Traffic Safety Research Articles

Analyzing the Suitability of Vehicle Telematics Data as a Surrogate Safety Measure for Short-Term Crashes

Traffic safety analysis relies heavily on the comprehensiveness of traffic crash data. However, crash occurrences are rare events. Researchers have sought alternative measures closely linked to crash occurrences, known as surrogate safety measures, to address this challenge. The advent of vehicle telematics has introduced a valuable source of data, consisting of time-stamped positional information of vehicles, facilitated through telecommunication networks. Wejo Data Services Inc. collects vehicle telematics data through collaboration with various automotive original equipment manufacturers (OEMs) to furnish connected vehicle data along with driver event variables, such as braking and acceleration. This study seeks to investigate the relationship between hard-braking events and the incidence of total crashes and peak period crashes across different facility types using analysis of variance and negative binomial models. The results show a strong correlation between hard braking and crash frequency. The ratio of hard-braking events to crashes exhibits notable variations across peak hours, off-peak periods, and nighttime. Further, the coefficient of hard braking is significantly higher on freeways compared with non-freeways. The model performance of different short period models improved better than the extent of improvement in total crash model after hard braking was introduced as an explanatory variable. The marginal effects of hard braking on crash frequency on freeways are consistently higher than those in non-freeway segments.

How do different seasonal environmental colors of highway tunnel entrances and dominant colors of portal walls affect the driver’s vehicle control ability?

The entrance section of the highway tunnel is a high-risk area for traffic accidents, and its traffic accident rate has significant seasonal differences. However, the dominant causes of this difference are still lacking in targeted research. On the other hand, due to the lack of a mature design system and normative guidance, the landscape design of tunnel entrances is more and more random, landscape elements tend to be complex, and color design tends to be diversified. The influence of tunnel entrance landscape design on driving behavior and its applicability in different seasonal environments needs to be studied. Given this, this study focused on two main variables − the typical seasonal environmental colors and different portal colors, and designed 12 schemes composed of three seasonal environments (green spring, golden yellow and red autumn, and white winter) and four portal colors (gray, blue, red, and tawny). Thirty subjects were invited to conduct a driving simulation experiment to study the influence of different seasonal environmental color characteristics and tunnel portal color on the amplitude and stability of driving behavior. The results show that the drivers’ behavior amplitude is large in spring and winter, and the traffic safety risk is relatively high. The typical color characteristics of the environment in different seasons are one of the reasons for the seasonal difference in the traffic accident rate. The driver’s operational stability is relatively poor in the autumn environment, which is related to the high scene complexity of the season. Compared with the grey portal scheme, the color design of the portal will increase the driver’s driving speed and the driving safety risk. However, its influence on the stability of driving behavior is bidirectional, and the color design coordinated with the environment can improve the stability of driving. When the portal is ready for landscape decoration, it is recommended to use a color scheme that is in harmony with the environment.

Magnitude and outcome of road traffic accidents among patients admitted in dessie town governmental hospitals, Northeast Amhara, Ethiopia, 2022

