交通信息与安全

2024, 42(4): .

Abstract(716) HTML (744) PDF(5)

Abstract:

Status and Prospects of Studies on Urban Rail Transit Resilience

ZHANG Xiran, LI Zhengzhong, ZHANG Xin, CHEN Shaokuan

2024, 42(4): 1-11. doi: 10.3963/j.jssn.1674-4861.2024.04.001

Abstract(835) HTML (806) PDF(25)

Abstract:
An urban rail transportation system with high-level resilience is able to effectively respond to emergencies caused by natural disasters, human error, equipment failure, and other factors. In order to fully grasp the research trends related to the resilience of urban rail transit, the keywords and hotspots are analyzed with bibliometric analysis. Early studies focuses on the resilience of track structure, while transportation service resilience has gradually gained attention in recent years. Based on the development process of resilience concept in the fields of physics, ecology, and urban management, the connotation of urban rail transit resilience is explained. Oriented to typical social events and natural disaster scenarios, the scope of resilience evaluation is extended from stations to lines and then to networks. However, under current technological conditions, there is a trade-off relationship between the scale and granularity of resilience evaluation. The linkage mechanisms between macro/micro and dynamic/static objects are not fully explored. In terms of resilience evaluation indicators, the indicators system based on topology, transportation capacity, comprehensive performance and operation flow are summarized. The existing indicator system is able to be further enriched in terms of spatial layout, engineering conditions, facilities, equipment, staffing, management and social forces. Four typical indicator measurement methods are sorted out, which include performance curve based on resilience modeling, big data analysis, simulation and numerical analysis. The measurement results based on a single method are easily affected by factors such as the amount of data, hypothetical conditions, and indicator weights. Multiple methods should be comprehensively used to measure indicators in different types and evaluation stages. Resilience enhancement strategies for pre-prevention, in-process adaptation and post-disaster recovery are also discussed. Most existing research primarily approaches from a perspective of operational management, and related studies on post-disaster recovery of infrastructure are still in an initial exploratory stage. Future research on urban rail transit resilience is expected from four aspects: ①Improving the authenticity of emergency scenarios modeling. ②Dynamic fine-grained analysis integrating spatial and functional division of the city. ③Exploring the propagation mechanism of emergencies to characterize system internal changes. ④Verifying effectiveness of resilience evaluation and improvement methods.

A Comprehensive Evaluation Model of Lane-based Risk for Urban Intersections Based on Bayesian Inference and XGBoost

WANG Yongjie, XU Yueying, SU Qian, LI Qiong, YOU Xinshang

2024, 42(4): 12-20. doi: 10.3963/j.jssn.1674-4861.2024.04.002

Abstract(91) HTML (41) PDF(22)

Abstract:
This study addresses the challenges of inadequate researches on lane-scale risk evaluation and the uncertainties in complex interactions at urban signal-controlled intersections. To this end, a comprehensive risk evaluation model is developed based on Bayesian inference and XGBoost. Specifically, this study is based on traffic video data from three intersections in Xi'an: Jixiang Village, Mingguang Road, and Qingsong Road. Two innovative risk-evaluation sets are constructed from the dimensions of temporal and spatial proximities, in which key indicators, including post-entrainment time, maximum speed, distance difference and speed difference are selected to capture dynamic risk characteristics of intersections. Further, Bayesian inference is used to develop a probabilistic evaluation method to address uncertainties in complex interactions at intersections. Next, SHAP value theory of the XGBoost model and Logistic regression are applied to analyze the significance and importance of factors influencing lane risk levels. The results show that: ①The proposed model outperforms baseline models in identifying medium and high-risk interactions. It also more accurately assesses extreme danger interactions, avoiding the overestimation observed in baseline models. ②Among the typical interactions, that between motor vehicle-bicycle, pedestrian-motor vehicle, and pedestrian-bicycle, only a small portion are classified as extreme risk, though medium-risk interactions account for 29.7%, 20.8%, and 34.3%, respectively. ③There are significant differences regarding the risk level across different lanes, with the first lane being more prone to traffic conflicts compared to the second, third, and fourth lanes. ④For all three interaction types, lane risk is mainly influenced by speed, acceleration, and traffic volume. In motor vehicle-bicycle interactions, the highest risk occurs in the first lane and on roads with narrow buffer zones, particularly during morning rush hours and on right-turn lanes. Pedestrian-motor vehicle interactions are primarily influenced by speed and traffic volume, with higher risks in the first lane. For pedestrian-bicycle interactions, narrower bicycle lanes increase the risk of conflicts.

