Lane Width Research Articles

Accident Prediction based on Integrated Design Consistency with the Lower Number of Vehicles/Traffic Volumes (due to Health Disaster/COVID-19)

Purpose: Malaysia is one of the developing countries that facing high road accident in Asia. The most common accident happens is between motorcyclist and cars. A motorcyclist is 17 times more dangerous than passenger cars. Analysis shows the three main types of accidents in Malaysia is collision with passenger cars, collisions with other motorcycles and single-motorcycle accidents. Road accidents keep rising in Malaysia, because lacks of road geometric design consistency where the drivers make mistakes errors due to the road geometric features. 
 Design/methodology/approach: The study was conducted at F0050 from km 21 until km 25. By using GPS (DG -200), the continuous speed profile data were using new analysis method to develop road design consistency profile of motorcycle and cars and determined the value of the index the integrated road design consistency (IC).
 Findings: The developments of consistency model are based on parameter: the bounded area between the profile and the average speed, standard deviation of speed along a segment. Hence, use ACAD software for develops consistency model’s profile. The integrated-consistency model is the impact of the speed profile on design consistency in traffic and safety evaluations. The highest areas of an accident at km 24 are justified based on the design consistency of the area and integrated design consistency model between of car and motorcycle is poor design. The study shows that crashes and fatalities are fully effected by traffic volumes but road design performances and driving behavior.
 Research limitations/implications: However, the traffic volumes are relatively low in the whole country regarding to the Movement Control Order (MCO) due to the pandemic COVID-19 since March 2020. In contrast to the reduction of the volumes, the number of crashes still shows relatively high. Therefore, it shows that crashes and fatalities are influenced by other factors. Geometric factors such as access point, tangent length, curve length, shoulder width, and lane width are influencing an operating speed profile for developed tangents and curves of motorcycle and cars.
 Practical implications: The result of this study can be used as references to solve accident problems in Malaysia.
 Paper type: This study is categorized as a case study.

Analisis Efektifitas Jalur Sepeda Berdasarkan Metode Bicycle Level Of Service (BLOS)

The bicycle lane policy on the Darmo Highway - Urip Sumoharjo Street - Basuki Rahmat Street - Suryo Governor Street - Panglima Sudirman Street Surabaya City for 8,789 Km has been implemented since 2014. The width of the bicycle lane on this corridor of 1.5 meters is on the road body. The application of bicycle lanes in the city of Surabaya is still less effective because there are still many motor vehicle users, the lack of bicycle users and the purpose of cycling is only for recreation (car free day). The purpose of this study was to analyze the level of effectiveness of the bicycle lanes in the Basuki Rahmat Jalan Surabaya corridor. The method used is quantitative descriptive analysis technique Bicycle Level Of Service (BLOS). Data collection is done by survey and documentation. The results of the BLOS analysis that Monday, Saturday and Sunday BLOS value is "F" more than 5.5 means that the environment is not safe for bicycles except Sunday at 06.00-08.00 BLOS value is "D" means less environment for bicycles.

Human-like driving behaviour emerges from a risk-based driver model

Current driving behaviour models are designed for specific scenarios, such as curve driving, obstacle avoidance, car-following, or overtaking. However, humans can drive in diverse scenarios. Can we find an underlying principle from which driving behaviour in different scenarios emerges? We propose the Driver’s Risk Field (DRF), a two-dimensional field that represents the driver’s belief about the probability of an event occurring. The DRF, when multiplied with the consequence of the event, provides an estimate of the driver’s perceived risk. Through human-in-the-loop and computer simulations, we show that human-like driving behaviour emerges when the DRF is coupled to a controller that maintains the perceived risk below a threshold-level. The DRF model predictions concur with driving behaviour reported in literature for seven different scenarios (curve radii, lane widths, obstacle avoidance, roadside furniture, car-following, overtaking, oncoming traffic). We conclude that our generalizable DRF model is scientifically satisfying and has applications in automated vehicles.

