Search Results (60)

Identifying road surface cracks by semantic segmentation is a difficult problem. This is because segmentation typically detects objects by area, whereas cracks are string-like. Conventional loss functions such as Binary Cross-Entropy (BCE), Dice, and IoU often fail to capture the fine, elongated features of cracks, as they rely on pixel-level, area-based overlap, leading to suboptimal performance. To address this, we investigate one of the skeleton-based losses, the Centerline Dice (clDice) loss, which emphasizes the preservation of tubular structures via soft skeletonization. We improve road crack segmentation by combining clDice with conventional loss functions, systematically evaluating its role by varying the weight parameter and skeletonization iterations. Experiments are conducted on the EdmCrack600 and CrackForest datasets using two segmentation models: a customized CNN-based U-Net++ and a transformer-based SegFormer. Performance is evaluated using the Dice coefficient, IoU, clDice, and Hausdorff Distance. Results show that combining clDice and IoU loss with customized U-Net++ achieves superior performance. Compared to a standard BCE baseline, it improves the Dice coefficient by 4.9 and 2.8 percentage points on EdmCrack600 and CrackForest and improves the clDice score by 3.9 and 1.7 percentage points. These results highlight improved segmentation of thin, linear cracks, supporting practical advancements in road monitoring and segmentation of linear structures. Full article

Current research on heritage corridors predominantly focuses on linear heritage in Europe and America, while studies in Asia urgently need to be expanded. This study investigates China’s linear heritage. Based on the minimum cumulative resistance (MCR) model, it conducts heritage corridor construction for the Henan section of the Sui–Tang Grand Canal, and reveals the following: (1) A total of 252 heritage sites were classified into three categories: canal hydraulic heritage (13.5%), canal settlement heritage (21.4%) and related heritage (65.1%), exhibiting a “local clustering under global dispersion” pattern with a core–secondary–edge structure. (2) The influence of natural–social resistance factors was ranked as follows: elevation > roads > land use > slope. Interwoven corridors were simulated by GIS and optimized to four primary corridors with multiple secondary corridors. (3) The transverse zone of the primary corridors was stratified into core area (0–10 km from the centerline), buffer area (10–25 km), and influence area (>25 km) with a total width of 25–30 km. The longitudinal section was partitioned into four subsections based on hydrological continuity and heritage density. Then, a tripartite conservation framework characterized by “heritage clusters–holistic corridor–transverse stratification and longitudinal section” was proposed. It aimed to provide insights into methodologies and content structuring for transnational linear heritage (e.g., the Silk Road and the Inca Trail). Full article

To reduce the weight of prefabricated cap beams, a new type of hybrid pier with a steel cap beam and single concrete column with an innovative flange–rebar–ultra-high-performance concrete (UHPC) connection structure is proposed in this paper. Focusing on the static performance of hybrid piers, a specimen with a geometric similarity ratio of 1:4 was fabricated for testing. The results showed that the ultimate load-bearing capacity reached 960 kN, and the failure mode was characterized by an obvious overall vertical displacement of 70.2 mm at the cantilever end, accompanied by local buckling in the webs between transversal diaphragms and ribs. Due to the varying-thickness design, longitudinal strains were comparable between the middle section (thin plates) and the root section (thick plates) of the cantilever beam, showing a trend of an initial increase followed by a decrease from the end of the cantilever beam to the road centerline. Meanwhile, the cross-sections of the connection joint and concrete column transformed from overall compression to eccentric compression during the test. At the ultimate state, their steel structures remained elastic, with no obvious damage in the concrete or UHPC, verifying good load-bearing capacity. Furthermore, the finite element analysis showed the new connection joint and construction method of hinged-to-rigid could reduce the column top concrete compressive stress by 18–54%, tensile stress by 11–68%, and steel cap beam Mises stress by 10%. Finally, based on the experimental and numerical studies, the safety reserve coefficient of the new hybrid pier was over 2.7. Full article

The complex climate in permafrost regions poses severe challenges to infrastructure, and freeze-thaw cycles accelerate the deformation and damage of road embankments. Conventional thermosyphon technology, though effective in lowering permafrost temperatures, has a limited range of effect, making it hard to meet the demand for large-scale temperature regulation. This paper proposes a V-shaped transverse thermosyphon design with bidirectional heat conduction. It connects at the embankment centerline and transversely penetrates the entire cross-section to expand the temperature regulation range. Using a hydro-thermal coupling model, the temperature regulation effects of vertical, inclined, and V-shaped thermosyphons were calculated. Results show that the V-shaped design outperforms the other two in temperature control across different embankment areas. Transverse temperature analysis indicates uniform cooling around the embankment center, while depth temperature analysis reveals more stable temperature control with lower and less fluctuating temperatures at greater depths. Long-term temperature analysis demonstrates superior annual temperature regulation, providing consistent cooling. This research offers a scientific basis for embankment temperature regulation design in permafrost regions and is crucial for ensuring long-term embankment stability and safety. Full article

