Skip to main content

2023 | Book

Green Transportation and Low Carbon Mobility Safety

Proceedings of the 12th International Conference on Green Intelligent Transportation Systems and Safety

Editors: Wuhong Wang, Jianping Wu, Xiaobei Jiang, Ruimin Li, Haodong Zhang

Publisher: Springer Nature Singapore

Book Series : Lecture Notes in Electrical Engineering


About this book

These proceedings gather selected papers from the 12th International Conference on Green Intelligent Transportation Systems and Safety, held in Beijing, China on November 17-19, 2021. The book contains cutting-edge research on Green Intelligent Mobility Systems, with the goal of achieving "green, intelligent, and safe transportation systems" as the guiding slogan. The contributions offered here can aid in the advancement of green mobility and intelligent transportation technology by increasing interconnectivity, resource sharing, flexibility, and efficiency. Researchers and engineers in the areas of Transportation Technology and Traffic Engineering, Automotive and Mechanical Engineering, Industrial and System Engineering, and Electrical Engineering will all benefit from the book's scope.

Table of Contents

A Scheduling Plan Model for Metro Crew Incorporating Fatigue and Biological Rhythms

Crew scheduling is one of the critical planning decisions in railway transportation. The existing scheduling and rostering methods usually take the lowest cost as the objective, ignoring the metrzzo crew members’ fatigue and biological rhythms. This paper proposed an optimization approach considering fatigue's impact on solving real-world metro crew scheduling and rostering problems. The shift work characteristics of the metro crew were analyzed firstly. The usability of the Ikeda formula for fatigue evaluation was verified and applied to the metro crew. Then the metro crew scheduling and rostering model were described, and the process of incorporating fatigue factors into the model was demonstrated. Moreover, using the genetic algorithm to solve the problems. Finally, this model was applied to the Beijing Metro Yanfang Line. The results illustrated that the method could significantly reduce the metro crew members’ fatigue value with optimized operating costs.

Yueyuan Chen, Weining Fang, Si Li, Jianxin Wang
Evaluation of Park and Ride Effect Based on Fuzzy Analytic Hierarchy Process

In order to reduce the complexity of the park-and-ride effect evaluation, enhance practicability, combining with the fuzzy hierarchy analysis model to evaluate park-and-ride effect analysis based on multiple park-and-ride point of field survey and access to relevant data, select more concerned about when the passengers to transfer to the three level indicators and nine secondary indicators, quantify the indexes, and the weight of each index is obtained by analytic hierarchy process, using the fuzzy analysis method for comprehensive evaluation In the end, this paper selects the parking and transfer point of Shengli South Street in Shenyang as an example to evaluate its implementation effect. The evaluation result is better, which is consistent with the results of the on-site passenger transfer satisfaction survey. Meanwhile, Suggestions are proposed for further optimization and improvement of the transfer point.

Yan Xing, Xin Rao, Weidong Liu, Wenhao Song
Exploiting Multi-Dec Net for Detecting Traffic Congestion in the Surveillance System

Congestion recognition is very important for traffic flow control. The current traffic monitoring system provides a large number of videos and pictures, which are not fully utilized. With the widespread application of deep learning in the transportation field, we proposed a congestion classification method based on Convolutional Neural Network. First, we obtain representative data from different camera angles, weather, and light conditions from the traffic monitoring system in Shaanxi Province, China. Drawing on the idea of dense blocks, we established a Multi-Dec Net to classify the congestion state of the images captured from the monitoring system. The Network consists of the first-stage identification Multi-Dec Net 1 and the second-stage identification Multi-Dec Net 2. We compare Multi-Dec Net with Alexnet, VggNet, and Resnet, the test results show that Multi-Dec Net has achieved an accuracy rate of 95.96% and an error rate of 4.78%. The model will also be deployed in the highway monitoring center in Shaanxi Province in the future, which will help early warning of traffic conditions, and rapid emergency response.

Jie Li, Jiaojiao Sun, Jun Wang, Yedi Zhuo, A. N. D. Yinli Jin
Signal Control Method for Modern Roundabouts with Waiting Areas to Meet High Traffic Demands

This paper aims to address a problem at signalized roundabouts, where the queuing overflow of left-turn vehicles can block vehicles that are travelling through the roundabout when the signal control method of Two-Stopline-for-Left-Turn control (TSLT) is adopted. A method for setting up the waiting area and coordinating the traffic signal between the approach lanes and the loop lanes is proposed. With the help of the loop lane space, the through and left-turn waiting areas are set up, and the traffic flow in each direction at the roundabout is separated by a coordinated traffic signal, in order to improve the utilization rate of the space resources of the roundabout. Under saturation constraints, delay calculation models are established based on the traffic states of the approach lanes and the loop lanes, and a signal control parameter optimization model is established with the goal of minimizing the delay. The simulation results show that the proposed method can satisfy the required conditions for roundabouts with high traffic volume.

Xiancai Jiang, Qingpeng Shang, Yao Jin
Short-Term Traffic Flow Prediction on a Freeway with Multiple Spatial Toll Data Via Temporal Convolutional Network

Accurate short-term traffic flow prediction plays an important role in traffic guidance and traffic safety. In order to solve the problem that traffic flow is difficult to predict because of the sharp nonlinearity and randomness, a deep learning framework, i.e., the temporal convolutional network (TCN) was explored to capture the nonlinear spatiotemporal characteristics of traffic flow. By adjusting the hyper-parameters of TCN, a traffic flow predictor is proposed. Besides, this paper adopts a method to calculate the designated cross-section traffic volume of freeways from toll data, which makes up for the inability to obtain traffic volume at some cross-section locations due to damage or lack of traffic detection equipment. Through data merging, data cleaning, data reconstruction, data filtering, a traffic flow data set which was selected as the data input with a time interval of 15 min was constructed based on the toll data of Shaanxi Province freeway network from December 2018 to April 2019 as the data source. The TCN model was compared with the SVM, SAE, LSTM and GRU models in terms of mean absolute error (MAE) and root mean square error (RMSE). The prediction results of multiple cross-sections showed the TCN model performs best with superior prediction accuracy, which indicate the TCN model has good robustness and generalization ability. The traffic volume calculation algorithm may provide a practical method for deriving the traffic volume without installing any additional regularly maintained detectors and equipment on the freeway. And the prediction results of TCN model can provide strong support for traffic control and traffic induction.

