Search Results (4,273)

To address the emergency obstacle-avoidance problem of intelligent vehicles on structured roads, this paper proposes an integrated planning and control method that combines an improved Artificial Potential Field (APF) with fuzzy Model Predictive Control (MPC). Different from a direct APF + MPC combination, the planning layer introduces a braking-distance threshold, an effective obstacle-influence boundary, and sinusoidal shape factors to reshape the obstacle repulsive field and alleviate local-minimum behavior. A seventh-order polynomial smoothing strategy is then adopted to generate a reference path with higher-order continuity. For trajectory tracking, a fuzzy adaptive MPC controller adjusts the prediction horizon and control horizon online according to lateral error, while a fuzzy PID controller regulates longitudinal speed. MATLAB/Simulink and CarSim co-simulation results in single-static, double-static, and double-dynamic obstacle scenarios show that the proposed method can generate smoother trajectories and achieve more stable tracking, thereby improving obstacle-avoidance safety and ride comfort. In the double-static scenario, the peak lateral error is reduced from about 0.7 m to within 0.1 m, while in the double-dynamic scenario the longitudinal speed is maintained within 78–80 km/h instead of dropping to about 67 km/h under the baseline controller. The study provides a practical technical framework for integrated decision-planning-control design in structured-road intelligent vehicles. Full article

The increasing need for energy-efficient and environmentally friendly electricity generation has led to the extensive use of hybrid electric systems. These systems integrate different energy sources in an effort to take advantage of the positives of each technology, as using a single source of energy comes with many limitations and disadvantages; hence, the popularity of hybrids has increased in recent times. In this regard, this paper proposes a lithium-ion battery (LIB) and ultracapacitor (UC)-based hybrid architecture considering an optimal energy management framework. In the transportation sector, hybrid vehicles (LIB and UC-based vehicles) effectively utilize the high energy density and power density of LIBs and UCs. This LIB and UC-based hybrid architecture provides an efficient power management solution considering the high power density of the LIB for smooth road profiles, and the high power density of the UC is driven during sudden spikes in load demand because the LIB will not function optimally during the sudden spikes due to lower power density. Furthermore, in order to achieve efficient utilization of the proposed hybrid system, an optimal energy management framework is used. In this regard, in this study, a fractional-order proportional–integral–derivative (FOPID) controller has been designed for effective and optimal energy management. Furthermore, the designed FOPID has been optimized using a metaheuristic technique, namely particle swarm optimization (PSO), to enhance LIB and UC-based hybrid electric vehicle energy management performance. Employing dynamic and optimal energy flow control, the FOPID-based system improves energy consumption, extends LIB life, and improves overall system performance and reliability. Full article

Road traffic accidents at intersections pose a persistent challenge in Riyadh, Saudi Arabia, contributing significantly to public health burdens and economic losses. Traditional statistical approaches often fail to capture the complex, non-linear interactions among geometric design, traffic parameters, and accident severity. This study develops a multi-methodological machine learning framework to predict intersection accident severity using the Equivalent Property Damage Only (EPDO) metric. Historical data (2017–2023) from Riyadh Municipality for 150 high-risk intersections were analyzed, incorporating predictors such as service road distance (SRD), U-turn distance (UTD), median width (MW), peak hour volume (PHV), heavy vehicle percentage (HV%), and injury/frequency counts. Six algorithms, i.e., Decision Tree, Random Forest, Gradient Boosting, Support Vector Machine, Linear Regression, and Artificial Neural Network, were compared using a 70/30 train–test split and k-fold cross-validation in this study. The Gradient Boosting model achieved superior performance (R² = 0.89 with MSE = 63.43 and RMSE = 7.96) and was selected for final deployment. SHAP feature importance analysis revealed minor injuries (MIs), serious injuries (SRIs), and fatalities (FAs) as the most important dominant predictors, with geometric factors (UTD, MW) and traffic composition (HV%) providing actionable infrastructure insights. The model ranked intersections and identified the “Jeddah Road with Taif Road” (predicted EPDO = 137.22) as the highest-risk location. Evidence-based recommendations include enforcing the minimum 300 m U-turn buffers with staggering service road exits ≥150 m and restricting heavy vehicles during peak hours. The scalable framework developed in this study supports the data-driven prioritization of safety interventions and aligns with sustainable urban mobility goals and offers transferability to other metropolitan contexts worldwide. Full article

