The EAISI Mobility Lab centers its research on three core themes: computer vision, sensor fusion, and world modelling for automated driving.
To drive these advancements, the lab leverages key technical AI components, focusing primarily on:
With a commitment to socially aware and responsible AI systems.
The EAISI Mobility Lab is dedicated to bridging the gap between scientific AI research and the deployment of inherently safe, automated driving applications in real-world environments.
The lab’s mission and vision are to build the next-generation machine-data driven intelligence and systems engineering tools necessary to accelerate the transition toward accident-free mobility. By advancing computer vision, sensor fusion, and world modelling, the lab enables mobile systems to move beyond mere sensing to deeply understand and anticipate complex situations.
The impact of the lab lies in solving key bottlenecks in open, dynamic environments, specifically improving generalization, conformity, and completeness in perception technology. This is achieved by validating scientific breakthroughs alongside NXP Semiconductors, engaging in co-located collaborative research, and advancing explainable, interpretable, and responsible AI frameworks to unlock socio-economic benefits and build trust in next-generation autonomous transportation.
Vulnerable road user path prediction – Focuses on predicting the paths and movements of vulnerable road users (such as pedestrians and cyclists) using deep inverse reinforcement learning.
Deep scene understanding generalizability – Focuses on enhancing the generalizability and completeness of deep scene understanding by utilizing semi-supervised training methods and improved network designs.
Embedded computational efficiency – Focuses on improving the computational efficiency of deep scene understanding models for embedded platforms using binary neural network accelerators.
Interpretable deep scene understanding – Focuses on improving the conformity and interpretability of deep scene understanding by applying graph neural networks and differentiable reasoning.
Safety estimation & path planning in mixed traffic – Focuses on improving on-road safety estimation and vehicle path planning by integrating prediction models of other road users to enable safer autonomous driving in mixed traffic environments.
Eindhoven Artificial Intelligence Systems Institute (EAISI) is an institute of Eindhoven University of Technology in the field of AI.
NXP Semiconductors is a Dutch-American semiconductor manufacturer for the secure identification, automotive and digital networking industries.