Xu Wei, Qiang Jing, YinLi Jin, HaoChen Wang
Fast Path Planning for Fixed-Wing Unmanned Aerial Vehicle with Multiple Constraints

Aiming at the problem of fixed-wing unmanned aerial vehicle (UAV) path planning, considering the actual flight conditions and flight performance of UAV, a multi-constraint UAV path planning model is constructed with the minimum flight range and correction times as the objective function. The improved A* algorithm is used to solve the problem: in order to adapt to the model, the objective function of the model is used as the evaluation function; in order to speed up the search efficiency, the branch and bound method is used for iterative search. The simulation results show that: the model can achieve bi-objective optimization, and it is reasonable. Compared with the traditional A* algorithm, the improved algorithm can better balance the optimization flight range and correction times, and save the algorithm planning time, and effectively complete the fast path planning of fixed-wing UAV with multiple constraints.

Gang Zhong, Yi Mao, Liandong Zhang, Shangwen Yang, Hao Liu
Urban Rail Transit and Economic Agglomeration: A Case Study in China

This research examines the impact of urban rail transit (URT) on economic agglomeration. Initially, forty cities that have opened URT by the end of 2019 in mainland of China are selected as the study subjects. Subsequently, this study gathers panel data of the cities from 2010 to 2019 including the extent of URT, economic agglomeration and control variables. Finally, a non-spatial panel data model and four spatial econometric models are established. The estimate results indicate that URT exerts a positive impact on economic agglomeration. Furthermore, the ability of URT to promote economic agglomeration is greater in the tertiary industry than in the secondary industry.

Zhibin Tao, Xuesong Feng, Kemeng Li, Ruolin Shi
Research on Intersection Signal Control Based on WIFI Probe Vehicle Detection

With the rapid development of China's economy and the accelerating process of urbanization, urban traffic problems are becoming more and more prominent. The purpose of this paper is to use wireless network information technology with WIFI as the core to solve related urban traffic problems. Based on WIFI probe technology can automatically collect open WIFI device's MAC address to locate the functional characteristics of the device, this article will combine WIFI probe technology with intersection traffic monitoring, design and test an intersection vehicle detection scheme based on WIFI probe technology, by determining probe placement testing area and data screening processing to the statistics, the intersection of the vehicle. And through real-time adjustment of signal timing do reduce the time of no vehicle with green light, shorten the delay time, the actual capacity maximum close to the lane on the road traffic capacity, thus achieve the purpose of improving the efficiency of traffic, which has good practical application value.

Yu-xin Hou, Rong-ze Yu, Ze-nan Yu, Wei-dong Liu
Traffic Lights Recognition Based on Position Feature

Due to the color and shape characteristics of traffic lights, the color model and shape detection are used to detect traffic lights in the work, the images were processed by the ROI (region of interest) extraction, image enhancement, grayscale binarization processing and morphological processing. Then the contour search and connected domain filtering algorithm were used to extract the traffic signal backlight backplane area, thus detecting and segmenting the traffic signal light backplane. Moreover, taking the traffic signal light backplane as positive sample, the other non-traffic light backplane was used as negative sample to build model library. HOG algorithm was used to extract the feature vectors of samples and exclude the false targets based on SVM classification algorithm. Finally, according to the positions of red and green signal lights on signal light, the pixel value accumulation in the area where the signal light is located was calculated as position feature to recognize the red and green signal lights.

Zhi-fa Yang, Xian-jun Fan, Zhuo Yu, Shi-wu Li, Ai-min liu, Chang-an song
Route Planning and Charging Navigation Strategy for Electric Vehicles Under the Mutual Assistance Trip System

Aiming at the range anxiety caused by battery capacity limitations on electric vehicles (EVs), a set of mutual assistance trip system was established, which used the mutual assistance information on EV drivers as a supplement to decision-making. The charging navigation and route selection optimization was carried out significantly, improving driver charging experience. The optimization model of the central control terminal aims at minimizing the total travel time costs of drivers, integrating real-time dynamic road conditions information, charging service information and EV mutual assistance information. It utilizes information entropy theory to quantify mutual assistance information on mutual assistance information risk factor θi, thus adjusts the weight of travel costs. The model is solved by the genetic algorithm based on the priority coding method. The results of the calculation example show that the mutual assistance trip system can significantly reduce the total travel costs of EV drivers. At the same time, the increase in mutual assistance information makes θi larger, which means the more reliable the corresponding travel cost ti, the greater the impact on the total time costs.

Zhaohui Zeng, Jiangfeng Wang, Dongyu Luo, Guojun Yang
Modelling and Simulation of Speed Guidance of Multi-Intersection in a Connected Vehicle Environment

In this paper, we design a cooperative speed guidance algorithm for the connected vehicles going through several continuous intersections to minimize vehicles’ delay or stopping time in the environment of connected vehicles. Using the vehicle’s position, speed, and real-time traffic signal information (signal light’s color and its’ remaining time) obtained by communication of infrastructure to connected vehicle, speed advisory system can produce an optimal speed value for vehicle (to accelerate to reduce delays or decelerate to reduce stopping time) with the algorithm when the connected vehicle enters a certain range of the first intersection. When the vehicle enters a certain range of the next intersection, speed advisory system does not calculate advisory speed (to avoid frequent changes in vehicle speed) if the vehicle can pass through the intersection at the limited maximum speed without any stop, otherwise, the speed advisory system produces an optimal speed (to reduce stopping time) using information of vehicle and traffic signal like when the vehicle enters the first intersection. Speed guidance of vehicle in rest intersections is same as mentioned above. Several simulation scenarios were established to test the proposed speed guidance strategy in the connected vehicle environment. Results show that improvements in efficiency and emission are obvious, however, safety has a slight decrease. A sensitivity analysis was conducted to compare the performance of different percentages of connected vehicles and the compliance rate of advisory speed, it was found that improvements of higher penetration rate and compliance rate are more obvious.

Ziwei Peng, Jiangfeng Wang, Zhijun Gao, Haitao Huang
Pattern Mining and Predictive Inference on Short-Term Weather and Collision Time Series Data

Collisions are rare random events. Many traffic safety indexes with a small-sized temporal or spatial unit, e.g., daily collisions of a city or a regional highway network, are highly random and fiercely fluctuated. The descriptive and inferential analyses for this type of short-term collision time series data, abbreviated as SCTS data in this paper, are still not well-established yet. This paper is to tackle this issue by a newly emerging approach—pattern mining combined with data mining methods. Based on a collision database, calendar information, and historical weather records, the approach of descriptive statistics was employed to illustrate correlations between all data items and to identify main affecting factors for a SCTS response, with respective to single variable pattern, variable pair and multiple variable correlations. Then the structure and flow-chart of the major attributes led to different SCTS outputs were further investigated by means of decision tree method. The established decision tree structure was then utilized to predict SCTS values of future days as consequence from their calendar characters and weather forecasts. The approaches of description and inference of SCTS data developed in this paper filled in the methodological vacancy of discovering SCTS data pattern and to infer their attributes. The study of this paper also provided a viable solution to predict SCTS and therefore help to pre-schedule safety countermeasures for practitioners.

