MEET THE SAFE-UP PARTNERS

Eindhoven University of Technology (TU/e)

Who belongs to your SAFE-UP team?

TU/e’s SAFE-UP team consists of researchers from the Control Systems Technology group of the Mechanical Engineering Department and the director of the Responsible Mobility Lab. Manuel Muñoz Sánchez is focusing on deep learning applications for the classification, prediction, and hybridization of AI methods with system knowledge, such as autonomous driving; Jos Elfring specialises in world modelling for autonomous systems; René van de Molengraft’s area of expertise is world modelling and control for autonomous robots, and Margriet van Schijndel manages the EAISI Mobility lab, linking running and new projects to international initiatives. She’s also a member of the Executive Board of the CCAM Partnership.

What does TUE bring to the project?

TU/e’s contribution to SAFE-UP is mostly within work package 3, where we are developing prediction models for vulnerable road users (VRUs) - such as pedestrians and cyclists - to anticipate any safety-critical situation and react in the safest manner possible. Training and optimising motion prediction models in potentially dangerous situations is challenging, due to the frequent lack of poor data and the complexity of the patterns and interactions. The ambition is to have better prediction models that not only look at the VRU but take into account the interaction between road users and context information (environment, roads, etc.), and are accurate in more complex on-road scenarios like bad weather.

Why did you want to take part in SAFE-UP?

TU/e is committed to making mobility safer for all users. In fact, our Responsible Mobility roadmap has a clear focus on this: “With AI working real-time to support the driver in critical situations, we believe in inherently safe vehicles using machine learning, world modelling, optimized human-machine cooperation, new vision and radar systems, and ultrafast actuators. The goal within TU/e is to significantly contribute to speed up the transition towards an accident-free and zero emissions mobility system”. The combination of passive and active safety, in combination with the future scenarios within SAFE-UP, can have a clear contribution, as was also shown by several of the project’s demonstrators. The future scenarios and the safety-critical situations they cover also provide essential insights for our other related activities.

Which other initiatives are you involved in?

Eindhoven University of Technology (TU/e) is involved in many EU projects and national research initiatives on Mobility and Artificial Intelligence. TU/e has also built the EAISI Mobility lab (EAISI Mobility Lab (tue.nl)), a joint program of NXP Semiconductors and TU/e to research novel perception and world modelling technologies for automated driving. For example, we completed the nationally-funded research project i-Cave, in which automated and cooperative driving technologies were developed, taking into account both psychological and social aspects. (i-Cave: the future of moving forward (tue.nl)). In the EU project C-MobILE, C-ITS solutions are investigated to improve safety and efficiency, especially in urban scenarios. Various cases have been investigated for cyclist-to-bus interactions, with simulations focusing on intersections known to be dangerous and complex, where around 30 to 40 buses cross the path of about 1000 cyclists. As road safety is at the core of our research roadmaps, we are actively contributing to the roadmaps of, for example, the European Technology Platform ERTRAC, on Safety, Urban Mobility, and Connected & Automated Driving.