A self-compacting polyurethane mixture (SPUM) forms at ambient temperature without external vibration and features superior mechanical performance, which has bene proposed as a promising steel bridge deck pavement (SBDP) material. This study experimentally evaluated thermal properties of SPUM and numerically simulated heat transfer of SBDP paving with SPUM as surface layer under five climates . For a binder content of 11.5%-13.5%, the pebble-based SPUM had a thermal conductivity of 1.503-1.460 W/(m K) and a solar reflectance of 19.6%-24.1%, while basalt-based SPUM had thermal conductivity of 1.148-1.158 W/(m K) and solar reflectance of 16.5%-17.3%. Compared to SBDP with SMA, the utilization of B135 (basalt-based SPUM with 13.5% binder) reduced the temperature by 1.6 degrees C in winter, and the paved P115 (pebble-based SPUM with 11.5% binder) lowered it by 4.0 degrees C in summer. SBDP with SPUM presented lower temperature difference between pavement surface and steel deck by 1.5 degrees C-2.0 degrees C in summer and 0.9 degrees C-1.1 degrees C in winter. Reducing thickness of SPUM from 3.5 to 1.0 cm decreased the vertical temperature gradient of SBDP by up to 10.2 K/m in summer and 4.5 K/m in winter. The results demonstrate that the utilization of SPUM is effective in mitigating thermal gradients of SBDP, which offers the potential for design of thermally resilient steel bridge deck pavement with thin layers.

Addressing the critical need for agile and comprehensive road assessment, this study introduces an intelligent monitoring system for pavement skid resistance based on a Robot Operating System (ROS) mobile platform. The system is engineered to overcome the inefficiencies and discrete nature of traditional point-based methods by enabling continuous, "line-based" dynamic friction profiling. Built upon an Ackermann-steering robot, the system integrates a high-precision force-sensing system and standard rubber sliders to dynamically calculate the friction coefficient from real-time horizontal and vertical force data. Field experiments conducted under a low load (14.7 N) and across a velocity gradient from 0.5 to 1.5 m/s demonstrated the device's stability and revealed a pavement-specific velocity dependence: on the smooth surface, the friction coefficient monotonically decreased with increasing velocity, aligning with classical theory, whereas on the asphalt pavement, it exhibited a complex and non-monotonic behavior, featuring a distinct peak at 1.0 m/s that experimentally highlights the friction-enhancement effect of dynamic vibrations at low speeds. Validation against conventional methods (BPN) confirmed a highly consistent overall trend. The primary contribution of this work lies in its ability to capture high-resolution, non-monotonic friction curves, providing indispensable physical insights for refining low-speed friction models. This automated and efficient approach offers an innovative paradigm for intelligent road inspection.

ABSTRACT Real-time sensing of the evolution of multi-source parameters within pavement structures is essential for effective construction, performance evaluation, and maintenance. This study develops a wireless smart aggregate (SA) sensor for real-time multiparameter monitoring within pavement mixtures. A strain transfer model for the sensing layer–encapsulation–medium system was derived, validated via finite element simulations, and applied to optimize the encapsulation design. The SA integrates strain gauges, accelerometers, magnetometers, and thermometers to capture stress, acceleration, rotation, and temperature simultaneously. Laboratory calibration verified that the sensor achieves high measurement accuracy, and gyratory compaction tests in cement-stabilized macadam demonstrated the sensor’s capability to track compaction-induced mechanical and kinematic changes. To mitigate material heterogeneity effects, a volatility index (Vt) combining stress and rotation variations was introduced, showing a strong correlation with compaction evolution (Pearson correlation coefficient ≥ 0.89). The results confirm the SA’s potential for multi-source in situ monitoring, offering an effective tool for pavement health assessment and intelligent construction control.