BackgroundRoad traffic accidents(RTA) are a major public health problem worldwide, accounting for almost 1.24 million deaths per year and it is the number one cause of death among those aged group 15–29 years. Even though there are great benefits from access to road transportation there also poses a great challenge in the individual’s daily activities ranging from minor injury to death.ObjectiveThis study aimed to assess the magnitude and outcome of road traffic accidents among patients admitted in Dessie Town Governmental Hospitals, Northeast Amhara, Ethiopia, 2022.MethodsA five-year hospital-based retrospective descriptive cross-sectional study design was conducted among 377 road traffic accident patients admitted to Dessie Town Governmental hospitals. Data were collected by simple random methods based on patient chart reviews from June 7/, 2022 to May 23/ 2017 using a checklist adapted from the WHO standard hospital-based road traffic accident questionnaires after obtaining consent from the concerned authority. EPI-Data software version 7.2 for data entry and SPSS version 25 for statistical analysis were used. Descriptive and inferential statistics were used. Statistical significance was declared at a p-value of < 0.05 with an adjusted odds ratio (AOR) and a 95% confidence interval (CI) in the final multinomial logistic regression model.ResultsThe magnitude of road traffic accidents was 59%, using of logistic multi nominal logistic regression we found results such that, road traffic victims who had unstable vital signs at admission (AOR = 6.4,95% CI; 2.5–16.6), didn’t get prehospital treatment (AOR = 9.3,95% CI; 4–20), and severe injury (AOR = 9, 95% CI;7-15.4), had a Glasgow coma scale of 3–5 (AOR = 5.2,95% CI; 1.4–20) were found predictors for death were as unstable vital signs at admission (AOR = 3.79,95%CI;2.1–6.8), Doesn’t get prehospital treatment (AOR = 2.8, 95% CI; 1.4–5.7), Hospital stay for one to two months duration (AOR = 6,95% CI;2.3–15), and greater than two months duration (AOR = 6.5,95%CI;2.5–17) were found predictors for disability among road traffic victims.Conclusions and recommendationsRoad traffic accidents constitute a major public health problem in our setting and contribute significantly to excessively high morbidity and mortality. Unstable vital signs at admission, Client doesn’t get prehospital treatment, severely injured client, and had a Glasgow coma scale of 3–5 were found predictors for death were as an unstable vital sign at admission, Client doesn’t get pre-hospital treatment, Hospital stays for one to two months duration, and greater than two months duration were found predictors for disability among road traffic victims.

Wait or Pass? Promoting intersection’s cooperation via identifying vehicle’s social behavior

Lack of communication between road users can reduce traffic efficiency and cause safety issues like traffic accidents. Researchers are exploring how intelligent vehicles should communicate with the environment, other vehicles, and road users. This study explores the impact of social information communication on traffic safety and efficiency at intersections through vehicle-to-vehicle (V2V) communication. The research examines how these factors influence drivers’ decision-making and cooperative behavior by incorporating social value orientation (SVO) and driving agent identity into V2V systems and automated vehicle (AV) decision-support systems. An experimental platform simulating intersection conflict scenarios was developed, and three studies involving 334 participants were conducted. The findings reveal that providing drivers with social information about opposing vehicles significantly promotes cooperative behavior and safer driving strategies. Specifically, the waiting rate for people facing proself vehicles (Mean = 0.22) is significantly higher than when facing prosocial vehicles (Mean = 0.79). When SVO is unknown, the waiting rate is around 0.5. Participants behaved more waiting when confronted with an AV than human-driven vehicles. With AV recommendations based on SVO, participants’ final waiting rate increases as the recommended waiting rate increases. The optimal recommended waiting rate for AV is most acceptable when it matches the average waiting rate of the other vehicle. This research underscores the importance of integrating social information into V2V communication to improve road safety, aiding in designing automated decision-making strategies for AV and enhancing user satisfaction.

The association of road, traffic, and environmental factors with wrong-way riding behavior on shared e-bikes

Wrong-way riding (WWR) behavior of electric bikes (e-bikes) is a hazardous practice that frequently threatens traffic safety. It is imperative to accurately detect instances of WWR and analyze the various factors contributing to this behavior to develop and implement effective management policies. The emergence of shared e-bikes and advancements in machine learning have opened up new avenues for accurately examining the determinants of WWR behavior. Therefore, this study utilizes a large-scale dataset of shared e-bike trajectory data to establish a framework for detecting WWR behavior. Subsequently, association rule mining investigates the correlation among road, traffic, environmental factors, and the likelihood of higher and lower WWR behavior. The results indicate that: (1) In the association rules for higher WWR frequency, the absence of central separation facilities and bike lane separation facilities on the tertiary road, as well as lower land use density and the lack of public transportation facilities, enhanced the likelihood of WWR occurrence; (2) In the association rules for lower WWR frequency, longer crossing facility distance, physical bike lane separation, along with higher land use mixing and the presence of public transportation facilities, reduce the probability of WWR occurrence; (3) Time conditions display spatial heterogeneity, and different land use and road condition modulate the impact of time factors on rider’ intention for WWR. Finally, based on the study findings, adequate policy recommendations are proposed for urban management agencies to reduce WWR behavior and enhance the safety of the e-bike traffic system.

