A Comprehensive Evaluation of State-of-the-Art Deep Learning Models for Road Surface Type Classification

In recent years,as intelligent transportation systems(ITS)such as autonomous driving and advanced driver-assistance systems have become more popular,there has been a rise in the need for different sources of traffic situation data.The classification of the road surface type,also known as the RST,is among the most essential of these situational data and can be utilized across the entirety of the ITS domain.Recently,the benefits of deep learning(DL)approaches for sensor-based RST classification have been demonstrated by automatic feature extraction without manual methods.The ability to extract important features is vital in making RST classification more accurate.This work investigates the most recent advances in DL algorithms for sensor-based RST classification and explores appropriate feature extraction models.We used different convolutional neural networks to understand the functional architecture better;we constructed an enhanced DL model called SE-ResNet,which uses residual connections and squeeze-and-excitation mod-ules to improve the classification performance.Comparative experiments with a publicly available benchmark dataset,the passive vehicular sensors dataset,have shown that SE-ResNet outperforms other state-of-the-art models.The proposed model achieved the highest accuracy of 98.41%and the highest F1-score of 98.19%when classifying surfaces into segments of dirt,cobblestone,or asphalt roads.Moreover,the proposed model significantly outperforms DL networks(CNN,LSTM,and CNN-LSTM).The proposed RE-ResNet achieved the classification accuracies of asphalt roads at 98.98,cobblestone roads at 97.02,and dirt roads at 99.56%,respectively.

DEPTH PREDICTION OF RAIN WATER ON ROAD SURFACE BASED ON ARTIFICIAL NEURAL NETWORK

A model based on the non-linear artificial neural network （ANN） is established to predict the thickness of the water film on road surfaces. The weight and the threshold can be determined by training test data, and the water film thickness on the road surface can be accurately predicted by the empirical verification based on sample data. Results show that the proposed ANN model is feasible to predict the water film thickness of the road surface.

SIMULATION ANALYSIS OF INFLUENCE OF ROAD SURFACE ROUGHNESS ON TYRE DYNAMIC PERFORMANCE IN AUTOMOTIVE BRAKING

Using a nonlinear time varying tyre model, this paper simulatively analyzes the influence of road surface roughness amplitude and road spatial frequency on automobile ground adhesion ability. The result shows that with the increase of road surface roughness, the tyre adhesion ability declines, and the automotive braking distance increases. Moreover, the reliability of the nonlinear time varying tyre model in reflecting the influence of the road surface roughness is validated. It is testified that this model is an effective dynamic one in the simulation of automotive braking performance on uneven road.

Automated extraction of expressway road surface from mobile laser scanning data

This paper presents a voxel-based region growing method for automatic road surface extraction from mobile laser scanning point clouds in an expressway environment.The proposed method has three major steps:constructing a voxel model;extracting the road surface points by employing the voxel-based segmentation algorithm;refining the road boundary using the curb-based segmentation algorithm.To evaluate the accuracy of the proposed method,the two-point cloud datasets of two typical test sites in an expressway environment consisting of flat and bumpy surfaces with a high slope were used.The proposed algorithm extracted the road surface successfully with high accuracy.There was an average recall of 99.5%,the precision was 96.3%,and the F1 score was 97.9%.From the extracted road surface,a framework for the estimation of road roughness was proposed.Good agreement was achieved when comparing the results of the road roughness map with the visual image,indicating the feasibility and effectiveness of the proposed framework.

Multi-objective genetic algorithm for the optimization of road surface cleaning process

The parameters affecting road surface cleaning using waterjets were researched and a fuzzy neural network method of calculating cleaning rate was provided. A genetic algorithm was used to configure the cleaning parameters of pressure, standoff distance, traverse rate and angle of nozzles for the optimization of the cleaning effectiveness, efficiency, energy and water con-sumption, and a multi-objective optimization model was established. After calculation, the optimized results and the trend of variation of cleaning effectiveness, efficiency, energy and water consumption in different weighting factors were analyzed.

New Method of Road Surface Identification in Anti-Lock Braking System

Based on the vehicle-road dynamic model, the road characteristic parameter, which depends on different types of road surfaces, is introduced and a new method of road surface identification for automotive anti-lock braking system (ABS) is proposed. According to the characteristics of vehicle-road dynamic model, a simple math resolution method of the model's factors is established. Only using the information of wheel speed, the vehicle reference velocity and the wheel slip ratio are estimated real-timely. And based on the wheel dynamic model, the road characteristic parameter is determined to identify the road surface for the determination of thresholds of ABS regulative parameters. With this new method, the road surface identification can be accurately obtained and calculation time is short that it can meet the ABS real time control need, and it also improves the performance of ABS.