Autonomous vehicle navigation requires low-latency and energy-efficient machine learning models capable of operating in dynamic and resource-constrained environments. Conventional deep learning approaches are often unsuitable for real-time deployment on embedded edge devices due to their high computational and memory demands. In this work, we propose a unified TinyML-optimized navigation framework that integrates a lightweight convolutional feature extractor (MobileNetV2) with a metric-based few-shot learning classifier to enable rapid adaptation to unseen driving scenarios with minimal data. The proposed framework jointly combines feature extraction, few-shot generalization, and edge-aware optimization into a single end-to-end pipeline designed specifically for real-time autonomous decision-making. Furthermore, post-training quantization and structured pruning are employed to significantly reduce the memory footprint and inference latency while preserving the classification performance. Experimental results demonstrate that the proposed model achieved a 93.4% accuracy on previously unseen road conditions, with an average inference latency of 68 ms and a memory usage of 18 MB, outperforming traditional CNN and LSTM models in efficiency while maintaining a competitive predictive performance. These results highlight the effectiveness of the proposed approach in enabling scalable, real-time navigation on low-power edge devices. Full article

University students face sustained academic, employment, and social pressures. Campus pathways, as central linear spaces in daily routines, hold significant potential to influence well-being, yet existing research has largely overlooked how their environmental characteristics affect multidimensional health. Using Shenyang Jianzhu University as a case, this study identified frequently used pathways through GPS tracking and surveys, and quantitatively analyzed how environmental features affect walking willingness, emotional experience, and social interaction. By comparing high- and low-benefit groups, the key environmental thresholds were identified to inform health-oriented design. Beyond verifying some established understandings (e.g., daily commuting paths prioritize efficiency, while leisure paths focus on experiential quality), the study further revealed several mechanisms through quantitative analysis. For example, “road accessibility”—an indicator of convenience—showed a significant negative correlation with emotional experience. The study established quantifiable prediction models and identified design thresholds for campus pathways. A high aesthetic greenery was key to achieving high overall benefits, while low building enclosure and vegetation complexity promoted social interaction. This achievement transforms health-oriented campus pathway design from qualitative principles into a measurable and optimizable scientific practice, thus providing an empirical basis and practical guidance for the construction of health-supportive campus environments. Full article

Microplastics have emerged as a relatively new type of pollutant and have attracted significant global attention. This study focuses on toxicology of microplastics in ambient PM_2.5 and road dustfall in Beijing. It utilizes the Plasmid Scission Assay to toxicologically evaluate the oxidative damage capacity of microplastics as a component of PM_2.5. The Pollution Load Index (PLI) method, based on the mass concentration of microplastics in ambient air, was employed to assess the ecological risk of atmospheric dustfall microplastics in Beijing. The results showed that both standard microplastic samples and mixed samples of microplastics with ambient PM_2.5 exhibited a dose–response relationship in DNA damage rates. At the same dose, microplastic samples with smaller particle sizes have a higher DNA damage rate. Based on the PLI results, most road dustfall microplastics in Beijing exhibit significant spatial variation. Analysis of road dustfall along the east–west main road across Beijing’s urban area revealed that microplastic pollution levels are higher in the eastern zone than in the western zone. Comparisons of pollution levels across functional areas in Beijing showed that university areas > residential areas > industrial areas > commercial areas > agricultural areas. In vertically collected samples, higher elevations (PLI_13.6m = 3.54) exhibit greater pollution levels than lower (PLI_1.5m = 1), which warrants special attention. These findings highlight the complex relationship between atmospheric microplastic accumulation and their oxidative capacity, providing essential insights for the design of targeted emission reduction strategies. Full article

