Search

• 1319 questionnaires were collected from cargo TTWs groups all China. • Most attributes of Cargo TTWs groups: young men, less experience, high workload. • The three-layer risk framework was verified and extent with SEM. • Mediating mechanism, "risk factors - violations - errors", have been discovered. • Anxiety has a negligible risk effect and Workload has a direct weak risk effect. Cargo two- or three-wheeled vehicles (TTWs), as a new form of micro-mobility, have become a popular mode of urban cargo transportation in China. Cargo TTW riders' psychological factors and risky behaviors lead to a number of accidents. A questionnaire is designed by comprehensively considering these factors and behaviors of cargo TTW riders that includes eleven risk factors to quantitatively analyze the risky behaviors based on structural equation modeling (SEM). One thousand three hundred nineteen participants reported using cargo TTWs on a questionnaire distributed across the country. The characteristics of riding behavior data are analyzed to verify the three-layer risk theoretical framework of "Psychological factors (Personality traits/specific factors) - Psychological acceptability of risks (confidence/perception/attitude) - Risky behaviors". The results show that anger has a strong direct effect on riding violations, while normlessness and altruism have a direct effect on riding errors. Workload has a weak but direct effect on risky behaviors, and riding feedback has a weak and mixed effect. In addition, high-risk groups are identified by analysis of variance (ANOVA) with rider population attributes. These quantitative analyses can help guide safety countermeasures to mitigate accidents involving cargo TTWs. [ABSTRACT FROM AUTHOR]

In recent years, the automated driving system has been known to be one of the most popular research topics of artificial intelligence (AI) and intelligent transportation system (ITS). The journey experience on automated vehicles and the intelligent automated driving system could be improved by individualization driving understanding. Although previous studies have proposed methods for driving styles understanding, the individualization driving classification has not been addressed thoroughly. Therefore, in this study, a supervised method is proposed to understand driving behavioral structure and the latent driving styles by incorporating the prior knowledge. Firstly, a novel method is established for driving behavioral encoding and raw driving data mining. Then, the Labeled Latent Dirichlet Allocation (LLDA) is proposed to understand the latent driving styles from individual driving with driving behaviors. Finally, the Safety Pilot Model Deployment (SPMD) data are used to validate the performance of the proposed model. Experimental results show that the proposed model uncovers latent driving styles effectively and shows good agreement to real situations, which provides theoretical guidance on driving behavior recognition for better individual experience on automated driving vehicles. [ABSTRACT FROM AUTHOR]

In recent years, the automated driving system has been known to be one of the most popular research topics of artificial intelligence (AI) and intelligent transportation system (ITS). The journey experience on automated vehicles and the intelligent automated driving system could be improved by individualization driving understanding. Although previous studies have proposed methods for driving styles understanding, the individualization driving classification has not been addressed thoroughly. Therefore, in this study, a supervised method is proposed to understand driving behavioral structure and the latent driving styles by incorporating the prior knowledge. Firstly, a novel method is established for driving behavioral encoding and raw driving data mining. Then, the Labeled Latent Dirichlet Allocation (LLDA) is proposed to understand the latent driving styles from individual driving with driving behaviors. Finally, the Safety Pilot Model Deployment (SPMD) data are used to validate the performance of the proposed model. Experimental results show that the proposed model uncovers latent driving styles effectively and shows good agreement to real situations, which provides theoretical guidance on driving behavior recognition for better individual experience on automated driving vehicles. [ABSTRACT FROM AUTHOR]

In recent years, the automated driving system has been known to be one of the most popular research topics of artificial intelligence (AI) and intelligent transportation system (ITS). The journey experience on automated vehicles and the intelligent automated driving system could be improved by individualization driving understanding. Although previous studies have proposed methods for driving styles understanding, the individualization driving classification has not been addressed thoroughly. Therefore, in this study, a supervised method is proposed to understand driving behavioral structure and the latent driving styles by incorporating the prior knowledge. Firstly, a novel method is established for driving behavioral encoding and raw driving data mining. Then, the Labeled Latent Dirichlet Allocation (LLDA) is proposed to understand the latent driving styles from individual driving with driving behaviors. Finally, the Safety Pilot Model Deployment (SPMD) data are used to validate the performance of the proposed model. Experimental results show that the proposed model uncovers latent driving styles effectively and shows good agreement to real situations, which provides theoretical guidance on driving behavior recognition for better individual experience on automated driving vehicles. [ABSTRACT FROM AUTHOR]