Yongsheng Chen, Chuanjiao Sun
The Residual Stress Modeling and Assessment Based on Fine Drilling

The diversity of fine drilling process factors results in the lack of detailed and effective modeling method for residual stress, and the sealing of the process limits the credibility assessment of the model. In order to solve these two problems, this paper studies the mechanism of stress formation and transfer in fine drilling, constructs the analytical model of residual stress on drilling surface, measures the time-varying stress of in-situ state by boundary method, and assesses the credibility of the model. The results show that the average credibility of the model is 91.30%, which verifies the efficiency of the model.

Kuikui Feng, Faping Zhang, Wuhong Wang, Zhenhe Wu, Haodong Zhang, Qian Cheng
Research on Dynamic Equation Construction Method with Clearance

The gap problem has always been a difficulty in dynamic analysis because it is mainly non-linear in the process of motion. Based on the L–N contact force model, the dynamic equation with clearance was built in this paper. At the same time, factors such as friction force and oil film force in the operation process were taken into account to obtain the dynamic model with comprehensive multiple factors, which laid a theoretical reference foundation for future motion analysis.

Yunhe Zhang, Faping Zhang, Wuhong Wang, Kai Wu, Yang Li
A Comprehensive Evaluation Approach for Vehicle-Infrastructure Cooperation System Using AHP and Entropy Method

The vehicle-infrastructure cooperation system (VICS) has been widely used and obtained the world-wide attention. Hence, it is imperative to effectively evaluate the safety, effectiveness, and service capability of this system. Existing research mainly focus on its functional evaluation, and seldom consider the influence of human–vehicle–road–cloud cooperation, decision-making control, and system adaptability on the evaluation result. In this condition, this study selected twenty-one evaluation indicators at the aspects of environmental perception and positioning accuracy, communication and transmission capability, application scenario, decision-making control effect, and system adaptability. Then, an evaluation index system of VICS was proposed. Next, the method combining analytic hierarchy process (AHP) and entropy method was adopted, and a framework of multi-mode communication VICS was built. Finally, a field testing was implemented, and the study results show that the proposed method can evaluate the VICS effectively and comprehensively.

Wanyu Niu, Xiaofeng Liu, Dongpeng Yue, Fan Zhang, Yonggang Yu
An Agent-Based Cellular Automata Model for Urban Road Traffic Flow Considering Connected and Automated Vehicles

Considering the development of the vehicle to vehicle (V2V) technology and the popularisation of connected and automated vehicles (CAVs), for an extended period, urban roads will be in a mixed traffic flow scene where CAVs and human-driven vehicles (HDVs) coexist. This paper uses an agent-based cellular automata model to establish a micro-traffic simulation framework for urban roads, called the ABCA-MS model. Considering the characteristics of the intermittent flow of urban roads and signal light control, corresponding car-following and lane-changing rules are established and applied to simulate mixed traffic flow containing CAVs. The simulation results show that the traffic efficiency and the permeability of CAVs show a positive correlation; under the given traffic volume condition, the critical CAVs penetration rate for a traffic state change from congestion to unblocked is 0.4. When the penetration rate of CAVs is in the range of 0–0.4, the improvement of road traffic efficiency is the most significant, and the effect of improvement gradually slows down with the increase of CAVs penetration. Even with a low penetration rate of CAVS, the road capacity can be effectively improved, and the traffic pressure can be alleviated.

Wang Jinghui, Lv Wei, Jiang Yajuan, Qin Shuangshuang, Huang Guangchen
Research on the Top-Level Design of Provincial Smart Service Areas Based on the Enterprise Architecture Approach

To comply with the development trend of service areas from single intelligence to provincial platform construction, this paper uses the enterprise architecture method to sort out the top-level design ideas of the provincial smart service area, and proposes business architecture, application architecture, data architecture and technical architecture of intelligent service area construction from a macro level. The research results can provide theoretical reference for the provincial service area management agencies to carry out intelligent service area planning and construction, which helps promote the construction of integrated, networked, and platform-based provincial intelligent service area system.

Rongjie Lin, Zhe Liu
Research on Investment Benefits Valuation Methods for Information Construction of Integrated Passenger Transportation Hubs

It is very important to give full play to the powerful accelerator and multiplier of informatization through the construction of smart transport hubs to improve the integration level of regional transportation and the quality of passenger travel service. However, Our country is currently in a period of economic structural adjustment, financial constraints, limited financial support for the informatization construction of integrated passenger transportation hub, so it is urgent to build a set of scientific and effective evaluation system for the investment benefits of informatization of integrated passenger transportation hub. Given the prominent problems existing in the current construction of smart transportation hub such as the low application level of information system, waste of resources, poor investment effect, the paper selects 15 indicators to evaluate the effect of information construction from six aspects, including investment scale, perception system, operation decision, emergency command, passenger service, and benefit reflection. On this basis, the paper introduces the data envelopment analysis method to construct the investment benefit evaluation model for information construction of integrated passenger transportation hub, and finally gives the solution idea of the model. The research has guiding significance and practical application value for the informatization project planning, construction, and investment effect evaluation of our country's integrated passenger transportation hub. Meanwhile, it can effectively avoid the waste of resources in the process of information construction.

Rongjie Lin, Zhe Liu
Learning Individual Travel Pattern by Using Large-Scale Mobile Location Data with Deep Learning

Factors found to be influencing individual travel patterns have been explored in several studies. A number of studies have suggested an association between mobile location data and individual travel patterns. This paper proposes a novel deep learning framework to extract individual travel patterns by using large-scale mobile location data. The proposed framework includes methods for extracting origin and destination points based on spatiotemporal thresholds, matching the origin and destination with traffic analysis zone, and predicting based on natural language processing methods, in which a neural network is constructed to vectorize the traffic area on the basis of spatiotemporal information. A case study is performed using mobile location data involving more than 3 million users in Beijing, and the results show that our proposed framework can effectively identify individual travel patterns. The results of this study can further provide insights into traffic demand identification, bus network optimization, and other related research.

