Traffic Safety Enhancement Research Articles

Ship visual trajectory exploitation via an ensemble instance segmentation framework

Maritime traffic safety is of utmost importance in the shipping industry. It is common to avoid potential traffic accident via the support of varied kinematic maritime data (e.g., ship position, ship speed). The study aims to fulfill maritime traffic safety enhancement by extracting multiple ship trajectories from maritime video data. The proposed framework is implemented with three steps. Firstly, the framework obtains initial ship imaging positions by proposing an ensemble YOLOX (You Only Look Once X) with CenterNet model. Secondly, ship contours from each maritime frame are explored with the deep snake module. Thirdly, consecutive ship imaging trajectory is identified with an enhanced Bytetrack multi-ship target tracking algorithm. Experimental results demonstrate that our model outperforms other ship trajectory extraction-like models considering that the identification F-Score (IDF1), IDP (identification Precision), IDR (identification recall), and MOTA (multi-object tracking accuracy) for our proposed framework were 0.965, 0.995, 0.933, and 0.925, respectively. The research findings can help ship crew better aware on-site traffic situations, and thus make optimal ship maneuvering operations for the purpose of enhancing maritime traffic system safety and sustainability.

A Dynamic Spatial-Temporal Attention Network for Early Anticipation of Traffic Accidents

The rapid advancement of sensor technologies and artificial intelligence are creating new opportunities for traffic safety enhancement. Dashboard cameras (dashcams) have been widely deployed on both human driving vehicles and automated driving vehicles. A computational intelligence model that can accurately and promptly predict accidents from the dashcam video will enhance the preparedness for accident prevention. The spatial-temporal interaction of traffic agents is complex. Visual cues for predicting a future accident are embedded deeply in dashcam video data. Therefore, the early anticipation of traffic accidents remains a challenge. Inspired by the attention behavior of humans in visually perceiving accident risks, this paper proposes a Dynamic Spatial-Temporal Attention (DSTA) network for the early accident anticipation from dashcam videos. The DSTA-network learns to select discriminative temporal segments of a video sequence with a Dynamic Temporal Attention (DTA) module. It also learns to focus on the informative spatial regions of frames with a Dynamic Spatial Attention (DSA) module. A Gated Recurrent Unit (GRU) is trained jointly with the attention modules to predict the probability of a future accident. The evaluation of the DSTA-network on two benchmark datasets confirms that it has exceeded the state-of-the-art performance. A thorough ablation study that assesses the DSTA-network at the component level reveals how the network achieves such performance. Furthermore, this paper proposes a method to fuse the prediction scores from two complementary models and verifies its effectiveness in further boosting the performance of early accident anticipation.

GAP-MM: 5G-Enabled Real-Time Autonomous Vehicle Platoon Membership Management Based on Blockchain

Autonomous Vehicle Platoon (AVP) is the most anticipated application of 5G ultrareliable and low latency communications. By joining the preexisting platoon to form the real-time AVP (RAVP), individual vehicles could gain benefits such as fuel consumption reduction and traffic safety enhancement. Unlike the scheduled AVP which only contains fixed vehicles, members in RAVP change frequently since individual vehicles would want to join or leave the platoon any time. Besides, malicious vehicles may attempt to sneak into the RAVP and try to manipulate the platoon. Public key cryptography and access control mechanisms can be adopted to create a relatively isolated area for communication inside the platoon. Nevertheless, there exist few works that take into account the regulation of the dynamic change members in RAVP. In this paper, we propose a membership management scheme for 5G-enabled RAVP by integrating revocable attribute-based encryption (RABE) and blockchain, namely, GAP-MM. It realizes fine-grained access control of key distribution and malicious vehicle’s key revocation efficiently. The sufficient evaluations and security analysis indicate that GAP-MM is practical for RAVP scenario in terms of both efficiency and security.