Whereas the development of ADAS seeks to improve driver's overall performance with a particular focus on traffic safety improvement; with it comes the requirement and opportunity to objectively evaluate the effectiveness of the technology in improving safety and overall road traffic efficiency. This study evaluates the effectiveness of ADAS in improving driver's risk perception in near-crash events using a novel metric from risk homeostasis theory (RHT) – Safety Margins, as an indicator. By designing a function that captures the initial and maximum risk-level of drivers when involved in a critical driving event while driving on a field operational test (FOT) route in Wuhan China; a comparison of the risk-level of drivers when ADAS was on as against when ADAS was off can be made, enabling an effective evaluation of the impact of ADAS on driver's risk perception and consequent risk mitigation. The results show that ADAS has a positive impact on the low-risk group and moderate-risk group for all drivers, but a negative impact on the high-risk group for skilled drivers. The impact evaluation is done under varying risk-levels of near-crash events with drivers of different driving experiences. [ABSTRACT FROM AUTHOR]

• A novel improved GRA and fuzzy theories based prioritizing model is proposed. • Comprehensive consideration for varying weight of each influencing factor and expert. • Rational fuzzy numbers of domain experts are utilized to the newly proposed technique. • A new framework for influencing factors analysis system of autonomous ship maneuvering. • A feasibility basis for autonomous ship maneuvering decision-making analysis system. Ship maneuvering decisions are influenced by several factors, and it is essential to prioritize the main influencing factors for efficient selection of the corresponding maneuvering decisions. Meanwhile, the autonomous ship maneuvering decision-making influencing factors constitute a typical grey system, which is suitable for research by grey relational analysis. Furthermore, in the fuzzy approach, linguistic assessment of factors is evaluated to obtain priorities numbers. Therefore, this study mainly focuses on the concept of human-like maneuvering for autonomous ships. Based on experimental data of experienced seafarers and using a simulation platform under the scenario of the Shanghai Waigaoqiao wharf, an inference model utilizing grey and fuzzy theories is proposed. The proposed model combined with expert linguistic terms in order to select the ship maneuvering decision-making main influencing factors from multi-source influencing factors (in overall and separated categories of natural environment, ship motion, force parameters, draft, and position), and to study the decision-making prioritization for maritime traffic safety for specific ship maneuvering scenarios. This method can prioritize the main factors which affect maneuvering decisions as well as guide an autonomous ship-assisted or automatic maneuvering evaluation system for the research of human-like maneuvering behavior. This study provides a new perspective on the identification of main ship maneuvering decision-making influencing factors in theory and in practice. It can be utilized for better decision-making concerning maritime traffic safety of autonomous ship maneuvering, which in turn makes shipping safer and promote the application and spreading of autonomous ships. [ABSTRACT FROM AUTHOR]

The prevention of severe injuries during crashes has become one of the leading issues in traffic management and transportation safety. Identifying the impact factors that affect traffic injury severity is critical for reducing the occurrence of severe injuries. In this study, the Fatality Analysis Reporting System data are selected as the dataset for the analysis. An algorithm named improved Markov Blanket was proposed to extract the significant and common factors that affect crash injury severity from 29 variables related to driver characteristics, vehicle characteristics, accidents types, road condition, and environment characteristics. The Pearson correlation coefficient test is applied to verify the significant correlation between the selected factors and traffic injury severity. Two widely used classification algorithms (Bayesian networks and C4.5 decision tree) were employed to evaluate the performance of the proposed feature selection algorithm. The calculation result of the correlation coefficient, accuracy of classification, and classification error rate indicated that the improved Markov Blanket not only could extract the significant impact factors but could also improve the accuracy of classification. Meanwhile, the relationship between five selected factors (atmospheric condition, time of crash, alcohol test result, crash type, and driver's distraction) and traffic injury severity was also analyzed in this study. The results indicated that crashes occurred in bad weather condition (e.g. fog or worse), in night time, in drunk driving, in crash type of single driver, and in distracted driving, which are associated with more severe injuries. [ABSTRACT FROM AUTHOR]