Frost heave in seasonally frozen soil surrounding natural gas pipelines (NGPs) can cause severe damage to adjacent infrastructure, including road surfaces and buildings. Based on the stratigraphic characteristics of seasonal frozen soil in Beijing, a soil–natural gas pipeline–heat pipe heat transfer model was developed to investigate the mitigation effect of the soil-freezing phenomenon by transferring shallow geothermal energy utilizing heat pipes. Results reveal that heat pipe configurations (distance, inclination angle, etc.) significantly affect soil temperature distribution and the soil frost heave mitigation effect. When the distance between the heat pipe wall and the NGP wall reaches 200 mm, or when the inclined angle between the heat pipe axis and the model centerline is 15°, the soil temperature above the NGP increases by 9.7 K and 17.7 K, respectively, demonstrating effective mitigation of the soil frost heave problem. In the range of 2500–40,000 W/(m·K), the thermal conductivity of heat pipes substantially impacts heat transfer efficiency, but the efficiency improvement plateaus beyond 20,000 W/(m·K). Furthermore, adding fins to the heat pipe condensation sections elevates local soil temperature peaks above the NGP to 274.2 K, which is 5.5 K higher than that without fins, indicating enhanced heat transfer performance. These findings show that utilizing heat pipes to transfer shallow geothermal energy can significantly raise soil temperatures above the NGP and effectively mitigate the soil frost heave problem, providing theoretical support for the practical applications of heat pipes in soil frost heave management. Full article

Maps of road layouts play an essential role in autonomous driving, and it is often advantageous to represent them in a compact form, using a sparse set of surveyed points of the lane boundaries. While lane centerlines are valuable references in the prediction and planning of trajectories, most centerline extraction methods only achieve satisfactory accuracy with high computational cost and limited performance in sparsely described scenarios. This paper explores the problem of centerline extraction based on a sparse set of border points, evaluating the performance of different approaches on both a self-created and a public dataset, and proposing a novel method to extract the lane centerline by searching and linking the internal maximal circles along the lane. Compared with other centerline extraction methods producing similar numbers of center points, the proposed approach is significantly more accurate: in our experiments, based on a self-created dataset of road layouts, it achieves a max deviation below 0.15 m and an overall RMSE less than 0.01 m, against the respective values of 1.7 m and 0.35 m for a popular approach based on Voronoi tessellation, and 1 m and 0.25 m for an alternative approach based on distance transform. Full article

Agricultural navigation, as an essential part of smart agriculture, is a crucial step in realizing intelligence and, compared with the structured features of urban roads, such as lane-keeping lines, traffic guidance lines, etc., the field environment is more complex. Especially at agricultural intersections, traditional navigation line extraction algorithms make it difficult to achieve the automatic prediction of multiple road navigation lines due to complex unstructured features such as weeds and trees. Therefore, this study proposed a field road navigation line prediction method based on an improved context guided network (CGNet), which can quickly, stably, and accurately detect intersection fields and promptly predict navigation lines for two different directional paths at intersections. Firstly, CGNet will be used to learn the local features of intersections and the joint features of video frames before and after the surrounding environment. Then, the CGNet with a self-attention block module is proposed by adding the self-attention mechanism to improve the semantic segmentation accuracy of CGNet in field road scenes, and the detection speed is not significantly reduced. The semantic segmentation accuracy mIoU is 0.89, and the processing speed is 104 FPS. Subsequently, a field road centerline extraction algorithm is proposed based on the partitioning idea, which can accurately obtain the centerlines of road intersections in the image. The average lateral deviation of each centerline is less than 4%. This study achieved the prediction of intersection navigation lines in mountainous field road scenes, which can provide technical support for field operation road planning of agricultural equipment such as plant protection and harvesting. At the same time, the research findings provide theoretical references for sustainable agricultural development. Full article

This study proposes and establishes a roadway longitudinal profile weighting model and innovatively develops a process and method for evaluating road surface roughness. Initially, the Gaussian model is employed to accurately fit the distribution frequency of vehicle centerlines recorded in British Standard BS 5400-10, and a generalized lateral distribution model of wheel trajectories is further derived. Corresponding model parameters are suggested for different types of lanes in this study. Subsequently, based on the proposed distribution model, a longitudinal profile weighting model for lanes is constructed. After adjusting the elevation of the cross-section, the equivalent longitudinal elevation of the roadway is calculated. Furthermore, this study presents a new indicator and method for assessing the roughness of the entire road surface, which comprehensively considers the elevations of all longitudinal profiles within the lane. To validate the effectiveness of the proposed new method and indicator, a comparative test was conducted using a vehicle-mounted profiler and a three-dimensional measurement system. The experimental results demonstrate significant improvements in measurement repeatability and scientific rigor, offering a new perspective and evaluation strategy for road performance assessment. Full article