Mini-roundabouts are increasingly implemented as compact, low-cost alternatives to conventional roundabouts and signalized intersections, especially at low-speed, space-constrained urban locations where safety is a concern. Their design emphasizes speed management, reduced conflict severity, and operational simplicity, contributing to safer mobility for all road users. This paper reviews U.S., German, and UK design guidelines and synthesizes empirical safety evidence from before-and-after studies of mini-roundabout conversions. In terms of design, the U.S. practice typically relies on a single large design vehicle and more permissive geometry, whereas the German guidance adopts a multi-vehicle approach with tighter curvature and stronger compactness to enforce lower speeds, affecting crash risk and driver behavior. The UK guidance is distinguished by its flush or slightly domed central marking and flexible application approach. Conversions from two-way stop-controlled (TWSC) or one-way stop-controlled (OWSC) intersections yield substantial reductions in injury and severe crashes, with total crash reductions of 17–42%. Conversions from all-way stop-controlled (AWSC) intersections present more variable outcomes, including increases in total crashes, because drivers are still reacting based on the previous control and may not adjust their expectations quickly. Modular roundabouts are also examined as alternative compact interventions for constrained or high-risk sites, with early evidence indicating reductions in severe crashes and improved speed control while minimizing construction costs and right-of-way impacts. Full article

With the rapid evolution of urbanization and artificial intelligence technology in China, small, intelligent road sweepers have emerged as a highly promising technical solution to address urban cleaning challenges. The development and breakthrough of high-performance dust collection trays (DCT) stand as the core prerequisite for the large-scale practical application of such sweepers. Although blowing–suction integration technology theoretically offers substantial potential for improving dust removal efficiency, it has not received adequate attention in the sweeper field, particularly in the research on its application in unmanned, small-sized models. In this study, a fresh concept of an efficient DCT was proposed, and its numerical method was verified by experiment. Then, the design work for this efficient DCT was efficiently carried out by combining computational fluid dynamics (CFD) numerical simulation with response surface methodology (RSM). Finally, the influence mechanisms of three key operational parameters of nozzle airflow velocity, suction negative pressure, and vehicle travel speed on the dust removal effect were numerically investigated. The results indicated that the parameter combination of DCT with an 18° blowing angle, 20° shoulder angle, and 0.2 diameter-to-length ratio was recommended, and its dust removal efficiency could reach a peak level of 98.7% when the nozzle blowing velocity, negative pressure at suction port, and travel speed were respectively 14 m/s, −1800 Pa, and 1.4 m/s. This research provides important theoretical support and a feasible technical pathway for the design of high-performance DCTs. Full article

Composite soil–steel corrugated bridges, which are widely used in road, railway, and civil engineering, are recognized as durable, sustainable, and cost-effective structures. Due to their interactions with the surrounding soil, relatively thin corrugated steel plates are usually used in these bridges. Larger spans are associated with larger cross-sections, and deep corrugations with a 500 mm pitch and a 237 mm depth are already in use worldwide. However, the behavioral benefits of high-strength steel and additional strengthening elements for CSS structures have rarely been investigated with regard to local buckling in the straight regions of the corrugation. This study analyzed the influence of high-strength steel and innovative corrugated cross-sections strengthened with circular steel pipes on the utilization ratio of steel plates in composite soil–steel structures. Two-dimensional numerical models of three bridges with spans of 26 m, 17.5 m, and 12 m and surrounded by soil were developed to identify internal forces from permanent and temporary actions. Plate utilization was designed according to the Swedish, Canadian, and American methods, considering local buckling in the 500 × 237 mm and 381 × 140 mm corrugation profiles. It was found that the use of higher-strength steel material, as well as the introduction of steel pipes, significantly reduced the plate thickness of regular corrugations. The results show that the use of higher-strength steel reduced the cross-section area of regular and innovative corrugations by 30–40%. Moreover, the cross-section area of the innovative profile was 5% to 36% lower than that of the regular corrugation profile. Nevertheless, the results show that the local buckling approach proposed by the Swedish design method could be considered conservative and should be revised. In addition, the method of preventing local buckling by reducing the plastic moment capacity could be neglected when using thicker plates and lower steel grades. Full article