• The effectiveness of ADAS on braking behavior is significant. • Drivers tend to increase braking times and reducing relative speed in the ADAS. • In any road conditions, when exposed in ADAS, the time headway has increased. • The impact of road categories and experience on driving behaviors is significant. • Drivers' acceptance on FCW function is higher than that of LDW. • Drivers' acceptance on expressway and freeway is higher than that on urban road. Various advanced driver assistance systems (ADAS) have been developed to improve drivers' behavior and perceptual ability; however, whether these ADAS have any measurable effect on driving performance needs to be verified by field operational tests. The purpose of this study was to evaluate the effectiveness of ADAS on Chinese drivers as well as any possible influences of roadway type, gender and experience on driving performance, which can be measured by several variables, including longitudinal, lateral and braking behavior. The ADAS used in this study was a Mobileye M630 with forward collision warning (FCW) and lane departure warning (LDW) functions. Thirty-two participants were recruited to drive a vehicle equipped with Mobileye M630. Participants drove the same test route twice. The route consisted of a 12 km urban road, 34 km urban expressway and 45 km freeway as well as a 14 km adaption road. Vehicle dynamics, environmental information and driving operational data was recorded by CAN (Controller Area Network) bus and video cameras. The results show that ADAS significantly affects braking behavior. Braking time increased and relative speed decreased when drivers were exposed to ADAS. The ADAS also significantly affects several longitudinal behaviors, including the longitudinal deceleration and time headway (THW). The occurrence of critically low THW decreased in the experiment. However, there was no significant effect on lateral behavior. Furthermore, driver acceptance of the FCW function was much higher than the LDW function, and acceptance on the expressway and freeway was much higher than on the urban road. The results also reveal the significant influence of road type and experience on driving behaviors. These findings support policy development and technology improvements for future development of ADAS. [ABSTRACT FROM AUTHOR]

• Effect of surrounding vehicles on target driver's behavior was described by force. • Vehicle cluster situation data was collected and analyzed. • Double variation process of cluster situation and driver's propensity was modeled. • The proposed model is verified by simulation, real and virtual vehicle experiment. The vehicle cluster situation is a kind of dynamic arrangement of a target vehicle and the surrounding vehicles during driving. Revealing the transfer mechanism of vehicle clustering in complex environments is of great significant for studying automated driving systems and driver assist systems. Taking three-lane scenario as an example, vehicle cluster situations changing with driver's evolving propensity were studied. To this end, the data of vehicle cluster situations were collected and analyzed through driving experiments in different environments. In addition, the dual random variations of the vehicle cluster situation and driver's propensity were modeled to explore the transfer mechanism. The verification results show that the predicted outcomes of vehicle cluster situations using the evolution rule of driver's propensity are consistent with the real-time recognition. Therefore, the transfer mechanism of vehicle cluster situations was found to be effective and reasonable. It is important for the research of intelligent driving command system of Internet of Things (IOT). [ABSTRACT FROM AUTHOR]

Highlights • A rollover speed model is established considering human, vehicle and road factors. • The model built into an smartphone can calculate safe speeds for heavy vehicles. • The smartphone based warning system is an human-machine interface for drivers. • The system is effective due to considering individual vehicles' characteristics. Abstract Inappropriate speed selection on a curved road is a main cause of rollover accidents for heavy vehicles due to their relative higher centers of gravity, comparing with those of passenger cars. Traditional driving safety improvement methods on curves include static/dynamic roadside speed limit signs that lack individual vehicle's characteristics, and the high-cost anti-rollover stability control systems that cannot take road geometric parameters like superelevation of a vehicle's upcoming curve into consideration. In this paper,a new rollover speed prediction model based on the derivation of three-degree-of-freedom vehicle dynamics and lateral load transfer ratio (LTR) index is presented. Through numerical experiments, the results show that this model could guarantee the vehicle roll stability with the calculated speed for entering a curve whose road radius is even 50 m, in which the vehicle's LTR never exceeds 0.72 and lateral acceleration is always less than 0.63 g. Moreover, the proposed model built in a mobile smartphone app can calculate curve radius at first, then provide an early alarming to the driver with an appropriate speed if rollover accident is imminent on the curve. The field tests on freeway off-ramps show that this smartphone-based rollover speed warning system can calculate the curve radii, and alert the driver with appropriate curve speeds that are partially equivalent to professional skilled drivers' speed choices. [ABSTRACT FROM AUTHOR]