Geopolymers, as advanced low-carbon binder materials, require a fundamental understanding of the composition-structure-mechanical property relationships in sodium aluminosilicate hydrate (N-A-S-H) gels for performance optimization. This study investigates the effects of varying Si/Al, Na/Al, and H2O/Na ratios on the atomic structure, dynamic behavior, and mechanical properties of N-A-S-H gels using molecular dynamics (MD) simulations with a non-reactive force field. Across all investigated compositions, the elastic modulus ranges from 59.23 to 92.84 GPa, while the uniaxial tensile strength varies from 2.90 to 4.95 GPa. Results indicate that an increased Si/Al ratio promotes the formation of stronger Si-O-Si bonds, enhancing crosslink density and structural rigidity, and consequently improving tensile strength and elastic modulus; specifically, increasing Si/ Al from 1.0 to 3.0 increases the elastic modulus from 59.23 to 92.84 GPa and the tensile strength from 2.90 to 4.88 GPa. Conversely, a higher H2O/Na ratio weakens the gel network by disrupting structural integrity through intensified hydration effects, resulting in reduced mechanical performance; increasing H2O/Na from 1.0 to 3.0 decreases the elastic modulus from 85.77 to 65.42 GPa and the tensile strength from 4.95 to 3.44 GPa. Additionally, an elevated Na/Al ratio promotes the conversion of bridging oxygen (BO) atoms to non-bridging oxygen (NBO) atoms, disrupting network connectivity and reducing structural polymerization, and thus impairing the mechanical properties; increasing Na/Al from 0.8 to 1.2 decreases the elastic modulus from 84.21 to 75.24 GPa and the tensile strength from 4.48 to 3.63 GPa. These findings provide essential atomic-scale insights for rational geopolymer design, supporting sustainable and high-performance binder development.

CMV and ks are the two most commonly used Intelligent Compaction Meter Values (ICMVs) during intelligent compaction of the cement-treated subbase layer. However, the CMV is susceptible to subgrade stiffness and roller operation modes, leading to low accuracy in evaluating the compaction quality of cement-treated subbase. ks is a stiffness ICMV, which is suitable for the single-layer subgrade structure at the completion stage of compaction and is not accurate for cement-treated subbase layers. In addition, the current ks calculation method cannot make full use of data from acquisitions, which makes high standard deviations in ks calculation. To address these issues, this paper firstly proposes a new ICMV kos derived from the governing equations of multi-layer dynamical model with three degrees of freedom, in which the potential influence from elastic-plastic behaviour of the cement-treated subbase layer and the elastic behaviour of the subgrade layer is successfully taken into consideration. Secondly, new acquisition methods for the displacement signal and the contact force signal are proposed for the field application of kos. Field validation results show that the determination coefficient between kos and compaction degree (R2 = 0.88) is significantly higher than that between CMV and compaction degree (R2 = 0.08) and that between ks and compaction degree (R2 = 0.68), indicating high accuracy and stability of kos.

Smart tires (STs) are transforming vehicle–road communication within Intelligent Transport Systems (ITS) by serving as mobile measurement platforms. This paper presents a comprehensive review of how advanced sensor integration, including multi-physical sensors and Vehicle-to-Everything (V2X) communication, is enhancing smart tire technology. We conceptualize smart tires not just as passive vehicle components, but as active sensing platforms that collect and transmit real-time data on vehicle dynamics and road conditions. The review explores key applications of ST, such as active vehicle control, infrastructure maintenance, and road condition assessment, while also highlighting the potential of STs in autonomous vehicle systems and sustainable transportation frameworks. Moreover, the review discusses challenges, future directions, and the integration of digital twin technologies for improving Vehicle-to-Infrastructure (V2I) communication and operational efficiency. The findings provide a comprehensive understanding of the evolving role of ST in creating safer, more efficient, and sustainable road transport systems.

AI algorithms have been found used in the modulus prediction task during Intelligent Compaction (IC) of asphalt pavement. Current prediction models have preliminarily established the black-box model relationship between the outputs (e.g. modulus of asphalt layer) and the inputs including roller dynamical parameters (e.g. acceleration, excitation frequency and amplitude) and pavement structural parameters (e.g. number and modulus of underlying layers). However, there is a lack of explicit mapping relationship between them, leading to poor generalization of current prediction models under varied field conditions. This work proposed a two-staged cascaded mechanistic data-driven model. In that, mechanical constraint is introduced in loss function for supplementing explicit mapping relationship to avoid unreasonable prediction results that significantly violate vibration mechanics. Moreover, a two-staged cascaded model structure is designed for introducing additional structural constraints to solve the multi-solution issue that the contribution of each structural layer to the dynamic response of vibrating system are coupled with each other encountered in the multi-layered pavement system. The prediction results show that, compared with the existing data-driven models, mostly adopting Huber loss function and end-to-end model structure, the proposed mechanistic data-driven model achieves superior performance.

