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

TL;DR

This work addresses the challenge of dilemma zones at signalized intersections by modeling personalized driver behavior. It uses a CARLA-enabled driving simulator to collect high-resolution trajectories and stop-go decisions from multiple drivers, and introduces a Personalized Transformer Encoder that leverages both common signals and driver-specific history. The results show that personalized modeling improves stop-or-go prediction accuracy by 3.7–12.6% over a Generic Transformer and 16.8–21.6% over Binary Logistic Regression. The work highlights heterogeneity in driver behavior, redefines time-to-stop-line metrics, and demonstrates practical benefits for personalized ADAS and traffic operation.

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.

Figures (6)

• Figure 1: Work flow of personalized driving behavior analysis and prediction.
• Figure 2: Illustration of Type II dilemma zone.
• Figure 3: Vehicle trajectories in dilemma zones.
• Figure 4: Distribution of speed profiles during yellow-light.
• Figure 5: Moments of "stop-or-go" decision-making by drivers.