An Assessment Model of Approach Risk Based on QAR Data and Mutual Information Method

WANG Lei, LI Ruijun, WANG Feiyin

2024, 42(4): 21-29. doi: 10.3963/j.jssn.1674-4861.2024.04.003

Abstract(95) HTML (40) PDF(7)

Abstract:
Unstable approaches can easily lead to typical consequence events. This study develops a quantitative risk assessment model to evaluate the risks associated with unstable approaches. Quick access recorder (QAR) data and approach risks are analyzed. Key QAR flight parameters indicative of unstable approaches are selected as monitoring indicators. Twelve monitoring indicators are identified to reflect the state of unstable approaches. The Borda count method is used to rank the monitoring indicators. Based on the ranking results, the study calculates how much each monitoring indicator influences unstable approach events. Potential severe consequences of unstable approaches are analyzed to identify typical consequence events. A risk assessment model is constructed based on the mutual information method from information entropy theory, incorporating the following improvements: ①The mutual information method and the Borda count method are integrated to define a weight that comprehensively reflects the monitoring indicators. This approach overcomes the limitations of using either method in isolation for weight determination. ②Laplace smoothing is utilized to handle the zero-frequency problem in the dataset. Information loss is mitigated, and a necessary complement is provided to the mutual information method, particularly for scenarios characterized by limited sample sizes. ③The correlation between consequence events is considered, and the base risk value is adjusted accordingly. The model is validated using a case study. The results show that using QAR data collected from Airline A in 2019, the model assesses the risk values of runway excursion, CFIT and hard landing, and loss of control in-flight as 4.609 5, 2.062 8, and 0.146 8, respectively. This risk ranking is consistent with the data proportion ranking published by the International Air Transport Association. Indicating that the model results align with actual operational situations. The model's risk rankings are found to be consistent across different aircraft type and years. This consistency is observed when comparing data from Airlines A and B. One hundred experiments are simulated under four different environments. The results show that the risk value trends and distributions share similar characteristics. The consistency between the risk rankings in the simulated and real environments reaches 90% overall. The risk of runway excursion fluctuates with changing conditions. The high-risk value of loss of control in-flight may indicate a serious safety event. The risk of CFIT and hard landing shows little fluctuation, with a uniform distribution, indicating a moderate and predictable risk.

A Driving Fatigue Model for Extra-long Tunnels Based on Multi-source Data

SHANG Ting, LIAN Guan, HUANG Xianlong, XIE Lei

2024, 42(4): 30-41. doi: 10.3963/j.jssn.1674-4861.2024.04.004

Abstract(71) HTML (28) PDF(17)

Abstract:
To investigate the evolution of driving fatigue in extra-long tunnels and its influencing factors, multi-source data from real-vehicle experiments are utilized to classify and identify driving fatigue, as well as to analyze the relationship between fatigue levels and influencing factors. Through significance tests of differences and correlation analysis, the percentage of eyelid closure over the pupil over time (PERCLOS) P80, the variable coefficient of pupil diameter, and acceleration are selected as key fatigue sensitivity indicators, and their changing patterns with accumulated driving time are examined. To construct a driving fatigue classification model, fatigue levels, based on the subjective fatigue detection results from the Karolinska sleepiness scale (KSS), are categorized into awake, semi-fatigued, and fatigued states. A multi-class classifier method is then employed to combine and classify these fatigue states. The grid-search method (GS) is utilized for parameter optimization, and the selected fatigue sensitivity indicators are used as input variables to establish a multi-class support vector machine model (GS-M-SVMs) for fatigue state classification. Following this, an ordinal multi-class Logistic model is developed to explore the relationship between driving fatigue levels and influencing factors in extra-long tunnels. The results indicate that the changing patterns of fatigue sensitivity indicators effectively capture the evolution of driving fatigue. The GS-M-SVMs model achieved a classification accuracy of 90.75%, indicating strong performance in fatigue level detection. Both accumulated driving time and tunnel length significantly influence driving fatigue levels, with regression coefficients of 2.634 and 0.395, respectively. This indicates that accumulated driving time is the primary factor contributing to increased fatigue in extra-long tunnels, while factors such as tunnel illumination and alignment do not significantly impact fatigue levels.