The present study aims at experimentally and numerically investigating the effect of an encapsulated healing agent on the mechanical characteristics of a stone mastic asphalt (SMA-15) after cyclic loading. As healing agents a thiol-containing urethane AR-polymer (ARP) and a sunflower oil (SfO) are used. The comparison of self-healing results in asphalts show that the use of encapsulated ARP allows to restore strength and stiffness up to 93

With the growing popularity of electric vehicles, the demand for robot-based unmanned automatic charging has become both urgent and challenging. Two key challenges need to be addressed: how to efficiently locate the charging port, and how to compliantly insert the connector into the port. In this paper, we propose an incremental learning method based on the broad learning system to address the visual positioning error of the charging port. This method allows the robot to transfer and generalize the search skills learned in simulation to real-world scenarios. As a result, the robot can rapidly locate the charging port in real-world environments without the need for complex contact state modeling, time-consuming data collection, or model retraining. Subsequently, a biomimetic admittance controller is designed to enable the robot to adapt its compliant behavior online during the plugging process. Finally, experiments are performed on a UR robot to verify the effectiveness of our method.

The continuous development of autonomous driving imposes increasing requirements on both vehicles and road infrastructure. However, traditional road marking materials cannot fully meet the demands of integrated sensing. To meet the perception needs of autonomous vehicles, this study proposes a Metal Modified Road Marking (MMRM) material solution based on broadband electromagnetic wave reflection. Due to the excellent reflection performance of aluminum beads in dual-band (380-780 nm and 76-81 GHz), the pre-mixed glass beads are partially replaced with aluminum beads to ensure the visibility of the road markings while enhancing their metallic reflective properties at the same time. As such, synergistic detection by camera and radar can be achieved via cooperative electromagnetic reflection across a broad spectrum. Performance tests were conducted to verify the stability of MMRMs, and the following conclusions are drawn: (1) MMRMs exhibit excellent reflection performance in dual-band. The retroreflective coefficient of MMRMs exceed 150 mcd/m2/lx, meeting the standards for new road markings. MMRMs with 40 % aluminum beads by volume exhibit optimal reflective properties, with the Radar Cross-Section (RCS) value fluctuating around -1 dB. (2) The coupling effect between adjacent aluminum beads reduces far-field reflectivity, leading to a nonlinear relationship between bead content and reflection performance, with RCS decreasing by 15 dB at 80 % bead content. (3) The cross-linked network structure of MMA enhances MMRMs' durability and stability, maintaining reflection performance under corrosion, temperature, and mechanical wear tests, ensuring their suitability for cooperative perception. (4) The penetration capability of millimeter waves allows MMRMs to remain effective under various interference conditions, meeting the requirements for all-weather road marking detection in autonomous driving.

Fog imposes adverse effect on driving safety. Traditional visibility measurement methods are expensive and limited to a short distance along the roadway. This study aims to identify visibility levels from foggy road images with deep learning methods. To address the shortage of foggy road image data set, a novel method is proposed to generate synthetic fog images based on point cloud and red, blue, & green (RGB) images. A synthetic foggy roadway image data set, kitti-foggy, containing 10,034 images was created with data from the kitti data set. Performance of the proposed method was compared with the traditional stereo-based method. Three typical image classification convolutional neural networks, including ResNet34, ResNet101, and Inception V4, were used to train the data set, and several evaluation matrices were used to evaluate their performances. The proposed method outputs more natural and authentic fog images. ResNet34 demonstrated the best performance among three algorithms with an overall accuracy of about 93%. Real data from a driving recorder and drones was used to verify the capability of ResNet34 to detect real fog. Findings of this study assist in the field of autonomous driving as well as intelligent transportation.