Risk Area Identification Model of Bus Bay Stops Based on Distribution of Conflicts

Effective identification of the risk area of the bus bay stop is a prerequisite for the enhancement of traffic safety. This study proposes a method of identifying the risk area based on the distribution of traffic conflicts. Firstly, the traffic flow data of the bus stop is collected by drones and video recognition software, and the traffic flow characteristics of the bus stop are analyzed by the mathematical and statistical methods. Secondly, using the gray clustering evaluation theory, on the basis of the rasterization of the functional area of the bus bay stop, a risk level model based on the index system of conflict rate, conflict severity, and potential conflict risk is proposed. Finally, take a bus stop in Guangzhou as an example to verify the solution. The results show that the constructed model can effectively identify the risk areas of bus bay stops. The risk areas of the bus bay stops are concentrated in the middle and lower reaches of the bus stop, which proves that the impact of bus exiting the stop on the surrounding traffic is greater than the process of bus entering the stop; the traffic risk areas of lanes near the bus stop are concentrated, and the severity of conflicts is low. The traffic risk zone of the lane far away from the bus stop is widely distributed, and the severity of conflict is higher. The research results can provide a basis for the micro safety performance evaluation and safety optimization of bus bay stops, which has strong theoretical and practical significance.

Cost–benefit analysis of traffic barrier geometric optimization, a hurdle machine learning‐based technique

AbstractAlthough it is well documented that optimized barrier dimensions would yield a benefit by enhancement of traffic safety, not many studies have been conducted quantifying how much money could be saved by enhancing barriers to their optimum heights, especially with the help of machine learning technique. Equivalent property damage only (EPDO) was considered as a response in the modeling to account for both aspects of crash frequency and severity. A hurdle model was developed to explore the impact of enhancing the geometric characteristics of barriers while accounting for zero inflation. The two components of the model, logistic regression and truncated negative binomial, could account for both abundance of zeroes and the skewed nature of non‐zero EPDO. The methodological approach provides a new understanding of the economic/monetary evaluation of an asset enhancement in the state with the help of machine learning technique. Our methodological approach provides the policy makers about the monetary benefit that they could expect by optimization of barriers heights. The analysis summarized in this paper will not only help Wyoming, but it provides other states with the justification of safety enhancement to their road barriers or assets.

Do the mobility patterns for city taxicabs impact road safety?

Recently, large investments have been made by cities such as Singapore, New York City, and London towards creating smart city initiatives in the areas of traffic safety enhancement and higher mobility. In this article, we investigate the impact of various network topology measures on the number of vehicle crashes in a city mobility network. Extant literature on mobility in city traffic networks has not studied the impact of network structure on road accidents. We fill this important gap by identifying the structural properties of critical zones in the city traffic network, which have a high risk of vehicle crashes. We use econometric methods to analyze a large dataset on city mobility from the New York Taxi and Limousine Commission, and a dataset on motor vehicle collisions from the New York Police Department; and derive various insights on the scope of traffic safety issues in a smart city. Our dataset has information on about 100,000,000 taxi trips over the year 2018. In this year, around 1,500,000 vehicle crash events were reported in New York City. One would expect that due to a large number of shortest paths, the number of accidents should be significantly more in the high betweenness centrality zones in the traffic mobility network. However, our analysis reveals that zones with high betweenness centrality tend to have a lower number of accidents. Furthermore, zones with a high degree centrality in the traffic mobility network are associated with a higher number of vehicle crash incidents. Our study reveals some crucial pointers for smart city policymakers and the operations managers of ride-sharing companies on how information on the mobility patterns of the high accident risk zones can be leveraged to reduce motor vehicle collisions.

Association between Truck Crashes due to Mechanical Failure and Truck Age

There have been efforts to restrict older trucks in many jurisdictions all over the world. The primary goal of the restrictions is to minimize greenhouse gas emissions. In addition to the environmental benefits, it is also possible that the truck age restriction could contribute to the enhancement of traffic safety. Older trucks are subject to longer travel-miles than newer trucks and tend to have higher mechanical failure rates. Extremely few studies have been done to explore the impact of trucks’ age on their crash occurrence due to mechanical problems. This study aims to investigate the association between the truck crashes due to mechanical issues and the truck age. Two approaches are adopted to achieve the objective. First, a chi-square test reveals that the proportions of the mechanical failures among older trucks are higher than those among newer ones ( χ 2 = 256.199, p < 0.0001 ). Second, the modeling results indicate that the number of truck crashes due to mechanical failures is significantly increased by the truck age. The findings suggest that policies restricting older trucks should consider not only environmental effects but also traffic safety benefits.