Hao-yang Yan, Yu-jie Li, Xiao-han Liu, Xi Chen, Xiao-lei Ma
Short-Term Traffic Flow Forecast Based on ARIMA-SVM Combined Model

Short-term traffic flow prediction is an important basis for traffic state discrimination and traffic congestion prediction. The ARIMA-SVM combined prediction model is used to forecast the urban short-term traffic flow. Firstly, the ARIMA model and SVM model are used to forecast the urban short-term traffic flow. Then, the ARIMA-SVM combined prediction model was obtained by using the updated dynamic weight weighted fusion method to forecast the urban traffic flow, and the results were compared with the separate ARIMA model and SVM model. The empirical results show that the ARIMA-SVM combined model can more accurately predict the city's short-term traffic flow.

Jiaxin Peng, Yongneng Xu, Menghui Wu
A Lightweight Fine-Grained VRU Detection Model for Roadside Units

Object detection of vulnerable road users (VRU) under low computing resources of roadside units is one of the key technologies to achieve vehicle-infrastructure cooperative perception. In this paper, a lightweight fine-grained VRU detection model is proposed. Analyzing the existing complex traffic environment, the traditional definition of VRU is no longer applicable. Our work includes two parts: One is to redefine the fine-grained VRU and construct a new dataset. This task makes the perceptual information obtained by detection more comprehensive and accurate. Another is to optimize YOLOv4 by using the channel pruning method in model compression. The optimized model is 60% lighter than the original model. Under the limitation of low computing resources at the roadside units, the real-time detection of VRU is realized while ensuring a certain detection accuracy.

Jian Shi, Dongxian Sun, Haodong Zhang, Haiqiu Tan, Yaoguang Hu, Wuhong Wang
Design of Evacuation Plan for Shenyang Metro Line 9 Based on Game Passenger Flow Distribution

In view of the increasingly complex road network structure and the long evacuation time of the subway, this paper studies the design of the multi-path evacuation passenger flow distribution ratio based on the game theory model and the matrix game method. The evacuation plan was designed in combination with the transfer station of the Olympic Sports Center of Shenyang Metro Line 9, and the plan was optimized from the three perspectives of facilities, passenger flow and station halls. Use Anylogic simulation software to construct the Olympic Sports Center subway station and simulate the evacuation plan. The simulation results show that the evacuation efficiency of personnel has been significantly improved after optimization; emergency treatment measures effectively guide personnel to escape from multiple evacuation exits, reducing the problem of personnel retention and improving the utilization efficiency of evacuation exits; the game path allocation method is used to clarify the number of personnel allocated in the complex subway network, and improve the utilization rate of roads and the efficiency of evacuation.

Weidong Liu, Quanbo Fu, Wenqi Sun, Rongze Yu
Drivers’ Visual Characteristics of Urban Expressway Based on Eye Tracker

In order to compare and analyze the visual characteristics of drivers in the congested and unblocked state of urban expressways, real vehicle tests were carried out on the eastern expressway in Changchun City using the German Dikablis eye tracker and its supporting D-Lab software. The test data was processed by using descriptive statistical analysis and non-parametric inspection methods to quantify the impact of congestion on the driver’s visual characteristics. The results show that drivers mainly obtain traffic information by gaze when driving on the expressway, and the gaze points are mostly concentrated on the road vehicles; the driver’s gaze duration and scan duration in the congested state account for the highest proportions in the 200–250 ms and 0–25 ms time periods, respectively; the average gaze duration and the average scan duration of the drivers in the congested state were higher than those in the unblocked state. The driver's gaze duration and saccade duration in the two states are significantly different, and the Mann–Whitney U test results are less than 0.05; the pupil area changes more drastically in the congested state, and the pupil area change rate is 38.67%.

Tianjun Feng, Ziwen Zhao, Xiujuan Tian
Hilbert-Huang Transform in Pavement Texture and Skid-Resistance Study

Insufficient skid-resistance of a pavement is a critical cause of traffic accidents. Pavement texture plays an important role in the skid performance. In this paper, the texture information was collected by a 3D laser scanner/device. The 3D texture features were restored by Gaussian filtering and 3D reconstruction. The texture signals were decomposed by Complementary Ensemble Empirical Mode Decomposition (CEEMD) algorithm. The correlation between Intrinsic Mode Function (IMF) and the original signal was analyzed. Hilbert transform on the most relevant IMF was performed to obtain the instantaneous frequency and instantaneous amplitude. The maximum correlation Hilbert transform index (Rcm) was proposed to characterize the coefficient of friction. In addition, the friction index was obtained on the same specimens. The linear fit R-square of Rcm and friction index is above 0.9. It is suggested that the finding of his study can accommodate friction management based on pavement texture collected by the 3D laser scanning system.

Yuan-shuai Dong, Yun Hou, Jia-lei Tian, Yu-xuan Cao, Chen-wei Guo, Tuo Fang, Jing Zhou
D-S Evidence Reasoning Based Transportation Project Investment Decision Model and Its Application

This research demonstrates five approaches of extracting degree of ignorance (DOI) coefficients of index systems based on the DOI factors of investment decision making systems. Based on Dempster -Shafer (D-S) evidence reasoning, A comprehensive decision-making model of transportation project investment (CDM-TPI) is constructed. Lastly, the case comparison studies are carried out to validate the effectiveness and practicability of the model when applying in the investment decision making systems with the DOI factors.

Ling Sui, Xiaoli Zhang
Optimization of Brake Judder Based on Dynamic Model of Disc-Pads Spring Contact

In order to predict the brake judder more accurately, the mechanism of brake judder was studied. First, a dynamic model of braking system based on surface-to-surface contact between disc and pads was constructed. The correctness of the model was verified by comparing the results of bench test and simulation. Then, the key parameters of brake judder were found from the calculation expressions of brake pressure variation (BPV) and brake torque variation (BTV), and two optimization directions were proposed to improve the structure of brake caliper and brake pad backplate. Finally, a method to determine parameters of the improved disc brake was proposed, which combined the finite element analysis with theoretical calculation. The parameters of the improved disc brake were substituted into the Simulink model of the dynamic model, and the time-domain responses of BPV and BTV were obtained. The results show that the maximum BPV and the maximum BTV of the optimized braking system are reduced by 30.7% and 34.2%, respectively, thus testifying to the correctness of the optimization method of brake judder proposed in this paper. The research method proposed in this paper had a certain contribution to the study of brake judder of disc brake. It not only improved the quantization accuracy of brake judder, but also reduced the probability of brake judder greatly. Therefore, this study has a certain engineering significance.