A Risk Analysis of Human-machine System of Civil Aircraft in Take-off Stage

WANG Yifan, SUN Youchao, LIU Xun, JIE Yuwen

2024, 42(4): 42-52. doi: 10.3963/j.jssn.1674-4861.2024.04.005

Abstract(64) HTML (34) PDF(3)

Abstract:
To assess the risk of a human-machine system of civil aircraft cockpits during the take-off phase and identify the dangerous nodes, this paper explores a resilience-based risk analysis method. The method employs the hierarchical task analysis (HTA) and the functional resonance accident model (FRAM) to identify the main functions of the human-machine system, and analyze their interconnections, thereby constructing a system function network model. A risk propagation network throughout the system function network is developed by analyzing the intrinsic and extrinsic influencing factors, and a susceptible-infectious-recovered (SIR) model is introduced to simulate the propagation of risk within this network. An improved cognitive reliability and error analysis method (CREAM) is developed to identify system failure modes and common performance conditions, thereby calculating probabilities of failure, transmission, and recovery within the proposed SIR model. Aiming at the dynamic propagation of risk, an enhanced resilience model is developed to accurately reflect system performance and resilience, which considers the timing of system disturbances and recovery. To validate the proposed method, an example of the take-off process is analyzed, and results show that: ①4 major risk nodes, 7 general risk nodes, 33 low-risk nodes, and 48 minimal-risk nodes are identified. ②In the first three categories of nodes, human errors account for 100%, 42%, and 45% respectively. ③Human factors, including pilot fatigue and visual load, are more likely to form incidents. ④These findings are corroborated with the statistical analysis results. Furthermore, the proposed method analyzes the performance change process of the human-machine system, which reveals challenges in system recovery and the tendency towards secondary risks. In summary, conclusions above confirm the effectiveness of the resilience-based analysis method proposed in this paper, emphasizing the need for risk management strategies.

Spatiotemporal Characteristics of Longitudinal Risky Driving Behaviors in Highway Tunnel Sections

HE Chaoqun, MA Sheqiang

2024, 42(4): 53-61. doi: 10.3963/j.jssn.1674-4861.2024.04.006

Abstract(67) HTML (25) PDF(8)

Abstract:
To pinpoint the patterns, locations, and timings of longitudinal risky driving behaviors in tunnel sections, enhancing the ability of traffic management departments to proactively prevent accidents, this study addresses the limitations of conventional separate spatio-temporal analysis dimensions by developing a spatio-temporal kernel density estimation model (STKDE). The model's optimal bandwidth is determined using least squares cross-validation (LSCV).A method for identifying longitudinal risky driving behaviors based on trajectory data is constructed, extracting spatio-temporal locations for four risky driving behaviors: speeding, extreme low speed, rapid acceleration, and rapid deceleration. By partitioning the spatio-temporal domain of the study area into units, STKDE is applied to compute the spatio-temporal kernel density estimation value, ψ, within each unit. The results of STKDE are visualized using a space-time cube model (ST-Cube). An empirical analysis based on high-precision trajectory data from the Xiaxiyao Tunnel reveals that high-speed driving behavior frequently occurs within 100 meters of the tunnel exit, with speeding peaking at 16:00 and 09:00. Low-speed driving behavior is frequent within 200 meters before the tunnel entrance, with extreme low speed peaking at 02:00 and 14:00. Within 100 meters before entry and throughout the first 1500 meters of the tunnel, the ψ values for rapid acceleration and deceleration remain above 0.5, indicating high-frequency occurrences.. Additionally, every 150~200 meters within the tunnel, these two types of sudden speed changes show simultaneous fluctuations, but significantly decrease and are no longer frequent once exiting the tunnel. A comparison with conventional spatio-temporal analysis methods shows that the STKDE method, combined with ST-Cube, achieves integrated spatio-temporal feature analysis and provides a quantifiable estimation of the likelihood of risky driving behaviors across the entire spatio-temporal domain, demonstrating a particular advantage in characterizing rapid acceleration and deceleration behaviors.

A Severity Analysis of Accidents of Delivery Riders Based on Partial Proportional Odds Model

ZHANG Jingyi, MA Jingfeng

2024, 42(4): 62-71. doi: 10.3963/j.jssn.1674-4861.2024.04.007

Abstract(78) HTML (41) PDF(7)

