Road Segments Research Articles

Modeling of Risk Factors for Vulnerable Road Users’ Crashes on National Highways Passing through Medium-sized Indian Cities

The safety of vulnerable road users, such as pedestrians, bicyclists and motorised twowheelers, on national highways passing through urban areas has been a significant concern in developing countries like India. This study analyzes Visakhapatnam police crash data (2014-2017) to identify high-risk road segments to vulnerable road users. Establishing negative binomial models helps analyze risk variables for fatal and injury crashes concerning several susceptible road users on mid-block sections. Further, this study focuses on site-specific information collected from road-safety audits, speed, average daily traffic, length of road section and percentage of two-wheelers, also considered to fully help understand the likelihood of a crash. The statistical analysis of this paper identified the risk variables, like length of road segment, presence of service roads, land-use type (i.e., commercial/mixed land use), number of curves and average daily traffic associated with the incidence of fatal and injury crashes of vulnerable road users in Visakhapatnam city, India. The conclusions highlight the crucial safety measures that transportation planners and policymakers must take to create a more secure environment for road users. Keywords: Vulnerable road users, Mid-block sections, Negative binomial models, Crash, Count-data models, Safety measures.

Fuzzy-Based Road Accident Risk Assessment

It is necessary to extensively investigate the causes of road accidents with the utmost precision to harness future technological advancements, such as autonomous driving and intelligent accident prevention systems. Nevertheless, since most accidents are attributed to simple human errors, unraveling the complex root-cause factors poses a considerable challenge. This is where fuzzy logic can offer a potential solution: it is essential to understand even seemingly straightforward errors, such as speeding, to identify external factors that could play a pivotal role in future accident prevention. A more in-depth examination and comprehension of elements like road curvature, slope, and their correlation with accidents are necessary. Additionally, it is crucial to explore how the frequency of accidents on specific road segments varies under diverse weather conditions. This article analyzes which curves can be considered more dangerous and the factors that render them risky. The fuzzy model presented in this article is primarily capable of estimating the risk of a given road segment based on its curvature characteristics. The model results presented in the article indicate that sections of the road can become more risky due to multiple curves and curves with a radius of less than 80 m. The model assesses risk based on the physical characteristics of road segments, primarily the curvature radius, while, typically, other road risk assessment models rely on traffic volume and accident counts.

Ride Comfort Prediction on Urban Road Using Discrete Pavement Roughness Index

The prediction of ride comfort holds significant potential for enhancing the driving experience of both human drivers and autonomous vehicles, as it is closely correlated with pavement roughness. However, in urban road scenarios, the presence of shorter road segments and local irregularities introduces added complexity to ride comfort prediction. To better capture and characterize the irregularities and short road sections’ unevenness, we adopt the discrete roughness index (DRI) instead of the commonly used international roughness index (IRI) for assessing road profile unevenness, which is more suitable for urban roads. Ride comfort prediction is developed through numerical simulations using an eight-degree-of-freedom full-car model. The maximum transient vibration value (MTVV) is adopted to assess ride comfort. Through comparing the correlations between the MTVV and pavement roughness indices, it is indicated that the fitting degree of MTVV-DRI outperforms that of MTVV-IRI on short sections. Then, a set of speed-related DRI thresholds to estimate ride comfort distribution on a given road section is proposed, with considerations of vehicle speed, time period, and wheel paths. A hyperbolic-tangent-based speed control strategy is also proposed to avoid abrupt speed and acceleration changes during deceleration. This prediction method can assist drivers or autonomous vehicles in generating driving control strategies and maintaining a high level of ride comfort.