Influence Factor Analysis of Bicycle Free-Flow Speed for Determining the Design Speeds of Separated Bicycle Lanes

To provide a knowledge basis for updating the design speed in bicycle facility codes, this paper examines factors that influence bicycle free-flow speed. We investigated six segments of Nanjing’s separated bicycle lane and established a generalized linear model of the relationship between bicycle free-flow speed and bicyclists’ gender, age, bicycle type, lane width, bicycle lateral position, and travel period. With the model, we determined the statistical significance of each factor and assessed each factor’s impact extent. Through comparing the 85th percentile speeds of different groups, we proposed the recommended values and a method for calculating the design speed of separate bicycle lanes. The following results and conclusions were obtained: (1) The significant influential factors of bicycle free-flow speed were bicyclists’ gender and age, bicycle type, lane width, and bicycles’ lateral position. (2) Bicycle type had the greatest impact on bicycle free-flow speed, following by bicycle lateral position, gender, age, and lane width in sequence. (3) The recommended design speeds for separate lanes of less than 3.5 m and the wider lanes were 25 km/h and 30 km/h, respectively.

Aging drivers and post delineated express lanes: Threading the needle at 70 miles per hour

Post-delineated express lanes represent a combination of driving complexities that are particularly difficult for older drivers to navigate. The narrow geometry and high speeds that are common to this treatment reflect a critical test for drivers whose depth perception, contrast sensitivity, and visual processing speed are reduced. The present study was designed to empirically examine the effects of age and color of express lane delineators on driver behavior. Three groups of participants (aged 18–39, 40–64, and 65 + years old) were required to complete a series of simulated driving scenarios consisting of combinations of single and dual lane configurations, with speed and lane position measured at the beginning and midpoint of each express lane. All drivers were pre-screened on various visual functioning abilities. Drivers in the 65 and older group show significant age-related declines in depth perception, contrast sensitivity, and phoria which were subsequently correlated with a wide range of driving measures including deceleration rate, brake time, jerk, speed, and lane position. Age related perceptual declines were statistically correlated with slower driving speed and wider lane deviations, including a statistically significant increase in the number of excursions beyond the typical 12-foot lane width. Based on these findings, the behavior of senior drivers was identified as a distinct design condition that should govern the design of high-speed, narrow geometric conditions. This age group requires wider lane widths, particularly at the beginning of single-lane post delimited sections, wider buffer areas around the post markers, and dual lane configurations wherever possible.

Comparison of LED-Embedded Pedestrian Crossing Signs with Rectangular Rapid Flashing Beacons and Pedestrian Hybrid Beacons

To date, the effectiveness of light emitting diode (LED)-embedded pedestrian crossing signs is not understood, despite such signs being installed and used at pedestrian crossings. A recent study by Texas Department of Transportation focused on examining driver yield rates, a surrogate safety measure, for signs installed across the state. A total of 13 sites were sampled and field observations and video footage were recorded at each site for a total of 681 staged crossings. The driver yield rate ranged from 5% to 88%, with an average of 40%, indicating that other factors affect the effectiveness of the signs. Statistical modeling results supported this finding with the main contributors to driver yield rates being vehicle volume, measured at 1 min before the staged pedestrian crossing (converted to an hourly volume for the model), operational speed (85th percentile), deviation from 12-ft lane width, and presence of sidewalk. High hourly volume, speeds 45 mph and greater, lack of sidewalks, and 12-ft lanes (no deviation from baseline 12-ft lane width) were found to affect yield probability adversely. Based on the findings from this study, LED-embedded signs would be a suitable candidate treatment at sites with sidewalks, lower operating speeds and traffic volumes, and narrow lanes. Lastly, a comparison between driver yield rates for rectangular rapid flashing beacons, pedestrian hybrid beacons, and LED-embedded signs for various posted speed limits demonstrated that, of the three treatments, LED-embedded signs have the lowest driver yield rates, especially for higher posted speed limits.