Abstract:
The existing studies of causal analysis on accidents severity of delivery riders mainly focus on partial fac-tors such as rider, behavior, time, space, road, environment, and accident characteristics. The differences are not quantified in the impacts of different factors on the severity of accidents from the seven aspects above, without con-sidering factors such as takeaway type, number of entrances, angle of intersection, comfort index, and other factors. In addition, when there are both unordered and ordered binary or multi-categorical variables in the independent vari-ables, the established models are limited by the parallel-lines (PL) assumption, and fail to own the flexibility of al-lowing some variables to comply, while others violate this assumption. A total of 1 473 accidents related with deliv-ery riders in Xi'an are selected to analyze the severity distribution and the spatiotemporal distribution. A total of 25 potential influencing factors are selected from the seven aspects. A partial proportional odds (PPO) model is developed to clarify the significant influence of various factors on the rider injury severities in delivery crashes and the vi-olation status of the PL assumption. The corresponding marginal effects are carried out to quantify the differences between and within the contributing factors. The results show that there is a"double peak"phenomenon in the tem-poral distribution of the delivery-involved crashes, and the crash density in urban areas is higher than in suburban ar-eas. The proportion of minor injuries to riders on road sections (35.57%) is higher than at intersections (31.76%). The PPO model performs better than the ordered Logit model and the generalized ordered Logit model. The deliv-ery type, season, location, number of entrances, intersection angle, road surface, weather, and comfort index all fol-low the PL assumption. There are significant differences in the crash severity among different significant factors. Traffic violations such as running red lights, going in the wrong direction, and speeding have the greatest impacts on the rider crash severity, with the maximum absolute value of the marginal effects exceeding 51%. However, some unexplored factors such as number of entrances, intersection angles, bicycle lane, delivery type, and comfort index significantly affect the severity (8% to 37%).

An Analysis of Risk Factors of Traffic Safety for Heavy Trucks Based on Model Group

KE Xingan, ZHAO Dan, WANG Qiuhong, HU Yuening, NIU Shuai

2024, 42(4): 72-80. doi: 10.3963/j.jssn.1674-4861.2024.04.008

Abstract(38) HTML (40) PDF(3)

Abstract:
To explore the risk factors and occurrence mechanisms behind the traffic accidents of heavy trucks in-depth, a model group, which comprises random forests (RF), Logistic regression (LR), geographically weighted Logistic regression (GWLR), and Bayesian network (BN), is established based on the data of heavy truck accidents in a certain province from 2016 to 2021. This model group allows for examining the impact magnitude, spatial het-erogeneity, as well as the causal pathways leading to accidents of the risk factors. The results reveal that: ①The driv-ing status of heavy trucks, collision patterns, and other eight factors have significant impacts on risks. The impact of rural traffic participants, as well as frontal and side collisions, varies slightly across different models, while the im-pact of rear-end collisions is higher in GWLR compared to that in BN.②When the heavy trucks are engaged in right turns, illegal behaviors or vulnerable road users, the risk of fatal accidents increases by 39.0%, 41.9% and 39.3%, respectively.③With the factor collision patterns serving as a mediator, the causal pathway analysis for the risk of fatal accidents indicates: the factors side collisions and vulnerable road users contribute to the increase of the risk of fatal accident by 64.4% compared to the impact of the factor scrapes involving other vehicles, therefore can be marked as the typical hazardous scenario. Furthermore, when the other participant of heavy truck accidents involves drivers not older than 30 years, the probability of rear-end collisions increases by 10.3% and 26.1% com-pared to those involving drivers aged 30-60 and above 60 respectively. ④Among the risk factors exhibiting spatial heterogeneity, the factor frontal collision exhibits the highest intensity, whereas the factor right turn shows the least. In conclusion, the analytical framework based on the model group can be used to identify the significant risk factors for traffic safety of heavy trucks, and to verify the differences of the impact across models and the notable spatial heterogeneity for these factors.

Impacts of Connected Warning Information on Driver Behavior in Pedestrian-vehicle Conflict at the Mid-block of Urban Roads

WANG Changshuai, SHAO Yongcheng, ZHU Tong, JIAO Yanli, XU Chengcheng

2024, 42(4): 81-89. doi: 10.3963/j.jssn.1674-4861.2024.04.009

Abstract(41) HTML (19) PDF(6)

Abstract:
This study investigates the impact of connected warning information on driver behavior during pedestri-an-vehicle conflicts on urban roads. Using a driving simulator, urban driving scenarios and connected warning infor-mation systems are designed, incorporating various types of visual blind areas to create six typical pedestrian-vehi-cle conflicts. Seventy participants are recruited and divided into experimental and control groups to complete the simulator tests, during which driver behavior and vehicle trajectory data are collected. Survival analysis is employed to examine the effects of different factors on drivers' reaction times and braking durations during conflicts. Addi-tionally, pedestrian-vehicle crash prediction models are developed to assess crash risk and evaluate the influence of connected warning information on driver behavior. Results indicated that blind spots caused by buses, trees and cars, and parked cars negatively impacted driver performance, resulting in longer reaction times and shorter braking durations. Contrarily, the presence of crosswalks reduced the mean avoidance reaction time by 0.90 s, increased the mean braking duration by 0.41 s, and lowered pedestrian-vehicle crash risk. Furthermore, connected warning infor-mation is found to positively affect driver behavior, reducing mean avoidance reaction times by 0.52 s and increas-ing braking duration by 0.40 s. This leads to smoother braking processes and significantly decreased crash risks, thereby enhancing pedestrian safety.