The efficient reuse of construction waste plays a vital role in fostering eco-friendly practices in the construction industry. However, the efficient application of recycled powder remains a significant challenge due to its low reactivity. In response to this issue, this study investigates the combination of recycled powder with slag to create a recycled powder-slag-based geopolymers (RPSG) through alkaline activation. To assess the performance of RPSG in erosion-prone environments, the effects of various sulfate types, concentrations, and erosion methods on its sulfate resistance were systematically evaluated. Advanced techniques, including SEM, XRD, FTIR, and MIP, were employed to analyze the microstructural changes. The results show that sulfate-induced expansive products initially improve, but eventually degrade, the performance of RPSG mortar. The stronger reactivity of Mg2 + causes decalcification, making the erosion effect of MgSO4 solution more severe. The wet-dry cycling process accelerates sulfate penetration during the wet phase, leading to faster generation of erosion products, while the crystallization of salts and shrinkage during the drying phase causes more significant degradation. These findings provide valuable theoretical insights for applying RPSG in sulfate-rich environments and serve as a practical reference for improving the durability design of geopolymer-based materials in engineering applications.

Traditional ultra-thin asphalt wearing course designs often oversimplify wheel loads as uniform pressures, neglecting critical non-uniform effects. This study establishes a 3D finite element model incorporating realistic non-uniform tire loading to reveal its mechanistic influence on pavement responses. Results demonstrate that non-uniform loading significantly alters stress states in ultra-thin layers, substantially elevating critical stresses compared to uniform assumptions. A novel Non-uniform Load Influence Factor (NLIF) accounting for thickness effects is developed to quantify these deviations. The analysis provides a foundation for revising material strength specifications and fatigue design criteria, contributing to improved performance and durability of ultra-thin pavement systems.

The dynamic response of vibrating roller is crucial for Intelligent Compaction (IC) technology. Due to the lack of suitable dynamical models, effective control for rocking motion cannot be achieved, which leads to inadequate compaction quality for asphalt pavement. In this work, a new roller-asphalt pavement rocking model with two-layer system is proposed. The influence on rocking motion by roller design, excitation, and pavement structural parameters are studied. The correlations between stiffness coefficient ks and compaction degree are obtained in field tests, to investigate the influence of rocking motion on Intelligent Compaction Measurement Values (ICMVs). The results suggest that the roller design parameters and pavement structural parameters notably influence the rocking motion, whereas roller excitation parameters have minimal impact. The correlations between compaction degree and ks at both ends of the roller are found weaker than that at the center of gravity of the roller, which proves that ICMVs is affected by rocking motion.

Geopolymer is promising to replace cement, thus reducing the CO2 emissions of concrete production. However, the brittle behavior of geopolymer under bending loads limits its engineering applications. This work adopted epoxy resin and polyurethane modified epoxy resin (PMER) to synthesize the epoxy resin-PMER (EP) emulsion, which was further added into metakaolin-based geopolymer matrix for toughening. Moreover, the co-toughening mechanism of epoxy resin and polyurethane on metakaolin-based geopolymer was revealed based on the changes of major chemical bonds and Si binding energy of geopolymer before and after modification. The results show that the introduction of PMER does contribute to improve the flexural strength of geopolymer matrix by enhancing EP toughness. However, PMER also plays an adverse role in the Si-O-C bond generation of EP-modified metakaolin-based geopolymer (EMG), thus weakening the interconnections between 3D mesh structures inside EMG. A balance between these two contrasting effects is achieved when PMER content within EP reaches 60wt.%. At this time, the addition of 10wt.% EP can increase the flexural strength of metakaolin-based geopolymer matrix by 2.6 times.

The pavement skid resistance is critical to the driving safety of a road. Automated vehicles, which have high requirements for environmental perception, are not yet commonly equipped with an estimation module that could monitor the real-time skid resistance, leading thus to instability in the steering and braking control. In this paper, we propose a data-driven approach incorporating vehicle dynamics implementing real-time pavement skid resistance estimation for autonomous driving. Feature data is collected through the automated vehicle dynamics calibration process, and the selection of features is essentially based on the use of the vehicle dynamics model. To extract the features of one-dimensional timing signals, a temporal one-dimensional convolutional neural network is proposed to improve the stability and robustness of the estimation and minimise the effect of small perturbations and noises. In addition, given the insufficient data that may be faced in practical situations, a transfer learning approach is further proposed to improve the model generalisation. Training data is collected in an autonomous driving simulation platform and the effectiveness of the method is then verified based on this platform. Satisfactory performance is demonstrated for different driving scenarios and under varied road conditions.