Mapping Road Surface Type of Kenya Using OpenStreetMap and High-resolution Google Satellite Imagery

Identifying road surface types (paved or unpaved) can ensure road vehicle safety, reduce energy consumption, and promote economic development. Existing studies identified road surface types by using sensors mounted on mobile devices and high-resolution satellite images that are not openly accessible, which makes it difficult to apply them to large-scale (e.g., national or regional) study areas. Addressing this issue, this study developed a dataset of road surface types (paved and unpaved) for the national road network of Kenya, containing 1,267,818 road segments classified as paved or unpaved. To accomplish this, this study proposes a method that integrates crowdsourced geographic data (OpenStreetMap) and Google satellite imagery to identify road surface types. The accuracy, recall, and F1 score of the method were all above 0.94, validating the effectiveness of the method. The data sources of the method are freely available, and the method may be applied to other countries and regions. The dataset developed based on the method can provide data support and decision support for local governments to improve road infrastructure.

Research on multitask model of object detection and road segmentation in unstructured road scenes

With the rapid development of artificial intelligence and computer vision technology, autonomous driving technology has become a hot area of concern. The driving scenarios of autonomous vehicles can be divided into structured scenarios and unstructured scenarios. Compared with structured scenes, unstructured road scenes lack the constraints of lane lines and traffic rules, and the safety awareness of traffic participants is weaker. Therefore, there are new and higher requirements for the environment perception tasks of autonomous vehicles in unstructured road scenes. The current research rarely integrates the target detection and road segmentation to achieve the simultaneous processing of target detection and road segmentation of autonomous vehicle in unstructured road scenes. Aiming at the above issues, a multitask model for object detection and road segmentation in unstructured road scenes is proposed. Through the sharing and fusion of the object detection model and road segmentation model, multitask model can complete the tasks of multi-object detection and road segmentation in unstructured road scenes while inputting a picture. Firstly, MobileNetV2 is used to replace the backbone network of YOLOv5, and multi-scale feature fusion is used to realize the information exchange layer between different features. Subsequently, a road segmentation model was designed based on the DeepLabV3+ algorithm. Its main feature is that it uses MobileNetV2 as the backbone network and combines the binary classification focus loss function for network optimization. Then, we fused the object detection algorithm and road segmentation algorithm based on the shared MobileNetV2 network to obtain a multitask model and trained it on both the public dataset and the self-built dataset NJFU. The training results demonstrate that the multitask model significantly enhances the algorithm’s execution speed by approximately 10 frames per scond while maintaining the accuracy of object detection and road segmentation. Finally, we conducted validation of the multitask model on an actual vehicle.

Risk Assessment of Road Blockage after Earthquakes

This paper presents a safety tool to assess the risk of road blockage during and after emergency situations, mainly due to earthquakes. This method can be used by public authorities to calculate the risk of road paths prone to blockage in case of seismic events. Typological classes of elements interfering with roads, such as unreinforced masonry and reinforced concrete buildings, unreinforced masonry and reinforced concrete bridges, retaining walls, and slopes, are considered. The mean annual frequency (MAF) of exceedance of a blockage limit state is calculated for a path with redundant road segments considering fragility curves from the literature. A practical example is presented for Amatrice, a town in Central Italy hit by the 2016 earthquake. After verifying that the MAF of exceedance demand is lower than the capacity for two roads, a strengthening solution is assumed for two buildings in the path, resulting in a reduction by more than 50% of the MAF demand. For a higher safety level, a bypass is proposed obtaining a demand/capacity ratios four orders of magnitude lower than that obtained with strengthening solutions, highlighting and quantifying the beneficial effect of removing vulnerable structures along the path.

Lane Reduction and Queue Length Relationship on Arterial and Collector Roads

Geometric diversity of roads for an example is the narrowing of road segments. The purpose of this research is to determine whether lane reduction leads to significant queue lengths, observe the behaviour of drivers on those road segments, and compare the actual queue length to the queue length generated by PTV Vissim. The data used in this study consists of direct observations during peak hours in the morning, afternoon, and evening, along with simulations using PTV Vissim. The queue results obtained from PTV Vissim were validated using two types of tests. The research findings indicate a significant relationship between queue length and lane reduction during peak hours on workdays. However, during the evening peak hours on weekends, the queue length is not significantly affected by lane reduction. As a potential solution, widening each road segment in PTV Vissim was tested, which increased road capacity and slightly reduced the queues that occurred.