Head-Up Display (HUD) systems are widely used in vehicles to overlay navigation prompts in the driver’s field of view, thereby reducing eyes-off-road time. However, suboptimal information presentation may impose extra cognitive demands and lead to driver distraction. To quantify the effects of key HUD navigation design factors on navigation task performance and cognitive workload, a 2 × 2 within-subjects experiment was conducted, manipulating display position (upper vs. lower visual field) and the presence of textual navigation information (with vs. Without text). Thirty university students with driving experience completed navigation tasks under four conditions in a single-lane urban driving simulation. Each task lasted 2–4 min and included six turning prompts. Task performance (accuracy, mean reaction time, and total driving time), subjective workload (PAAS), and eye-tracking measures (mean fixation duration, mean pupil diameter, fixation count, and fixation count proportion) were collected and analyzed using repeated-measures ANOVA. Results showed that display position significantly affected driving efficiency and subjective workload: lower-field displays produced shorter reaction times and lower PAAS scores, while accuracy and total driving time showed no significant differences. Eye-tracking results indicated higher fixation counts and fixation ratios for lower displays. A significant interaction between display position and text was observed for mean fixation duration, whereas mean pupil diameter showed no significant effects. These findings indicate that display position is a critical factor in HUD navigation design, while textual information primarily influences visual inspection patterns rather than overall navigation task performance. Full article

As autonomous driving technology progresses, efficient and accurate object detectors are able to detect pedestrians, vehicles, road signs, and obstacles in real time, thereby enhancing driving safety and serving as a part of autonomous driving. However, the performance of such object detectors is limited and cannot be leveraged to satisfy modern autonomous driving systems. To address this issue, we develop an object detection network for autonomous driving scenarios, SST-YOLO, which is based on YOLOv8. First, we propose a Sobel Convolution & Convolution (SCC) module to enhance the backbone, which incorporates a SobelConv branch to explicitly model gradient-based edge information and improve structural feature representation. In addition, we replace the original path aggregation feature pyramid network (PAFPN) with a Small Object Augmentation Pyramid Network (SOAPN), which integrates SPDConv and CSP-OmniKernel modules to strengthen multi-scale feature fusion and enhance small object representation. Finally, a Task-Adaptive Decomposition & Alignment Head (TADAHead) is designed, which employs task decomposition, dynamic deformable convolution, and classification-aware modulation to decouple tasks and achieve adaptive spatial alignment, thereby improving detection accuracy and robustness in complex scenarios. Experiments on the public autonomous driving dataset KITTI show that our proposed method outperforms the baseline YOLOv8 model. Compared with the baseline results, mAP@0.5:0.95 ranges from 65.1% to 69.2%, which indicates that the proposed SST-YOLO network can achieve object detection for autonomous cars. Full article

To develop a refined traffic diversion scheme for expressway reconstruction and expansion, this study establishes generalized link impedance functions for cars and trucks, considering their differences in road travel time, time value, and toll costs. Subsequently, a traffic diversion model is constructed based on user equilibrium theory, taking the heterogeneity between cars and trucks into consideration. A path-based solution algorithm using the method of successive averages is designed to solve the model. To evaluate the environmental impact of the traffic diversion, a vehicle exhaust emission (including CO₂, CO, HC, and NO_x) estimation method based on the COPERT model is proposed. The results of a case study show that the optimized traffic diversion scheme significantly reduces the average V/C ratio while increasing the average velocity of both cars and trucks on the reconstructed links, without substantially compromising the traffic efficiency of other links. Additionally, the diversion scheme reduces the exhaust pollutant emissions, but increases the CO₂ emissions within the network. The findings justify the effectiveness of the traffic diversion approach on alleviating the traffic congestion on the reconstructed expressway and its mixed impacts on the environment. Full article

