Niclas Vödisch

Niclas Vödisch is a Ph.D. student and researcher at the Autonomous Intelligent Systems Lab in the Department of Computer Science at the University of Freiburg. Supervised by Prof. Dr. Wolfram Burgard and co-supervised by Prof. Dr. Abhinav Valada, he is actively contributing to the field of robotics and AI as a member of the ELLIS Society. His research focuses on enhancing machine perception and SLAM systems using deep learning methods, with primary applications in mobile robotics and autonomous driving.

Education:

• Ph.D. Student at University of Freiburg (June 2021 - June 2025, thesis submitted, defense pending)
• Visiting Ph.D. Student at University of Zurich (June 2024 - December 2024)
• M.Sc. in Computational Science and Engineering at ETH Zurich (September 2018 - May 2021)
• Visiting Undergraduate Student at Carnegie Mellon University (August 2016 - May 2017)
• B.Sc. in Computational Engineering Science at RWTH Aachen University (September 2014 - June 2018)

Vödisch's research interests center around Continual Learning for Robotics, Machine Perception, and Simultaneous Localization and Mapping (SLAM). His work addresses the critical challenge of reducing dependency on extensive annotated training data in robotic perception systems. Through innovative approaches leveraging foundation models, he has developed methods that achieve high performance with minimal supervision, making robotic systems more adaptable to real-world environments. His research spans theoretical advances in deep learning and their practical implementation in autonomous systems.

An analysis of his publication record reveals a clear progression from foundational SLAM techniques to sophisticated continual learning frameworks and foundation model applications. A unifying theme across his work is data efficiency in robotic perception, with increasing emphasis on collaborative multi-agent systems and cross-modal information integration for robust performance in challenging conditions. His recent publications demonstrate growing influence in the robotics community, evidenced by the IROS 2024 Best Paper Award for his BEVCar work.

Scientific Awards:

• IROS 2024 Best Paper Award on Cognitive Robotics (BEVCar)
• IROS 2024 Best Student Paper Award Finalist (BEVCar)
• Dean's List at Carnegie Mellon University (fall 2016)
• DAAD full scholarship for CMU studies (2016-2017)

Niclas has mentored numerous Master's students on projects spanning LiDAR panoptic segmentation, collaborative scene graph generation, and autonomous driving systems. His teaching portfolio includes co-organizing seminars on Robot Learning and Learning with Limited Supervision, as well as leading the FreiCAR practical autonomous driving course across multiple semesters. His research has received substantial funding from the German Research Foundation (DFG) Emmy Noether Program, NVIDIA academic grants, and Qualcomm Technologies Inc., reflecting the significance and potential impact of his work.

As a key contributor to the Autonomous Intelligent Systems group at Freiburg, Vödisch collaborates closely with the Robot Learning group and has extended his research network through his visiting position at the University of Zurich's Robotics and Perception Group. His interdisciplinary approach bridges computer vision, robotics, and machine learning, positioning him at the forefront of research in AI-powered autonomous systems with practical real-world applications.