Pavement Damage Research Articles

Detection and classification of pavement damages using wavelet scattering transform, fractal dimension by box-counting method and machine learning algorithms

Pavement condition assessment is essential for roadway maintenance and rehabilitation processes. Image-based inspection methods provide information regarding the surface of the pavement and allow quantitative analyses of pavement conditions. A methodology for the detection of damages in the pavement, by applying pattern recognition and image analysis and machine learning algorithms, is presented in the paper. This methodology consists of image acquisition, image processing using the wavelet scattering transform (WST), feature extraction employing the fractal dimension by box-counting method, and finally classification. The methodology was applied for the detection of three common types of damages: potholes, longitudinal and alligator cracks. Two different supervised learning algorithms, Artificial Neural Networks (ANN) and Support Vector Machine (SVM), were used for classification and results are compared training Convolutional Neural Networks (CNN). The multilayer ANN an overall accuracy of 98.36% and an F-score of 98.33%, while the SVM an accuracy of 97.22% and an F-score of 97.22%.

Fatigue life prediction of orthotropic steel deck strengthened with UHPC under stochastic traffic load

In recent years, the ultra-high performance concrete (UHPC) has been increasingly used to strengthen orthotropic steel decks (OSD) to solve the cracking problems at fatigue-prone details and pavement damage. In this paper, the fatigue life of a cable-stayed orthotropic steel decks bridge under stochastic traffic loads is calculated before and after the orthotropic steel decks strengthened with the ultra-high performance concrete layer. The traffic data of the real bridge for 1 week is first obtained based on the weigh-in-motion system. Then, a stochastic traffic load on the bridge is simulated for its service life by the Monte Carlo method. A fatigue life analysis framework, which includes the traffic load simulation, a refined finite element model, the S-N curve and Miner linear cumulative damage criterion, is proposed for fatigue life prediction of orthotropic steel decks. For the bridge before reinforcement, the predicting results for the fatigue life of three fatigue-prone details, including the scallop cutout, rib-to-diagram and rib-to-deck joint are basically consistent with that of the actual bridge inspection results. After strengthening by ultra-high performance concrete, the fatigue life of the three structural details are increased from 15.87, 13.89, and 32.26 years to more than 100 years, respectively, as compared with the original orthotropic steel decks structure.

YOLOv5s-M: A deep learning network model for road pavement damage detection from urban street-view imagery

Road pavement damage affects driving comfort markedly, threatens driving safety, and may even cause traffic accidents. The traffic management department conventionally captures pavement damage information mainly using manual and vehicle-mounted equipment, which is not conducive to the detection of large-scale road pavement distress. Street-view images can provide full-view images of urban roads where the data is updated regularly by navigation map service companies, making it possible to rapidly detect pavement damage in urban areas. This paper presents a new pavement damage detection approach that is built upon an improved YOLOv5 network and street-view images. The proposed model can deal with a large-scale detection layer to improve the detection precision of large distress targets, achieving thus both cross-layer and cross-scale feature fusion by using the Generalized Feature Pyramid Network (Generalized-FPN) structure. The improved network also involves a diagonal Intersection over Union loss for regression calculation of the boundary box and builds the decoupled Head structure to achieve the decoupling detection of prediction and regression. As a result, the fusion of the weak feature information in feature layers is enhanced at different spatial scales, a more suitable method is achieved for pavement damage detection in the complex context of multi-scale street-view images, and the accuracy of the modified network is much improved in the detection of pavement distress from street-view imagery. Furthermore, We created a large image sample set for model training and testing, and a total of 156,304 street-view images, obtained from Fengtai District, Beijing, China was used for demonstrating the usefulness of the proposed network. The findings indicated that the proposed approach could effectively achieve pavement damage detection of urban roads from street-view images, with a precision average of 79.8% on the test samples. Moreover, the developed model was applied for pavement damage detection for all the roads in Fengtai District, Beijing, indicating that our method can offer viable damage data for road maintenance planning.

