Ralph Grewe

Dr. Ralph Grewe started his career at Continental in 2011 with a PhD position in the field of environment modeling at the Advanced Engineering of Systems & Technology in Lindau. In 2014 he moved to Continental Frankfurt where he took over the position as group leader for perception and environment model in 2016. In 2018 he became the Domain Owner and in 2022 Product Owner for CEM and Localization Domains in the Program Automated Driving. Today he Release Train Architect at Innovation Line Driverless.

---

Presentation

Evaluating End2End AD for the Real World

Building a fully trainable AD stack is a promising approach to efficiently build and scale an AD system. Transformer Neural Networks using tokenization of diverse information into a common representation and attention mechanisms to efficiently combine data in time and space foster understandable and flexible End2End architectures. An enabler to follow the fast-evolving research in End2End is a toolchain allowing quick application of such Architectures from Training over simulation into closed-loop real world vehicles. We give an overview of the Toolchain and present insights we gained from tests in simulation and closed-loop application in the real world.