Using machine leaning techniques for evaluation of motorcycle injury severity

There is a growing interest in the application of the machine learning techniques in predicting the motorcycle crash severity. This is partly due to a progress in autonomous vehicles technology, and machine learning technique, which as a main component of autonomous vehicle could be implemented for traffic safety enhancement. Wyoming's motorcycle crash fatalities constitute a concern since the count of riders being killed in motorcycle crashes in 2014 was 11% of the total road fatalities in the state. The first step of crash reduction could be achieved through identification of contributory factors to crashes. This could be accomplished by using a right model with high accuracy in predicting crashes. Thus, this study adopted random forest, support vector machine, multivariate adaptive regression splines and binary logistic regression techniques to predict the injury severity outcomes of motorcycle crashes. Even though researchers applied all the aforementioned techniques to model motorcycle injury severities, a comparative analysis to assess the predictive power of such modeling frameworks is limited. Hence, this study contributes to the road safety literature by comparing the performance of the discussed techniques. In this study, Wyoming's motorcycle crash injury severities are modeled as functions of the characteristics that give rise to crashes. Before conducting any analyses, feature reduction was used to identify a best number of predictors to be included in the model. Also to have an unbiased estimation of the performance of different machine learning techniques, 5-fold cross-validation was used for model performance evaluation. Two measure, Area under the curve (AUC), and confusion matrix were used to compare different models' performance. The machine learning results indicate that random forest model outperformed the other models with the least misclassification and higher AUC. It was also revealed that a dichotomous response variable, with fatality and incapacitation injury in one category, along with all other categories in another group would result in a lower misclassification rate than a polychotomous response variable. This might result from the nature of motorcycle crashes, lacking a protection compared with passenger cars, preventing machine learning technique to get trained properly. Moreover, the most important variables identified by the random forest model are those related to the operating speed, resentful other party, traffic volume, truck traffic volume, riding under the influence, horizontal curvature, wide roadway with more than two lanes and rider's age.

Ship Trajectory Data Compression Algorithms for Automatic Identification System: Comparison and Analysis

With the development of Internet of Things (IoT) technology and its vast applications in ship transportation systems, such as the Automatic Identification System (AIS), a large quantity of ship trajectory data have been recorded and stored. Nowadays ship transportation has also entered the age of big data, which can support IoT applications in Intelligent Transportation System (ITS), e.g. traffic monitoring, fleet management and traffic safety enhancement. However, the redundancy of ship trajectory data considerably reduces the effectiveness and efficiency of large scale traffic data storage, mining and visualization. Therefore, compression processing of the data becomes a very important issue for these applications. Because ship trajectory is a type of vector data, employing the vector data compression algorithms is an efficient way to solve the data redundancy problem. In this paper, the pseudo-code of five typical vector data compression algorithms for ship trajectory data compression is introduced. The performances of these algorithms were tested by the compression experiments of actual ship trajectories in the Qiongzhou Strait. The results show that ships’ speeds and rate of turns, the requirement of real time processing can affect the option of the most appropriate algorithm, and the algorithm selection in different applications is suggested. The results and conclusions lay the foundation for the future development of ship transportation intelligentization.

A Trial of Existing Data Utilization for Traffic Safety Enhancement in Trucking Offices

A Trial of Existing Data Utilization for Traffic Safety Enhancement in Trucking Offices

Demonstrating Innovative Technologies for the Flemish Asphalt Sector in the CyPaTs Project

In 2015, the ROAD_IT project initiated the development and demonstration of an integrated and coherent IT process control system for the Flemish asphalt sector in order to modernize existing asphalt paving technologies and to obtain real-time data to monitor pavement behavior. One of the demonstration test tracks is CyPaTs, the construction of a bicycle path built in September 2017, using innovative technologies (www.uantwerpen.be/cypats). Five technologies are described in this contribution. An asphalt solar collector (PSC) with a piping system was installed directly in the asphalt. Cold water during summer season and hot water in winter season running through the pipes, keep the asphalt structure in a better temperature interval, avoiding rutting and cracking. Other advantages of this system are: energy gain, the prevention of damage to asphalt and the enhancement of traffic safety. The expected energy gain per year varies between 0.5 and 0.8 GJ/m2. About 20% of this energy is used for the operation of the asphalt collector itself. The remaining 80% can be used in nearby buildings. Fiber Bragg Grating (FBG) monitoring system was integrated in all three asphalt layers for the first time in Belgium. Two novel approaches of FBGs installation in asphalt layers were elaborated in this bicycle path: installation of FBGs in prefabricated asphalt specimens at the bottom of base layer and installation of FBGs in a saw cut of approx. 2mm in the previously constructed asphalt layer. The results proved a survival rate of the FBGs of 100%. The obtained strain and temperature data from FBG monitoring system has proved to be an excellent approach to establish and reflect the real condition of the asphalt pavement behaviour in time at different temperatures. The temperatures of the asphalt pavement during construction were followed up by the infrared thermography measurement techniques: a thermographic line-scanner (PAVE-IR by Moba AG) which was mounted at the back of the finisher and a hand-held IR camera (FLIR T640) was used for taking pictures every 2 meters. A real-time temperature contour plot of the pavement during construction was created to monitor asphalt pavement temperatures for quality inspection during the paving process or for later assessment. Two other non-destructive technologies for quality assessment were applied during this project. At first, the thickness was measured using aluminium plates and the MIT-SCAN T3. The obtained values were compared with topographic height measurements. Secondly, the density was measured with the PQI-380 non-nuclear density meter at several spots. The objective here is to check the density of the bicycle path, as well as the accuracy and investigate different parameters that influence the variations of the results, in particular the temperature dependency.

