Advertisement

ATZelectronics worldwide

, Volume 14, Issue 3, pp 26–29 | Cite as

Aleatory Function Validation - AI-based Method for the Validation of Driver Assistance Systems

  • Josef BaumgartnerEmail author
  • René Honcak
Cover Story
  • 1 Downloads

The ASAP Group has developed a validation method for complex system functions, such as driver assistance systems, based on reinforcement learning: the aleatory function validation. The use of artificial intelligence also allows the comprehensive, time-saving validation of even complex effect chains with interactions between various control units. Self-learning algorithms specifically seek out critical stimulations which could lead to errors in the effect chain - amongst other things an important step in the direction of autonomous driving.

Scenario

Automotive manufacturers and their development partners, including the ASAP Group, are currently focussing on US regulation FMVSS111. It makes reversing cameras mandatory in all new cars sold in the USA as of May 2018. One critical requirement is that the image from the reversing camera must be displayed no later than 2 s after the reverse gear is engaged and at no point must other images be superimposed over it. The validation of...

Copyright information

© Springer Fachmedien Wiesbaden GmbH, ein Teil von Springer Nature 2019

Authors and Affiliations

  1. 1.ASAPMünchenGermany

Personalised recommendations