Influence of Mesoscopic Structural Characteristics of Asphalt Mixture on Damage Behavior of Asphalt Pavement

Mechanical responses and damage behaviors of asphalt pavement are closely associated with the mesostructure characteristics of the asphalt mixture. The aim of this research was to investigate the effect of the different mesostructure characteristics of asphalt mixtures on pavement damage behavior through a multiscale finite-element (FE) analysis method. A two-dimensional (2D) random aggregate generation algorithm was used to construct the mesostructures of asphalt mixtures with different characteristics, including aggregate distribution uniformity, shape, volume fraction, and size range. Bilinear cohesive elements were used to model damage initiation within asphalt pavement. Also, the nonuniformity coefficient (NUC) was applied to evaluate the uniformity of aggregate distribution within asphalt pavement. The results indicated that the NUC could well characterize the uniformity of coarse aggregate distribution in pavement, and the damage initiation zone became larger as the NUC increased. Compared with regular aggregates, irregular shape aggregates led to more damage initiation in asphalt pavement. For regular shape aggregates, the damage initiation zone became smaller as the aggregate shape approached the circle. In addition, the decrease of aggregate size and volume fraction could increase the possibility of “top-down” cracking of pavement.

Comparison of Asphalt Concrete-Wearing Course (AC-WC) characteristics using 60/70 asphalt penetration and Elvaloy modified asphalt

The performance of the asphalt mixture is affected by the asphalt characteristics and the gradation of the aggregate. Polymer-modified asphalt has been widely used to overcome road pavement damage such as cracking, spalling, yielding, and damage due to deformation due to grooves, curling, or rutting. Polymer-modified asphalt has been widely used, both from plastomers and elastomeric types. Elvaloy modification asphalt resulted from the conventional asphalt mixing process of 60/70 with Elvaloy additives from Dupont that are elastomer using RET (Reactive Elastomer Terpolymer) technology. This study aimed to compare the characteristics of the AC-WC mixture using 60/70 penetration asphalt and Elvaloy modified asphalt. The study began by making a paved mixture using 60/70 penetration asphalt with the middle-graded aggregate. The Elvaloy modified asphalt has three aggregate gradations: lower limit, middle, and upper limit gradation. Test objects were made as many as 156 test objects to obtain Optimum Asphalt Content (OAC) values, then using OAC values, made 36 test objects for the Marshall test; 4 test objects for Wheel Tracking Machine (WTM) testing. The Marshall and WTM tests determine the performance of 60/70 penetration asphalt and Elvaloy modified asphalt on the AC-WC mixture. The research results of the properties test showed that Elvaloy modified asphalt has a better penetration value, softening point and ductility than 60/70 penetration asphalt. Marshall’s test results provided the highest Stability and flow value in the Elvaloy modified asphalt mixture. From the results of the WTM test, only the middle grade of Elvaloy modified asphalt meets the specification.

Damage Detection and Localization of Bridge Deck Pavement Based on Deep Learning.

Bridge deck pavement damage has a significant effect on the driving safety and long-term durability of bridges. To achieve the damage detection and localization of bridge deck pavement, a three-stage detection method based on the you-only-look-once version 7 (YOLOv7) network and the revised LaneNet was proposed in this study. In stage 1, the Road Damage Dataset 202 (RDD2022) is preprocessed and adopted to train the YOLOv7 model, and five classes of damage were obtained. In stage 2, the LaneNet network was pruned to retain the semantic segmentation part, with the VGG16 network as an encoder to generate lane line binary images. In stage 3, the lane line binary images were post-processed by a proposed image processing algorithm to obtain the lane area. Based on the damage coordinates from stage 1, the final pavement damage classes and lane localization were obtained. The proposed method was compared and analyzed in the RDD2022 dataset, and was applied on the Fourth Nanjing Yangtze River Bridge in China. The results shows that the mean average precision (mAP) of YOLOv7 on the preprocessed RDD2022 dataset reaches 0.663, higher than that of other models in the YOLO series. The accuracy of the lane localization of the revised LaneNet is 0.933, higher than that of instance segmentation, 0.856. Meanwhile, the inference speed of the revised LaneNet is 12.3 frames per second (FPS) on NVIDIA GeForce RTX 3090, higher than that of instance segmentation 6.53 FPS. The proposed method can provide a reference for the maintenance of bridge deck pavement.

