Advertisement

ATZ worldwide

, Volume 121, Issue 12, pp 46–49 | Cite as

Machine Learning for Automated Driving

  • Peter Schiekofer
  • Yusuf Erdogan
  • Stefan Schindler
  • Markus Wendl
Development

Automated and autonomous driving are among the core topics of future mobility. In its innovation project "Park and Charge", Bertrandt shows how important environment recognition and precise trajectory planning using artificial intelligence are. Machine learning is used to improve localization, networking and cloud applications.

Current status of object detection

Autonomous driving functions are already available in some production models. They allow the car to fulfill the driver's role in a range of situations, but they must be able to take reliable decisions very quickly. As a result, the use of Artificial Intelligence (AI) and, in particular, of Machine Learning (ML) is essential. The decisions are made by pre-trained deep learning algorithms and the input consists of measurements from a variety of sensors, including camera, radar, and lidar systems. The main function of the algorithms is to reconstruct the car's environment using object recognition and to make this information...

Copyright information

© Springer Fachmedien Wiesbaden GmbH, part of Springer Nature 2019

Authors and Affiliations

  • Peter Schiekofer
    • 1
  • Yusuf Erdogan
    • 2
  • Stefan Schindler
    • 2
  • Markus Wendl
    • 3
  1. 1.BertrandtEhningenGermany
  2. 2.BertrandtFrankfurt am MainGermany
  3. 3.BertrandtKölnGermany

Personalised recommendations