The Mediating Role of Traffic Safety Awareness on Road Safety Attitude and Behavior of Drivers

This research study aimed to determine the mediating role of traffic safety awareness on the relationship between road safety attitude and behavior of drivers. The researcher surveyed 378 licensed drivers selected using a stratified random sampling technique. The study utilized Mean, Pearson Product Moment Correlation, Medgraph using Sobel z-test and Regression. The study found a very high level of Traffic Safety Awareness, a moderate level of Driving behavior, and high Road Safety Attitudes among licensed drivers. A significant correlation was observed between road safety attitude and drivers’ behavior. There is also a high significance on the relationship between road safety attitude and traffic safety awareness, as well as between traffic safety awareness and drivers’ driving behavior. The mediation analysis examined the role of traffic safety awareness as a mediator in the relationship between road safety attitude and driver behavior. The results point towards partial mediation, where the mediator variable partially explains the relationship between the independent and dependent variables.

Trend, disparities, and projection analysis of public data on foot fractures in Sweden: a retrospective analysis of 179 129 fractures

IntroductionOrthopedic injuries to the foot constitute a significant portion of lower extremity injuries, necessitating an understanding of trends for effective preventive strategies and resource allocation. Demographic shifts, improved traffic safety, and lifelong physical activity may alter incidence rates, trauma mechanisms, and fracture distribution. This study explores the prevalence of foot fractures in Sweden using publicly available data.MethodsUtilizing data from the Swedish National Board of Health and Welfare (SNBHW) spanning 2008–2022, retrospective study focuses on foot fractures in Sweden. Analysis includes calculating annual incidence rates per 100,000 person-years, assessing temporal trends, and exploring seasonal variations. Poisson regression analysis was used for projections into 2035.ResultsBetween 2008–2022, the average annual foot fracture incidence was 11,942, with notable fluctuations influenced by the COVID-19 pandemic. Age and sex disparities impact rates, and seasonal variance highlights increased incidence in summer. By 2035, foot fractures will decreasae amongst several demographic groups.ConclusionThis study provides insights into temporal trends, sex differences, and seasonal variations foot fracture patterns in Sweden. The identified trends suggest the utilization of targeted preventive strategies, efficient resource allocation, and informed healthcare planning. Despite limitations, this research offers valuable insights into foot fractures within the Swedish population, utilizing publicly aggregated data.

Handover for V2V communication in 5G using convolutional neural networks

Vehicle communication is one of the most vital aspects of modern transportation systems because it enables real-time data transmission between vehicles and infrastructure to improve traffic flow and road safety. The next generation of mobile technology, 5G, was created to address earlier generations' growing need for high data rates and quality of service issues. 5G cellular technology aims to eliminate penetration loss by segregating outside and inside settings and allowing extremely high transmission speeds, achieved by installing hundreds of dispersed antenna arrays using a distributed antenna system (DAS). Huge multiple-input multiple-output (MIMO) systems are accomplished via DASs and huge MIMO systems, where hundreds of dispersed antenna arrays are built. Because deep learning (DL) techniques employ artificial neural networks with at least one hidden layer, they are used in this study for vehicle recognition. They can swiftly process vast quantities of labeled training data to identify features. Therefore, this paper employed the VGG19 DL model through transfer learning to address the task of vehicle detection and obstacle identification. It also proposes a novel horizontal handover prediction method based on channel characteristics. The suggested techniques are designed for heterogeneous networks or horizontal handovers using DL. In the designated surrounding regions of 5G environments, the suggested detection and handover algorithms identified vehicles with a success rate of 97 % and predicted the next station for handover.

