Modeling dispositional and initial learned trust in automated vehicles with predictability and explainability

Ayoub, Jackie; Yang, Xi; Zhou, Feng

doi:10.1016/j.trf.2020.12.015

Cited by 68 publications

(27 citation statements)

References 54 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Daily commuters showed a lower level of TiAD, compared to people that drive few times a week, and these weekly drivers also had a lower level of TiAD than casual drivers. As reported by Ayoub et al (2021), the number of years of driving also seemed to have a negative impact on TiAD. It is possible that frequent drivers have developed a high level of confidence in their own driving skills and thus tend to trust driving automation less, in line with the seminal findings reported by Lee and Moray (1994).…”

Section: Determinants Of the Initial Learned Tiadmentioning

confidence: 55%

“…This effect of the gender is usually absent in experiments, and this finding raises a few questions. It might exist a gender effect in the overall population, as indicated by Ayoub et al (2021)'s model, that fades away in experimental sessions because some females that distrust HAD are not interested in participating in this kind of experiments. This finding should be examined thoroughly before even being considered.…”

Section: Determinants Of the Dispositional Tiadmentioning

confidence: 99%

“…They were able to show respondents' initial TiAD was strongly influenced by the brand reputation and the level of expected reliability of the level 3 driving automation. Ayoub et al (2021) understanding of the initial level of TiAD, by assessing the numerous factors related to individuals (i.e., dispositional trust and initial learned trust, Hoff & Bashir, 2015), and proposing a model integrating the respective weights of these factors.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Why do they trust it? Determinants of the initial Trust in Automated Driving

Manchon¹,

Beaufort²,

Bueno³

et al. 2022

Preprint

View full text Add to dashboard Cite

Because driving automation is quickly increasing, humans’ relation to automated driving evolves accordingly. Trust in automated driving has been shown to strongly correlate with drivers’ acceptance of such systems. Nonetheless, there is currently a lack of specific questionnaires to assess drivers’ level of trust in Highly Automated Driving (HAD). Moreover, little is known about which factors determine trust before any interaction with HAD systems. Theoretical trust in automation models propose several factors to explain how trust builds before and during interactions with automation. Using 844 answers collected through an online survey, this paper aims to propose a new scale designed to evaluate initial trust in automated driving. Moreover, we also measure other factors linked to drivers’ initial level of trust in HAD by operationalizing the trust components proposed by Hoff and Bashir (2015)’s theoretical trust model. In order to better describe what determines trust in HAD, a linear model based on the collected data is proposed. The model not only weights the factors determining trust but is also able to predict the level of trust considering these factors.

show abstract

Section: Determinants Of the Initial Learned Tiadmentioning

confidence: 55%

Section: Determinants Of the Dispositional Tiadmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Why do they trust it? Determinants of the initial Trust in Automated Driving

Manchon¹,

Beaufort²,

Bueno³

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…To break through this limitation, Lundberg et al [ 46 ] proposed the TreeExplainer, which is suitable for tree-based machine learning models, such as LightGBM and CatBoost. The TreeExplainer can calculate the accurate Shapley value and correctly estimate the Shapley value when there is correlation between features [ 47 ]. The SHAP interaction values can be calculated as the difference between the Shapley values of feature

with and without feature

, as shown in Equation (12).…”

Section: Methodsmentioning

confidence: 99%

Insights into Factors Affecting Traffic Accident Severity of Novice and Experienced Drivers: A Machine Learning Approach

