Devices with sensing, actuating and computational capabilities are nowadays embedded in all kinds of physical systems (such as smart grids), making our living and working places smarter. However, engineering those systems – which interlace physical and software components – and ensuring performance and reliability requirements is a challenging task. The goal of the Visual Analytics for Engineering Smarter Systems (VAESS) research area is to explore and generate new knowledge to better understand and engineer such complex cyber-physical systems, based on the data collected from the systems and their environments in order to better predict and control their performance and behavior.
Visual analytics (VA) enables us to analyze those large and complex information spaces that are collected from cyber-physical systems. Various types of models (e.g., statistical models or machine learning algorithms) are used to examine the collected data and to predict/control the system behavior. Those models even depend on suitable parameter settings and might be integrated into a larger context. All these aspects need to be clearly understood by human experts in order to make those systems better. For this and for providing reliable trust into the models, we will develop foundational visual analytics principles, techniques and tools for analyzing both data and models.
Concrete research questions and aims are, for instance:
- What are the concrete tasks and aims of the analyses?
- What is the overall analysis process (workflow) and how can it be described?
- How is this process reflected in the future VA tools to be developed?
- How to make the black boxes of the analysis methods and resulting models transparent for the user (for increasing trust into them)?
- Build novel or combine existing visual representations and interaction techniques that support better understanding of the data/models to solve the defined tasks. These approaches should scale for large data sets and improve the trustworthiness of the analysis results.
- Develop VA solutions for model curation, simulation result analysis, and prediction.
- Validate the VA approaches qualitatively and quantitatively.
For doing this research, we combine the expertise of people coming from several fields: from interactive visualization and visual analytics to machine learning and individual domain knowledge. We are open to collaborate with external partners who provide challenging data sets from cyber-physical systems and interesting questions.
Visual Analytics for Engineering Smarter Systems is an application area of the Linnaeus University Centre of Excellence (LNUC) for Data Intensive Sciences and Applications.
- Andreas Kerren Professor
- Angelos Chatzimparmpas Doctoral student
- Ilir Jusufi Senior lecturer
- Marcelo Milrad Professor, Vice-dean
- Rafael Messias Martins Postdoctoral Fellow