An Inspection Method of Urban Road Parking Based on UAV Image

WANG Chen, ZHANG Lingyun, LIU Bo, ZHANG Hang

2024, 42(4): 90-101. doi: 10.3963/j.jssn.1674-4861.2024.04.010

Abstract(43) HTML (19) PDF(4)

Abstract:
Efficient and accurate inspection of parked vehicles on urban roads is of significant importance for smart city management. Addressing the issues of low efficiency, high cost, and inaccuracy in current inspection methods, a drone-based image inspection approach is investigated. To enable all-weather inspection, an illumination enhance-ment algorithm is employed to boost images captured under low-light conditions, while a deblurring algorithm is uti-lized to improve the quality of blurred images. To tackle the limitations of the existing YOLOv5 algorithm, includ-ing insufficient detection accuracy and real-time performance, several modifications are introduced. The Fo-cal-EIOU Loss function is optimized to accelerate model convergence. The C3 module is replaced with the C2F module, utilizing varying sizes of convolutional kernels to extract features, enhancing adaptability to targets of dif-ferent sizes and shapes. Furthermore, the SimAM attention mechanism is incorporated to improve the network's ro-bustness and anti-interference capability, predicting 3D attention weights for feature maps without increasing model parameters. The CARAFE operator is adopted for upsampling to expand the receptive field, comprehensively lever-aging semantic information from feature maps. Experimental results demonstrate that the modified YOLOv5 model achieves 5.1% increase in accuracy, 5.9% improvement in recall, and 3.6% enhancement in mean Average Precision (mAP). Secondly, the SVTR character recognition network is utilized to identify license plate numbers, accomplish-ing both feature extraction and text transcription tasks within a single vision model. Finally, field tests conducted on a drone-based engineering application platform reveal that this inspection method can accurately, rapidly, and intelli-gently complete inspections, achieving an accuracy of 90% and a detection speed of 170 frames per second, essentially meeting inspection precision and real-time requirements.

An Expressway Traffic Flow Prediction Method Considering Inter-Lane Differences and Upstream and Downstream Cross-Section Correlations

LI Chun, ZHANG Cunbao, CHEN Feng, FU Dingjun

2024, 42(4): 102-109. doi: 10.3963/j.jssn.1674-4861.2024.04.011

Abstract(36) HTML (14) PDF(6)

Abstract:
In the existing traffic flow prediction studies, the differences of traffic flow between different lanes within a cross-section road and the correlation of traffic flow between upstream and downstream cross-sections are not fully considered. This research develops a framework for expressway traffic flow prediction by integrating principal component analysis (PCA) and long short-term memory (LSTM) neural networks, which satisfies the real-time and accuracy requirements of intelligent connected technologies. The traffic flow data of urban expressways are collected, and the fast Fourier transform (FFT) method is applied for data preprocessing to improve the predictability of the original data. PCA is used to fuse the features of lateral and longitudinal traffic flow between lanes, establishing correlated traffic flow data between lanes to reduce data dimensionality. The correlated data is then integrated into the LSTM model for lane-level traffic flow prediction, and the prediction results aggregated to obtain the cross-section traffic flow estimates. The proposed method is validated using traffic data from expressway checkpoint detectors along the Third Ring Road in Wuhan. The results demonstrate that the model considering inter-lane differences and upstream and downstream cross-section correlations can improve the accuracy of cross-section traffic flow prediction. Compared to prediction results that only consider temporal features, the mean absolute error, root mean square error, and mean absolute percentage error can be reduced by 6.66%, 6.23%, and 17.51%, respectively. When compared with models that account for either upstream-downstream correlations or inter-lane differences alone, the proposed model demonstrates superior performance, with reductions in mean absolute error, root mean square error, and mean absolute percentage error ranging from a minimum of 1.53% to a maximum of 12.88%. Additionally, the proposed model demonstrates higher prediction accuracy than support vector regression (SVR) and random forest (RF) algorithms. In time-segmented predictions, the model performs particularly well during the evening peak and off-peak hours.

