The study outline of our ICRA 2021 study.

The work resulting from the collaboration with my co-lead author Jesper Karlsson on encoding human driving styles in motion planning for autonomous vehicles has been accepted to ICRA2021. It extends recent prior work by Jesper Karlsson, Fernando Barbosa, and Jana Tumova [1].

We present a new approach to encode human driving styles through the use of signal temporal logic and its robustness metrics. Specifically, we use a penalty structure that can be used in many motion planning frameworks, and calibrate its parameters to model different automated driving styles. We combine this penalty structure with a set of signal temporal logic formula, based on the Responsibility-Sensitive Safety (RSS) model, to generate trajectories that we expected to correlate with three different driving styles: aggressive, neutral, and defensive.

Contributions of this work:

  • We provide STL formulas for the RSS model [2] as a specification to encode perceived driving styles;
  • We show that parameterizing the cost function results in different driving behaviors that are perceived as distinct driving styles
  • We show that people prefer different AV generated motion styles, which correlates to their self-reported driving style.

[1] Karlsson, J., Barbosa, F. S., & Tumova, J. (2020). Sampling-based motion planning with temporal logic missions and spatial preferences. In IFAC, International Federation of Automatic Control virtually from Tuesday to Friday, May 25-28, 2021.

[2] Shalev-Shwartz, S., Shammah, S., & Shashua, A. (2017). On a formal model of safe and scalable self-driving cars. arXiv preprint arXiv:1708.06374.

@inproceedings{karlsson2021encoding, title={Encoding Human Driving Styles in Motion Planning for Autonomous Vehicles}, author={Karlsson, Jesper and van Waveren, Sanne and Pek, Christian and Torre, Ilaria and Leite, Iolanda and Tumova, Jana}, booktitle={ICRA International Conference on Robotics and Automation}, year={2021}, }