Arrangement of Access to the Religious Tourism Area of The Tomb K.H. Abdurrahman Wahid Jombang Regency

Jombang Regency is one of the regencies in the East Java region that boasts numerous tourist destinations. One of the religious attractions in Jombang Regency is the Tomb of K.H. Abdurrahman Wahid. Suboptimal road traffic conditions occur in the tourism area, particularly on K.H. Hasyim Asyari Street. This road segment frequently experiences high traffic volume due to tourism activities, numerous obstacles such as roadside parking, and non-motorized vehicles moving around the tourism area. This research adheres to the Guidelines for Indonesian Road Capacity (PKJI) 2014 regarding Urban Road Capacity. The primary data required for this study include geometric road data, traffic volume data, and roadside obstacle data. Meanwhile, secondary data in this study comprise road network data and Average Daily Traffic (ADT) data from the relevant authorities, namely the Jombang Regency Transportation Agency. From the traffic performance calculations, the saturation degree values for the existing conditions on K.H. Hasyim Asyari Street towards the south (DJ) = 0.7624 and towards the north (DJ) = 0.7075 were obtained. Based on the survey results, K.H. Hasyim Asyari Street towards the north has a capacity (C) = 2465, and towards the south, it also has a capacity (C) = 2465. The parking space requirements in the Religious Tourism Area of the Tomb of K.H. Abdurrahman Wahid (Gus Dur) are planned to be 171 parking spaces for motorcycles, 54 parking spaces for cars, and 53 parking spaces for buses.

Multi-Vehicle Collaborative Trajectory Planning for Emergency Vehicle Priority at Autonomous Intersections

Emergency vehicle (EV) prioritization plays an important role in improving rescue efficiency and saving lives and property. Most existing studies have been confined to either investigating signal prioritization for EVs at intersections or focusing on clearing emergency lanes on road segments, with limited consideration for an integrated approach that combines both methods. In this study, we have developed a bi-level trajectory planning model aimed at optimizing the trajectories of EVs at intersections and road segments located within the communication range, using vehicle-to-everything technology. The upper-level model is designed to plan the entry times, speeds, and internal trajectories of vehicles within the intersection while avoiding any conflicts that may arise during their movements. In particular, to address the dimensionality of the proposed large-scale optimization problems, we introduce the equidistant discretization method to discretize the entry speeds of vehicles into a finite set of selectable values. The lower-level model focuses on optimizing the longitudinal and lateral trajectories on roadways to ensure that vehicles arrive at the stop line punctually and promptly. Both the upper- and lower-level models are formulated as mixed-integer linear programming models. The A Mathematical Programming Language and the Gurobi solver are employed for optimization. The case study demonstrates that the proposed model effectively optimizes vehicle trajectories at intersections and road segments, leading to a significant reduction in delays for EVs.

Prediction Intervals for Bus Travel Time Based on Road Segment Sharing, Multiple Routes’ Driving Style Similarity, and Bootstrap Method

Providing accurate information about bus travel times can help passengers plan their itinerary and reduce waiting time. However, due to various uncertainty factors and the sparsity of single-route data, traditional travel time predictions cannot accurately describe the credibility of the prediction results, which is not conducive to passengers waiting based on the predicted results. To address the above issues, this paper proposes a bus travel time prediction intervals model based on shared road segments, multiple routes’ driving style similarity, and the bootstrap method. The model first divides the predicted route into segments, dividing adjacent stations shared by multiple routes into one section. Then, the hierarchical clustering algorithm is used to group all drivers in multiple bus routes in this section according to their driving styles. Finally, the bootstrap method is used to construct a bus travel time prediction interval for different categories of drivers. The travel time data sets of Shenyang 239, 134, and New Area Line 1 were selected for experimental verification. The experimental results indicate that the quality of the prediction interval constructed using a data set fused with multiple routes is better than that constructed using a single-route data set. In the two cases studied, the MPIW of the three time periods decreased by 101.04 s, 151.72 s, 33.87 s, and 126.58 s, 127.47 s, 17.06 s, respectively.

