Beyond Grand Theft Auto V for Training, Testing and Enhancing Deep Learning in Self Driving Cars

@article{Martinez2017BeyondGT,
  title={Beyond Grand Theft Auto V for Training, Testing and Enhancing Deep Learning in Self Driving Cars},
  author={Mark Martinez and Chawin Sitawarin and Kevin Finch and Lennart Meincke and Alex Yablonski and Alain L. Kornhauser},
  journal={CoRR},
  year={2017},
  volume={abs/1712.01397}
}
As an initial assessment, over 480,000 labeled virtual images of normal highway driving were readily generated in Grand Theft Auto V's virtual environment. Using these images, a CNN was trained to detect following distance to cars/objects ahead, lane markings, and driving angle (angular heading relative to lane centerline): all variables necessary for basic autonomous driving. Encouraging results were obtained when tested on over 50,000 labeled virtual images from substantially different GTA-V… CONTINUE READING
Recent Discussions
This paper has been referenced on Twitter 30 times over the past 90 days. VIEW TWEETS

References

Publications referenced by this paper.
Showing 1-10 of 19 references

Tesla’s self-driving system cleared in deadly crash, Jan 2017

  • Neal E. Boudette
  • 2017
1 Excerpt

Uber suspends tests of self-driving vehicles after arizona crash, Mar 2017

  • Mike Isaac
  • 2017
1 Excerpt

End - toend learning of driving models from large - scale video datasets

  • Trevor Darrell.
  • 2016

Autonomous cars through the ages

  • Tom Vanderbilt
  • 2012
1 Excerpt

Similar Papers

Loading similar papers…