Geomatics techniques for evaluation of road pavement rutting

Rutting is one of the severe pavement distresses. It is defined as longitudinal depression under wheel path, caused by repeated traffic load, and it is an indicator of the structural failure as well as having effect on road user’s safety and riding quality. The objectives of this study are to use geomatics techniques such as terrestrial laser scanner, precise positioning system (RTK) and cellular phone for rutting measurements in minor roads in northern part of Jordan, in addition to develop rutting and lateral displacement models using different parameters such as annual average daily traffic (AADT), truck percent, lane width, pavement age and pavement thickness. Manual measurement of rutting was used as reference method for verifying the results obtained from laser scanning, RTK and cellular phone. The study showed that the used methods produced accurate and reliable results compared with the manual method based on root-mean-square (RMSE) which was 0.557 cm for the RTK measurements, 0.577 cm for cellular phone measurements and 0.592 cm of the laser scanner system, respectively. On the other hand, the consistency of accuracy of measurements was slightly better for the cellular phone measurements with a mean average error (MAE) of 0.415 cm, while it was 0.422 cm for the RTK system and 0.442 cm for the laser scanner measurements. The finding of this research will support the development of using geomatics techniques for the measurement of pavement rutting which facilitate the processes and give reliable surface measurement in short time.

Study of potential implementations for motorcycle lanes in Padang city

Currently, two-wheeled vehicles or motorcycle dominate more than 80% of motorized vehicles in Indonesia. In addition to low prices, these motorized vehicles are fuel efficient, have easier and faster mobilization. However, the large number of motorcycles results in complicated traffic operations and reduced levels of safety. Lanes separation is the best engineering solution to overcome this problem. The purpose of this study was to review the potential application of motorcycle lanes in the city of Padang, a case study on the Raya Ampang and By Pass roads. Collecting the primary data were conducted by Video Image Processor. Each road was carried out recording for 1 day during peak hours in the morning, afternoon, and evening. Analysis of road performance is based on the Highway Capacity Manual of Indonesia (MKJI) 1997 with the degree of saturation (DS) parameter. Road performance analysis is reviewed by comparing the existing conditions and the conditions in which the inclusive motorcycle lanes are applied. The results of the study show that both the Raya Ampang and By Pass streets is feasible to provide an inclusive lane or shared lanes for motorcycles—with a lane width of 3 meters. With motorcycle lane application scenarios, on the Raya Ampang street for non-motorcycles lanes, there is a 10.7% increase in road performance, while for By Pass road, the increase in road performance is 31.5%.

Performance Analysis for Uplink Transmission in User-Centric Ultra-Dense V2I Networks

User-centric ultra-dense network (UDN) has been considered a key enabler of high data rate and high reliable vehicle-to-infrastructure (V2I) communication. In this paper, an analytical framework is developed based on stochastic geometry to characterize the uplink V2I transmission performances, in terms of the outage probability, the average vehicle throughput and the average spatial throughput, particularly resolving two important but not well-understood factors in such networks: the spatial correlation in the interference powers among the cooperative roadside units (RSUs) as caused by the shared set of co-channel vehicles, and the combined vertical distance (CVD) owing to the lane width and the antenna height difference between the vehicle and the RSUs. Exact analytical expressions of the outage probability are derived for the general multiple association cases, a far-distance approximation method is introduced to reduce the computational complexity, and lower and upper bounds are obtained corresponding to two extreme assumptions of the spatial interference correlation. We show that the impacts of the spatial interference correlation can be significant and ignoring the correlation can lead to non-optimal system deployment, and that the far-distance approximation can approach well with the exact results. Meanwhile, although CVD diminishes the V2I uplink transmission performances, the diminution can be mitigated through RSU cooperation.