A Study on the Visual Behavior Characteristics of Drivers at the Entrance Area of Freeway Tunnels in Foggy Weather

JIANG Wenyi, DU Zhigang, HE Shiming, MEI Jialin, HAN Lei

2024, 42(4): 110-117. doi: 10.3963/j.jssn.1674-4861.2024.04.012

Abstract(51) HTML (16) PDF(7)

Abstract:
The tunnel entrance area and foggy environments significantly impact drivers' vision, causing changes in drivers' visual behaviors that lead to alterations in driving behaviors, thus affecting traffic safety. To investigate the influence of fog on characteristics of drivers' visual behavior at the tunnel entrance area, a real-vehicle test is conducted in Sanhu Tunnel and Hankou Tunnel in Henan Province, and 24 participants are recruited to analyze the differences in drivers' visual behavior characteristics inside and outside the tunnels under foggy and clear weather, with drivers' fixation and saccade behaviors as the main objects of the study. The results indicate significant differences in drivers' visual behavior characteristics between foggy and clear weather. ①Before entering the tunnel in foggy weather, fixation distribution becomes more concentrated, the percentage of fixation durations exceeding 1 000 ms increases, with foggy conditions at night reaching 29%, average fixation duration increases, the frequency of fixation decreases, the range and efficiency of visual search decrease, and saccades within the range from 0° to 5° accounting for 88.89%, drivers pay less attention to the environments on both sides and try to search for the tunnel portal, demonstrating more cautious driving behavior. ②After entering the tunnel in foggy weather, the range of fixation points expands to varying degrees, average fixation time and average saccade duration increase, the range and efficiency of visual search improve, and drivers pay more attention to the tunnel sidewalls and contour. ③In foggy weather, nighttime has a greater impact on drivers' visual behaviors than daytime, and the level of driving caution increases further. ④During the daytime, fog reduces the visibility of the visual-guiding facilities at the tunnel portal, leading to poor recognition of the tunnel portal, drivers experience a dramatic change in their visual environment, which increases visual load and psychological pressure, significantly increasing driving risk.

A Study on Parallel Routes Lateral Separation for Urban Logistics UAV

ZHANG Jian, ZHAO Yifei, LU Fei, LUO Xinyue, LI Zongxiao

2024, 42(4): 118-124. doi: 10.3963/j.jssn.1674-4861.2024.04.013

Abstract(47) HTML (15) PDF(3)

Abstract:
To improve the operational safety of urban logistics with unmanned aerial vehicles (UAV) in high-density regions, lateral separation (LS) for UAVs in parallel routes is studied. Specifically, the conflict frequency and collision probability are taken into account in the LS model and the LS for UAVs meeting the safety requirements is determined. The target level of safety (TLS) is set and the criteria of conflict is defined as the lateral distance of adjacent UAVs is less than the required LS. Then, the frequency of conflict can be counted by using the real operation data of UAVs. An urban logistics system using UAVs is simulated, comprehensively incorporating the UAVs' performances, flight flow, airspace volume, deviation probability, deviation angle, conflict detection and resolution, etc. The Monte Carlo method is introduced to assess the probability of collision by simulating the scenarios where a UAV has a random accidental deviation causing conflict with the adjacent UAVs under a given probability. A total of 5.1 million rounds of simulations are performed and the LS ranges from 1 meter to 51 meter. The results show that: ①50 302 violations are observed; ②the probability density function (PDF) of collision and the LS value fit an exponential distribution, well; ③applying a proper lateral separation would be a potential risk mitigation strategy for reaching TSL. In summary, the required lateral separation for UAVs in the urban city is suggested as 33 m by comparing the accepted TSL with the residual risk calculated by the fitted PDF.

An Energy Management Strategy for Hybrid Ship Based on SVM and MPC

LIANG Tianchi, YUAN Yupeng, TONG Liang

2024, 42(4): 125-135. doi: 10.3963/j.jssn.1674-4861.2024.04.014

Abstract(35) HTML (35) PDF(3)

Abstract:
To improve the energy efficiency and fuel economy of hybrid ships, energy management strategy for hybrid ships based on support vector machine (SVM) and model predictive control (MPC) is proposed, incorporating the working condition recognition into the strategy management. A working-condition recognition model is developed based on the SVM theory, the kernel function type and key parameters are optimized by using the operation data from the hybrid powered cruise ships called"MEIWEI KEYUE", and the one of four working conditions is recognized by feeding the real-time operation data. An energy management MPC model (EM-MPC) is proposed, allocating the output powers of main engine and composite ESUs for minimizing the consumption of fuel and maintaining the state of charge (SOC) stabilization of energy storage unit (ESU), which is constrained by the power demand prediction (PDP) model. Then, a prediction model based on the multi-step Markov model is proposed to improve the accuracy of PDP under different working conditions, and the PDP constraint in the EM-MPC model is updated based on the recognized working condition, which contributes to the real-time power allocation. The optimal solution is decomposed into high frequency signal and low frequency signal by wavelet transform method, and these signals are assigned to the super capacitor with high power density and the battery with high energy density, respectively. To validate the proposed strategy, a simulation via Matlab is introduced and the results show that: ① the cumulative fuel consumption of the proposed method is 4 404.556 1 g and the average fuel consumption rate is 202.9737 g/kWh; ② under the same working condition, the fuel consumption can be saved by 4.55% via the proposed EM-MPC, comparing with the traditional method.

