Investigating Personalized Driving Behaviors in Dilemma Zones: Analysis and Prediction of Stop-or-Go Decisions

TL;DR

This paper investigates how individual drivers make stop-or-go decisions in dilemma zones at intersections, using a simulator and a personalized transformer model to improve prediction accuracy over generic models.

Contribution

It introduces a personalized transformer encoder that captures individual driving behaviors, enhancing prediction accuracy of driver decisions in dilemma zones.

Findings

Personalized transformer improves prediction accuracy by up to 21.6%.

Simulator data reveals diverse driver responses in dilemma zones.

Personalized models outperform generic and logistic regression approaches.

Abstract

Dilemma zones at signalized intersections present a commonly occurring but unsolved challenge for both drivers and traffic operators. Onsets of the yellow lights prompt varied responses from different drivers: some may brake abruptly, compromising the ride comfort, while others may accelerate, increasing the risk of red-light violations and potential safety hazards. Such diversity in drivers' stop-or-go decisions may result from not only surrounding traffic conditions, but also personalized driving behaviors. To this end, identifying personalized driving behaviors and integrating them into advanced driver assistance systems (ADAS) to mitigate the dilemma zone problem presents an intriguing scientific question. In this study, we employ a game engine-based (i.e., CARLA-enabled) driving simulator to collect high-resolution vehicle trajectories, incoming traffic signal phase and timing information, and stop-or-go decisions from four subject drivers in various scenarios. This approach allows us to analyze personalized driving behaviors in dilemma zones and develop a Personalized Transformer Encoder to predict individual drivers' stop-or-go decisions. The results show that the Personalized Transformer Encoder improves the accuracy of predicting driver decision-making in the dilemma zone by 3.7% to 12.6% compared to the Generic Transformer Encoder, and by 16.8% to 21.6% over the binary logistic regression model.