Research on optimized design of road space in mixed sections of historical district: a case study of Xi’an, China

To solve the problems existing in most Chinese historic districts, such as incomplete traffic network planning and unreasonable spatial layout, this paper analyzes the spatial characteristics of motor vehicles in the mixed road section of a historical district and summarizes the application conditions and shortcomings of the model for calculating the Bolankerf lane width and re-calibrates the parameters of the model with the survey data, to determine the formula for calculating the width of motor lanes in different sections of roads. Furthermore, this paper establishes a game model of road space optimization for mixed sections of motor vehicles and non-motor vehicles with the goal of minimizing the generalized travel cost. Finally, by taking the Xiangzimiao historic district as the verification object, this paper draws from the Bolankerf model and the game model to determine the most reasonable road space division plan within blocks.

Effects of peripheral transverse line markings on drivers’ speed and headway choice and crash risk in car-following: A naturalistic observation study

Rear-end crashes are closely related to car-following situation of vehicles. Speeding and insufficient headway are the major reasons as the drivers have not enough time to react to a sudden brake from the leading vehicle. Perceptual countermeasures, like speed reduction markings, are widely used in practice for accident prevention, and are verified with substantial effectiveness. However, compared with its practical application, the perceptual countermeasures are rarely analyzed in depth from the perspective of drivers’ visual perception where the meaning of “perceptual” actually dwells. In addition, its effect on drivers’ headway (distance) choice is almost ignored in previous research. Given this, the present study explored the effects of a certain type of perceptual treatment, i.e., the peripheral transverse line markings (PTLMs), on drivers’ choice of speed and headway (distance) in car-following by a series of on-road experiments. In the on-road experiments, temporary line markings were installed on a real-world freeway in China to shape the PTLMs. The intersection angle (α) and the longitudinal spacing (λ) of the PTLMs were manipulated to attempt to associate the line markings with drivers’ visual perception. Results of general and sectional relative differences of time headway (ηh, θh), speed (ηv, θv), and distance (ηd, θd) suggests that 1) the speed was reduced, the distance and time headway were increased significantly after the installation of PTLMs when compared with the original condition; 2) a larger intersection angle (α) and a smaller longitudinal spacing (λ) of PTLMs could lead to a greater variations in speed and headway (distance); in particular, the PTLMs in a form of α=150°, λ=2m resulted in 0.44 s increase in time headway, 1.33 m/s reduction in speed, and 4.07 m increase in distance in maximum; 3) the real-time crash risk variations under the influence of PTLMs were evaluated by two modified and extended surrogate safety indicators. The effects of PTLMs were discussed and explained considering the influences of optical illusion on lane width narrowing, edge rate on speed and “discontinuity effect” on distance, respectively. The findings of this study provide theoretical support for the perceptual countermeasures and suggest comparative advantages of PTLMs in dealing with rear-end crashes by intervening drivers’ speed and headway choice.

Factors Influencing Crash Frequency on Colombian Rural Roads

Traffic crashes in Colombia have become a public health problem causing about 7,000 deaths and 45,000 severe injuries per year. Around 40% of these events occur on rural roads, taking note that the vulnerable users (pedestrians, motorcyclists, cyclists) account for the largest percentage of the victims. The objective of this research is to identify the factors that influence the frequency of crashes, including the singular orography of the country. For this purpose, we estimated Negative Binomial (Poisson-gamma) regression, Zero-inflated model, and generalized the linear mixed model, thus developing a comparative analysis of results in the Colombian context. The data used in the study came from the official sources regarding records about crashes with consequences; that is, with the occurrence of fatalities or injuries on the Colombian roads. For collecting the highway characteristics, an in-field inventory was conducted, gathering information about both infrastructure and operational parameters in more than three thousand kilometres of the national network. The events were geo-referenced, with registries of vehicles, involved victims, and their condition. The results suggest that highways in flat terrain have higher crash frequency than highways in rolling or mountainous terrain. Besides, the presence of pedestrians, the existence of a median and the density of intersections per kilometre also increase the probability of crashes. Meanwhile, roads with shoulders and wide lanes have lower crash frequency. Specific interventions in the infrastructure and control for reducing crashes risk attending the modelling results have been suggested.