RESUSPENSION METHOD FOR ROAD SURFACE DUST COLLECTION AND AERODYNAMIC SIZE DISTRIBUTION CHARACTERIZATION

Traffic-generated fugitive dust is a source of urban atmospheric particulate pollution in Beijing. This paper introduces the resuspension method, recommended by the US EPA in AP-42 documents, for collecting Beijing road-surface dust. Analysis shows a single-peak distribution in the number size distribution and a double-peak mode for mass size distribution of the road surface dust. The median diameter of the mass concentration distribution of the road dust on a high-grade road was higher than that on a low-grade road. The ratio of PM2.5 to PM10 was consistent with that obtained in a similar study for Hong Kong. For the two selected road samples, the average relative deviation of the size distribution was 10.9% and 11.9%. All results indicate that the method introduced in this paper can effectively determine the size distribution of fugitive dust from traffic.

Robust Identification of Road Surface Condition Based on Ego‑Vehicle Trajectory Reckoning

The type of road surface condition(RSC)will directly affect the driving performance of vehicles.Monitoring the type of RSC is essential for both transportation agencies and individual drivers.However,most existing methods are solely based on a dynamics-based method or an image-based method,which is susceptible to road excitation limitations and interference from the external environment.Therefore,this paper proposes a decision-level fusion identification framework of the RSC based on ego-vehicle trajectory reckoning to accurately obtain the type of RSC that the front wheels of the vehicle will expe-rience.First,a road feature extraction model based on multi-task learning is conducted,which can simultaneously segment the drivable area and road cast shadow.Second,the optimized candidate regions of interest are classified with confidence levels by ShuffleNet.Considering environmental interference,candidate regions of interest regarded as virtual sensors are fused by improved Dempster-Shafer evidence theory to obtain the fusion results.Finally,the ego-vehicle trajectory reckoning module based on the kinematic bicycle model is added to the proposed fusion method to extract the RSC experienced by the front wheels.The performance of the entire framework is verified on a specific dataset with shadow and split curve roads.The results reveal that the proposed method can identify the RSC with accurate predictions in real time.

Road surface condition sensor based on scanning detection of backward power

A method of detecting dry, icy and wet road surface conditions based on scanniag detection of single wavelength backward power is proposed in this letter. The detector is used to receive the backward scattered power which changes with the incidence angle. The relationship between backward power and incidence angle is used to find out the effective angle range and distinguish method. Experiment and simulation show that it is feasible to classifv these three conditions within incidence angle of 5.3 degree.

Sustainability of lime stabilized road subgrade in mountainous regions of Türkiye

In this paper the use of lime stabilized subgrade for low volume roads in two regions with high mountains and different frost penetration conditions in Türkiye was investigated in terms of design,performance,and cost.Pavements on unstabilized and stabilized subgrade were designed for two regions(Izmir and Van),covering all climate variations.The resilient modulus of the lime stabilized subgrade with different soil pulverization levels for non-freezing and freezing conditions were taken from a previous laboratory study.Frost effects were considered in pavement design using two different approaches,including limited subgrade frost penetration method and reduced subgrade strength method.Detailed application and evaluation were performed for all steps.Lime stabilized subgrades significantly reduced the thickness of base courses,and the benefit of lime stabilization was highly dependent on soil pulverization level.A detailed cost analysis on the unstabilized and stabilized cases found that the use of lime stabilization in the subgrade provided significant initial cost savings.Comparative analysis by using the AASHTO(1993)method and KENPAVE software,and quantity effect of soil pulverization level on the performance of low volume roads from a service life perspective,show that subgrade resilient modulus can be estimated.It is also possible to make correct performance estimation in the field.The results of the study show that lime stabilization is a good solution for low volume roads in the mountainous regions of Türkiye.

Use of generated artificial road profiles in road roughness evaluation

In the evaluation of road roughness and its effects on vehicles response in terms of ride quality, loads induced on pavement, drivers＇ comfort, etc., it is very common to generate road profles based on the equation provided by ISO 8608 standard, according to which it is possible to group road surface profiles into eight different classes. However, real profiles are significantly different from the artificial ones because of the non-stationary fea- ture of the first ones and the not full capability of the ISO 8608 equation to correctly describe the frequency content of real road profiles. In this paper, the international roughness index, the frequency-weighted vertical acceleration awz according to ISO 2631, and the dynamic load index are applied both on artificial and real profiles, highlighting the different results obtained. The analysis carried out in this work has highlighted some limitation of the ISO 8608 approach in the description of performance and conditions of real pavement profiles. Furthermore, the different sensitivity of the various indices to the fitted power spectral density parameters is shown, which should be taken into account when performing analysis using artificial profiles.