Effect of 3D laser point spacing on cement concrete crack width measurement

The objective of this paper is to examine the reliability of three-dimensional (3D) laser detection technology data density for crack width detection results of cement slabs. Four groups of cement concrete crack elevation data with a laser data density of 0.5–1.5 mm were obtained using an indoor 3D laser detection system, and 3D models were established. The nonlinear least squares method was applied to fit the fracture section, and the crack width was determined by the peak value analysis. The results demonstrate that the lateral spacing of laser points exerts a large impact on the mean and discrete degrees of cement concrete crack width detection results. The laser point spacing is positively correlated with crack identification errors. Insufficient laser accuracy leads to an overestimation of crack severity level and affects the accuracy of pavement damage condition evaluation. High-precision laser equipment exhibits certain reliability for detecting cement concrete crack width above 3 mm. In the actual pavement crack width detection process, the appropriate transverse spacing of laser points can be selected according to different error limit requirements to fulfill the requirements of both detection reliability and data processing efficiency. Suggestions for future research include expanding the experimental conditions, increasing the 3D laser point spacings, and selecting more road lanes and pavement materials to further examine the influential factors of pavement crack width measurement.

Extended Weight Systems Pavement Analysis

Kentucky established its Extended Weight Coal or Coal By-products Haul Road System (EWCHRS) in 1986. This road network includes segments of the Coal Haul Highway System on which more than 50,000 tons of coal or coal byproducts were transported by motor vehicles during the previous calendar year. In the last few years, the state has introduced extended weight (EW) systems that focus on petroleum products and metal commodities. Trucks can operate above posted weight limits on these networks by paying an annual fee ranging from $160 to 2,000. Although EW networks benefit motor carriers by improving efficiency, vehicles that exceed the weight limits can damage pavement and reduce its service life. This study examined the relationship between pavement age and condition on EWCHRS and non-EWCHRS routes to understand the implications of EW networks for infrastructure life-cycles. Using data from 2008 through 2020 (with the exception of 2016) to perform regression analysis, researchers found routes that spent at least 20% of the study period on the EWCHRS saw pavement life decline by 1.4 to 2 years over an average 12-year life cycle. In general, a positive correlation was observed between exposure on the EWCHRS and loss of pavement life, which can increase maintenance, repair, and rehabilitation costs.

Flexible Pavement Resiliency and Mitigation Strategies Following Adverse Environmental Events

Water accelerates flexible pavement damage in various ways, including subgrade and granular pavement materials strength reduction, frost heave, swelling of expansive subgrade soils, and asphalt binder stripping of aggregate in asphalt concrete (AC) mixes. This paper presents case studies of AC pavement sections in the U.S. that withstood high moisture events with limited effect on pavement performance and no immediate need for major maintenance or rehabilitation actions. The case studies comprise pavement sections in Florida and Louisiana that were subjected to catastrophic flooding events, a pavement section in Delaware that is constantly exposed to tidal water level variations, and pavement sections in Montana and Minnesota that experience seasonal freeze-thaw. These cases illustrated that flexible pavements could withstand climatic changes when appropriately designed and constructed. Mechanically or chemically stabilized subgrade, proper drainage, and/or minimizing plastic fines in unbound layers are all techniques that could effectively control moisture damage.

Incorporation of Pavement Dynamic Loading in Mechanistic–Empirical Design Frameworks

A framework for incorporating dynamic loading in pavement design is introduced. The framework is composed of five main steps. Firstly, input parameters, such as traffic characteristics, material properties, and the pavement configuration, are determined. Both primary and secondary data were obtained from the literature in this step. Secondly, a database of critical pavement responses, used by AASHTOWare software’s transfer functions to calculate pavement damage, was built using advanced pavement finite element models after the loading input was defined. The database includes various pavement configurations, axle configurations (single and tandem), pavement material properties, temperatures, loading conditions, and tire types (single steering, dual, and wide-base). AASHTOWare’s transfer function predicts pavement damage from pavement responses and is used to calculate the international roughness index (IRI) in the third and fourth steps, respectively. Finally, the load spectrum is adjusted to incorporate roughness-induced dynamic loading once the IRI exceeds 95 in./mi. A simple analytical dynamic-loading model was developed, based on mechanistic truck–trailer results. The model is a function of the tire configuration, speed, and IRI. Multiple case studies were performed, considering various pavement configurations, material properties, average annual daily traffic values, and dynamic-loading percentiles. Results showed that dynamic loading had the most significant impact on fatigue life, followed by rutting potential. The AASHTOWare’s transfer functions were found to be insensitive to an increase in loading when considering thick-pavement structures. The outcome of this effort was validated using the measured and predicted IRI from the Specific Pavement Study sections, which are part of the Long-Term Pavement Performance program.

