
Many researchers argue that the modular perception–planning–control design of autonomous vehicles (AVs) inevitably introduces delays, yet how such delays influence car-following behaviors and what their magnitudes are remain uncertain due to limited empirical data and ineffective calibration methods. Using large-scale naturalistic autonomous vehicle trajectory data, we develop an optimization-based calibration framework that jointly estimates feedback delay in perception/decision and parasitic lag in actuation, and accommodates both linear constant-time-headway and nonlinear intelligent-driver-model controllers. The method combines Bayesian optimization (BO) for globally searching delay parameters with simultaneous perturbation stochastic approximation (SPSA) for efficiently tuning controller parameters. Results show that incorporating both delays substantially improves model accuracy, reducing the root mean square error (RMSE) by up to an order of magnitude compared with models without delays, while linear controllers reproduce AV trajectories more faithfully and local calibration proves more accurate for short horizons whereas global calibration is more reliable for long horizons.
Abstract Urban water-supply systems depend on elevated water tanks for storage, pressure support, firefighting, and post-earthquake service continuity. This study evaluates the component-level seismic fragility and reliability of three 250 m 3 Intze-type reinforced-concrete frame-staging tanks with normal, radial, and cross bracing. Incremental dynamic analysis was performed using 44 records divided into far-fault, near-fault directivity with low and high peak ground velocity (PGV)/peak ground acceleration (PGA) ratios, and fling-step ensembles. Five structural and hydrodynamic demand parameters were examined. High-PGV/PGA directivity motions produced the greatest fragility, reducing the normal frame without diagonal bracing (WT1) median collapse PGA from 0.992 g under far-fault motions to 0.442 g. Over PGA = 0.2 g–0.8 g, the maximum inter-storey drift ratio (MIDR)-based near-fault demand amplification factor ranged from 2.486 to 2.583 across the three configurations under high-ratio directivity motions. Robustness was assessed using 2,000 bootstrap resamples, leave-one-record-out checks, and alternative formulations, confirming that the numerical factor remains configuration- and dataset-specific. The corresponding impulsive-mass-acceleration factors were 0.706, 1.200, and 1.194, indicating no systematic attenuation. At Zone V, the WT1 50-year collapse probability increased from 0.75% under far-fault motions to 14.66% under fling-step motions. The findings support near-fault-sensitive component-level assessment without implying complete hydraulic-network functionality.
Autonomous vehicles (AVs) have the potential to transform future mobility systems, yet their adoption remains limited in many developing and transitional economies. This study extends the Decomposed Theory of Planned Behavior (DTPB) by incorporating Knowledge of Autonomous Vehicles (KAV) and Perceived Economic Impact (PEI) to explain behavioral intention toward AV adoption. Survey data were collected from 430 licensed drivers in Armenia and analyzed using structural equation modeling and hierarchical regression. The results showed that attitude, subjective norms, and perceived behavioral control significantly predicted behavioral intention, with perceived behavioral control emerging as the strongest determinant. Attitude was positively influenced by perceived usefulness, perceived ease of use, compatibility, and KAV, while subjective norms were shaped by peer and superior influence, and perceived behavioral control by self-efficacy and facilitating conditions. PEI also exerted a significant positive effect on behavioral intention. Moderation analysis indicated that KAV strengthened the relationships between perceived usefulness and attitude and between perceived ease of use and attitude, whereas PEI moderated the effects of subjective norms and perceived behavioral control on behavioral intention. These findings validate the extended DTPB and highlight the importance of cognitive awareness and economic perceptions in promoting AV adoption within transitional economies.
Abstract Subway tunnels are essential components of urban transportation systems, and water leakage can threaten their structural safety. Accurate detection and segmentation of leakage regions are therefore critical. This study develops a tunnel water leakage dataset from point-cloud data collected using a mobile measurement platform and converts the point clouds into grayscale images. We propose DWCF-YOLO, an enhanced YOLOv11-based network for water leakage detection and segmentation. The model incorporates multi-level wavelet transform convolution to extract multi-frequency features and promote local–global feature fusion, improving the identification of leakage areas at different scales. In addition, content-guided attention provides channel- and spatial-directional guidance for multi-scale feature maps, enhancing sensitivity to leakage regions. Experimental results show that DWCF-YOLO achieves 96.8% accuracy, 97.8% recall, 98.4% F1-score, 98.9% precision, 96.8% mIoU, and 97.8% mPA. Compared with mainstream models, it obtains the best performance in four of the six evaluation metrics, demonstrating its effectiveness and reliability for tunnel water leakage detection.
Forecasting urban water demand requires approaches that account for both environmental variability and human behavior. Traditional models rely primarily on climatic drivers and assume stable consumption patterns, overlooking the rapid behavioral shifts triggered by social events, consumers’ attention, and environmental stress. The objective of this study is to develop and evaluate a behaviorally informed, event-aware framework for short-term urban water demand forecasting that explicitly integrates physical and social drivers. The proposed framework utilizes data from weather records, Meta mobility distributions, Google Trends search activity, and demand records from the North Okanagan Valley in British Columbia, Canada. By coupling data-driven modeling with interpretable Machine Learning (ML), the framework identifies environmental thresholds and behavioral triggers influencing consumption. The results underscore that integrating digital behavior indicators with climate ones improves forecasting performance, increasing R2 from 0.79 to 0.85, while behavioral indicators alone achieve an R2 of 0.70, highlighting their independent explanatory power. These findings emphasize the value of integrating real-time behavioral proxies into operational forecasting models, supporting more adaptive and socially aware water resource management. The framework is demonstrated using a real-world case study and is designed to be transferable to other urban systems where human activity strongly influences short-term demand dynamics.