Andreas Look

Andreas Look

Professor for Autonomous Systems · Coburg University of Applied Sciences

About

I am a Professor for Autonomous Systems at Coburg University of Applied Sciences (HS Coburg). My research is driven by the vision of creating intelligent, safe, and scalable autonomous vehicles. I focus on the intersection of Artificial Intelligence and Robotics, specifically addressing the challenges of operating in highly dynamic, unpredictable, and partially observable environments.

To achieve this, my work explores the frontiers of Reinforcement Learning, Trajectory Prediction, and Uncertainty Modeling. By leveraging modern machine learning techniques—from information-theoretic approaches to the integration of foundation models—my goal is to develop algorithms that don't just succeed in simulation, but can safely handle complex real-world edge cases such as heavy occlusions and stochastic human behaviors.

My academic philosophy is deeply informed by my industry background. Before joining HS Coburg, I was a Research Scientist at the Bosch Center for Artificial Intelligence (BCAI). There, I worked on the full pipeline of applied AI, translating cutting-edge theoretical research into robust software for automotive systems. This experience cemented my core belief: true algorithmic breakthroughs must be validated through real-world deployment on actual hardware.

Today, I am dedicated to advancing this research frontier while educating the next generation of robotics and AI engineers. I am always open to academic collaborations, industry partnerships, and discussions with motivated students.

Interests

Autonomous Driving Prediction & Motion Planning Uncertainty Modeling Bayesian Deep Learning Reinforcement Learning End-to-End (E2E) Models Robotics Real-World Deployment

Selected Publications

Uncertainty Matters paper preview
Uncertainty Matters: Structured Probabilistic Online Mapping for Motion Prediction in Autonomous Driving Preprint
P. Gogoi, F. Janjos, B. Yang, A. Look · 2026
Don't double it paper preview
Don't double it: Efficient Agent Prediction in Occlusions Preprint
A. Rothenhäusler, M. Mazzola, A. Look, R. Rajan, J. Bödecker · 2026
Stochasticity in Motion paper preview
Stochasticity in Motion: An Information-Theoretic Approach to Trajectory Prediction IROS 2025
A. Distelzweig, A. Look, E. Kosman, F. Janjoš, J. Wagner, A. Valada
Motion Forecasting paper preview
Motion Forecasting via Model-Based Risk Minimization ICRA 2025
A. Distelzweig, E. Kosman, A. Look, F. Janjoš, D. K. Manivannan, A. Valada
Can you text paper preview
Can you text what is happening? Integrating pre-trained language encoders into trajectory prediction models IROS Workshop
A. Keysan, A. Look, E. Kosman, G. Gürsun, J. Wagner, Y. Yao, B. Rakitsch