Determinants of Pedestrians' Psychological Boundary in the Coexistence of Pedestrian and Non-motorized Vehicles

TANG Tianpei, YUAN Quan, YUAN Meining, CHEN Zhiyu, LIN Xinrong

2024, 42(4): 136-143. doi: 10.3963/j.jssn.1674-4861.2024.04.015

Abstract(52) HTML (19) PDF(7)

Abstract:
Investigating the mechanisms influencing pedestrians' psychological boundary in the coexistence of pedestrians and non-motorized vehicles is crucial for improving pedestrian safety and comfort in traffic design and management. This study addresses the limitations of previous studies, which mainly focus on pedestrians'interpersonal spatial expectations. To this end, this study redefines psychological boundaries between pedestrians and non-motorized vehicles based on the mental envelop theory from the two key dimensions: the subject mental envelope (SME) anticipated by pedestrians and the object mental envelope (OME) imposed by pedestrians concerning non-motorized vehicles. Next, the proposed model incorporates perceived threat, perceived closeness, and personal characteristics, and priority as a potential influencing factor. The structural equation model with partial least squares is employed to assess model fit and conduct path analysis. The estimation results show that the SRMR value is 0.035, and R² values for SME and OME are 0.728 and 0.773, respectively, indicating good fit and strong explanatory power. Further, results show that perceived threat and closeness significantly affect both SME and OME, with stronger effects on OME, which suggests that pedestrians perceive non-motorized vehicles as a threat and demand stricter boundaries for their riding areas. Priority demonstrates a greater positive impact on OME than that on SME, suggesting that pedestrians, when feeling prioritized, expect stricter constraints on non-motorized vehicles. The traffic volume of non-motorized vehicles has a substantial impact on OME. OME also positively influences SME, suggesting that the expanded OME can further elucidate pedestrians' requirements for their safe-activity space. Notably, gender and height of a pedestrian significantly impact both SME and OME, while average weekly cycling frequency significantly affects OME exclusively.

A Game Theory-based Lane Change Decision Model for Automated Vehicle Leaving the Freeway Dedicated Lane for Automated Driving

LU Chunyi, HE Shanglu, GAO Binbin, CAO Congyong, FAN Zhengwen

2024, 42(4): 144-153. doi: 10.3963/j.jssn.1674-4861.2024.04.016

Abstract(57) HTML (25) PDF(6)

Abstract:
Under the condition where there is dedicated lane for autonomous vehicles on the freeway, autonomous vehicles need to conduct a consequence lane-changing maneuvers to leave the dedicated lane as well as the freeway mainline. These series of actions would bring the conflicts between autonomous vehicles and human-driving vehicles, which would increase the risk of collisions but decrease the traffic efficiency of diversion area. This study developed a lane-changing decision model for autonomous vehicles based on game theory and model predictive control, considering the impact of different types of lane and mixed traffic conditions. A measure, lane changing crisis, is proposed for autonomous vehicles, which is calculated based on the distance from automated vehicle to the off-ramp and the distribution of acceptable lane change gap in adjacent lanes. The driving style of human-driving vehicle is valued by the variation of acceleration. The cost functions in the model predictive control method for different types of vehicles are formulated. The optimal acceleration of autonomous vehicles at the next time interval could be predicted based on the current traffic state. The interaction between autonomous vehicle and human-driving vehicle is described by the Stackelberg game. The lane changing decision is made to maximize the benefit of autonomous vehicle when the benefit of human-driving vehicle is the largest. And the optimal acceleration of autonomous vehicle at the next time interval is also achieved. A simulation platform integrating Python and SUMO is constructed. Four experiment scenarios with different traffic densities under different lane types and mixed traffic conditions are set. And the proposed model has been also compared with the other two lane-changing models, e.g. the default model within SUMO. The results indicate that the proposed model could consistently make optimal lane-changing decision to minimize the loss of speed. And compared with the other lane-changing models, the proposed model could increase the average speed during lane changes increasing by 1.44% and 11.81%, respectively, which validate the outperformance of the proposed method.