Gongyu Pan, Yaqi Feng, Peng Liu, Qizhao Xu, Lin Chen
Optimal Study of Bus Priority Signal Control Based on Service Reliability

With continuous development in society and economy, a series of problems like traffic congestion are also emerging. In solutions of alleviating these traffic problems, bus priority control is proved to be effective. However, the implementation of bus priority measures needs to meet certain conditions, for bus passengers, time is what they most concern. Therefore, to improve the time effect of passengers, bus priority control method is investigated, then real-life traffic data at the intersection of Huju Road-Beijing West Road in Nanjing City is utilized to quantitatively analyze the priority control method. The results show that, under certain conditions, the method can bring time effects to passengers both at the intersection and downstream stops, improving the service reliability of buses.

Mengqi Wang, Rui Li, Sulan Meng
Electric Vehicle Charging Demand Forecast Based on Residents’ Travel Data

The charging demand of electric vehicles is closely related to residents’ travel behavior, this paper proposes a charging demand prediction method based on the travel behavior of electric vehicle users. By studying the probability distribution of temporal and spatial characteristics in travel behavior, constructing a charging decision model for electric vehicle users, use Monte Carlo method to simulate the travel activities of each electric car in one day, get the charging demand of the whole city. Take Nanjing as an example to verify the model, this method can effectively predict the temporal and spatial distribution of charging demand, and can provide a basis for subsequent site selection of charging facilities.

Zhule Jin, Yongneng Xu, Zheng Li
Traffic Flow Prediction Based on GM-RBF

In order to provide more reliable data for traffic control and guidance system, traffic flow prediction is very important. This paper proposes a traffic flow prediction method based on GM-RBF combined model. Firstly, build a GM(1, 1) prediction model to predict multiple sequences. Secondly, the GM model and RBF neural network are combined in application, and the residual feedback is established by the RBF model, so as to solve the problem of low accuracy of the grey model in the prediction. Taking the data every 15 min of the A12 highway in Suffolk, UK from January 1, 2019 to January 20, 2019 as a sample, taking into account the influence of weather factors on road traffic flow, the traffic flow per 15 min in a day on January 21, 2019 is predicted. The experimental results show that MAPE of GM-RBF model is 2.302911%, MAE is 3.625, RMSE is 5.6807. Compared with GM model and RBF prediction model, the error evaluation index of GM-RBF combination model has been significantly improved in accuracy. Therefore, the combined model has good applicability in traffic flow prediction.

Yaxin Chen, Yongneng Xu, Hui Cheng
Quantitative Analysis of Characteristics and Influencing Factors’ Correlation of Electric Bicycle Traffic Accidents

With the rapid growth of the electric bicycles ownership in P.R.C, while facilitating people's traveling, illegal behaviors of electric bicycles has become common. The frequent occurrence of traffic accidents has attracted widespread attention from all sectors of the society. In order to improve the level of electric bicycle traffic safety, based on the 10-year traffic accident data of electric bicycles in a southern city in China, the electric bicycle traffic accident characteristics was analyzed, through data screening and standardizing, on the characteristics of cyclists, temporal and spatial characteristics, accident types and causes, etc. On this basis, 16 influencing factors such as time, location, illegal behavior, accident form, and cause of accident identification of electric bicycle traffic accidents were used to construct multinomial logit model for accidents without casualties, accidents with injuries, and accidents with fatalities; and the correlation analysis of the main influencing factors was carried out for these three types of accident information of electric bicycles. Following that, electric bicycle traffic accidents shall be prevented and reduced by strengthening targeted prevention and control measures.

Yumeng Zhang, Fengchun Han
Study on the Operation Scheme of Standby Trains with Large Tidal Passenger Flow Under Full-Length and Short-Turn Mode

The train operation scheme is the foundation of the urban rail transit organization. In order to improve the transportation efficiency of urban rail lines under the characteristics of tidal passenger flow, alleviate the congestion phenomenon of large passenger flow stations, and shorten the travel time of passengers, this paper focuses on the morning peak passenger flow stations on working days, and proposes a kind of urban rail backup vehicle for the characteristics of tidal passenger flow. Scheme to quickly relieve congested passenger flow. First, in the form of the upper and lower two-level full-length and short-turn mode, through the analysis and research on the conditional nature of the line site points, it is concluded that on the basis of determining the placement site of the standby train on the existing line, the heuristic algorithm particle swarm with constraints is used The optimization algorithm takes the minimum total waiting time at a large passenger flow station as the objective function, and considers the constraints of the number of trains, the cross-section full load rate, the train tracking interval and the minimum train departure frequency, and establishes the optimization model of the reserve train placement scheme. Based on the case analysis of Nanjing Metro Line 3, the results show that Liuzhoudonglu Station is a large passenger flow station during the morning peak hour on this line. There are two ready-to-deliver stations in the upstream direction. The capacity analysis shows that during the morning peak hour A total of 43 trains are operated. On the basis of the original 21 pairs of large and small roads, a standby train can be added to the No. 4 Taifunglu station to shorten the total waiting time of passengers by 721 s during the morning peak hour and alleviate the problem of tidal passenger congestion. The research results can be used for rail transit operation strategies and improve the efficiency of passenger flow relief under the condition of large passenger flow.

Hui Cheng, Mao Ye, Yaxin Chen
Classification of Driving Tendency of Commercial Truck Drivers Based on AdaBoost Algorithm

In order to study the differences in the driving behavior of truck drivers, a classified management of truck drivers is implemented. Obtain the vehicle driving data of 51 commercial truck drivers in natural driving conditions through the on-board OBD device, and preprocess the original data, including detecting abnormal values, time processing, filling missing values, deleting parking data, etc. On the basis of standardizing the data of commercial cargo vehicles, the index is reduced by factor analysis to obtain the speed control behavior clusters of target vehicle drivers. By extracting variable speed factors and acceleration factors and clustering them according to factor scores, three types of driver's light, medium and heavy driving behaviors are obtained. Based on the K-means cluster analysis of the data, the AdaBoost algorithm is used to establish a classification model for the safety tendency of commercial truck drivers, and the truck drivers are divided into radical drivers and conservative drivers. First, the factor analysis method is used to extract the indicators of the two directions of speeding and acceleration, and then the K-means algorithm is used to classify from two perspectives, and finally the driver's different driving conditions can be analyzed. In addition, through further screening of all driving behavior indicators through K-means clustering, the adaboost algorithm is finally used to verify and analyze the clustering results to determine driver styles with different tendencies. Data verification classification results show that the average accuracy of the driving tendency classification model of commercial truck drivers based on the AdaBoost algorithm can reach 95.74%, which can effectively distinguish radical truck drivers from conservative truck drivers.

