Characterizing Within-Driver Variability in Driving Dynamics During Obstacle Avoidance Maneuvers

Kendric R. Ortiz, Adam J. Thorpe, AnaMaria Perez, Maya Luster, Brandon J. Pitts, Meeko OishiarXiv 2022 · DOI · arXiv

Abstract

This study presents a data-driven framework for quantifying within-driver variability during complex driving tasks. By leveraging Kernel Distribution Embedding and the Maximum Mean Discrepancy (MMD) metric, we characterize individual learning profiles in a simulated obstacle-avoidance scenario. Our results demonstrate that driving trajectories can be mapped into a high-dimensional functional space to identify patterns of convergence, consistency, or disengagement, providing a foundation for more personalized human-centric autonomous systems.

Introduction

Autonomous and driver-assistance systems (ADAS) typically target the "average" human driver. However, human behavior is inherently inconsistent, exhibiting significant variability not only between individuals but also within a single driver across repeated trials. Traditional control-theoretic models often fail to capture this "naturalistic" decision-making because they rely on rigid assumptions like the Markov property or Gaussian noise. To develop truly human-centric autonomy, systems must distinguish between a driver becoming more proficient, remaining erratic, or losing focus.

Methodology

We propose a nonparametric machine learning approach to quantify driving variability:

  1. Kernel Distribution Embedding: Driving trajectories (comprising position, velocity, and heading) are treated as realizations of a stochastic process and projected into a Reproducing Kernel Hilbert Space (RKHS). This allows a complex trajectory to be represented as a single "embedding" point.
  2. Maximum Mean Discrepancy (MMD): We utilize MMD as a distance metric to calculate the separation between probability distributions within the RKHS.
  3. Convergence Benchmarking: For each participant, a "final distribution" is defined using the last four trajectories of the experiment. The MMD is then calculated between each initial trajectory and this benchmark to track convergence toward a consistent driving style.
Figure 1: Conceptual diagram of the obstacle avoidance task phases, including normal driving and adjusted maneuvers.

Experimental Setup

A pilot study was conducted with six senior engineering students using an advanced driving simulator. Participants drove at 45 mph on a nighttime rural highway and encountered a hidden obstacle (a tire) behind a blind S-curve. To induce a learning curve and increase task difficulty, participants were required to steer using only their non-dominant hand. Data collected included 12 trajectories per participant and eye-tracking fixation durations.

Results and Discussion

The MMD metric effectively visualized "learning" as a decrease in distance to the final benchmark over time.

Figure 2: Maximum Mean Discrepancy (MMD) across eight trajectories for six participants, illustrating distinct learning profiles.

The study identified three distinct behavioral profiles:

  • Learners (Participants 1, 3, 4): Showed a steady decrease in MMD, indicating successful adaptation and increased consistency.
  • Consistent/Ceiling Group: Exhibited flat MMD curves, suggesting they either reached peak performance immediately or lacked motivation for further improvement.
  • Disengaged (Participant 2): Showed a "U-shaped" curve. Post-experiment observation revealed this participant switched to two-handed driving against instructions before losing focus.

The Eye-Tracking Paradox

Correlating MMD with eye-tracking data revealed divergent strategies. Some "Learners" reduced road fixation as they became more efficient, while others increased fixation to maintain focus. This suggests that ADAS monitoring eye-tracking alone may misinterpret a driver's state without considering individual strategy differences.

Conclusion

This research demonstrates that MMD is a sensitive metric for capturing the interplay between physical action and cognitive attention. While limited by a small sample size, the framework provides a path toward real-time, customized interventions. Future systems could provide steering assistance to "Learners" showing high variability while withholding it once the driver's MMD stabilizes, thereby improving user acceptance of safety technologies.

Summary updated

Abstract

Variability in human response creates non-trivial challenges for modeling and control of human-automation systems. As autonomy becomes pervasive, methods that can accommodate human variability will become paramount, to ensure efficiency, safety, and high levels of performance. We propose an easily computable modeling framework which takes advantage of a metric to assess variability in individual human response in a dynamic task that subjects repeat over several trials. Our approach is based in a transformation of observed trajectories to a reproducing kernel Hilbert space, which captures variability in human response as a distribution embedded within the Hilbert space. We evaluate the similarity across responses via the maximum mean discrepancy, which measures the distance between distributions within the Hilbert space. We apply this metric to a difficult driving task designed to elucidate differences across subjects. We conducted a pilot study with 6 subjects in an advanced driving simulator, in which subjects were tasked with collision avoidance of an obstacle in the middle of the road, around a blind corner, in a nighttime scenario, while steering only with the non-dominant hand.
BibTeX
@misc{ortiz2022characterizing,
  author = {Kendric R. Ortiz and Adam J. Thorpe and AnaMaria Perez and Maya Luster and Brandon J. Pitts and Meeko Oishi},
  title = {{Characterizing Within-Driver Variability in Driving Dynamics During Obstacle Avoidance Maneuvers}},
  year = {2022},
  eprint = {2206.01331},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2206.01331},
}
Figure