Based on the conditional heteroskedasticity of following vehicle acceleration fluctuation in car-following behavior, this paper combines the idea of driving force, Intelligent Driving Model (IDM) and Generalized Autoregressive Conditional Heteroskedasticity (GARCH model) to establish a novel car-following model. First, the concept of safety driving force and efficiency driving force are recommended, and both of them are defined by the idea of IDM, and the dynamic balance of them are the reason for acceleration fluctuation. Then, according to the heteroskedasticity of acceleration sequence, the classic GARCH model is introduced to establish the relationship among the variance of acceleration fluctuation items, driving force items and fluctuating memory items. On this basis, a novel car-following model is established, and the relational properties are studied. Finally, an example is used to verify our model, the results show that the novel car-following model is more accurate than IDM model, and the forward prediction result is close to the actual acceleration change value, which can predict the occurrence of dangerous driving behavior. [ABSTRACT FROM AUTHOR]

• A hybrid traffic flow model of micro and macro is proposed. • The Maximum Likelihood Estimation of stop and go coefficients is deduced. • The mechanism of traffic flow oscillation is explored with Brownian noise. To investigate the stop-and-go phenomenon triggered by car-following, a hybrid model of micro and macro is proposed. Firstly, with the stochastic nature of driving behavior, a Brownian noise is added into the velocity difference to give a modification to the car-following model. Then, to combine the macroscopic traffic flow characteristics, the traffic stream model is introduced, and the two models are fused in the context of high traffic flow. Next, the theoretical validity of the fusion model is verified by the existence and uniqueness of the solution. The coefficients of the hybrid model are extracted as the stop coefficient and go coefficient which reflect the micro driving behavior. The Maximum Likelihood Estimations of the two coefficients are also given, as well as the boundedness of traffic oscillation in stop-and-go. Finally, numerical simulations show that the model has good fitting and prediction ability for traffic flow in high flow period. Moreover, the greater the stop coefficient, the stronger the traffic flow oscillation intensity, the larger the upper boundedness of the oscillation, the more severe the stop-and-go. [ABSTRACT FROM AUTHOR]

Although trucks only account for approximately 4% of all the vehicles based on the Federal Railway Administration (FRA) database, about 25% involved truck accidents happen at highway-rail grade crossings. This study applied an ordered probit model to explore the determinants of injury severity of truck drivers at highway-rail grade crossing in the United States. Given the importance of trucking to the economics of a country and the safety concerns posed by the trucks (as a result of their large size and weight making them difficult to control, maneuver, and stop), a comprehensive research on truck accidents is critical. Based on data analysis results, the strong effects of driver-, environmental-, weather- characteristics on the injury severities in truck accidents happened at highway-rail grade crossings are found. The findings reveal that better speed control for trucks will significantly reduce driver injury severity in accidents occurring at highway-rail grade crossings. In addition, several truck driver behavior characteristics (such as driving under influence of fatigue during peak hour) were found to be statistically significant predictors of high-level injury severity. Thus, education and enforcement targeted to truck drivers could facilitate safety improvements. Moreover, environmental factor (such as area type and roadway pavement) is found to be statistically significant. Truck drivers are more likely to have severe injury in open space area with low traffic volume compared with other areas. The bad weather and visibility condition is found to increase the probability of truck drivers’ high level injury severity. [ABSTRACT FROM AUTHOR]

Correct ego-lane index estimation is essential for lane change and decision making for intelligent vehicles, especially in global navigation satellite system (GNSS)-challenged environments. To achieve this, we propose an ego-lane index estimation approach in an urban scenario based on particle filter (PF). The particles are initialized and propagated by dead reckoning with inertial measurement unit (IMU) and odometry. A lane-level map is used to navigate the particles taking advantage of topologic and geometric information of lanes. GNSS single-point positioning (SPP) can provide position estimation with meter-level accuracy in urban environments, which can limit drift introduced by dead reckoning for updating the weight of each particle. Light detection and ranging (LiDAR) is a common sensor in an intelligent vehicle. A LiDAR-based road boundary detection method provides distance measurements from the vehicle to the left/right road boundaries, which provides a measurement for importance weighting. However, the high precision of the LiDAR measurements may put a tight constraint on the distribution of particles, which can lead to particle degeneration with sparse particle sets. To deal with this problem, we propose a novel step that shifts particles laterally based on LiDAR measurements instead of importance weighting in the traditional PF scheme. We tested our methods on an urban expressway at a low traffic volume period to ensure road boundaries can be detected by LiDAR measurements at most time steps. Experimental results prove that our improved PF scheme can correctly estimate ego-lane index at all time steps, while the traditional PF scheme produces wrong estimations at some time steps. [ABSTRACT FROM AUTHOR]