Zhaofei Wang, Qiuping Wang, Shiqing Wang, Jianfeng Xi, Jian Tian
An Evaluation of the Effect of Urban Tunnel Lighting on Driving Comfort: A Driving Simulation Study

Urban tunnel lighting is significant for driving safety and comfort. In order to investigate the relationship between the lighting parameters and driving comfort under different situations, this paper established a tunnel lighting system model based on Shenzhen Henglongshan tunnel by UC-win/Road. By designing tunnel lighting driving simulation experiments, the authors changed the parameters of the tunnel lighting system model and conducted a series of tests by the driving simulator under different lighting conditions. Through analysis of the driver’s eye movement data, this paper proposes that the minimum brightness of the tunnel lighting in the middle section is 0.5 cd/m2, and the lighting brightness threshold for safe driving of the driver is 2.2 cd/m2. The research results can provide a reference for the construction of tunnel lighting projects, and a theoretical basis for optimizing the design of tunnel lighting and improving the quality of tunnel lighting.

Yanwei Zang, Zihai Yan, Huojun Wu, Penglu Gan, Mingwei Hu, Wenlin Wu, Peng Liu, Guoqing He, Jinghang Xiao
The Influence of Speed Limit Value of High-Grade Highway in High Altitude Area on Running Speed and Traffic Safety

This article collected relevant data on two high-grade highways in Tibet. Through the statistical analysis of the speed limit value, running speed and traffic accident data, the impact of the increase of the highway speed limit value on the running speed and traffic safety was studied. The regression method was used to analyze the functional relationship between the speed limit increase value and the running speed increase value and the number of traffic accidents, which provides a reference for the speed limit value setting of high-grade highways in high altitude areas. The results show that using the design speed as the speed limit cannot meet the driving expectations of most drivers, and to a certain extent reduces the capacity of high-grade highways. After the speed limit was increased, the original accident-prone road was still the key area of ​​accidents. If no measures are taken, the number of traffic accidents may increase. In engineering practice, the determination of the speed limit value should not only consider the actual running speed of the vehicle, but also analyze the traffic accidents. The increase of the speed limit value on the accident-prone sections should be fully demonstrated and effective engineering measures should be taken.

Research on Interference-Free Monitoring of Driver's Steering Behavior

The study of driver's driving behavior is not only an important part of traffic safety, but also of great significance to promoting the development of autonomous vehicles. In this paper, a Triboelectric Nanogenerators (AK-TENG) with an aluminum (AI)-Kapton friction layer structure is developed for driving behavior monitoring of drivers. AK-TENG has high sensitivity and certain flexibility. Compared with ordinary Triboelectric Nanogenerators, the structure is stronger and more stable, and less damaged. It is used to obtain the driving behavior data of the driver on the steering wheel. This paper uses AK-TENG to obtain driving behavior information data to monitor the driver's steering behavior, and also provides a new idea for the design of high-sensitivity self-powered sensors in intelligent transportation systems.

Haodong Zhang, Haiqiu Tan, Kuikui Feng, Jian Shi, Dongxian Sun, Jie Zhang, Wuhong Wang
Freeway Design Consistency Evaluation Model Based on Alignment and Traffic Characteristics

In order to identify the segments of China's highways that may have traffic safety hazards in the design phase, reduce the traffic accident rate and improve the operational safety of vehicles. This paper constructs a safety evaluation model based on design consistency applicable to China's highways, including the operating speed prediction model and design consistency evaluation indexes. Combined with actual cases, the effectiveness of the model is verified by comparing the constructed model, IHSDM and the evaluation method of “Specifications for Highway Safety Audit”. The results show that the accuracy of the model is 42% higher compared with IHSDM, and it can effectively evaluate the design safety of China's highways. It has important theoretical and practical significance for the study of highway design safety evaluation in China.

Zhang Sufeng, Ma Yanli, Zhou Nianfa, Li Ping, Tian Jia-jia
Driving Risk Identification Considering Coupling Coordination Degree of Primary and Secondary Tasks

To study the influence of the coupling relationship between primary and secondary tasks on driving safety, four driving experiments are designed, including Bluetooth calling, conversation, screen touch operation and in-vehicle radio operation. The vehicle operation data such as standard deviation of horizontal speed, standard deviation of longitudinal speed and steering entropy are collected. The driver visual data such as entropy rate of fixation area, standard deviation of horizontal viewing angle and vertical viewing angle, and average glance speed are collected. The coupling model of primary and secondary tasks in the vehicle is established to determine the coupling coordination degree between them. The driving proportion threshold of driver's secondary tasks and the risk level of various secondary tasks are obtained. The results show that the coupling degree of normal driving and Bluetooth communication can reach high quality coordination level with high safety. Similarly, the coordination level between normal driving and talking is also high. The coupling degree of normal driving and operating radio can reach medium coordination level with general safety. The coupling degree of normal driving and touch screen operation can only achieve low coordination, which is a relatively dangerous driving state.

Ma Yanli, Zhu Jieyu, Yining Lou, Dong Fangqi
Passenger Flow Organization Optimization of Xiamafang Station Based on AnyLogic

The article takes Xiamafang Station as an example. First, AnyLogic simulation software is used to construct the passenger flow organization model of Xiamafang Station, and the congestion points in the scene of large passenger flow in the morning peak are extracted. Then, takes the average queue length and the density of passengers as evaluation indicators to propose optimization plans. Finally, the simulation results show that the optimized scheme shortens the average queue length of passengers in the station, reduces the density, and makes trip more convenient. It is hoped that the research in the article can provide reference for the optimization of passenger flow organization in subway stations, so that rail transit can provide better services for passengers.

Xueyan Kong, Yongneng Xu, Zhe Li
A New Car Following Model Considering the Multi-headway Variation Forecast Effect

An extended car following model is presented by considering the effect of Multi-Headway Variation Forecast (MHVF) effect in the real world. The model’s linear stability criterion was obtained by employing the linear stability theory. Theoretical analysis result shows that the new consideration leads to the stabilization of traffic systems. By means of nonlinear analysis method, the modified Korteweg-deVries (mKdV) equation near the critical point was derived, thus the propagation behavior of traffic jam can be characterized by the kink-antikink soliton solution for the mKdV equation. Numerical simulation is carried out and its results is in good agreement with the aforementioned theoretical analysis. Both of them show that the MHVF effect can suppress the emergence of traffic jamming and stabilize the vehicular system.