In complex urban road conditions, stationary or illegally parked vehicles present a considerable risk to the overall traffic system. In safety-critical applications like autonomous driving, the detection of stopped vehicles is of utmost importance. Previous methods for detecting stopped vehicles have been designed for stationary viewpoints, such as security cameras, which consistently monitor fixed locations. However, these methods for detecting stopped vehicles based on stationary views cannot address blind spots and are not applicable from driving vehicles. To address these limitations, we propose a novel deep learning-based framework for detecting stopped vehicles in dynamic environments, particularly those recorded by dashcams. The proposed framework integrates a deep learning-based object detector and tracker, along with movement estimation using the dense optical flow method. We also introduced additional centerline detection and inter-vehicle distance measurement. The experimental results demonstrate that the proposed framework can effectively identify stopped vehicles under real-world road conditions. Full article

Promoting the deployment and application of embedded systems in complex forest scenarios is an inevitable developmental trend in advanced intelligent forestry equipment. Unstructured roads, which lack effective artificial traffic signs and reference objects, pose significant challenges for driverless technology in forest scenarios, owing to their high nonlinearity and uncertainty. In this research, an unstructured road parameterization construction method, “DeepLab-Road”, based on tight coupling of multisource information is proposed, which aims to provide a new segmented architecture scheme for the embedded deployment of a forestry engineering vehicle driving assistance system. DeepLab-Road utilizes MobileNetV2 as the backbone network that improves the completeness of feature extraction through the inverse residual strategy. Then, it integrates pluggable modules including DenseASPP and strip-pooling mechanisms. They can connect the dilated convolutions in a denser manner to improve feature resolution without significantly increasing the model size. The boundary pixel tensor expansion is then completed through a cascade of two-dimensional Lidar point cloud information. Combined with the coordinate transformation, a quasi-structured road parameterization model in the vehicle coordinate system is established. The strategy is trained on a self-built Unstructured Road Scene Dataset and transplanted into our intelligent experimental platform to verify its effectiveness. Experimental results show that the system can meet real-time data processing requirements (≥12 frames/s) under low-speed conditions (≤1.5 m/s). For the trackable road centerline, the average matching error between the image and the Lidar was 0.11 m. This study offers valuable technical support for the rejection of satellite signals and autonomous navigation in unstructured environments devoid of high-precision maps, such as forest product transportation, agricultural and forestry management, autonomous inspection and spraying, nursery stock harvesting, skidding, and transportation. Full article

The safe and efficient operation of highways minimizes the environmental impact, reduces accidents, and promotes the reliability of the transportation infrastructure, all in support of sustainable transportation. The horizontal alignment of highways holds particular importance as it directly impacts driver behavior, vehicle stability, and overall road safety. In many cases, highway inventory data held by infrastructure operators may contain inaccurate or outdated information. The accuracy of the variables used in crash prediction models eliminates possible bias in the variable estimators. This research proposes a methodology to obtain accurate horizontal geometric features from digital imagery based on the analysis of the planimetry, feature geolocation and centerline azimuth sequence. The reliability of the method is verified by means of numerical and statistical procedures. This methodology is applied to 150 km of motorway segments in Portugal. Although it is found that the geometric characteristics of most of the inventory segments closely matched the extracted alignments, very significant differences are found in some sections. The results of the proposed procedure are illustrated with several examples. Finally, the propagation of error in the determination of the geometric design independent variables in the fitting of the statistical models is discussed based on the results. Full article

Traditional DeepLabV3+ image semantic segmentation methods face challenges in pitaya orchard environments characterized by multiple interference factors, complex image backgrounds, high computational complexity, and extensive memory consumption. This paper introduces an improved visual navigation path recognition method for pitaya orchards. Initially, DeepLabV3+ utilizes a lightweight MobileNetV2 as its primary feature extraction backbone, which is augmented with a Pyramid Split Attention (PSA) module placed after the Atrous Spatial Pyramid Pooling (ASPP) module. This improvement enhances the spatial feature representation of feature maps, thereby sharpening the segmentation boundaries. Additionally, an Efficient Channel Attention Network (ECANet) mechanism is integrated with the lower-level features of MobileNetV2 to reduce computational complexity and refine the clarity of target boundaries. The paper also designs a navigation path extraction algorithm, which fits the road mask regions segmented by the model to achieve precise navigation path recognition. Experimental findings show that the enhanced DeepLabV3+ model achieved a Mean Intersection over Union (MIoU) and average pixel accuracy of 95.79% and 97.81%, respectively. These figures represent increases of 0.59 and 0.41 percentage points when contrasted with the original model. Furthermore, the model’s memory consumption is reduced by 85.64%, 84.70%, and 85.06% when contrasted with the Pyramid Scene Parsing Network (PSPNet), U-Net, and Fully Convolutional Network (FCN) models, respectively. This reduction makes the proposed model more efficient while maintaining high segmentation accuracy, thus supporting enhanced operational efficiency in practical applications. The test results for navigation path recognition accuracy reveal that the angle error between the navigation centerline extracted using the least squares method and the manually fitted centerline is less than 5°. Additionally, the average deviation between the road centerlines extracted under three different lighting conditions and the actual road centerline is only 2.66 pixels, with an average image recognition time of 0.10 s. This performance suggests that the study can provide an effective reference for visual navigation in smart agriculture. Full article