RESULTS OF STUDYING ROAD CONSTRUCTION PARAMETERS' CONDITION

A system for the accounting of road infrastructure facilities' parameters specified during design and construction of motor roads. A study of parameters' condition was performed. Analysis of road accident rate statistics and identification of parameters to assess the efficiency of the proposed road construction activities aimed at the reduction in the number of road accidents were carried out. Actual data on condition parameters with the use of modern automated multi-functional diagnostic equipment were obtained. Actions aimed at the improvement of road infrastructure parameters through traffic safety enhancement, as well as actions aimed at the reconstruction of the motor road or project-supported construction of road segments were developed. Efficiency of the proposed actions was assessed.

A Rapid Prototyping Environment for Cooperative Advanced Driver Assistance Systems

Advanced Driver Assistance Systems (ADAS) were strong innovation drivers in recent years, towards the enhancement of traffic safety and efficiency. Today’s ADAS adopt an autonomous approach with all instrumentation and intelligence on board of one vehicle. However, to further enhance their benefit, ADAS need to cooperate in the future, using communication technologies. The resulting combination of vehicle automation and cooperation, for instance, enables solving hazardous situations by a coordinated safety intervention on multiple vehicles at the same point in time. Since the complexity of such cooperative ADAS grows with each vehicle involved, very large parameter spaces need to be regarded during their development, which necessitate novel development approaches. In this paper, we present an environment for rapidly prototyping cooperative ADAS based on vehicle simulation. Its underlying approach is either to bring ideas for cooperative ADAS through the prototyping stage towards plausible candidates for further development or to discard them as quickly as possible. This is enabled by an iterative process of refining and assessment. We reconcile the aspects of automation and cooperation in simulation by a tradeoff between precision and scalability. Reducing precise mapping of vehicle dynamics below the limits of driving dynamics enables simulating multiple vehicles at the same time. In order to validate this precision, we also present a method to validate the vehicle dynamics in simulation against real world vehicles.

Implementing road safety measures in conditions limited by budget

Abstract Adopted in 2018 was the Strategy of Traffic Safety for 2018 – 2024 years which stipulates the social risk index making not more than 4 fatalities per 100.000 people, as well as the craving for zero fatality in road vehicle accidents by 2030, as a benchmark in 2024. A level of the tasks at hand requires significant increase in efficiency of traffic safety management which is possible by application of the expert system for traffic safety assurance. Given in the present article is an evaluation method of activities implementation investment attractiveness which allows determining the importance of some or other activity depending on the selected production rules in a limited budget. The fuzzy-set theory (Kofman, 1982), the expert evaluation method, the expert system functioning principles form the basis of this method. In order to illustrate this method the activities on enhancement of pedestrian traffic safety are selected. Considered as an example are two variants of production rules which make it possible to evaluate the selective potentials of the method. The proposed method can be used not only as an instrument of the expert system for traffic safety, but also independently. The use of this method allows most effectively spend budget funds allocated for enhancement of traffic safety.

Method of Traffic Safety Enhancement with Use of RFID Technologies and its Implementation

A new solution for the problem of traffic safety regulation with use of RFID technologies is suggested. Traffic safety enhancement is achieved by polling with a modulated radio signal of all electronic analogues of traffic signs and traffic lights representing RFID markings duplicating main traffic signs and traffic lights and located on them directly or nearby. Polling is performed with a radio signal from a transmitter/receiver unit (radar) focused on a narrow zone of radiation action. The suggested method as distinct from the known ones allows for active control of traffic speed in high-risk areas, elimination of the risk of traffic movement along tram rails, pavements and bus lanes as well as for stopping a car when police chases an offender.