CO2 emissions during airport pavement induction heating for ice and snow melting are primarily determined by the electrothermal behavior of electrically conductive layer (ECL). A single type of conductive filler is difficult to significantly improve the electrothermal performance of ECL while maintaining its mechanical properties. This work introduces nano carbon black (NCB) into carbon fiber-reinforced geopolymer to prepare NCB-carbon fiber co-modified geopolymer composites (NMGC) with great potential to dramatically improve the electrothermal performance of ECL and reduce CO2 emissions during induction heating. Furthermore, in order to provide guidance support for the ECL design, the evolution mechanisms of the electrothermal and mechanical properties of NMGC were inquired through electrochemical impedance spectroscopy (EIS) analysis and geopolymerization rate characterization, respectively. Results show that introducing 1 wt% NCB into the carbon fiber-reinforced geopolymer can effectively increase the reaction rate within NMGC and improve its electrothermal performance due to the enhanced electron transition. Indoor induction heating tests further indicate that the optimized NMGC can reduce the CO2 emissions by 13.7 kg/m2 comparing to the NMGC without NCB during 30 min of induction heating.

The integrity of road markings has a significant impact on driving safety, especially in the emerging automated driving scenario. Road markings are susceptible to wearing damage; it is therefore important for road agencies to examine and maintain markings periodically. This study presents a novel method for detection of road marking defect via an unmanned aerial vehicle (UAV)-laser radar (LiDAR) platform. The key idea is to evaluate the damage rate of road markings by point reflectivity. The method consists of three major steps: point cloud registration, marking segmentation, and defective marking detection. A high-intensity prioritized raster method is proposed to extract whole marking regions within defect area, and the damage rate of markings is defined to evaluate damage severity. Field test results show that the precision and recall of marking extraction are over 90%, and the precision of marking defect detection is over 93%. The failed detections were attributed to subjectivity of ground truth and neglect of small defects.

Permanent deformation in ballast layers is a major contributing factor to the railway track geometry deterioration. In spite of a considerable amount of research on understanding and predicting performance of ballast layers, accurately capturing their settlements remains a challenge. In order to contribute to solving this important issue, a new numerical method for predicting ballast settlements is presented in this paper. This method is based on the finite element (FE) method combined with a constitutive model that captures permanent deformation accumulation in unbound materials under cyclic loading. This allows predicting permanent deformations of large structures and at large number of load cycles in a computationally efficient manner. The developed constitutive model is validated based on triaxial test measurements over wide range of loading conditions. Stress state in ballast layers has been examined with a 3D FE model, for several embankment structures and traffic load magnitudes. The determined stress distributions and loading frequencies were used as an input of the constitutive model to evaluate permanent strains and settlements of ballast layer. The influence of embankment structural designs and traffic loading magnitudes on the ballast layers settlements is examined and the results obtained are compared with the existing empirical performance models.

Compaction construction is one of the most critical factors affecting road performance and durability. To improve the construction quality and service life of road, intelligent compaction (IC) technology has received continuous attention. In contrast to traditional compaction construction, IC improves the compaction quality and uniformity of each road layer through real-time monitoring and adjustment of the compaction process. Intelligent compaction measurement values (ICMVs) were proposed to provide quantitative indexes for compaction quality monitoring and basis for dynamic regulation of compaction conditions. However, the current ICMVs have problems such as low accuracy and stability, and poor applicability under different working conditions, which limit the application of IC technology. To address these shortcomings, this paper conducts a state-of-the-art review of ICMVs, outlining the past research achievements and discussing future development directions. In this paper, the vibrating dynamical models of compaction was firstly summarized and the future research directions are proposed. Next, the existing calculation models of ICMVs are systematically investigated and the significant factors that affected ICVMs calculations are summarized. Solutions to eliminating influences from those factors are discussed in further, e.g., by a combined effort in improving calculations models for ICMVs and advancing the multiple regression analysis method.