Road identification method based on angular acceleration of vehicle driving wheel

The fundamental principle of road identification by using angular acceleration of driving wheels was demonstrated in this paper. Based on the analysis of energy conversion and parameters variation during the vehicle drive slip process, the change of adhesion coefficient relative to the an- gular acceleration were theoretically studied experimentally validated. The variation shows that the change of adhesion coefficient relative to the angular acceleration and the change of slip ratio in the drive slip process have same trend-both of them exist an only optimal angular acceleration corre- sponding to the peak value of adhesion coefficient. The peak adhesion coefficient of the prototype vehicle is about 0. 14 on the ice-covered road surfaces, with the corresponding optimal angular accel- eration of about 23.5 rad/s2 and optimal slip ratio of about 9. 4%.

Evaluation of alternative criteria for determining the optimal location of RWIS stations

This article presents a comprehensive framework for determining the location of road weather information system （RWIS） stations over a regional road network. In the proposed methodology, the region is divided into a grid of equal-sized zones which are considered as the minimum spatial unit for allocating a candidate set of RWIS stations. These zones are ranked according to a set of pre-specified criteria that reflect the needs for, and potential benefits from, real-time RWIS, including road surface temperature variability, precipitation, network traffic, and collision patterns. A case study based on the existing RWIS network in the province of Ontario was conducted to illustrate the major features of the proposed method and evaluate the implications of alternative loca- tion selection criteria. The findings of the study suggest that it is feasible to develop a systematic process for locating RWIS stations using an integrated location criterion to capture multiple factors being considered in prac- tice. The study has also revealed the need to establish quantitative models for estimating the benefit of real-time information from RWIS stations, which is the foundation of a cost-benefit-based RWIS location optimization model.

Particulate matter exchange between atmosphere and roads surfaces in urban areas

The deposition and the re-suspension of particulate matter(PM) in urban areas are the key processes that contribute not only to stormwater pollution, but also to air pollution. However, investigation of the deposition and the re-suspension of PM is challenging because of the difficulties in distinguishing between the resuspended and the deposited PM. This study created two Bayesian Networks(BN) models to explore the deposition and the re-suspension of PM as well as the important influential factors. The outcomes of BN modelling revealed that deposition and re-suspension of PM10 occurred under both, high-traffic and low-traffic conditions, and the re-suspension of PM2.5 occurred under low-traffic conditions. The deposition of PM10 under low-volume traffic condition is 1.6 times higher than under highvolume traffic condition, which is attributed to the decrease in PM10 caused by relatively higher turbulence under high-volume traffic conditions. PM10 is more easily resuspended from road surfaces compared to PM2.5 as the particles which larger than the thickness of the laminar airflow over the road surface are more easily removed from road surfaces. The increase in wind speed contributes to the increase in PM build-up by transporting particulates from roadside areas to the road surfaces and the airborne PM2.5 and PM10 increases with the increase in relative humidity. The study outcomes provide a step improvement in the understanding of the transfer processes of PM2.5 and PM10 between atmosphere and urban road surfaces, which in turn will contribute to the effective design of mitigation measures for urban stormwater and air pollution.

Highway icing time prediction with deep learning approaches based on data from road sensors

In harsh climates,highway icing poses a hazard to traffic safety and increases road maintenance costs.It is of great significance to predict when the highway icing may occur and take a preventive plan.However,there are few studies on highway icing time prediction due to the scarcity and complexity of data.In this study,variables of icing temperature,friction,ice percentage,road surface temperature,water film height,saline concentration,and road condition were collected by road sensors distributed on a highway in China.A large-scale time series highway surface information dataset called HighwayIce is formed.Furthermore,a deep learning approach called IceAlarm,composed of long short-term memory neural network(LSTM),multilayer perceptron(MLP),and residual connection,has been developed to predict when the highway will ice.The LSTM is used to process dynamic variables,the MLP is used to process static variables,and the fully-connected layers with residual connections are used to make a deep fusion.The experimental results show that the average mean absolute error before icing using the IceAlarm model is about 6min and outperforms all baseline models.The HighwayIce dataset and IceAlarm model can help improve the prediction accuracy and efficiency of forecasting real-world road icing time,therefore reducing the impact of icy road conditions on traffic.