Congruency of Crash- and Hospital- Reported Injuries Among Child Passengers

Introduction: Prior work has found incongruencies in injury information reported by crash and hospital records. However, no work has focused on child passengers. The objective of this study was to compare crash scene and hospital-reported injury information for crash-involved child passengers. This study also explored injury location and severity by child age and restraint type. Methods: Utilizing linked New Jersey data from 2017 through 2019, the authors identified crash-involved child passengers <13 years old and their injuries in crash and hospital reports. Then, they characterized the congruency of injury frequency, severity, and location, as well as the frequency of injuries by child age and restraint type. Analyses were conducted from December 2023 through February 2024. Results: Of 84,060 crash-involved child passengers, crash reports documented 7,858 (9%) children with at least “possible” injuries, while 2,577 (3%) had at least one injury in hospital events. Crash report and hospital data were incongruent for both body region of injury and injury severity. The proportion of children injured increased as children's ages increased and as restraint type progressed. Conclusions: Crash reports overestimated the number of injured child passengers and misrepresented injury severity and locations. Child restraint systems mitigated a child's injury risk. Importantly, injury information documented on crash reports currently informs the allocation of traffic safety resources. These results highlight the importance of improving these reports’ accuracy and underscore calls to link administrative datasets for public health efforts.

Evaluating Friction Characteristics of High Friction Surface Treatment Application Under Varied Polishing and Slippery Conditions

The frictional attributes of high friction surface treatment (HFST) play a crucial role in ensuring optimal traffic safety, particularly in wet weather conditions. Friction consists of two important components: adhesion and hysteresis. This research focuses on evaluating these essential factors in HFST using two different aggregates and two distinct sizes by considering various abrasion and polishing methods. To isolate these components for assessment, testing was carried out under various slippery conditions, including dry, wet with water, and wet with water + soap. The inclusion of liquid hand soap in the test procedure effectively minimized or even eliminated the adhesion component’s influence, making it possible to primarily focus on the hysteresis component. Consequently, the British Pendulum Number (BPN) measured in this research predominantly reflected the hysteresis-related friction. The analysis of variance results emphasized the substantial impact of different abrasion and polishing methods on the BPN values obtained under various surface conditions. Notably, Micro-Deval Abrasion (MDA) with 105, 180, and 240 min abrasion exhibited the most pronounced influence on BPN variation and the higher F-value for the MDA 105 min indicated that this specific abrasion time exerted a more significant influence on the variation in BPN values than other factors. Furthermore, the utilization of the Aggregate Image Measurement System yielded valuable insights into the micro-texture of the aggregates. It revealed that the calcined bauxite HFST size is anticipated to provide the rougher surface morphology (texture) on a pavement surface compared with other sources in this study, thereby contributing to high pavement surface friction. The findings from this study contribute to a deeper understanding of the frictional characteristics of HFST under different scenarios, providing valuable insights for optimizing HFST applications to enhance road safety and skid resistance.

Pemberdayaan Guru Sekolah Mengenai Keselamatan Berlalu Lintas

The behavior of drivers in the Tegal City school community often violates traffic rules because they do not understand and are not aware of traffic safety. Increased knowledge, understanding, awareness and skills of teachers regarding traffic safety need to be increased and empowered more massively. Empowering school teachers regarding traffic safety includes 4 stages (preparation, initial test, community empowerment, and final test) with traffic safety community results. 240 PAUD/TK/RA, SD/MI, SMP/MTs, and SMA/SMK/MA teachers as participants for 2 days with 7 topics. The majority of participants consisted of bachelor's degrees (89.6%), women (55.8%), ages 36-46 (35%), and PAUD teachers (25.8%). The initial test for 7 topics is still between 20%. Activities on the first day are stages 1-4 and on the second day stages 5-7. The final test of 7 topics is above 75%. 7 topics had an increase of more than 55%. Increasing school teachers' knowledge has an impact on imparting knowledge to students about traffic safety which can be a step to reduce the occurrence of accidents on the road.

Road car image target detection and recognition based on YOLOv8 deep learning algorithm

In this paper, target detection of car images in roads is performed based on the YOLOv8 model of YOLO family of models, which improves the accuracy and generalisation of the target detection task by combining multi-scale prediction, CSPNet structure and optimisation techniques such as BoF and BoS. The input images contain five types of vehicles such as Ambulance, Bus, Car, Motorcycle and Truck, which are analysed and learnt to have a classification accuracy of 75.4% on Ambulance, 53.5% on Bus, 55.1% on Car, 51.1% on Motorcycle and 42.5% on Truck. Despite the gap in specific classification accuracy, the YOLOv8 model can detect 100% of vehicles on the road, demonstrating good target detection capability. This research is of great significance for improving road traffic safety, intelligent traffic management, and the development of future autonomous driving technology. By optimising the deep learning model to achieve more accurate and efficient vehicle target detection, it can help to improve road safety and traffic efficiency, and promote the progress of intelligent transportation systems.

