Driver Behavior Modelling at the Urban Intersection via Canonical Correlation Analysis

TL;DR

This paper introduces a CCA-based framework for modeling driver behavior at urban intersections, utilizing feature selection and probabilistic models, validated through simulated and real-world data, outperforming existing methods.

Contribution

The paper proposes a novel CCA-based approach combined with GMM and GPR for driver behavior modeling at intersections, improving performance over previous attention-based methods.

Findings

The framework achieves better accuracy than existing models.

Experimental results align well with drivers' judgments.

Validated with both simulated and naturalistic driving data.

Abstract

The urban intersection is a typically dynamic and complex scenario for intelligent vehicles, which exists a variety of driving behaviors and traffic participants. Accurately modelling the driver behavior at the intersection is essential for intelligent transportation systems (ITS). Previous researches mainly focus on using attention mechanism to model the degree of correlation. In this research, a canonical correlation analysis (CCA)-based framework is proposed. The value of canonical correlation is used for feature selection. Gaussian mixture model and Gaussian process regression are applied for driver behavior modelling. Two experiments using simulated and naturalistic driving data are designed for verification. Experimental results are consistent with the driver's judgment. Comparative studies show that the proposed framework can obtain a better performance.