Influence of Lane Width on Semi- Autonomous Vehicle Performance

In the medium-term, the number of semi-autonomous vehicles is expected to rise significantly. These changes in vehicle capabilities make it necessary to analyze their interaction with road infrastructure, which has been developed for human-driven vehicles. Current systems use artificial vision, recording the oncoming road and using the center and edgeline road markings to automatically facilitate keeping the vehicle within the lane. In addition to alignment and road markings, lane width has emerged as one of the geometric parameters that might cause disengagement and therefore must be assessed. The objective of this research was to study the impact of lane width on semi-autonomous vehicle performance. The automatic lateral control of this type of vehicle was tested along 81 lanes of an urban arterial comprising diverse widths. Results showed that the semi-autonomous system tended to fail on narrow lanes. There was a maximum width below which human control was always required—referred to as the human lane width—measuring 2.5 m. A minimum width above which automatic control was always possible—the automatic lane width—was established to be 2.75 m. Finally, a lane width of 2.72 m was found to have the same probability of automatic and human lateral control, namely the critical lane width. Following a similar methodology, these parameters could be determined for other vehicles, enhancing the interaction between autonomous vehicles and road infrastructure and thus supporting rapid deployment of autonomous technology without compromising safety.

Linear Parameter Varying Observer for Lane Estimation Using Cylinder Domain in Vehicles

In this paper, to address the vehicle state estimation problem, we propose an approach that can overcome the practical problems associated with vehicle lane changes on highways. The vehicle state with respect to the road, which is called the lateral offset, jumps in camera vision sensors when the vehicle changes lanes. Thus, we solve this state jump problem by translating the state estimation problem into a new domain called the cylinder domain. We propose a linear parameter varying (LPV) observer to estimate the vehicle states by considering the lane width variation during a lane change. The lane information is estimated by the proposed LPV observer in the cylinder domain; thus, the estimated lane information does not jump. Consequently, the lane estimation performance can be improved during the transient response when the lateral position of the autonomous vehicle crosses over between the original and target lanes in the event of a lane change. The proposed algorithm is validated with real vehicle driving data obtained using a camera vision sensor.

Behavioral factors of motorcyclists in right-turn movements at unsignalized intersections: An insight from Yogyakarta, Indonesia

Conflicting actions of motorcyclists making right-turn movements (RTM) pose a considerable risk of accidents, which is of great concern when attempting to augment the level of safety at unsignalized intersections. This study was conducted to identify the habits of motorcyclists in RTMs, including normal RTM, weaving/merging RTM, and opposite indirect RTM when crossing unsignalized intersections. The behavioral factors were affected by gender, passenger status, motorcycle type, STOP factor, speed, travel time, and lane width of major and minor roads, which were tested using multinomial and nested logit models on 2358 motorcyclists. The results indicated that all of the observed variables significantly influenced motorcyclists' behaviors at unsignalized intersections, excepting the variable of travel time. The factors of speed and lane width on minor roads significantly correlated to motorcyclist behaviors during weaving and merging RTM only. The logit model reveals that female motorcyclists tend to make normal RTMs. The study also proposes ideal and practical recommendations, such as stop and speed regulations, channelization, number of lanes and lane widths, and maneuver markers, in order to address dangerous behaviors, in particular weaving and merging and opposite indirect-type RTMs.

Safety effects of traffic lane and shoulder widths on two-lane undivided rural roads: A matched case-control study from Norway

This study estimates the effects of lane and shoulder widths on occurrence of head-on and single-vehicle accidents on rural two-lane undivided roads in Norway while considering the differences between winter and non-winter accidents and their severity levels. A matched case-control method was applied to calculate the odds ratios for lane and shoulder width categories, while controlling for the effects of AADT and adjusting for the effects of region, speed limit, segment length, share of long vehicles in AADT and horizontal alignment. The study used a sample of 71,999 roadway segments identified in GIS and 1886 related accidents recorded by the police in five-year period. The results suggest that it is relevant to consider winter and non-winter accidents as well as severe and slight accidents separately when studying the effects of lane and shoulder widths on the occurrence of head-on and single-vehicle accidents. When examining lane and shoulder widths for all related accidents, the lane widths 1.50–2.50 m and shoulder widths 0.50−0.75 m were relatively safer than other categories on Norwegian two-lane rural undivided roads.