Inference Method for Abnormal Traffic Events in Highway Monitoring Blind Spots

Highway monitoring is critical for traffic safety and efficiency. However, monitoring blind spots—areas between surveillance points—poses significant challenges. This study proposes an inference method to detect abnormal traffic events in highway blind spots by analyzing vehicle counts from adjacent monitored regions. Utilizing Vissim and LumenRT for simulation, we create realistic highway scenarios to collect time series data. Cross-correlation is used to determine the time lag between vehicle counts, and Pearson's correlation coefficient evaluates the correlation between time-shifted data. The results show significant correlation under normal conditions, which decreases noticeably during abnormal events, confirming the method's effectiveness. This approach enhances the ability to monitor and respond to traffic anomalies in blind spots, improving overall highway safety and efficiency.

Driving Characteristics of Drivers Avoiding Lateral Obstacles in Mountain Highway Tunnels: A Dynamic Lateral Safety Distance Prediction Model

A natural driving real-vehicle test was conducted in nine tunnels in Chongqing, China. The driving trajectories and speeds of 40 drivers were recorded to better understand drivers’ driving characteristics under the mutual influence of various tunnel lengths and numerous lateral obstacles on mountainous highways. According to the driver’s need for lateral safety distance in different scenarios, the conditions and scope of the occurrence of the shy away effect are defined, and a dynamic avoidance prediction model is established in multiple scenarios. It was demonstrated that, when there were no lateral dynamic obstacles, drivers were influenced by the tunnel sidewalls to drift away from them. However, when lateral dynamic obstacles were present, drivers’ driving trajectories were corrected to be closer to the inside of the lane. The influence of vehicles of the same type resulted in the trajectory offset to the left by 0.2 m (extra-long tunnel), 0.7 m (long tunnel), 0.5 m (medium tunnel), and 0.7 m (short tunnel) compared with the driving trajectory without the influence of the obstacle. The influence of trucks causes the trajectory offset to the left compared with the driving trajectory without obstacles by 0.2 m (extra-long tunnel), 0.5 m (long tunnel), 1 m (medium tunnel), and 1.75 m (short tunnel). There is some variation in the degree of risk to the driver from different obstacles under different tunnels. The lateral distance between the driver and the dynamic obstacle is related to the nature of the available driving distance ahead. A sufficiently long driving maneuver distance enables the subject vehicle to gradually overtake the lateral dynamic obstacle in a more stable state. Dynamic avoidance models with fitting accuracies of 0.6–0.99 were implemented to predict vehicle trajectories in different tunnels and under the influence of various lateral dynamic obstacles. The study of the correlation of vehicle trajectories under the influence of various factors in tunnels is completed by the research findings, which can serve as a foundation for the design of driving behavior regulation, an improvement in traffic safety facilities, traffic management, and monitoring of lateral safety distances of intelligent vehicles in tunnels.

Traffic safety implementation at local level in line with 2030 Agenda for Sustainable Development: synergies, challenges and enablers