A unified dataset for the city-scale traffic assignment model in 20 U.S. cities

City-scale traffic data, such as traffic flow, speed, and density on every road segment, are the foundation of modern urban research. However, accessing such data on a city scale is challenging due to the limited number of sensors and privacy concerns. Consequently, most of the existing traffic datasets are typically limited to small, specific urban areas with incomplete data types, hindering the research in urban studies, such as transportation, environment, and energy fields. It still lacks a city-scale traffic dataset with comprehensive data types and satisfactory quality that can be publicly available across cities. To address this issue, we propose a unified approach for producing city-scale traffic data using the classic traffic assignment model in transportation studies. Specifically, the inputs of our approach are sourced from open public databases, including road networks, traffic demand, and travel time. Then the approach outputs comprehensive and validated citywide traffic data on the entire road network. In this study, we apply the proposed approach to 20 cities in the United States, achieving an average correlation coefficient of 0.79 in average travel time and an average relative error of 5.16% and 10.47% in average travel speed when compared with the real-world data.

Recursive logit-based meta-inverse reinforcement learning for driver-preferred route planning

Driver-preferred route planning often evaluates the quality of a planned route based on how closely it is followed by the driver. Despite decades of research in this area, there still exist nonnegligible deviations from planned routes. Recently, with the prevalence of GPS data, Inverse Reinforcement Learning (IRL) has attracted much interest due to its ability to directly learn routing patterns from GPS trajectories. However, existing IRL methods are limited in that: (1) They rely on numerical approximations to calculate the expected state visitation frequencies (SVFs), which are inaccurate and time-consuming; and (2) They ignore the fact that the coverage of GPS trajectories is skewed toward popular road segments, causing difficulties in learning from sparsely covered ones. To overcome these challenges, we propose a recursive logit-based meta-IRL approach, where (1) We use the recursive logit model to capture drivers’ route choice behavior so that the expected SVFs can be analytically derived, which substantially reduces the computational efforts; and (2) We introduce meta-parameters and employ meta-learning techniques so that the learning on sparsely covered road segments can benefit from that on popular ones. When training our IRL model, we update the rewards of road segments with the expected SVFs by solving several systems of linear equations and update the meta-parameters through a two-level optimization structure to ensure its fast adaption and versatility. We validate our approach using real GPS data in Chengdu, China. Results show that our planned routes better match actual routes compared with state-of-the-art methods including the recursive logit model, Deep-IRL and Dij-IRL: the F1-Score increases by 4.17% with the introduction of the recursive logit model and further increases to 5.19% after meta-learning is employed. Moreover, we can reduce training time by over 95%.

Maintenance Strategy for the Muliama-Wa2e-Makki Road Section in Jayawijaya Regency, Papua Province, Highlands

In Jayawijaya Regency, there are several prioritized road sections, including national, provincial, and district roads. The Muliama-Wame-Makki road section is classified as a provincial road serving as an alternative route, with a total length of 22 kilometers connecting Jayawijaya Regency and Lanni Jaya Regency. The traffic on this road is quite heavy with a diverse composition (Mixed Traffic). An analysis using the Pavement Condition Index (PCI) was conducted based on field survey data, including measurements of the extent, types, and quantities of surface damage per 200 meters along each road segment. The types of damage observed on the Muliama-Wame-Makki road section include patching, longitudinal cracks, alligator cracking, and pavement subsidence, with a total damage percentage of 1.684%. The maintenance strategy is determined based on the PCI calculation and Bina Marga method, suggesting that routine maintenance is sufficient for addressing the observed issues.

