The junior research group “Computational Interaction and Mobility” (CIAO) wants to develop computationally rational models that interact with technology in a human-like manner and make human-like decisions. For that reason, the group will build these models using reinforcement learning. User interfaces shall be evaluated by using them. Additionally, a better understanding on how humans and intelligent systems can collaborate shall be created. Especially, this junior research group is focusing on following key topics:

• Analysis of large naturalistic driving data
• Driver distraction research
• Computational models of human-computer Interaction
• Data-driven evaluation of user interfaces

In fact, the importance of the research lies in the necessity to concentrate on the needs of users. To create systems and interfaces that are enjoyable to use, the design process must focus on the user. Nevertheless, the quick advancement of technology makes it more and more difficult (and costly) to conduct studies with actual users. By creating user models that can effectively simulate human interaction, the junior research group can enhance our understanding of interaction behavior. Besides, it will be possible to automate the evaluation process to create interactive systems that meet user needs and are user-friendly.

Projects

Currently, the group is working on computational models of multitasking while driving. Furthermore, they will build a mixed-reality driving simulator that will allow to evaluate the models that the group develops.

Additionally, together with colleagues from institutions in Finland, the Netherlands, and the UK, they are in the first planning steps to apply for Marie Skłodowska-Curie Doctoral Network.

In general, the group is always interested in collaboration. On the one hand, they would like to further collaborate with automotive OEMs. As a result, they could push the boundaries of data-driven decision-making in the design process of automotive user interfaces. On the other hand, computational models of user interaction and decision making play a vital role in the whole mobility area. This junior research group sees a huge potential of these models to be applied in the design of mobile devices such as smartphones. Whereas smartphones are designed to be used in single task scenarios, they are constantly used in multitasking scenarios. Here, user models that simulate our interaction behavior could be of great use to better understand how humans allocate their attention between the competing tasks.

Team

Lead

The junior research group leader, Patrick Ebel, started at ScaDS.AI Leipzig on July 1st 2023. Ebel studied Mechatronics in Karlsruhe for his Bachelor’s and Automotive Systems with focus on machine learning in Berlin for his Master’s. He started his PhD in Computer Science at the TU Berlin and the University of Cologne.

Previously, Ebel worked at the University of Cologne at the Chair of Software and Systems Engineering together with Prof. Andreas Vogelsang. In 2023, Ebel defended his dissertation “On the forces of driver distraction: Explainable predictions for the visual demand of in-vehicle touchscreen interactions” with summa cum laude. For his dissertation, Ebel won the Early Career Best Paper Award from Human Factors and Ergonomics Society Europe. Back in 2019, Ebel was placed 4th at the research competition “Forum junge Spitzenforscher”.

Ebel’s research focus is concentrated at the intersection of machine learning and human-computer interaction. The main goal is to develop computational models that simulate human-like interaction behaviour. Currently, the focus of his work is on the interaction between drivers and other road users with (semi-)automated cars.

Publications

By now, no papers have been publicated by the junior research group.