Abstract: This research presents a comprehensive analysis of motor vehicle–bicycle crashes using 4 years of reported crash data (2004–2007) in Beijing. The interrelationship of irregular maneuvers, crash patterns and bicyclist injury severity are investigated by controlling for a variety of risk factors related to bicyclist demographics, roadway geometric design, road environment, etc. Results show that different irregular maneuvers are correlated with a number of risk factors at different roadway locations such as the bicyclist age and gender, weather and traffic condition. Furthermore, angle collisions are the leading pattern of motor vehicle–bicycle crashes, and different irregular maneuvers may lead to some specific crash patterns such as head-on or rear-end crashes. Orthokinetic scrape is more likely to result in running over bicyclists, which may lead to more severe injury. Moreover, bicyclist injury severity level could be elevated by specific crash patterns and risk factors including head-on and angle collisions, occurrence of running over bicyclists, night without streetlight, roads without median/division, higher speed limit, heavy vehicle involvement and older bicyclists. This study suggests installation of median, division between roadway and bikeway, and improvement of illumination on road segments. Reduced speed limit is also recommended at roadway locations with high bicycle traffic volume. Furthermore, it may be necessary to develop safety campaigns aimed at male, teenage and older bicyclists. [Copyright &y& Elsevier]

Frequent water traffic accidents pose severe threats to human life and property safety, the water environment, and adverse effects on social stability. Understanding historical accidents is essential for accident prevention and risk mitigation. However, at present, research on the comprehensive statistical analysis of characteristics and causes of ship accidents that occurred in the inland water areas of the Yangtze River, especially in the fluctuating backwater area (FBA) of Three Gorges Reservoir (TGR) region, is still scanty, even less a hierarchical and systematic analysis framework. Therefore, this paper proposes a comprehensive ship accident characteristics and causes analysis framework, and summarizes and visualizes the characteristics of ship accidents through the statistical and comparative analysis of historical data in terms of categories and severity of accidents, ship types involved in accidents, spatial and temporal distribution characteristics, ship accident losses, and root causes and lessons learned from the related accidents. In order to demonstrate the value of the proposed framework, based on the official sources, ten years of ship accident data from 2009 to 2018 in the FBA of TGR region are collected and analyzed in detail. On the basis of the results, this paper summarizes accident prevention and supervision guidelines to provide decision support for maritime safety. This research is of significance to the water traffic accident precaution and risk mitigation in this region and can be useful for other similar specific scenarios worldwide. • A hierarchical and systematic investigation framework is proposed for characteristics and causes analysis of ship accidents. • Human errors, unsafe ship conditions, and harsh traffic environment conditions are the main risk categories for water traffic accidents. • Ship accidents threaten human life safety and cause substantial economic losses, and accident prevention is the core of safety management. • Comprehensive case study manifests the proposed framework is applicable and significant to ship accidents analysis and risk mitigation. [ABSTRACT FROM AUTHOR]

• A deep learning method is proposed for autonomous ship-oriented small ship detection. • A modified Generative Adversarial Network is applied for training data augmentation. • An improved YOLO v2 algorithm is used for small ship detection. • Extensive experiments are conducted to show the effectiveness of the proposed method. Small ship detection is an important topic in autonomous ship technology and plays an essential role in shipping safety. Since traditional object detection techniques based on the shipborne radar are not qualified for the task of near and small ship detection, deep learning-based image recognition methods based on video surveillance systems can be naturally utilized on autonomous vessels to effectively detect near and small ships. However, a limited number of real-world samples of small ships may fail to train a learning method that can accurately detect small ships in most cases. To address this, a novel hybrid deep learning method that combines a modified Generative Adversarial Network (GAN) and a Convolutional Neural Network (CNN)-based detection approach is proposed for small ship detection. Specifically, a Gaussian Mixture Wasserstein GAN with Gradient Penalty is utilized to first directly generate sufficient informative artificial samples of small ships based on the zero-sum game between a generator and a discriminator, and then an improved CNN-based real-time detection method is trained on both the original and the generated data for accurate small ship detection. Experimental results show that the proposed deep learning method (a) is competent to generate sufficient informative small ship samples and (b) can obtain significantly improved and robust results of small ship detection. The results also indicate that the proposed method can be effectively applied to ensuring autonomous ship safety. [ABSTRACT FROM AUTHOR]