Herein, to address the challenges faced by Automatic Guided Vehicles (AGVs) in construction site environments, including heavy vehicle loads, extensive road search areas, and randomly distributed obstacles, this paper presents a hierarchical trajectory planning algorithm that combines coarse planning and precise planning. In the first-level coarse planning, lateral and longitudinal sampling is performed based on road environment constraints. A multi-criteria cost function is designed, taking into account factors such as deviation from the road centerline, shortest path cost, and obstacle collision safety cost. An efficient dynamic programming algorithm is used to obtain the optimal path. Considering nonholonomic constraints of vehicles, eliminating inflection points using improved B-Spline path fitting, and a quadratic programming algorithm is proposed to enhance path smoothness, completing the coarse planning algorithm. In the second-level precise planning, the coarse planning path is used as a reference line, and small-range sampling is conducted based on AGV motion constraints, including lateral displacement and longitudinal velocity. Lateral and longitudinal polynomials are constructed. To address the impact of randomly appearing obstacles on vehicle stability and safety, an evaluation function is designed, considering factors such as jerk and acceleration. The optimal trajectory is determined through collision detection, ensuring both safe obstacle avoidance and AGV smoothness. Experimental results demonstrate the effectiveness of this method in solving the path planning challenges faced by AGVs in construction site environments characterized by heavy vehicle loads, extensive road search areas, and randomly distributed obstacles. Full article

The complexity inherent in navigating intricate traffic environments poses substantial hurdles for intelligent driving technology. The continual progress in mapping and sensor technologies has equipped vehicles with the capability to intricately perceive their exact position and the intricate interplay among surrounding traffic elements. Building upon this foundation, this paper introduces a deep reinforcement learning method to solve the decision-making and trajectory planning problem of intelligent vehicles. The method employs a deep learning framework for feature extraction, utilizing a grid map generated from a blend of static environmental markers such as road centerlines and lane demarcations, in addition to dynamic environmental cues including vehicle positions across varied lanes, all harmonized within the Frenet coordinate system. The grid map serves as the input for the state space, and the input for the action space comprises a vector encompassing lane change timing, velocity, and vertical displacement at the lane change endpoint. To optimize the action strategy, a reinforcement learning approach is employed. The feasibility, stability, and efficiency of the proposed method are substantiated via experiments conducted in the CARLA simulator across diverse driving scenarios, and the proposed method can increase the average success rate of lane change by 6.8% and 13.1% compared with the traditional planning control algorithm and the simple reinforcement learning method. Full article

The operation of the automatic emergency braking (AEB) system may lead to a significant increase in lateral offset of vehicles in curved road conditions, which can pose a potential risk of collisions with vehicles in adjacent lanes or road edges. In order to address this issue, this study proposes an integrated longitudinal and lateral control strategy for collision avoidance during emergency braking, which utilizes a control algorithm based on Time to Collision (TTC) for longitudinal control and a control algorithm based on yaw angle and preview point lateral deviation for lateral control. On one hand, the AEB system facilitates proactive longitudinal intervention to prevent collisions in the forward direction. On the other hand, the Lane Keeping Assist (LKA) system allows for lateral intervention, reducing the lateral offset of the vehicle during braking. To evaluate the effectiveness of this integrated control strategy, a collaborative simulation model involving Matlab/Simulink, PreScan, and CarSim is constructed. Under typical curved road conditions, comparative simulations are conducted among three different control systems: ➀ AEB control system alone; ➁ independent control system of AEB and LKA; and ➂ integrated control system of AEB and LKA. The results indicate that although all three control systems are effective in preventing longitudinal rear-end collisions, the integrated control system outperforms the other two control systems significantly in suppressing the vehicle’s lateral offset. In the scenario with a curve radius of 60 m and an initial vehicle speed of 60 km/h, System ➀ exhibits a lateral offset from the lane centerline reaching up to 1.72 m. In contrast, Systems ➁ and ➂ demonstrate significant improvements with lateral offsets of 0.29 m and 0.21 m, respectively. Full article