Real-time semantic segmentation based on BiSeNetV2 for wild road

Abstract State-of-the-art segmentation models have shown great performance in structured road segmentation. However, these models are not suitable for the wild roads, which are highly unstructured. To tackle the problem of real-time semantic segmentation of wild roads, we propose a Multi-Information Concatenate Network based on BiSeNetV2 and construct a segmentation dataset Dalle Molle institute for artificial intelligence feature segmentation (IDSIAFS) based on Dalle Molle institute for artificial intelligence. The proposed model removes structural redundancy and optimizes the semantic branch based on BiSeNetV2. Moreover, the Dual-Path Semantic Inference Layer (TPSIL) reduces computation by designing the channel dimension of the semantic branch feature map and aggregates feature maps of different depths. Finally, the segmentation results are achieved by fusing both shallow detail information and deep semantic information. Experiments on the IDSIAFS dataset demonstrate that our proposed model achieves an 89.5% Intersection over Union. The comparative experiments on Cityscapes and India driving dataset benchmarks show that proposed model achieves good inference accuracy and faster inference speed.

AutoLTS: Automating Cycling Stress Assessment via Contrastive Learning and Spatial Post-processing

Cycling stress assessment, which quantifies cyclists' perceived stress imposed by the built environment and motor traffics, increasingly informs cycling infrastructure planning and cycling route recommendation. However, currently calculating cycling stress is slow and data-intensive, which hinders its broader application. In this paper, We propose a deep learning framework to support accurate, fast, and large-scale cycling stress assessments for urban road networks based on street-view images. Our framework features i) a contrastive learning approach that leverages the ordinal relationship among cycling stress labels, and ii) a post-processing technique that enforces spatial smoothness into our predictions. On a dataset of 39,153 road segments collected in Toronto, Canada, our results demonstrate the effectiveness of our deep learning framework and the value of using image data for cycling stress assessment in the absence of high-quality road geometry and motor traffic data.

Road Accessibility during Natural Hazards Based on Volunteered Geographic Information Data and Network Analysis

Natural hazards can present a significant risk to road infrastructure. This infrastructure is a fundamental component of the transportation infrastructure, with significant importance. During emergencies, society heavily relies on the functionality of the road infrastructure to facilitate evacuation and access to emergency facilities. This study introduces a versatile, multi-scale framework designed to analyze accessibility within road networks during natural hazard scenarios. The first module of the framework focuses on assessing the influence of natural hazards on road infrastructure to identify damaged or blocked road segments and intersections. It relies on near real-time information, often provided by citizen science through Volunteered Geographic Information (VGI) data and Natural Language Processing (NLP) of VGI texts. The second module conducts network analysis based on freely available Open Street Map (OSM) data, differentiating between intact and degraded road networks. Four accessibility measures are employed: betweenness centrality, closeness centrality, a free-flow assumption index, and a novel alternative routing assumption measure considering congestion scenarios. The study showcases its framework through an exemplary application in California, the United States, considering different hazard scenarios, where degraded roads and connected roads impacted by the hazard can be identified. The road extraction methodology allows the extraction of 75% to 100% of the impacted roads mentioned in VGI text messages for the respective case studies. In addition to the directly extracted impacted roads, constructing the degraded network also involves finding road segments that overlap with hazard impact zones, as these are at risk of being impacted. Conducting the network analysis with the four different measures on the intact and degraded network, changes in network accessibility due to the impacts of hazards can be identified. The results show that using each measure is justified, as each measure could demonstrate the accessibility change. However, their combination and comparison provide valuable insights. In conclusion, this study successfully addresses the challenges of developing a generic, complete framework from impact extraction to network analysis independently of the scale and characteristics of road network types.

The Impact of Urban Design on Utilitarian and Leisure Walking—The Relative Influence of Street Network Connectivity and Streetscape Features

Road network connectivity determines the accessibility of urban activities for pedestrians, while streetscape characteristics have an impact on route attractiveness. Methods used to measure the influence of connectivity and streetscape characteristics on walking behavior differ substantially, while trip purpose is a key factor. This paper explores the respective contributions of network connectivity and streetscape features to explain leisure walking and utilitarian trips on 740 street segments in Santarém (Portugal). The indicators cover the most commonly used factors in walkability indexes, such as density, diversity, design, and accessibility. The streetscape features measure imageability, enclosure, human scale, transparency, and complexity. The walking trip information was collected via survey. The results show that connectivity measures have a greater overall explanatory power for both trip motives. However, the findings highlight the need to consider a variety of design indicators to explain walking behavior due to the higher explanation power of the model with two types of indicators.