• A novel sparse representation method for pedestrian trajectory abnormal analysis. • Utilizing Lp-regularization (0 < p < 1) to get sparser solutions. • An effective solver for the proposed method with EM algorithm and entropy. Pedestrian abnormal trajectory understanding based on video surveillance systems can improve public safety. However, manually identifying pedestrian abnormal trajectories is usually a prohibitive workload. The objective of this study is to propose an automatic method for understanding pedestrian abnormal trajectories. An improved sparse representation model, namely information entropy constrained trajectory representation method (IECTR), is developed for pedestrian trajectory classification. It aims to reduce the entropy for trajectory representation and to obtain superior analyzing results. In the proposed method, the orthogonal matching pursuit (OMP) is embedded in the expectation maximization (EM) method to iteratively obtain the selection probabilities and the sparse coefficients. In addition, the lower-bound sparser condition of L p -minimization (0 < p < 1) is applied in the proposed method to guarantee salient solutions. In order to validate the performance and effectiveness of the proposed method, classification experiments are conducted using five pedestrian trajectory datasets. The results show that the identification accuracy of the proposed method is superior to the compared methods, including naïve Bayes classifier (NBC), support vector machine (SVM), k-nearest neighbor (kNN), and typical sparse representation-based methods. [ABSTRACT FROM AUTHOR]

To date, the increasing density of water traffic has caused the ship's navigation environment to deteriorate, resulting in frequent water traffic accidents. In addition, a majority of maritime accidents are caused by human factors, and one of the important ways to solve the ship accidents caused by human factors is to utilize intelligent maneuvering of ships. Based on the actual crews' operational data from full-task handling simulation platform, this study combines a 30,000-ton bulk carrier inbound navigation scenario and uses the decision tree method to propose a knowledge learning model under multiple environmental constraints to give intelligent ships the ability to make decisions like a human: An intelligent ship Human-like Decision-making Maneuvering Decision Recognition (HDMDR) model. The decision-making mechanism for the maneuvering behavior of Officer On Watch (OOW) under the influence of the specific water traffic environment in the inbound scenario is analyzed, and the OOW's decision-making knowledge is automatically acquired and represented. The validation tests and the comparative analysis with the classic classification algorithms of k-Nearest Neighbours (k-NN) and Support Vector Machine (SVM) are performed to demonstrate the accuracy of the proposed HDMDR model. This paper provides a feasible basis for the human-like decision-making analysis of intelligent ships. • Enable a ship to make decisions like a human under multiple environmental constraints. • Propose a novel intelligent ship human-like maneuvering decision recognition model. • Develop the new standardization principle of maneuvering decision-making factors. • Automatic acquisition and representation of the OOW's decision-making knowledge. • Obtain the actual crews' operational data from full-task handling simulation platform. [ABSTRACT FROM AUTHOR]

With the development of online cars, the demand for travel prediction is increasing in order to reduce the information asymmetry between passengers and drivers of online car-hailing. This paper proposes a travel demand forecasting model named OC-CNN based on the convolutional neural network to forecast the travel demand. In order to make full use of the spatial characteristics of the travel demand distribution, this paper meshes the prediction area and creates a travel demand data set of the graphical structure to preserve its spatial properties. Taking advantage of the convolutional neural network in image feature extraction, the historical demand data of the first twenty-five minutes of the entire region are used as a model input to predict the travel demand for the next five minutes. In order to verify the performance of the proposed method, one-month data from online car-hailing of the Chengdu Fourth Ring Road are used. The results show that the model successfully extracts the spatiotemporal features of the data, and the prediction accuracies of the proposed method are superior to those of the representative methods, including the Bayesian Ridge Model, Linear Regression, Support Vector Regression, and Long Short-Term Memory networks. [ABSTRACT FROM AUTHOR]