Efficient and Safe Vehicle Navigation Based on Driver Behavior Classification

Abstract

We present an autonomous driving planning algorithm that takes into account neighboring drivers' behaviors and achieves safer and more efficient navigation. Our approach leverages the advantages of a data-driven mapping that is used to characterize the behavior of other drivers on the road. Our formulation also takes into account pedestrians and cyclists and uses psychology-based models to perform safe navigation. We demonstrate our benefits over previous methods: safer behavior in avoiding dangerous neighboring drivers, pedestrians and cyclists, and efficient navigation around careful drivers.

Cite

Text

Cheung et al. "Efficient and Safe Vehicle Navigation Based on Driver Behavior Classification." IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, 2018. doi:10.1109/CVPRW.2018.00149

Markdown

[Cheung et al. "Efficient and Safe Vehicle Navigation Based on Driver Behavior Classification." IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, 2018.](https://mlanthology.org/cvprw/2018/cheung2018cvprw-efficient/) doi:10.1109/CVPRW.2018.00149

BibTeX

@inproceedings{cheung2018cvprw-efficient,
  title     = {{Efficient and Safe Vehicle Navigation Based on Driver Behavior Classification}},
  author    = {Cheung, Ernest and Bera, Aniket and Manocha, Dinesh},
  booktitle = {IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops},
  year      = {2018},
  pages     = {1024-1031},
  doi       = {10.1109/CVPRW.2018.00149},
  url       = {https://mlanthology.org/cvprw/2018/cheung2018cvprw-efficient/}
}