Geometric road design factors affecting the risk of urban run-off crashes. A case-control study.

ObjectiveSingle vehicle run-off crashes in urban areas constitute a growing problem that deserves more attention from authorities and researchers. This study aims to detect geometric road design risk factors characterizing places where urban run-off crashes might happen.MethodsA case-control study was performed in the urban area of Valladolid (Spain) with data corresponding to a four-year period. Logistic regression models were used to analyze data, considering different variables related to design parameters in the models: type of intersection, radius of curvature, width of the pavement, width of the traffic lane, number of lanes for traffic in the same direction, direction of the traffic, length of the previous straight section, distance to the previous traffic light, slope, and finally, priority regulation. Two different scenarios were investigated: intersections and curves.ResultsThe Adjusted Odds-Ratio of a run-off crash was five times higher in double direction roads with median strip than in one-way urban roads, for both curves and intersections, and almost nine times higher on road sections with previous straight lengths greater than 500 meters. Specific risk factors for intersections are “number of lanes for traffic in the same direction” (the odds of a run-off crash are more than five times higher on a road with two or more lanes), “length of preceding straight section” (the odds on road sections with lengths greater than 500 meters are more than nine times that of road sections with a length of less than 150 meters). For curves, specific factors are “width of the traffic lane” (the odds of a run-off crash on curves with lanes wider than 3.75m are more than six times higher) and “priority regulation” (the odds of a run-off crash increases more than twelve times on road sections with traffic light regulation over those without any regulation).ConclusionsThe current study identifies urban road configurations that might require redesigning with the aim of decreasing the odds of a run-off crash, or the implementation of passive protective systems to mitigate their consequences. Specifically, intersections in two direction roads with median strip, more than two lanes per direction and a long preceding straight section, as well as curves with wide lanes and traffic light regulation, are the places that require attention.

A cellular automata traffic flow model with spatial variation in the cell width

This paper addresses a practical problem in road transportation regarding how the arrangement of three-dimensional facilities on a road (e.g., toll stations, construction guardrails and safety cones) will affect driving behavior. According to literature research and site investigations, in most cases, such facilities cause drivers to perceive a certain level of variation in the lane width, which will ultimately affect their driving behavior. For example, a driver will control a vehicle more carefully when approaching such a facility and usually reduce the passing speed according to his/her perception and experience. To the best of our knowledge, however, few related numerical studies and corresponding practical test cases have been reported at present. This paper uses a cellular automata traffic flow model to analyze this traffic problem. Notably, the existing models of this kind do not consider the cell width and its effect on vehicle speed. Motivated by the above problems and based on the human eye imaging principle, the dynamic visual field mechanism and test data, the present work first proposes a numerical relationship among the vehicle speed, the distance to the target and the perceived width of a fixed facility alongside the lane. Then, a new cellular automata traffic flow model with spatial variation in the cell width is designed to correlate the cell shape variable with the perceived width of the lane. Simulations based on this new model are conducted to simulate and compare the statuses of vehicles in a single lane with and without fixed facilities. The simulation results show that the speed reduction caused by cell deformation has a significant influence on the flow of traffic and that a “gate” effect occurs in deformed cells, resulting in behavior close to actual traffic phenomena. However, as the traffic density increases, the influence of deformed cells on a vehicle’s movement lessens. The proposed model may be a meaningful extension of cellular automata traffic flow models.