Yi-rong Kang, Shu-hong Yang
Optimum Design of Urban Road Intersection Signal Timing Based on VISSIM Simulation

In the urban road network, road intersections have always been an important node of the entire urban traffic system, and they are also the areas with the densest traffic volume on urban roads. At present, there are still many problems in the traffic of some urban road intersections. On the one hand, it leads to lower traffic efficiency of vehicles, and on the other hand, it also poses a threat to the safe driving of vehicles. Based on this, in order to improve the traffic capacity, operation efficiency and safety of urban road intersections, this paper takes the Xinghai Road-Shuangqi Road intersection in Nanjing as an example, and uses VISSIM micro-simulation software to analyze. The average queuing length, maximum queuing length, and number of stops for the entrance, south entrance, and west entrance have been improved, and the index corresponding to the north entrance has become worse. If joint optimization is used in the future, the channelization of the south entrance will be changed to a straight lane and left and right shared lanes. And design the signal timing through the Webster method, and its simulation effect will be greatly improved. It can provide an effective reference for the optimal design of signal timing at intersections of the same type of urban roads.

Zhe Li, Congyong Cao, Chaoqun Kong
Speech-Based Driver Emotion Recognition

The premise that vehicles bring convenience to human life is to ensure the safety of people in vehicles. However, driver’s negative emotions are an important cause of risky driving, road rage, and traffic crashes, which seriously endangers traffic safety. In this paper, we proposed a driver emotion recognition method based on driver’s speech using audio features. Firstly, we extracted 6 features for speech gender recognition. After gender recognition, a combination of gender and MFCCs features were used for negative emotion recognition. Finally, a driver emotion recognition application was developed for function display.

Haiqiu Tan, Haodong Zhang, Jian Shi, Dongxian Sun, Jie Zhang, Xiaobei Jiang, Wuhong Wang
Research on OD Estimation of Public Transit Passenger Flow Based on Multi-source Data

OD calculations based on big data of public transport passenger flow can effectively improve the service quality of public transport and increase its attractiveness. In this paper, the bus GPS data and station GIS data are processed and time–space correlation is established to obtain the time interval of bus station, clustering analysis of bus IC card transaction data, and matching to get bus passengers’ boarding station and time. Then, based on the probability of passenger travel behavior, the station's attraction power and transfer capacity are used to calculate the passenger's alighting station. Use actual cases to verify the algorithm, and finally calculate the OD matrix and passenger flow statistics of the bus line, so that the public transportation department can accurately obtain the bus travel information of urban residents, and provide data support for traditional public transportation planning.

Chaoqun Kong, Tangyi Guo, Liu He
Research on Metro Vehicles Allocation Based on Capacity and Maintenance Model

With the continuous acceleration of urban rail transit construction mileage and the opening of existing extension lines, the demand for metro vehicles on the extended lines has also increased accordingly. However, in the actual vehicle configuration, there is room for optimization of the number of vehicles allocation on each line. By analyzing the relevant factors affecting the assignment of metro vehicles, including signal system, passenger flow conditions, maintenance mode, passenger flow organization and passenger flow development theory, comprehensively consider the length of the line, the speed of the train, the turn-back time of the train, and the interval between trains, construct a metro vehicles allocation model based on capacity demand and vehicle maintenance. Taking the extension of subway line M in a certain city as the research object, through the “four-stage method”, predict the passenger flow, and analysis the key factors including the maximum cross-section passenger flow in the peak hour of metro and the mixed running of long and short routing in the case city, the allocation plan is obtained according to the allocation model. Compared with the original plan allocation plan, the train line rate is increased and the vehicle cost is reduced. Verifies the effectiveness and practicability of the vehicle model based on capacity and maintenance.

Fang Gao, Yongneng Xu, Jue Zhang, Zhonglin Tan
Research on Traffic Flow Model Based on Lattice Hydrodynamics

The operating state of road traffic flow directly affects the stability of the entire urban transportation system. Once the urban transportation system becomes unstable, it will cause problems such as aggravation of traffic congestion. In the intelligent transportation environment, this paper establishes a lattice hydrodynamics model considering the density difference and the flow difference at the same time, adopts the linear stability theory, and conducts theoretical analysis and simulation research on the traffic flow characteristics through numerical simulation. The results show that the stability of traffic flow can be improved by considering both the density difference and the flow difference in the two-lane system.

Jie Yang, Tangyi Guo
Regionally Differentiated Real-Time Energy Consumption Prediction of Electric Vehicles Oriented to Travel Characteristics

Real-time prediction of electric vehicle energy consumption is of great significance to users’ travel planning and charging decisions. This paper analyzed the influence of travel characteristics and regional differences on the power consumption of electric vehicles, and built a regional electric vehicle energy consumption model based on travel characteristics prediction: In this paper, a large number of travel samples are obtained by preprocessing the real-time operation data of electric vehicles, and the influencing factors of power consumption in the travel samples are analyzed to determine that the most relevant characteristic parameters are travel mileage and time, which are used as the main characteristic indicators of energy consumption prediction. On this basis, a single-region BP neural network energy consumption prediction model was built, and the optimal network model structure was adjusted and determined through error feedback, which achieved a prediction accuracy of 93.2%; then, the travel samples of different cities are modeled and cross predicted, and established a multi-regional energy consumption prediction model; finally, the prediction results of different models are compared. The results show that this model has the highest accuracy in the energy consumption prediction of the actual operation of urban electric vehicles, which can reach 92% and above. Combining the existing electricity with the predicted energy consumption results can provide effective support for users to make reasonable charging decisions before travel.

Cheng Wang, Ya-nan Wang, Ji-yuan Tan, Fu-yu Liu, Yuan-yuan Jiang, Zhen-po Wang
Road Traffic Accident Prediction Based on BP Neural Network

In order to predict road traffic accident indicators scientifically and accurately, this paper established a road traffic accident prediction model based on the basic theory of BP neural network. The 14 main influencing factors of traffic accidents were selected by using correlation analysis theory as the input variable of the prediction model and traffic accident deaths was taken as the output variable of the prediction model. Data related to road traffic accidents in China from 2000 to 2017 were selected as training samples of the model, MATLAB nntool was used to train the prediction model by using traingdm, traingda, trainlm and traingd training functions, and predict the deaths of road traffic accidents in 2018 in China. It is verified that the relative error of the constructed road traffic accident prediction model is within 1%, which can be used for road traffic accident prediction.

