Fostering young female researchers from the Girls' Day Academy Dresden

Where does data come from in the world of transportation, and what do you do with it? What is transportation science, and what does the work of a transportation scientist look like? 

On December 7, 2022, our chair invited interested young female researchers to answer these and other questions and get involved in

• machine learning (ML),
• traffic research,
• route planning, and
• autonomous driving.

At the event, the female students from the Girls' Day Akademie Dresden worked with our team to train an ML model to predict shoe size based on body size and weight. The students analyzed influencing factors for their choice of transportation on their way to school. They also investigated how data analytics can help better understand urban transportation and make it more environmentally friendly. 

What does it mean to study transportation?

Anke Richter-Baxendale (Public Relations & Communications, Deanery of the "Friedrich List" Faculty of Transport and Traffic Sciences) and Lukas Unterschütz (Master's student and student assistant at the Chair of Big Data Analytics in Transportation) shared their perspectives on studying at the faculty -- from a content and student perspective. 

After this short interlude, the students planned an optimal route through Dresden to visit different sights. For this purpose, they interactively dealt with the traveling salesperson problem, as it is used by parcel delivery companies, for example. 

Finally, the participants competed in a car racing simulation against an autonomously driving car, which they defeated already on the first try. Reinforcement learning (an emerging branch from the research domain of artificial intelligence) was used for the simulation, in which the car gradually learned to drive solely by exploring the environment. Similar to human behavior: The longer the car was allowed to practice driving, the better the expected driving skill.  Only an advanced model with many thousands of practice laps outperformed the students. In this way, they demonstrated in practice how efficiently people could learn and transfer their knowledge to new tasks. 

We want to thank the participants and hope to have spurred interest in our work!