Potential of AI for Highly Automated Driving Functions

Authors

Peter Gronerth, MSc., Julian Bock, MSc., Univ.-Prof. Dr.-Ing. Lutz Eckstein,
Institute for Automotive Engineering, RWTH Aachen University, Aachen;
Benjamin Altpeter, MEng.,
fka Forschungsgesellschaft Kraftfahrwesen mbH Aachen, Aachen

Summary

This paper provides an introduction into the topic of Artificial Intelligence, achieved by using Neural Networks and their potential applications in highly automated driving. This includes an overview on their key concepts, as well as an on current usages found in literature and other sources.

Furthermore, an approach to the creation of semi artificial datasets is presented. Training results of different sets, both newly created and already publicly available against one another are given. This is done to show the importance of correct training data as well as the potential of our approach to increase the amount of such data.

Members Login