Pedestrians are amongst the most vulnerable road user groups. Efforts to enhance pedestrian safety have mainly focused on intersections and midblock crossings. This study investigated the effect of bus stop environments on pedestrian safety in Kumasi, an area with a high incidence of pedestrian fatalities in Ghana. Crashes within a 50 m radius of bus stops were extracted using a spatial join. The Negative Binomial regression model was applied to model pedestrian crashes around bus stops as a function of three distinct non-collinear independent variable groups: road design features, bus stop characteristics, and pedestrian exposure measures. Formal bus stops were associated with higher crash rates than informal ones. The presence of medians and crosswalks was associated with lower crash rates, whereas wider carriageways were associated with higher crash rates. Higher crashes were linked to passing pedestrians and waiting pedestrians, while crossing pedestrians were associated with reduced crashes. These findings suggest that the combined effects of infrastructure and behavioural factors influence pedestrian safety at bus stops. Prioritising low-cost safety treatments, such as guard-railed waiting areas, marked crosswalks, medians, and raised crossings, around bus stops will yield substantial safety benefits for resource-constrained contexts and advance sustainable urban mobility. Full article

Increasingly frequent extreme rainfall commonly leads to water accumulation on the road surface, elevating vehicle tire hydroplaning to a major threat to driving safety. Existing research mainly focused on tire model optimization or predicting critical hydroplaning speed features based on empirical formulas and numerical simulations. However, there is a lack of systematic validation of the tire–water–pavement coupling interaction under realistic pavement conditions, with particular insufficient attention paid to pavement dynamic responses. In this study, numerical simulation and field full-scale accelerated loading methods were applied to investigate dynamic response characteristics of the tire–water–pavement coupling interaction system. Parametric analyses were first performed to investigate the influences of vehicle speed, vehicle load, water-film thickness, and tire lateral position on the mechanical behaviors of the fluid–structure interaction for a moving vehicle tire. Subsequently, field-measured dynamic responses’ features were used to validate the numerical model, which was then further applied to predict critical conditions of vehicle tire hydroplaning. Finally, the mechanisms of hydroplaning and corresponding mitigation measures were discussed. The study revealed that increasing vehicle speed and water-film thickness, as well as decreasing vehicle load, would reduce the pavement supporting force. The tire–pavement contact stress and strain decreased from the vehicle tire’s center position towards its shoulders. The predicted critical hydroplaning condition suggested that increasing vehicle load mitigated hydroplaning by reducing the proportion of water-induced hydrodynamic lifting force relative to the total vehicle load. When the water depth is relatively shallow, the hydroplaning risk increases rapidly with water depth, while the water’s adverse impact on tire–pavement contact force gradually diminishes as water depth continues to increase. It implies that a vehicle with a relatively low axle load driving on the pavement with a thin thickness of retained water in light rain will still face the hydroplaning risk, as the pavement’s supporting force may be substantially reduced in this weather. The findings provide theoretical foundations and experimentally supported insights on driving safety assessment and anti-skid design of water-covered pavement. Full article

Cognitive or visual distraction caused by drivers using mobile phones, operating the central console, or conversing with passengers while driving is a significant contributing factor to road traffic accidents. Aiming to solve the problem that existing driving behavior monitoring systems exhibit insufficient recognition accuracy and low real-time detection performance in complex driving environments, this study proposes a dual-flow driver distraction detection model based on Sobel edge detection (DFSED-Model). The model is designed with a collaborative learning framework: the first flow adopts a lightweight pre-trained backbone network to achieve efficient semantic feature extraction. The second flow utilizes Sobel edge detection to extract the driver’s driving contours and enhances the model’s spatial sensitivity to driving movements and hand movements. Through the feature learning process of the first-flow-guided auxiliary branch, collaborative optimization of knowledge transfer and attention focusing is realized, thereby improving the model’s convergence speed and discriminative performance. The proposed model is evaluated on three widely used public datasets: the State Farm Distracted Driver Detection (SFD) dataset, the 100-Driver dataset, and the American University in Cairo Distracted Driver Dataset (AUCDD-V1). Under the premise of maintaining low computational overhead, the accuracy of the DFSED-Model reaches 99.87%, 99.86%, and 95.71%, respectively, which is significantly superior to that of many mainstream models. The results demonstrate that the proposed method achieves a favorable balance between accuracy, parameter count, and efficiency, and possesses strong practical value and deployment potential. Full article