Challenges of Data-Driven Simulation of Diverse and Consistent Human Driving Behaviors

TL;DR

This paper introduces a mathematical framework for creating more realistic data-driven simulations of human driving behavior, emphasizing diversity and consistency to improve autonomous vehicle testing environments.

Contribution

The paper presents a novel mathematical framework for data-driven human driving behavior modeling, enhancing realism over traditional physics-based simulations.

Findings

Validation with NGSIM dataset supports the importance of diversity and consistency in behavior modeling.

The proposed model outperforms physics-based models in realism.

Increased realism improves autonomous vehicle simulation testing.

Abstract

Building simulation environments for developing and testing autonomous vehicles necessitates that the simulators accurately model the statistical realism of the real-world environment, including the interaction with other vehicles driven by human drivers. To address this requirement, an accurate human behavior model is essential to incorporate the diversity and consistency of human driving behavior. We propose a mathematical framework for designing a data-driven simulation model that simulates human driving behavior more realistically than the currently used physics-based simulation models. Experiments conducted using the NGSIM dataset validate our hypothesis regarding the necessity of considering the complexity, diversity, and consistency of human driving behavior when aiming to develop realistic simulators.