A Model of Associations Between Expressions of Driving Anger and Inducements for Extreme Commuting Group

WAN Ping, QIU Gangao, CHEN Peidong, MA Xiaofeng

2024, 42(4): 154-163. doi: 10.3963/j.jssn.1674-4861.2024.04.017

Abstract(33) HTML (15) PDF(7)

Abstract:
The extreme commuting population (individuals with commuting time exceeding 60 minutes) is susceptible to driving anger due to prolonged and high-stress commuting environments, which can adversely affect traffic safety. This study focuses on the phenomenon of "road rage" among extreme commuters and develops a model to quantify the associations between driving anger expressions and driving anger inducements within this group. The driving anger scale for extreme commuting group (EC_DAS) and the driving anger expression inventory for extreme commuting group (EC_DAX) are designed and surveyed to a cohort of 450 commuters traveling between Yanjiao and Beijing, China. Based on the survey data, scales are revised through exploratory factor analysis and tests of reliability and validity. Next, a model of association between expressions of driving anger and inducements for extreme commuting group is developed with discourtesy, traffic obstructions, slow driving, extreme commuting, and illegal driving as exogenous latent variables, and use of the vehicle to express anger and verbal aggression as endogenous latent variables. The impact of these anger triggers on the expression of driving anger in the extreme commuting group is quantified using a structural equation model. The results are as follows: ①In the EC_DAS, the highest score is observed for slow driving (3.37), followed by extreme commuting (3.07), with illegal driving receiving the lowest score (2.95). In the EC_DAX, verbal aggression scored higher (2.99) than the use of the vehicle to express anger (2.90). ②The structural equation model exhibits a strong goodness of fit, whose results show that use of the vehicle to express anger and verbal aggression are significantly and positively influenced by driving anger inducements including discourtesy, traffic obstructions, slow driving, extreme commuting, and illegal driving. Moreover, it is noted that these factors explain a higher variance in verbal aggression (38%) than in use of the vehicle to express anger (37%). ③Additionally, slow driving, extreme commuting, and traffic obstructions emerge as the three most significant inducements of use of the vehicle to express anger, with standardized effect coefficient of 0.221, 0.169 and 0.162, respectively, while traffic obstructions, slow driving, and discourtesy are identified as the three most significant inducements of verbal aggressive, with standardized effect efficient of 0.215, 0.189, and 0.148, respectively. ④Gender and monthly income do not have significant impacts on anger levels under different driving inducements or anger expression. However, age is significantly negative-correlated with anger levels induced by discourtesy, traffic obstructions, and illegal driving. Driving experience is significantly negative-correlated with anger levels induced by extreme commuting, education level is significantly negative-correlated with anger levels induced by slow driving, and job position is significantly negative-correlated with anger levels induced by traffic obstructions.

Carbon Emission Prediction Method for Inland Ports Based on an Improved ASIF Model

LI Zhihao, CHEN Guang, MA Xiaofeng, LAI Hongjia, GAO Jie, ZHONG Ming

2024, 42(4): 164-174. doi: 10.3963/j.jssn.1674-4861.2024.04.018

Abstract(43) HTML (12) PDF(9)

Abstract:
Addressing the complexities and low accuracy of prediction associated with medium-to-long-term forecasting of port carbon emissions, this study proposes a carbon emission prediction (CEP) model for inland container ports based on an improved activity-modal structure-energy intensity-emission factor (ASIF) method. The objective is to quantify the impact of primary factors on long-term port carbon emissions, thereby providing a basis for targeted carbon neutrality strategies. By considering port container throughput, equipment structure, energy consumption intensity, and emission factors as influential factors of port carbon emissions, and accounting for the "multi-process, multi-equipment" characteristics within the container transportation chain, an improved ASIF model is established, which enables CEP from macro to micro levels. A scenario prediction indicator system is developed based on the explanatory variables of the ASIF model. Taking a container port on the Yangtze River as an example, predictions are made for its throughput, equipment composition, and transportation structure under the business-as-usual (BAU) scenario and the low-carbon (LC) scenario. Subsequently, carbon emissions from ship navigation, ship berthing, quay cranes, internal container trucks, yard cranes, and external container trucks are calculated. Lastly, to analyze the emission reduction potential under different low-carbon development strategies, a single-factor experimental approach is employed. The results indicate that: ①compared to existing prediction models, the deviations of carbon emission by the improved ASIF are within 10%. ②Under BAU and LC scenarios for the case port, with the continuous growth of container throughput, carbon emissions have not yet peaked by the year 2060 in the BAU scenario, whereas they are projected to the peak around the year 2055 in the LC scenario. ③Ship emission control, energy efficiency improvement, energy structure optimization, and collection and distribution system optimization are all effective low-carbon development strategies, albeit with decreasing effectiveness. ④Between the year 2020 to 2060, these strategies could achieve cumulative carbon reductions of approximately 190 000, 170 000, 144 000, and 11 000 t, respectively.

Current Issue 2024, Volume 42, Issue 4

Current Issue
2024, Volume 42, Issue 4