How do drivers negotiate intersections with pedestrians? The importance of pedestrian time-to-arrival and visibility

Forward collision warning (FCW) and autonomous emergency braking (AEB) systems are increasingly available and prevent or mitigate collisions by alerting the driver or autonomously braking the vehicle. Threat-assessment and decision-making algorithms for FCW and AEB aim to find the best compromise for safety by intervening at the “right” time: neither too early, potentially upsetting the driver, nor too late, possibly missing opportunities to avoid the collision.Today, the extent to which activation times for FCW and AEB should depend on factors such as pedestrian speed and lane width is unknown. To guide the design of FCW and AEB intervention time, we employed a fractional factorial design, and determined how seven factors (crossing side, car speed, pedestrian speed, crossing angle, pedestrian size, zebra-crossing presence, and lane width) affect the driver’s response process and comfort zone when negotiating an intersection with a pedestrian. Ninety-four volunteers drove through an intersection in a fixed-base driving simulator, which was based on open-source software (OpenDS). Several parameters, including pedestrian time-to-arrival and driver response time, were calculated to describe the driver response process and define driver comfort boundaries.Linear mixed-effect models showed that driver responses depended mainly on pedestrian time-to-arrival and visibility, whereas factors such as pedestrian size, zebra-crossing presence, and lane width did not significantly influence the driver response process. Drivers released the accelerator pedal in 99.8 % of the trials and braked in 89 % of the trials. Forty-six percent of the drivers changed their negotiation strategy (proportion of pedal braking to engine braking) to minimize driving effort over the course of the experiment. In fact, 51 % of the of the inexperienced drivers changed their response strategy whereas only 40 % of the experienced drivers did; nevertheless, all drivers behaved similarly, independent of driving experience. The flexible and customizable driving environment provided by OpenDS may be a viable platform for behavioural experiments in driving simulators.Results from this study suggest that visibility and pedestrian time-to-arrival are the most important variables for defining the earliest acceptable FCW and AEB activations. Fractional factorial design effectively compared the influence of seven factors on driver behaviour within a single experiment; however, this design did not allow in-depth data analysis. In the future, OpenDS might become a standard platform, enabling crowdsourcing and favouring repeatability across studies in traffic safety. Finally, this study advises future design and evaluation procedures (e.g. new car assessment programs) for FCW and AEB by highlighting which factors deserve further investigation and which ones do not.

Intensity Thresholding and Deep Learning Based Lane Marking Extraction and Lane Width Estimation from Mobile Light Detection and Ranging (LiDAR) Point Clouds

Lane markings are one of the essential elements of road information, which is useful for a wide range of transportation applications. Several studies have been conducted to extract lane markings through intensity thresholding of Light Detection and Ranging (LiDAR) point clouds acquired by mobile mapping systems (MMS). This paper proposes an intensity thresholding strategy using unsupervised intensity normalization and a deep learning strategy using automatically labeled training data for lane marking extraction. For comparative evaluation, original intensity thresholding and deep learning using manually established labels strategies are also implemented. A pavement surface-based assessment of lane marking extraction by the four strategies is conducted in asphalt and concrete pavement areas covered by MMS equipped with multiple LiDAR scanners. Additionally, the extracted lane markings are used for lane width estimation and reporting lane marking gaps along various highways. The normalized intensity thresholding leads to a better lane marking extraction with an F1-score of 78.9% in comparison to the original intensity thresholding with an F1-score of 72.3%. On the other hand, the deep learning model trained with automatically generated labels achieves a higher F1-score of 85.9% than the one trained on manually established labels with an F1-score of 75.1%. In concrete pavement area, the normalized intensity thresholding and both deep learning strategies obtain better lane marking extraction (i.e., lane markings along longer segments of the highway have been extracted) than the original intensity thresholding approach. For the lane width results, more estimates are observed, especially in areas with poor edge lane marking, using the two deep learning models when compared with the intensity thresholding strategies due to the higher recall rates for the former. The outcome of the proposed strategies is used to develop a framework for reporting lane marking gap regions, which can be subsequently visualized in RGB imagery to identify their cause.