Continuous remote monitoring of a motorway section using geophones

Pavement monitoring has been a widespread area of research. Monitoring pavements using embeddable sensors and a data acquisition system is an active research topic, with important questions about the selection of appropriate sensors and data acquisition systems, and interpretation of data. This work presents results of the instrumentation of a French motorway section, using geophones and temperature probes. The sensors are connected to a remote monitoring system which is compact, has a low power consumption, and real-time data processing capabilities. This system performs continuous monitoring, under a heavy motorway traffic and transmits the data wirelessly to a remote server. This work focuses on the treatment and analysis of the sensor signals. It indicates a good reliability of the system and embedded sensors, and presents different methods of interpretation of the data, to monitor pavement deflections, but also traffic composition and traffic loads, which could be used for pavement management, for detection of pavement damage and preventive maintenance.

Swelling of sulfate-bearing soil: A case study of A1 highway pavement failure

Lime stabilization is one of the most commonly used methods for improving weak road subgrades, but it becomes very problematic in sulfate-bearing soils. This article presents results of a case study in which the sulfate attack was the most probable reason for a significant deformation of a highway pavement founded on lime-stabilized Doggerian clay. Macroscopic analysis revealed partial degradation of the concrete structure, a larger extent of stabilization than designed, and large gypsum crystals in the native soil. Chemical tests indicated a very high content of sulfates in the soil (> 10,000 mg/kg of SO4) and in the groundwater (> 3,000 mg/l of SO42-). The maximum free swelling strain (εvolmax = 3–6%) and the swelling pressure (pcmax = 56 kPa) of the clay, determined in the Wasiliew apparatus and the oedometer tests, were not large enough to be considered as the only cause of the pavement damage. Sand-cement and clay-lime mixtures, which simulated stabilized soil layers in the road profile, were observed for several months in oedometric tests, under various loading conditions. The swelling of Sa-Cem was negligible, but the Cl-Lim mixtures turned out to be much more expansive than the natural soil (εvolmax up to 21%, pcmax = 230 kPa) with a much longer duration of volume increase. The samples were flooded with distilled water or hard groundwater to clarify whether the constant access of the material to sulfates is important in the swelling process, but the results were unequivocal. X-ray diffraction revealed that the addition of lime resulted in the formation of secondary minerals: butlerite, portlandite, and ettringite. The presence and swelling of ettringite in the stabilized sub-base soil layers was eventually agreed to be the most probable reason for the pavement deformation. These results indicate the need to introduce an obligatory investigation of the level of sulfates in clayey subbase soils subject to lime stabilization.

Evaluating the impact of overweight trucks on rigid pavement performance using AASHTOWare pavement ME design

ABSTRACT The purpose of this study is to evaluate the impact of overweight vehicles on pavement performance over its design life, with Jointed Plain Concrete Pavements (JPCPs) particularly considered for analysis. The Long-Term Pavement Performance database was used to provide information on pavement construction, age, overall structure, specific traffic inputs, and recorded historical pavement performance. AASHTOWare Pavement ME Design (PMED) was used to analyze data extracted from LTPP based on legislation by the Federal Highway Administration (FHWA) regulating gross vehicle weight (GVW) and single and tandem axle weights of overweight trucks (OWTs) to estimate pavement performance and service life. The overall findings indicated that although increasing overweight by 20% on single or tandem axles could considerably reduce pavement service life by up to 20% in the various scenarios examined in this study, OWT exceeding the GVW limit legally-allowed by FHWA could damage pavement more than OWT exceeding axle weight limits.

Evaluating pavement performance in bus rapid transit systems: Lessons from Seoul, South Korea

The proposed research aims to systematically investigate the condition of bus stops throughout Seoul, a densely populated metropolitan area. The study involves conducting a survey of multiple bus stops, taking into consideration traffic volume and pavement damage type. Field tests using the Falling Weight Deflectometer are conducted in heavily damaged bus stop sections to evaluate the performance of the surface, subbase, and subgrade layers. Additionally, this research proposes a polymer-modified stone matrix asphalt mixture with a nominal maximum size of 10 mm (10 mm PSMA) to enhance the rutting resistance of bus stop pavements while ensuring compliance with dynamic modulus acceptance values. Laboratory tests including Hamburg wheel tracking and dynamic modulus tests are conducted to achieve this objective. The results reveal that inadequate quality of the surface layer is the main cause of bus stop pavement failure in Seoul, with the average FWD modulus of the asphalt layer falling below the standard requirement of 725 ksi (28 % lower), indicating a 'low' structural condition based on evaluation standards. This can be attributed to inappropriate gradation and overload issues during rush hour. The dynamic modulus test results show that the PSMA 10 mm mixture outperforms the conventional 13 mm mixture in the low-frequency range and shows comparable performance in the high-frequency range. The Hamburg wheel tracking test indicates that all proposed mixtures have settlement values below 5.5 mm and do not exhibit the stripping point phenomenon, indicating excellent rutting resistance. Notably, the 10 mm PSMA mixture demonstrates superior permanent deformation resistance compared to the 13 mm PSMA and 13 mm PMA mixtures (1.26–185 times lower). Overall, the proposed mixtures exhibit promising performance in mitigating fast degradation at bus stops.