Potential for Enhancing Traffic Safety on Highways of Russia

Abstract The article investigates the correlation between the factors of enhancing traffic safety on highways and parameters of the economic growth in Russia and in countries with transition economy; such correlation does not always lead to traffic safety enhancement. As a rule, the population motorization (car density) level growth in such countries is not accompanied by efficient actions on accident rate decrease. Among the factors enhancing traffic safety on highways are the efficiency of traffic infrastructure planning (changing and removing traffic flows from cities’ centers, creation of a large amount of pedestrian and bicycle lanes separated from traffic flows) as well as harmonization of physical infrastructure and safety goals, ensuring conjugation of the developed transport infrastructure with the decrease of the share of private vehicle fleet and increase of the share of public transport.

Standardization of cooperation of intelligent transport systems in vehicles

Communications between road vehicles and with their environment (V2X) is important for cooperative ITS (C-ITS) application implementations including those which enhancing road traffic safety. Formal and de facto V2X communications standards are key to application development requiring cooperation of distributed components with maintenance of all interoperability conditions. Experiences proof that ITS solutions support road traffic safety enhancement and this is reasoning of European Union priorities of this subjects and thus many research, innovation and deployment actions.

Discovering temporal and spatial patterns and characteristics of pavement distress condition data on major corridors in New Mexico

Roadway networks, as part of transportation infrastructure, play an indispensable role in regional economies and community development. The high-quality pavement serviceability of these networks is essential to ensure safe, cost-effective daily traffic operations. In-depth analyses of network-wide pavement surface condition data are necessary inputs for optimal pavement design and maintenance, traffic safety enhancement, and sustainable traffic infrastructure system development. This study aims to investigate various pavement distress condition performance measurements and their correlations to better understand temporal–spatial characteristics of roadway distress based on pavement distress condition data collected in New Mexico from 2006 to 2009. Eight major corridors across various urban and rural areas were selected for analyzing pavement surface-distress conditions and discovering their intrinsic characteristics and patterns across both temporal and spatial domains. The results show that there are not strong correlations among different distress measurements, implying the rationality of the current pavement performance measurement protocol used by the state transportation agencies. Regression models were established and GIS-based spatial analyses were performed to extract temporal and spatial patterns of Distress Rate (DR) data. The model results illustrate significant correlations of the DR data on the same route between two consecutive years, which can be partially characterized by a Markov process. GIS-based spatial investigations also show unique features of pavement condition deterioration attributed to diverse geometric characteristics and traffic conditions, such as vehicle compositions and volumes and urban and rural areas. The research findings are helpful to understand the characteristics of pavement distress conditions more clearly and to optimize traffic infrastructure design and maintenance.

Multi-Criteria Optimization of Bridge Management

Optimizing investment funding levels and combinations of treatment types and timings, as an aid to management decision making, are vital functions of any bridge management system (BMS). Bridge management is always involved the bridge performance assessment which is essentially a multiple criteria decision making (MCDM) problem, which involves multiple assessment criteria such as traffic safety enhancement, preservation of bridge condition, protection from extreme events, agency cost minimization and user cost minimization. In this paper, a multi-criteria optimization methodology was developed not only to take due cognizance of costs but also to include a wide range of performance criteria to facilitate balanced decision-support practices in bridge management. The proposed methodology uses the AHP to determine the weights of criteria, linguistic terms to assess bridge performance under each criterion and aggregate bridge performance under different criteria into an overall performance score for each bridge structure. A numerical example illustrated the integrated methodology and the optimization process.

Nonlinear vehicle distance control in longitudinal direction

Autonomous intelligent cruise control design is a very important aspect of automation systems in future traffic patterns. A strategy for automated vehicle tracking is described, which keeps a velocity-dependent distance between the vehicles in order to provide a constant time-headway. As it is based on the principle of feedback linearization, highly structured and organized algorithms can be provided, without using special preferred operating points. Essential nonlinearities are compensated for in a direct way, and an arbitrary linear behaviour can be imposed on the remaining system. In order to provide a fast transient behaviour and insensitivity to disturbances, the control laws include, besides a proportional part, derivative as well as integral terms. Furthermore, the system is prevented from having an overshoot, and comfort parameters can be adjusted. Owing to its high performance, this control system directly contributes to the enhancement of traffic safety.