Chen

Shao

2021

IJERPH

View full text Add to dashboard Cite

Traffic accidents have significant financial and social impacts. Reducing the losses caused by traffic accidents has always been one of the most important issues. This paper presents an effort to investigate the factors affecting the accident severity of drivers with different driving experience. Special focus was placed on the combined effect of driving experience and age. Based on our dataset (traffic accidents that occurred between 2005 and 2021 in Shaanxi, China), CatBoost model was applied to deal with categorical feature, and SHAP (Shapley Additive exPlanations) model was used to interpret the output. Results show that accident cause, age, visibility, light condition, season, road alignment, and terrain are the key factors affecting accident severity for both novice and experienced drivers. Age has the opposite impact on fatal accident for novice and experienced drivers. Novice drivers younger than 30 or older than 55 are prone to suffer fatal accident, but for experienced drivers, the risk of fatal accident decreases when they are young and increases when they are old. These findings fill the research gap of the combined effect of driving experience and age on accident severity. Meanwhile, it can provide useful insights for practitioners to improve traffic safety for novice and experienced drivers.

show abstract

“…In recent years, XGB models have been proven to have good performance in traffic flow prediction(Mahmoud et al, 2021), rail defects prediction(Mohammadi et al, 2019), driving behavior prediction(Ayoub et al, 2021) and road risk identification and prediction(Das et al, 2020). (3)Logistic regression (LR):LR is generally used to model the relationship between a categorical dependent variable and categorical/dichotomous/ continuous independent variables.…”

mentioning

confidence: 99%

Using naturalistic driving data to identify driving style based on longitudinal driving operation conditions

Lyu

Wang

et al. 2021

JICV

View full text Add to dashboard Cite

Purpose An individual’s driving style significantly affects overall traffic safety. However, driving style is difficult to identify due to temporal and spatial differences and scene heterogeneity of driving behavior data. As such, the study of real-time driving-style identification methods is of great significance for formulating personalized driving strategies, improving traffic safety and reducing fuel consumption. This study aims to establish a driving style recognition framework based on longitudinal driving operation conditions (DOCs) using a machine learning model and natural driving data collected by a vehicle equipped with an advanced driving assistance system (ADAS). Design/methodology/approach Specifically, a driving style recognition framework based on longitudinal DOCs was established. To train the model, a real-world driving experiment was conducted. First, the driving styles of 44 drivers were preliminarily identified through natural driving data and video data; drivers were categorized through a subjective evaluation as conservative, moderate or aggressive. Then, based on the ADAS driving data, a criterion for extracting longitudinal DOCs was developed. Third, taking the ADAS data from 47 Kms of the two test expressways as the research object, six DOCs were calibrated and the characteristic data sets of the different DOCs were extracted and constructed. Finally, four machine learning classification (MLC) models were used to classify and predict driving style based on the natural driving data. Findings The results showed that six longitudinal DOCs were calibrated according to the proposed calibration criterion. Cautious drivers undertook the largest proportion of the free cruise condition (FCC), while aggressive drivers primarily undertook the FCC, following steady condition and relative approximation condition. Compared with cautious and moderate drivers, aggressive drivers adopted a smaller time headway (THW) and distance headway (DHW). THW, time-to-collision (TTC) and DHW showed highly significant differences in driving style identification, while longitudinal acceleration (LA) showed no significant difference in driving style identification. Speed and TTC showed no significant difference between moderate and aggressive drivers. In consideration of the cross-validation results and model prediction results, the overall hierarchical prediction performance ranking of the four studied machine learning models under the current sample data set was extreme gradient boosting > multi-layer perceptron > logistic regression > support vector machine. Originality/value The contribution of this research is to propose a criterion and solution for using longitudinal driving behavior data to label longitudinal DOCs and rapidly identify driving styles based on those DOCs and MLC models. This study provides a reference for real-time online driving style identification in vehicles equipped with onboard data acquisition equipment, such as ADAS.

show abstract

Modeling dispositional and initial learned trust in automated vehicles with predictability and explainability

Cited by 68 publications

References 54 publications

Why do they trust it? Determinants of the initial Trust in Automated Driving

Why do they trust it? Determinants of the initial Trust in Automated Driving

Insights into Factors Affecting Traffic Accident Severity of Novice and Experienced Drivers: A Machine Learning Approach

Using naturalistic driving data to identify driving style based on longitudinal driving operation conditions

Contact Info

Product

Resources

About