Automatic Detection of Road Cracks using EfficientNet with Residual U-Net-based Segmentation and YOLOv5-based Detection

The main factor affecting road performance is pavement damage. One of the difficulties in maintaining roads is pavement cracking. Credible and reliable inspection of heritage structural health relies heavily on crack detection on road surfaces. To achieve intelligent operation and maintenance, intelligent crack detection is essential to traffic safety. The detection of road pavement cracks using computer vision has gained popularity in recent years. Recent technological breakthroughs in general deep learning algorithms have resulted in improved results in the discipline of crack detection. In this paper, two techniques for object identification and segmentation are proposed. The EfficientNet with residual U-Net technique is suggested for segmentation, while the YOLO v5 algorithm is offered for crack detection. To correctly separate the pavement cracks, a crack segmentation network is used. Road crack identification and segmentation accuracy were enhanced by optimising the model's hyperparameters and increasing the feature extraction structure. The suggested algorithm's performance is compared to state-of-the-art algorithms. The suggested work achieves 99.35% accuracy.

Evaluation of Material Composition on the Shear Performance of Asphalt Mixtures under Different Test Conditions

Although shear strength plays an important role in the performance of asphalt mixtures, it is still not adopted as a control index in traditional asphalt pavement structure design. Among most shear strength tests, the shape of specimen damage in the uniaxial penetration test and circle shear test proved to be more accurate in reflecting the practical asphalt pavement damage shape. To explore the impact of material composition on the shear performance of asphalt mixtures under different test conditions, uniaxial penetration tests, circle shear tests, and unconfined compressive strength tests were conducted to evaluate shear strength with considerations of asphalt mixture composition (asphalt binders, aggregate, and mineral powder). Experimental results demonstrate that the SBS-modified asphalt mixtures have a higher shear strength than conventional 70# asphalt mixtures, and the shear performance of mixtures is positively correlated with softening point of asphalt binder. For the same gradation, the shear strength of asphalt mixtures increases with the asphalt-aggregate ratio first, then decreases with the ratio increases. The shear performance of mixtures can be increased by properly increasing the maximum nominal aggregate size and reasonably adjusting the aggregate gradation. Mineral powder replaced by 20% cement or 10% PSP (phosphorus slag powder) can also satisfy the requirement. Both coarse aggregate and fine aggregate containing silt impact the shear performance of mixtures; it is recommended that the silt content of coarse aggregates is controlled within 3%, and that of fine aggregate should be within 1%.

Characterization Of Soil For Road Shoulders Mixed With Reclaimed Asphalt Pavement Waste

The use of recovered asphalt pavement can reduce the amount of new bitumen and aggregates used in pavement construction and rehabilitation (RAP). RAP is a waste product that results from the removal of an old or damaged pavement surface. Although it has been used since the 1970s, and numerous recommendations for using RAP in the new mixture have been made, there are only a few research available. Because the materials used are recycled, it is also cost-effective and environmentally friendly. The purpose of this study is to characterize soil that has been blended with reclaimed asphalt pavement (RAP) for use on the local road shoulder. RAP is one of the rehabilitation procedures used to repair a deteriorated surface by removing the upper pavement and replacing it with new pavement. According to earlier study, mixing dirt with different materials improves the finding. The goals of this study are to establish the material qualities of soil and RAP, determine the optimum moisture content of material and degree of compaction for road shoulders, and determine the appropriate mix proportion of soil and RAP for road shoulders using the California Bearing Ratio (CBR) test. The Atterberg limit, liquid limit, and plastic limit for soil were evaluated on a sample of soil and RAP. For soil and RAP, a sieve analysis was performed. Compaction tests for soil and RAP were carried out with a mixture of 10S, 2S8RAP, 4S6RAP, 6S4RAP, 8S2RAP, and 10RAP, as well as CBR tests for soil and RAP. . According to the results of the laboratory test, 20% of RAP (8S2RAP) had a better-mixed proportion for the road shoulder. As a result, repurposed materials like RAP can be used as road shoulder material. 
 Keywords: reclaimed asphalt, reuse of Asphalt,

