
A new challenge that arises with the complexity of modern systems is the huge amount of data we need to deal with. This encourages the involvement of new tools and methods such as system analysis and smart decision making based on ML and AI methods.
This panel session brings senior engineers and scientists from several SE disciplines to discuss current issues and challenges they face with.The panel will facilitate an open discussion how they see the evolvement of SE in the future:
Will smart algorithms help a system learn from its failures and make a better version of itself without human interactions?

Panelist:
Alain Girault (alain.girault@inria.fr), INRIA, Grenoble
Edward Lee (eal@berkeley.edu), University of California Berkeley
Scott Walker (sjiw@soton.ac.uk), University Southampton
Moderator:
Carna Zivkovic (zivkovic@cs.uni-kl.de), TU Kaiserslautern

Abstract:

The term Industry 4.0 was born in 2011 at the Hannover Fair in Germany as a vision of “smart factory” in which all machines are connected to one another and manufacturing processes are automated to shorten time-to-market, reduce costs and provide high quality products.
However, the deployment of Industry 4.0 has its own price; dealing with numerous challenges is required to deploy and realize such complex solution.
And this is the point where we come to Modeling as one of key enablers for structuring and understanding the complexity of systems and processes in early development phases.
The goal of this panel to discuss and try to provide answers to the following key questions:

• Which concerns in Industry 4.0 can be addressed with modeling?
• What techniques and languages are applied for this purpose?
• To which extent are model-based techniques successful in Industry 4.0?

Panelist:
Frank Schirmeister (franks@cadence.com), Cadence
Ronald Jancke (roland.jancke@eas.iis.fraunhofer.de), Fraunhofer IIS / EAS
Sara Vinco (sara.vinco@polito.it), Politecnico di Torino
Julio Medina (julio.medina@unican.es), University of Cantabria
Peter Wilson (prw30@bath.ac.uk) Bath University, UK

Moderator:
Franco Fummi (franco.fummi@univr.it), University Verona

1^st Tutorial: Integrating ontological development with software engineering trends

Abstract:

Ontologies are defined in Computer Science as formal, explicit specifications of shared conceptualisations. The discipline of Ontology Engineering can be understood as the one that works on methods, tools and techniques to facilitate the development of ontologies. This discipline has been active for over more than two decades, and has witnessed an important evolution during its lifetime. As ontologies are being increasingly adopted in information systems, the ontological engineering methodologies are moving from heavy-weight and traditional life-cycles to embrace agile and software development practices. In this context, some ontological engineering supporting tools are being developed to integrate and orchestrate activities that are usually carried out repeatedly during ontological engineering projects. This is the case for OnToology, an on-line application that exploits GitHub capabilities to ease collaborative ontology development focusing on ontology documentation, evaluation and publication. This tutorial will cover both methodological and technical aspects of collaborative ontology development. It will include practical activities that will be supported by OnToology in combination with other systems. At the end of the tutorial attendees would be able to carry out and adapt common ontology development activities by incorporating OnToology into their workflows. The examples used in the tutorial will be taken from the VICINITY ontology network. However, the participants can follow the tutorial activities applying the acquired knowledge to their own use cases.

Bio:

Dr. María Poveda-Villalón is a lecturer in the Artificial Intelligence Department of the Universidad Politécnica de Madrid and is also part of the Ontology Engineering Group research lab. Her research activities focus on Ontological Engineering, Ontology Evaluation, Knowledge Representation and the Semantic Web. Previously she finished her studies in Computer Science (2009) by Universidad Politécnica de Madrid, and then she moved to study the Artificial Intelligence Research Master finished in 2010 in the same university. She has worked in the context of Spanish research projects as well as in European projects such as ETSI STF for SAREF extensions, VICINITY (H2020-688467), READY4SmartCities (FP7-608711), and NeOn (FP6-027595). She has contributed to the organization of the "Linked Data in Architecture and Construction Workshop" since 2015 edition, the "13th OWL: Experiences and Directions Workshop and 5th OWL reasoner evaluation workshop" in 2016, the "Linked Energy Data Vocamp" in 2015 and the "Catching up with ontological engineering: To git-commit and beyond" tutorial at EKAW2018. Finally, she is part of the W3C Web of Things Working Group.

Alba Fernández Izquierdo is a Ph.D. Candidate in the Ontology Engineering Group (Universidad Politécnica de Madrid, Spain). She received her Bachelor Degree in Computer Engineering from Universidad de Santiago de Compostela (Spain) in 2015 and her MSc Degree in Artificial Intelligence from Universidad Politécnica de Madrid (Spain) in 2016. She is currently working on ontology testing through ontological requirements. She is focused on ontology quality and ontological collaborative engineering.

- Web page: Maria Poveda
- Web page: Alba Fernandez Izquierdo

2^nd Tutorial: TTool AMS extension

Abstract:

Embedded systems are commonly built upon heterogeneous digital and analog integrated circuits, including sensors and actuators. Model-driven approaches for designing software and hardware have been proposed, yet they are generally limited to the digital parts of systems. This tutorial for the verification and virtual prototyping of embedded systems. These systems are described at different abstraction levels, starting from a SysML-like high-level model down to software design and mapping, from which a cycle-bit accurate prototype is generated. This prototype is based on the SoCLib library of virtual components written in SystemC. In the second part of the tutorial we will show how TTool has been extended to model and simulate analog/mixed-signal systems, and how the simulation handles the resulting synchronization issues. Last but not least, demonstration will be performed on a rover, as a case study, which will first be modeled by a purely digital design, then the sensors’ models will be added and the SystemC-AMS part of the prototype for co-simulation with SoCLib will be generated.

Bio:

Daniela Genius is an assistant professor at Paris Sorbonne University and member of the "Embedded Systems" team at LIP6 laboratory in Paris, France. She obtained her PhD in Karlsuhe, Germany in 2000, then spent six months as a PostDoctoral researcher at the University of Chapel Hill, North Carolina, before joining Philips Research France as a PostDoctoral researcher during two years, in a project with INRIA, dealing with compilation issues for high resolution video processors. Since 2002, she is assistant professor in the System on Chip department of LIP6. Her research interests are virtual prototyping for heterogeneous MP-SoC, hardware/software co-design, and design space exploration for NoC-based multi-processor systems on chip.

- Web page: Daniela Genius