This study explores how traffic safety implementation at local level in line with the 2030 Agenda for Sustainable Development is characterised. By investigating interactions between traffic safety and other sustainability goals (synergies), as well as identifying contradictions between goals and conflicting interests (challenges), the study aims to pinpoint the main enablers for implementing traffic safety as part of sustainability. The emphasis lies on local authorities in Sweden representing a mature traffic safety context when viewed globally. To delve into this issue, interviews were conducted with 37 municipal employees from the city of Gothenburg, encompassing diverse roles that exemplify the significant influence Swedish local authorities wield in shaping traffic safety: as road manager and planner, in maintenance and construction phases, as employer and purchaser, and as influencer of behaviour. The study reveals several synergies between traffic safety and other sustainability goals, as well as contradictions between goals and conflicting interests posing challenges to traffic safety implementation. Main enablers for implementation of traffic safety as part of sustainability at the local level concern, for example, an integrated approach on strategic level where traffic safety could be handled as either a prerequisite (e.g. for active mobility) or a framework condition (e.g. for green cars). The necessity to bridge gaps between strategic and operational levels is also emphasised. Contradicting goals and interests are inevitable and must be managed. The study identifies the traffic strategy and strategic analyses, in the European context referred to as Sustainable Urban Mobility Planning, as a platform for traffic safety implementation as part of sustainability. Such platform highlights implementation areas in common for different target areas. The significance of integrating traffic safety in relevant activities and processes, alongside fostering horizontal collaborations across target areas, becomes apparent when striving to embed traffic safety in a wider sustainability context. While Swedish municipalities may not always recognize traffic safety as part of the sustainability agenda, existing platforms within municipalities serve as a foundation for adopting an integrated approach to traffic safety implementation. However, there is a need for a central coordinating function to ensure robust leadership in integrating traffic safety as a vital component of sustainability.

Estimating Urban Traffic Safety and Analyzing Spatial Patterns through the Integration of City-Wide Near-Miss Data: A New York City Case Study

City-wide near-miss data can be beneficial for traffic safety estimation. In this study, we evaluate urban traffic safety and examine spatial patterns by incorporating city-wide near-miss data (59,277 near-misses). Our methodology employs a grid-based method, the Empirical Bayes (EB) approach, and spatial analysis tools including global Moran’s I and local Moran’s I. The study findings reveal that near-misses have the strongest correlation with observed crash frequency among all the variables studied. Interestingly, the ratio of near-misses to crashes is roughly estimated to be 1957:1, providing a potentially useful benchmark for urban areas. For other variables, an increased number of intersections and bus stops, along with a greater road length, contribute to a higher crash frequency. Conversely, residential and open-space land use rates show a negative correlation with crash frequency. Through spatial analysis, potential risk hotspots including roads linking bridges and tunnels, and avenues bustling with pedestrian activity, are highlighted. The study also identified negative local spatial correlations in crash frequencies, suggesting significant safety risk variations within relatively short distances. By mapping the differences between observed and predicted crash frequencies, we identified specific grid areas with unexpectedly high or low crash frequencies. These findings highlight the crucial role of near-miss data in urban traffic safety policy and planning, particularly relevant with the imminent rise of autonomous and connected vehicles. By integrating near-miss data into safety estimations, we can develop a more comprehensive understanding of traffic safety and, thus, more effectively address urban traffic risks.

A simulation‐based impact assessment of autonomous vehicles in urban networks

AbstractThe behavioural differences between autonomous vehicles (AVs) and human‐driven vehicles (HDVs) can significantly impact traffic efficiency, safety, and emissions. Simulation‐based impact assessments using microscopic traffic models often modify car‐following (CF) and lane‐changing (LC) configurations to differentiate AVs from HDVs. Typically, researchers adjust CF model parameters to replicate AV driving behaviour, but these assumptions can lead to varying conclusions on AV impacts. The scope of each study (e.g., freeways, highways, urban links, intersections) also influences the outcomes. This research conducts an impact assessment utilizing optimized AV driving behavior rather than assumptions on a city network level (Munich) using a simulation‐based platform. The particle swarm optimization (PSO) algorithm is used to calibrate the base model and run simulation experiments under various penetration rates (PRs) and demand scenarios. Results show significant safety improvements throughout the network under higher PRs, while lower PRs might lead to deteriorating safety. At 100% AV PR, the total number of conflicts decreased by around 25% compared to a fully HDV environment. Considering AVs' sensing capabilities, additional safety improvements are found in almost any AV PR. However, AVs might not improve traffic efficiency; in some cases, they may slightly increase average network travel time, though this change is minimal.

