We warmly congratulate our colleagues Laura Gregorc and Kailin Tong on receiving the Hans List Fonds 2024, which was awarded to them for their outstanding scientific work. The ceremony took place on October 4, 2024, at AVL List GmbH in Graz.

Laura Gregorc successfully completed her master’s thesis titled “In-Cabin Gaze Detection using Neural Networks on Mobile Devices.” Under the supervision of Mag. Dr. Andreas Windisch from FH JOANNEUM and Elem Güzel Kalayci from VIRTUAL VEHICLE, she developed an innovative system for in-cabin gaze detection. Her approach uses a combination of Convolutional Neural Networks and a Vision Transformer to determine the gaze vector.
This research was conducted as part of the DriveIns project, which aims to proactively detect driver distractions to prevent accidents and improve communication with other vehicles.
The models were executed directly on a smartphone via an app, ensuring that sensitive camera data remains on the device and protecting privacy.

This work represents a significant advancement and offers a practical solution for enhancing road safety.

Procedure on the edge device.

We also congratulate Kailin Tong for his successful dissertation “Safe Motion Planning for Autonomous Vehicles“.
His work introduces two key innovations in the field of autonomous driving: The research presents a trajectory repairing framework that improves the way autonomous vehicles respond to changes in the behavior of other vehicles. Typically, autonomous vehicles react immediately to any change, which can lead to unnecessary braking or steering. With the Trajectory Repairing framework, the vehicle monitors behavior changes that could lead to a potential collision. Instead of reacting immediately, it calculates the maximum time available to react safely.

This allows the vehicle to observe the behavior of other vehicles during this time, allowing it to make a more informed decision and reduce unnecessary braking or steering.

Additionally, he focused on risk assessment for autonomous vehicles, developing detailed risk metrics and assessment frameworks. Simulations demonstrated that this method is more efficient, robust, and safer than conventional approaches.
This was tested in the EU project ArchitectECA2030 Demonstrator, and it effectively manages sensor and perception failures, builds confidence in the real-world deployment of autonomous vehicles and reduces potential hazards.

Both scientific works make significant contributions to the advancement of mobility technologies. We warmly congratulate Laura and Kailin on their impressive achievements and wish them continued success in their future endeavors!