SS-GAN based road surface crack region segmentation and forecasting

Automatic segmentation of crack regions is the major and complex task in an intelligent transportation system. In this proposed work, we have developed a Semantic Segmentation based Generative Adversarial Network model (SS-GAN). Various conditional factors such as the number and type of vehicles traveled during the months and environmental conditions like temperature, and precipitation will be additionally fed into the proposed SS-GAN model to segment and forecast the respective months’ road crack images. In-order to forecast similar crack images of appropriate months with better performance, we have introduced new loss functions additionally. A semantic segmentation model is developed to extract the required semantic features of the crack regions which helps to segment road crack images more accurately. Finally, the proposed model concatenates the extracted features of both discriminator and semantic segmentation models to classify types of cracks with better performance. SS-GAN model is trained using the Crack ForeCast (CFC) dataset in an end-to-end and paired translation model. A comprehensive analysis of experimental results is carried out to measure the performance of the proposed SS-GAN model. Various metrics are used to evaluate the quality of semantic feature-based segmentation for the forecasted crack images. The obtained results prove that the SS-GAN model showcases better performance than other state-of-the-art models.

Bicycle Infrastructure Design Principles in Urban Bikeability Indices: A Systematic Review

Bicycling is a sustainable form of micromobility and offers numerous health and environmental benefits. Scientific studies investigating bikeability have grown substantially, especially over the past decade. This paper presents a systematic literature review of the developed urban bikeability indices (BIs). The paper provides insight into the scientific literature on bikeability as a tool to measure bicycle environment friendliness; more importantly, the paper seeks to know if the BIs consider bicycle infrastructure design principles. Data extraction included identifying the geographical location, essential indicators, sample size and distribution, data source, the unit of analysis, measurement scale, methods used to weigh indicators, and identification of studies using bicycle design principles in BIs. The database search yielded 1649 research articles using different keywords and combinations, while 15 studies satisfied the inclusion criteria. The studies were found to be conducted in various geographical locations. The unit of analysis for developing the index varied across studies, from street segments or bicycle lanes to zones within the city or even the entire city. The most commonly utilized method in developing urban BIs was a scoring and weighting system to weigh the indicators. The weighting methods include an equal weight system, survey-based and literature review-based methods, expert surveys, the analytic hierarchy process, and a weighted linear combination model. The essential criterion is bicycle infrastructure, such as bike lanes, routes, and bicycle paths as 14 studies considered it for the construction of the BIs. The review findings suggest a lack of consideration of all five bicycle infrastructure design principles, as only three studies considered them all, while others only included a subset. Safety and comfort are the most commonly considered principles, while coherence is the least considered principles in the BIs. It is crucial to consider all five bicycle infrastructure design principles to create a bicycle-friendly environment and attract more people to this sustainable mode of transportation.

MGL2Rank: Learning to rank the importance of nodes in road networks based on multi-graph fusion

The identification of important nodes with strong propagation capabilities in road networks is a vital topic in urban planning. Existing methods for evaluating the importance of nodes in traffic networks only consider topological information and traffic volumes, the diversity of the traffic characteristics in road networks, such as the number of lanes and average speed of road segments, is ignored, thus limiting their performance. To solve this problem, we propose a graph learning-based framework (MGL2Rank) that integrates the rich characteristics of road networks to rank the importance of nodes. This framework comprises an embedding module containing a sampling algorithm (MGWalk) and an encoder network to learn the latent representations for each road segment. MGWalk utilizes multi-graph fusion to capture the topology of road networks and establish associations between road segments based on their attributes. The obtained node representation is then used to learn the importance ranking of the road segments. Finally, a synthetic dataset is constructed for ranking tasks based on the regional road network of Shenyang City, and the ranking results on this dataset demonstrate the effectiveness of our method. The data and source code for MGL2Rank are available at https://github.com/ZJ726.