Research on Driver Style Recognition Based on GA-K-Means and PSO-SVM

&lt;div&gt;This article aims to address the challenge of recognizing driving styles, a task that has become increasingly complex due to the high dimensionality of driving data. To tackle this problem, a novel method for driver style clustering, which leverages the principal component analysis (PCA) for dimensionality reduction and an improved GA-K-means algorithm for clustering, is proposed. In order to distill low-dimensional features from the original dataset, PCA algorithm is employed for feature extraction and dimensionality reduction. Subsequently, an enhanced GA-K-means algorithm is utilized to cluster the extracted driving features. The incorporation of the genetic algorithm circumvents the issue of the model falling into local optima, thereby facilitating effective driver style recognition. The clustering results are evaluated using the silhouette coefficient, Calinski–Harabasz (CH) index, and GAP value, demonstrating that this method yields more stable classification results compared to traditional clustering methods. In the final stage, a particle swarm optimization-SVM (PSO-SVM) algorithm is applied to classify the clustering results, which are then compared with results from other machine learning algorithms such as decision tree, naive Bayes network, and K-nearest-neighbor (KNN). This comprehensive approach to driver style recognition holds promise for enhancing traffic safety and efficiency. The accurate recognition of driving style can lay the foundation for further optimization of advanced driver assistance systems (ADAS).&lt;/div&gt;

ANALISIS KESELAMATAN JALAN DENGAN PENDEKATAN AUDIT KESELAMATAN JALAN PADA JALAN LOKAL DI KOTA TEGAL

Jalan Sultan Hasanudin – KH. Abdul Ghoni, which is located in Tegal City, Central Java, is a road with a secondary local function which is an alternative way to get to primary roads. By being an alternative road, this road section must meet adequate road safety standards. To improve traffic safety, it is necessary to carry out a road safety audit. This research aims to see the level of road safety in Tegal City, specifically on the Jalan Sultan Hasanudin - Jalan KH section. Abdul Ghoni. The method used in this research is descriptive quantitative obtained from the Hawkeye survey and processed using the Hawkeye processing toolkit software to analyze the data. The research results show that the Sultan Hasanudin – KH. Abdul Ghoni, in the aspect of road equipment facilities, there are 7 road equipment facilities in a damaged condition, in the aspect of the level of road unevenness in the medium category it has a percentage of 45% of the total length of the section, and for the aspect of the transverse slope on the Jl. Sultan Hasanuddin – KH. Abdul Ghoni has not met the standard for the transverse slope of asphalt pavement. Improved safety on the Jl. Sultan Hasanudin carried out further handling and improvements to road equipment and road pavement facilities, so that the Jl. Sultan Hasanuddin – KH. Abdul Ghoni does not have the potential to cause traffic accidents due to deficiencies in road infrastructure or dangerous road conditions.

A real-time and high-accuracy railway obstacle detection method using lightweight CNN and improved transformer

With the sustained development of railway transportation, the urgent necessity to improve train operation safety makes obstacle detection become the research focus. However, existing railway obstacle detectors still face challenges in balancing detection accuracy and speed during the shunting process. In addition, they are not robust enough in real-world railway environments, especially in complex scenes involving small obstacles. To address these problems, this paper presents a real-time and high-accuracy railway obstacle detection model using lightweight CNN and improved transformer (RH-Net) for detecting railway obstacles efficiently to guarantee traffic safety. First, the Lightweight Feature Extraction Module (LEM) is designed to minimize the model’s computational load while maintaining its feature extraction ability. Then, the Improved Transformer Module (IFM) is developed to boost the model’s ability about stably extract global contextual information. Finally, the Enhanced Multi-Scale Feature Fusion Module (EFM) is proposed to optimize the detection of obstacles with different sizes, especially small objects. In the experiments on railway dataset, RH-Net achieves optimal detection performance on GeForce GTX 1080Ti (96.99% mAP and 135 FPS) and Jetson Xavier NX (97.02% mAP and 43 FPS), which is significantly superior to the existing detection models. Experimental results show that RH-Net has excellent detection ability, which can accurately and efficiently detect obstacles in complicated railway environments. Moreover, the experiments on MS COCO indicate that RH-Net can achieve more satisfactory detection performance than existing state-of-the-art methods. Therefore, the proposed model can be well-applied to more complex real-world scenes for multiple object detection.