The team Autonomous Driving (AD) team conducts research on AI-based environment perception and trajectory planning for autonomous vehicles (Level 4 >).
We consider both subsymbolic AI-techniques used in learning systems (machine learning, deep learning) as well as symbolic techniques such as reasoning or constraint-based methods.
With respect to learning systems, we work with both real and synthetic data in an idealized implementation of the Digital Reality Principle, which is the thematic guideline of the research area ASR.
In addition to our own research contributions (see below), we cooperate with other ASR teams and apply the findings to autonomous driving, such as human behavior models and hybrid deep learning. This cooperation is currently taking place primarily within the framework of the research project REACT.
In addition, the team is part of the Autonomous Driving Competence Center, which is also headed by Dr. Ing. Christian Müller. We also cooperate with our external partners as part of joint projects.
Our own technical / scientific contributions focus on the following core topics:
- radar simulation
- context-dependent sensor fusion (camera, lidar, radar) at the feature level using a learning system (e.g., deep learning
target-oriented generation and preprocessing of synthetic data for camera, lidar and radar.
- adaptation of synthetic data (domain adaptation). This includes the complete removal or reduction of biases.
- learning with explicit ethics models (in external cooperation)
- AI for the protection of autonomous systems (validation and verification) in cooperation with TÜV Süd.