Nagahama AKIHITO (長濱 章仁)

The increase in mixed traffic with weak lane discipline (2D mixed traffic) has attracted significant research attention. To better replicate and understand traffic with weak lane discipline, this study examined the variation in response time relative to the position of the leading vehicle, including lateral shifts. Through experiments conducted using a driving simulator and functional fitting, we demonstrated that changes in response time due to longitudinal and lateral locational shifts are well represented by linear and exponential functions, respectively. Additionally, we proposed an extended formulation of the 2D optimal velocity model (2D OVM) that incorporates variable response times, termed the 2D OVM with varying sensitivities (2D OVMVS). The stability condition was derived using a linear approximation. A comparative analysis of the phase diagrams of the 2D OVM and 2D OVMVS, along with a sensitivity analysis, revealed that the proposed 2D OVMVS exhibited a larger unstable region in the phase diagram and lower stability in stable regions than the 2D OVM. As a result, in 2D traffic with weak lane discipline, the equilibrium formation of vehicles was more susceptible to disruption. Our findings indicate that variable response times, as observed in this study, substantially influence the stability of no-lane traffic. Unlike fixed-response models, incorporating response time variability accentuates unstable tendencies. This underscores the necessity of accounting for non-uniform response time distributions in future traffic models.

This study examines the relationships between self-organized vehicle groups and remaining vehicles (referred to as "remains") within heterogeneous, disordered traffic flows, and compares their characteristics. The findings reveal that leader–follower relationships are less prevalent among the remains, whereas connections with grouped vehicles are more frequent in both groups and remains. Additionally, groups form longer leader-follower networks with diverse pathways for the propagation of acceleration and deceleration waves. Furthermore, the results suggest that a typical vehicle platoon comprises a sparse distribution of remains interspersed around longer groups. Moreover, owing to their extended network lengths and varied densities, groups are likely to feature amplified acceleration and deceleration waves. The findings also suggest that some remains may gradually disperse, hindering the backward propagation of waves. Thus, this study provides novel insights into the formation and dynamics of groups and remains in disordered traffic, with the aim of enhancing traffic-flow modeling.

BACKGROUND: Concerns persist regarding the potential reduction in driving performance due to taking second-generation antihistamines or performing hands-free calling. Previous studies have indicated a potential risk to driving performance under an emergency event when these two factors are combined, whereas a non-emergency event was operated effectively. Currently, there is a lack of a discriminative index capable of detecting the potential risks of driving performance impairment. This study aims to investigate the relationship between driving performance and eye movements under combined conditions of taking second-generation antihistamines and a calling task, and to assess the usefulness of eye movement measurements as a discriminative index for detecting potential risks of driving performance impairment. METHODS: Participants engaged in a simulated driving task, which included a calling task, both under taking or not taking second-generation antihistamines. Driving performance and eye movements were monitored during both emergency and non-emergency events, assessing their correlation between driving performance and eye movements. The study further evaluated the usefulness of eye movement as a discriminative index for potential driving impairment risk through receiver operating characteristic (ROC) analysis. RESULTS: In the case of a non-emergency event, no correlation was observed between driving performance and eye movement under the combined conditions. Conversely, a correlation was observed during an emergency event. The ROC analysis, conducted to assess the discriminative index capability of eye movements in detecting the potential risk of driving performance impairment, demonstrated a high discriminative power, with an area under the curve of 0.833. CONCLUSIONS: The findings of this study show the correlation between driving performance and eye movements under the concurrent influence of second-generation antihistamines and a calling task, suggesting the usefulness of eye movement measurement as a discriminant index for detecting potential risks of driving performance impairment.

Abstract

In practical deep-learning applications, such as medical image analysis, autonomous driving, and traffic simulation, the uncertainty of a classification model’s output is critical. Evidential deep learning (EDL) can output this uncertainty for the prediction; however, its accuracy depends on a user-defined threshold, and it cannot handle training data with unknown classes that are unexpectedly contaminated or deliberately mixed for better classification of unknown class. To address these limitations, I propose a classification method called modified-EDL that extends classical EDL such that it outputs a prediction, i.e. an input belongs to a collective unknown class along with a probability. Although other methods handle unknown classes by creating new unknown classes and attempting to learn each class efficiently, the proposed m-EDL outputs, in a natural way, the “uncertainty of the prediction” of classical EDL and uses the output as the probability of an unknown class. Although classical EDL can also classify both known and unknown classes, experiments on three datasets from different domains demonstrated that m-EDL outperformed EDL on known classes when there were instances of unknown classes. Moreover, extensive experiments under different conditions established that m-EDL can predict unknown classes even when the unknown classes in the training and test data have different properties. If unknown class data are to be mixed intentionally during training to increase the discrimination accuracy of unknown classes, it is necessary to mix such data that the characteristics of the mixed data are as close as possible to those of known class data. This ability extends the range of practical applications that can benefit from deep learning-based classification and prediction models.