Wear Rate By Utilizing Asphalt Starbit PG 70 Due To Peat Water Soat

Asphalt Starbit PG 70 is a polymer (elastomer) based modified asphalt. The advantages of Asphalt Starbit PG 70 are resistant to water and weather, thus affecting the construction of highway infrastructure. Construction of highways in peat water areas, where the characteristics of peat water are that it contains high levels of organic matter and iron and has a low pH < 4 – 5, causing performance disturbances and causing pavement damage. The study aimed to analyze the wear value of Starbit asphalt due to peat water immersion for 30 minutes, 24 hours, and 48 hours. The research method is wear testing based on the 2018 Revision 2 Bina Marga Specifications. The results of soaking in peat water for 48 hours experienced an increase in weight loss of 7.37% due to having a high level of acidity which contains lignin compounds in organic matter. Meets the specifications of Highways 2018 Revision 2, namely for a weight loss value of ≤ 20%.

Road performance and thermal-insulation effect in construction of asphalt mixture containing phase-change thermal-insulation agent

Highway construction strategies impact the early damage of asphalt pavement and its sustainability. The pavement quality is directly influenced by the temperature of the asphalt mixture during paving and rolling, but delaying the cooling of the mixture at this stage has rarely been investigated. An asphalt mixture modified with a phase-change thermal-insulation agent (PCTIA) was prepared potentially improve the temperature of the asphalt mixture during construction. The high-temperature, low-temperature, and water-stability performance of asphalt mixtures with different PCTIA contents were analysed. The temperature regulation test was carried out, and the temperature field model of the asphalt pavement paving process was established. Based on the laboratory test and FEM models, the influence of different construction environmental conditions on the thermal-insulation performance of PCTIA and its thermal-insulation mechanism were analysed by using the distribution of temperature, average temperature, and equivalent temperature. The thermal-insulation effect of PCTIA was quantified by latent heat accumulated temperature value (LHATV). Results show that PCTIA improves the high and low-temperature performance of asphalt mixture, and has little effect on water stability. The best overall road performance is achieved at 2.5% PCTIA content. PCTIA can significantly reduce the cooling rate of asphalt mixture, and the distribution of temperature difference and time difference is obviously different in different positions of loose asphalt mixture layer and different external environments. Latent heat accumulated temperature value (LHATV) can characterize the phase-change insulation effect of PCTIA well. The wind speed increases and air temperature-bottom temperature decreases reduced LHATV indicating that too rapid cooling of asphalt mixture would weaken the thermal-insulation performance of PCTIA. Both average temperature and equivalent temperature can reflect the thermal-insulation effect of PCTIA, where the average temperature can be used to analyse the specific phase-change process of PCTIA in asphalt mixture and then analyse the insulation mechanism of PCTIA. Combining average temperature and equivalent temperature, PCTIA had the most application value in low temperature and low wind speed environments. This study aims to explore the effective method of delaying the cooling of high-temperature asphalt mixture from the perspective of phase-change energy storage and explain its mechanism.

Pavement surface defect recognition method based on vehicle system vibration data and feedforward neural network

ABSTRACT The pavement surface will inevitably develop cracks, subsidence, pits, ruts, and other defects due to the effects of traffic and the environment. In road engineering, vibration-based pavement condition monitoring has been widely adopted. Due to differences in vehicle dynamics, acquisition equipment, and road quality, the application of these techniques to the identification of minor pavement damage is limited. With the widespread adoption of the inertial navigation system (INS) in autonomous vehicles, INS-based pavement evaluation has emerged as a promising new technique. In this paper, the acceleration sensor, gyroscope, GPS, and other INS-integrated devices were used to collect data on vehicle attitude changes. For the detection of pavement apparent millimetre disease, a new method utilising INS data and machine learning was proposed. The method analyses the original vibration signal in time domain, determines the degree of sensing parameter influence, and extracts the index that can characterise the signal change. Multiple machine learning recognition models have been built to effectively classify road conditions, with the best-performing model achieving an F1 score of 99.61% and precision of 99.33%. The recall rate, accuracy rate, and F1 score of disease height classification were all above 0.7 on a macro and micro scale.