Wolfram Burgard is a Professor of Robotics and Artificial Intelligence at the University of Technology Nuremberg. Before this, he was a Professor of Autonomous Intelligent Systems at the University of Freiburg and also Vice President for Automated Driving Technology at Toyota Research. His areas of interest lie in artificial intelligence and mobile robots. His research mainly focuses on the development of robust and adaptive techniques for robot state estimation and control. Over the past years, his group has developed a series of innovative probabilistic and machine-learning techniques for robot navigation and perception. Wolfram Burgard has coauthored over 400 publications in robotics and artificial intelligence. He is a Fellow of the European European Association for Artificial Intelligence (EurAI), the Association for the Advancement of Artificial Intelligence (AAAI), and the Institute of Electrical and Electronics Engineers (IEEE). In 2009, he received the Gottfried Wilhelm Leibniz Prize, the most prestigious German research award. In 2010, he earned an Advanced Grant from the European Research Council. He is a member of the Heidelberg Academy of Science and the National Academy of Science Leopoldina. In 2022, he received the Technical Field Award of the IEEE Robotics and Automation Society.
KEYNOTE: Probabilistic and Machine Learning Approaches for Autonomous Robots and Automated Driving
For autonomous robots and automated driving, the capability to robustly perceive their environments and execute their actions is the ultimate goal. The key challenge is that no sensors and actuators are perfect, which means that robots and cars need the ability to properly deal with the resulting uncertainty. In this presentation, I will introduce the probabilistic approach to robotics, which provides a rigorous statistical methodology to solve the state estimation problem. I will furthermore discuss how this approach can be extended using state-of-the-art technology from machine learning to bring us closer to the development of truly robust systems that serve us in our everyday lives.
Dengxin Dai is a Senior Researcher at the MPI for Informatics, heading the research group Vision for Autonomous Systems. In 2016, he obtained his Ph.D. in Computer Vision at ETH Zurich and has been a group leader from 2016 to 2021 working on autonomous driving. His research interests lie in autonomous driving, robust perception in adverse weather and illumination conditions, domain adaptation, sensor fusion, multi-task learning, and object recognition under limited supervision. He has been area chair of multiple major computer vision conferences (e.g. CVPR21, CVPR22, ECCV22), has organized multiple international workshops, is on the Editorial Board of IJCV, and is an ELLIS member. His team has won multiple awards including 1st Place at Waymo Open Dataset Challenge 2022 and 2nd Place at nuScenes Tracking Challenge 2021. He has received the Golden Owl Award at ETH Zurich in 2021 for his exceptional teaching.
KEYNOTE: Adaptable and Scalable Machine Perception for Autonomous Driving
While steady progress has been made in machine perception, the performance is mainly benchmarked under good weather and favorable lighting conditions. Even the best performing algorithms on the existing benchmarks can become untrustworthy in a new domain or under adverse conditions. The ability to robustly cope with new domains such as bad weather or lighting conditions is absolutely essential for robotic applications such as autonomous driving. In this talk, I will first present our work on semantic scene understanding under adverse weather conditions and under general unseen domains. This involves multiple methods on the topic including weather phenomenon simulation, incremental domain adaptation, novel network architectures, and new training strategies. I will then present a few other recent works on reducing human annotations for scalable machine perception. This includes supervision transfer across modalities and learning with weak supervision. Our methods contain multiple contributions to increasing the adaptability and scalability of visual perception systems.
Jan Lange is a Senior Defensive Security Expert at CARIAD SE, a Volkswagen Group Company. He is currently focused on building an end-to-end solution for defensive security operations spanning both the vehicle fleet and connected backend systems, bringing much-needed defensive security capabilities into the automotive world. Before joining CARIAD in January 2021 he worked for 5 years in the insurance industry for Allianz Deutschland’s Incident Response and Vulnerability Management teams, one of Germany’s largest companies. Parallel to his employment at Allianz he completed his master’s degree in IT Security, focusing on memory forensics in his thesis and contributing to the Volatility memory analysis framework. He is a committed member of the Blue Team by heart and is supporting CARIAD’s vision to build the most secure mobility products.
KEYNOTE: How to defend connected intelligent vehicles: Transferring established Information Security best practices to the vehicular world
With each passing year vehicles get closer and closer to “normal” IT devices such as computers and smartphones, forcing the automotive industry’s Information Security capabilities to improve as well. The same threat landscape enterprise IT has had to deal with for decades is now slowly expanding its attention towards automotive mobility products and the likelihood of sophisticated attacks against vehicles and their infrastructure will only increase. There is much to learn from the methods and controls used by “classical” enterprise IT, ranging from attack detection and Incident Response over Cyber Threat Intelligence to active defense combining automated response with human interaction. Based on our experiences in the automotive industry we will share our vision of secure mobility, our strategy on how to achieve it, and the challenges we have faced during its implementation so far.