Engineering Application and Development of Anti-skid Sand

In view of the increasing cement concrete pavement in China,the proportion of road non-slip surface layer is large,the winter slippery performance is insufficient and the later non-slip treatment is difficult. Through the concrete construction and post-application and development of the anti-skid sand in the road and bridge,the feasible anti-skid optimization measures are put forward.

Why Dirt Roads Develop A Washboard Surface

在沙土路面上行驶的车辆往往会使路面形成象搓衣板一样硌硌楞楞的痕迹,无论是小型私家车还是大型卡车经过时都因为颠簸而苦不堪言,而路政部门似乎除了定期用推土机将路面铲平之外别无他法。这是轮胎压出的印迹还是由于别的原因出现的呢?什么情况下可以避免使路面成为搓衣板呢?英国剑桥的一组研究人员最近通过实验室模拟和电脑分析找出了答案。

Determination of riding comfort on cycleways using a smartphone application

Well maintained cycleways will encourage more people to cycle,as the condition of cycleways is important for the safety,accessibility and riding comfort of cyclists.Despite that,only a few models used to describe the quality of service for cyclists take the surface condition into account.Objective measuring methods are needed to enable reliable and effective assessment of surface conditions,and measurable performance criteria related to the needs of cyclists should be developed.The purpose of this study has been to test the reliability and validity of using accelerometers in smartphones to assess the riding comfort on cycleways.A smartphone application converting three-dimensional accelerometer measurements into a single indicator for cycleways has been used to assess road surfaces in two field studies,in Sweden and Norway,respectively.Both studies assessed test sections of varying quality.To relate the measurements to subjective riding comfort assessments by cyclists,recruited cyclists collected quantitative data using the app,whilst also rating their perceived riding comfort by completing a survey.Measurements were also related to standard road surface condition indicators,generated from a road surface tester equipped with 19 laser sensors:international roughness index(IRI),mega-and macrotexture.The results show that it is possible to describe the unevenness of a cycleway using the technology present in smartphones.A software application can be used to collect and analyse data from the acceleration sensors in the phone,which can then be used to describe the riding comfort of cyclists.It is mainly the unevenness in the 50-1000 mm sizerange that create the greatest discomfort for cyclists,and intermittent vibrations are perceived as more uncomfortable than more evenly distributed vibrations.Therefore,IRI is not a relevant measurement for describing the riding comfort of cyclists.

Characteristics and application of road absorbing solar energy

If the heat of road surface can be stored in summer, the road surface temperature will be decreased to prevent permanent deformation of pavement. Besides, if the heat stored is released, it can supply heat for buildings or raise the road surface temperature for snow melting in winter. A road-solar energy system was built in this study, and the heat transfer mechanism and effect of the system were analyzed according to the monitored solar radiant heat, the solar energy absorbed by road and the heat stored by soil. The results showed that the road surface temperature was mainly affected by solar radiation, but the effect is hysteretic in nature. The temperature of the solar road surface was 3~C-6~C lower than that of the ordinary road surface. The temperature of the solar road along the vertical direction was 2~C-5~C lower than that of the ordinary road. The temperature difference increased as the distance to the heat transfer tubes decreased. The average solar collector efficiency of the system was 14.4%, and the average solar absorptivity of road surface was 36%.

Characterization of road surfacing aggregates based on their mineralogical fingerprint

To ensure the safety of infrastructure users,the long-term skid resistance is a crucial factor and is determined in large by the mineralogical and morphological characteristics of surfacing aggregate.Most studies have investigated these aggregate properties separately without considering the interrelation between one another.The objective of this study is to consider the morphological characteristics as well as the mineralogical fingerprint of aggregate to develop an innovative approach to optimize the aggregate selection process.The investigations are based on 11 different aggregate types with a broad range of mineralogy,commonly used in Germany.The long-term influence of polishing and wearing on the surface aggregate was simulated by means of the Aachen Polishing Machine and the MicroDeval test respectively.To evaluate the impact of these tests,the aggregate shape was characterized by means of an imaging system called Aggregate Image Measurement System while the skid resistance of aggregates was evaluated with the British Pendulum Test.The test results show that the quartz and calcite are the key crystals to determine the anti-wear resistance of aggregates.A correlation between the skid resistance,morphological properties and mineralogy is derived,which proves the mineralogical fingerprint technology is practical for characterization of aggregates used in pavement surface layers.