Yan Xing, Wen-hao Song, Wei-dong Liu, Shu-shida Gao
Autonomous Vehicle Path Planning Based on Improved Ant Colony Algorithm

Path planning is one of the key technologies for autonomous vehicles. Ant Colony Algorithm can effectively achieve the goal of path planning for autonomous vehicles, but the algorithm has the problems of low search efficiency and local optimal solution in path planning. Therefore, this paper improves the classical ant colony algorithm, using adaptive initial pheromone distribution range build initial pheromone distribution, at the same time improve stimulating factor enhanced heuristic search efficiency, and introduces the rollback strategy self-locking and deadlock problem and adopt preferential set limit to update pheromone strategy, help reduce blind ant search path, and reduce the redundancy of map information. The simulation results show that the improved ant colony algorithm can greatly improve the global search ability and convergence speed, and can help the autonomous vehicle to find the optimal path quickly.

Yan Xing, Xin Rao, Weidong Liu, Wenhao Song
Optimized Component Learners Diversity of Traffic State Forecasting Model with Multimode Perturbation

Based on optimizing the diversity of component learners, this paper puts forward a method of traffic state prediction NNPDAP. In this paper, the perturbation training data set, perturbation input attribute and perturbation learning parameter are used to construct eight perturbation modes for optimizing the diversity of component learners. There have built three groups of experiments respectively for comparing the accuracy of traffic state prediction, error distribution, and time efficiency. The experimental results show that, by enhancing the diversity of component learners can improve the accuracy of prediction, so this method has a stronger competitiveness compared with no perturbation method.

Qingchao Liu, Tianyu Xu, Chun Li, Shiqi Nie
Snow Depth Inversion and Analysis of Temporal-Spatial Snow Distribution Along Tianshan Highway Based on MODIS Data

Most of Tianshan Highway is located in the high and cold mountain areas over 2,000 m above the sea level in Xinjiang. Highway distress in late stage is exacerbated and frequently occurs especially avalanche, wind blowing snow, geohazard, flood damage, and other disasters along the highway, which seriously affects the traffic. Restricted by the climate and topography and other factors, conventional ground monitoring is hard to be applied into dealing with the emergency situation. In contrast, remote sensing technology can monitor snows in large-scale and real-time. Based on the remote sensing data sources and studied areas, MODIS data is selected to reversely derive snow depth and analyze the temporal-spatial snow distribution in Dushanzi-Kuqa region of Tianshan Highway of northern Xinjiang in recent years. Analysis revealed that the snowfall was mainly concentrated on late October to the early April in the next year. The snow area in July is the least in summer, and reaches largest in winter, accounting for about 35.7% of the entire studied area. Most of the snow depth is at 12 to 15 cm, and the maximum snow cover depth is about 23 cm in the Duzishan region of Wusu County. The study conclusion is of great theoretical and practical significance for distress monitoring and controlling.

Bo Yin
Study on Echo Model of Saturated Traffic Flow in Subsea Tunnel

In order to solve the traffic flow changes caused by accidents in subsea tunnel, the theoretical echo model is established. The characters of the subsea is considered such as the different reaction time and the broadcast process. The speed of the echo model in the subsea is given after the incident. The study focused on the influence of traffic flow changes caused by accidents on subsequent vehicles in the saturated state of subsea tunnel. The result will provide technical support for the operation of the subsea tunnel.

Chuanjiao Sun
Prediction for Taxi-Hailing Demand—An Adaptive Multi-view Deep Learning Model

The emergence of taxi requesting service have changed the situation of traditional taxis. Nowadays, it has gained great popularity all over the world. As more people use this service, some problems spring up gradually and the imbalance between supply and demand is one of the most serious ones. It is urgent for researchers to solve this problem because it affects greatly the service quality of taxi system. In summary, traditional papers involving the demand prediction of taxi-hailing are mainly divided into two directions. Some researchers tend to use regional attributes (e.g., land use variables) for spatial modeling and analysis (e.g., using geographically weighted regression, GWR) and then interpret the model. The others mainly depend on spatiotemporal correlation to predict the demand (e.g., autoregressive integrated moving average, ARIMA) and then focus on the improvement of prediction. Based on previous researches, this paper proposes an adaptive multi-view deep learning model which comprehensively integrates the focuses of previous studies. The framework of the model includes four views, namely feature views (including various features of zones such as the conditions of weather, transportation, land use and so on), semantic views (division of zones that functionally similar), spatial views (searching for adaptive spatial neighbors), temporal views (searching for adaptive time windows). By embedding these views into our model, the accuracy and extension of prediction are greatly improved. To evaluate the prediction performance of our model, it is respectively compared with models which use different algorithms or have different model frameworks. Validation based on large amounts of data shows the superiority of our proposed model.

Xin Tang, Yongfeng Ma, Zhuopeng Xie, Shuyan Chen
Recognition and Comparison of Driving Styles of Heavy-Duty Truck Drivers Under Different Scenarios

With the acceleration of urbanization, the demand for heavy-duty trucks has increased and transportation safety and management issues are facing large challenges. The heavy-duty truck driver’s behavior is characterized by his or her driving style and plays an important role in driving safety. Consequently, this paper proposes a novel framework to classify driving styles of heavy-duty trucks and make comparision under different scenarios. On rural road and urban road, 11 heavy-duty truck drivers were chosen to conduct experiments under no load or full load. VBOX device was applied to collect data including speed, acceleration and location information. K-means clustering was used to divide driving style into three categories including aggressive, normal and calm. The results show that load and road environment have a great influence on the driving style of heavy-duty truck drivers. It is worth noting that heavy-duty truck drivers are more aggressive with full load than no load when on urban roads. The empirical results demonstrate that the proposed method has efficiency in recognizing the driving style and reveal the variations of the driving style of heavy-duty truck drivers under different scenarios. Moreover, it is meaningful and practical to analyze the driving style in improving road construction, traffic safety and reducing energy consumption.

Linghua Yu, Yongfeng Ma, Shuyan Chen, Hong Yao, Muxiong Zhou
Green Transportation and Low Carbon Mobility Safety
Wuhong Wang
Jianping Wu
Xiaobei Jiang
Ruimin Li
Haodong Zhang
Copyright Year
Springer Nature Singapore
Electronic ISBN
Print ISBN