Search Results (3,898)

Point cloud scene flow estimation aims to capture the three-dimensional motion of each point in a sequence of point clouds. Although progress has occurred in this field, existing methods often face significant challenges. In particular, two key issues persist: the absence of corresponding local information from the source point cloud to the target, preventing correct feature matching, and the presence of highly similar adjacent structures in target regions, which leads to ambiguous correspondences due to indistinguishable point features. To address these problems, this paper introduces a novel self-supervised method for point cloud scene flow estimation. Theoretically, we establish a new framework that integrates discriminative feature learning with probabilistic flow refinement. A new network architecture, SetConv++, is designed to learn more discriminative point feature representations, enhancing differentiation in similar structures. Additionally, a refinement module uses the random walk algorithm to adjust initial flow estimates. This approach reconstructs low-confidence flows with high-confidence surrounding ones, reducing missing correspondence issues. Crucially, a new flow smoothing loss term ensures local consistency while suppressing error propagation—a fundamental limitation in existing methods. Through comprehensive experiments on the KITTI Scene Flow dataset, our method demonstrates superior performance. It significantly outperforms existing self-supervised approaches across multiple standard evaluation metrics. Specifically, on the KITTI Scene Flow dataset, our method reduces the Endpoint Error (EPE) by 13.6% (from 0.0411 to 0.0355) and improves Accuracy Strict (AS) by 2.43 percentage points (from 92.68% to 95.11%) compared to baseline self-supervised approaches, while also reducing the outlier rate (Out) by 1.5 percentage points. This advancement not only provides a robust theoretical framework for handling ambiguous correspondences but also enables more reliable and efficient downstream applications—such as autonomous driving perception systems requiring real-time motion accuracy in complex scenes. Full article

Autonomous driving technology is widely recognized as a key solution for enhancing future road safety by preventing traffic accidents caused by human error. However, the widespread adoption of autonomous vehicles (AVs) has not yet been achieved, and traffic accidents involving autonomous vehicles in mixed traffic conditions continue to be reported. This study analyzed conflict events using real-world autonomous driving data and identified AV conflict hotspots. A two-dimensional Time to Collision was employed as a surrogate safety indicator to comprehensively capture various types of conflicts in urban interrupted traffic flow. Analysis of approximately 1000 h of driving data revealed 958,011 conflict events, which were distributed along major AV trajectories. The Network Kernel Density Estimation was applied to identify AV conflict hotspots based on conflict events. The optimal hotspot identification model was determined by evaluating various parameter combinations using the Predictive Accuracy Index validated against real-world accident data. Several hotspots were identified on arterial roads with signalized intersections, nearby bus stops, and frequent access points to roadside facilities such as restaurants, stores, gas stations, and residential complexes. Differences in hotspot patterns by conflict type reveal distinct risk characteristics across road sections, emphasizing the necessity of customized safety countermeasures for each conflict type. The findings of this study are expected to accelerate the deployment and wider adoption of autonomous driving technology, promoting the sustainable operation of AVs. Full article

Brake-induced steering wheel shimmy is a critical nonlinear dynamic phenomenon that severely compromises vehicle handling stability and driving safety. While clearances in steering mechanism kinematic pairs are widely recognized as a primary cause of shimmy instability, the coupling effect of multiple concurrent clearances remains poorly characterized, particularly under transient braking conditions. In this work, a 5-degree-of-freedom non-autonomous dynamic model of brake-induced shimmy is developed using Lagrange’s equations. The model comprehensively incorporates the non-smooth contact behavior of multiple clearance joints, transient braking axle load transfer, and the longitudinal–lateral coupling nonlinearity of tires. The nonlinear dynamic evolution of the system is investigated through phase portraits, Poincaré sections, and continuous wavelet transform analysis. Numerical results demonstrate that multi-clearance coupling increases the peak shimmy angle by more than 40% compared to the single-clearance case. As the clearance magnitude increases from 0.05 mm to 0.40 mm, the system undergoes a transition from stable periodic motion to high-dimensional chaos, accompanied by a 67% reduction in vibration energy concentration at the 0.4 mm clearance level. This study elucidates the nonlinear mechanism underlying clearance-induced brake shimmy, providing a robust theoretical foundation for steering system parameter optimization and shimmy mitigation strategies. Full article

LiDAR point clouds provide accurate and direct representations of spatial locations and geometric structures of objects in 3D space, making them essential for applications such as target recognition in autonomous driving and 3D reconstruction in smart cities. However, large-scale point clouds pose challenges, including massive data volume, uneven density distribution, and complex object structures. Existing point-based and graph-based semantic segmentation networks often suffer from limitations such as loss of local contextual information, over-reliance on local graph construction, and insufficient modeling of relationships between neighboring points. To address these issues, we propose PG-Net, a novel network that integrates discrete point distribution features with local graph structural information. The framework includes: (1) a point branch equipped with a Local Adaptive Feature Augmentation (LAFA) module to extract efficient local features; (2) a graph branch featuring a Dynamic Graph Feature Aggregation (DGFA) module, which explicitly models relationships among points in local graphs and adaptively balances a point’s intrinsic features with its neighborhood context; and (3) fuses local features from both branches, allowing their complementary strengths to enhance feature representation, a process further promoted by a New Aggregation Loss Function. Experiments on the Toronto3D and S3DIS datasets show that PG-Net achieves overall accuracy (OA) of 97.69% and 89.87%, and mean Intersection-over-Union (mIoU) of 83.51% and 73.22%, respectively. Comparative and ablation studies against advanced methods such as RandLA-Net, BAAF-Net, and LACV-Net demonstrate the effectiveness and robustness of our approach. By jointly exploiting discrete point distribution and local graph structural relationships, PG-Net effectively leverages the complementary strengths of its dual-branch design, offering a reliable solution for efficient and accurate large-scale point cloud semantic segmentation. Full article

This study examines the spatiotemporal evolution and driving mechanisms of urban expansion in the Guangxi Zhuang Autonomous Region, China, from 2013 to 2023. Using Suomi-NPP VIIRS nighttime light (NTL) data, we combine Standard Deviational Ellipse (SDE) analysis, centroid migration, kernel density estimation (KDE), landscape metrics, Local Moran’s I (LISA), and system Generalised Method of Moments (system-GMM) estimation. The results show that the centroid of urban development remained within Binyang County while moving overall toward the southeast with recurrent north–south oscillations. The SDE results indicate a stable northeast–southwest orientation, with secondary expansion in other directions. The urban structure is dominated by a strong Nanning core, accompanied by secondary clusters in Liuzhou, Guilin, and other prefecture-level cities. Nanning recorded the largest absolute expansion, followed by secondary centres, including Liuzhou, Guilin, Yulin, Wuzhou, Fangchenggang, Qinzhou, and Beihai, whereas western and northern Guangxi expanded more slowly. The system-GMM results indicate that financial deepening has a marginally significant positive effect on built-up area expansion and fiscal pressure has a marginally significant constraining effect, both at the 10% level; land finance dependency does not emerge as an independent driver in this small panel. We interpret these findings through a Source–Channel–Valve framework, in which financial deepening provides the capital source, land finance represents a hypothesised institutional channel, and fiscal pressure acts as a regulatory constraint. The study provides empirical evidence for sustainable and regionally coordinated urban development in Guangxi and comparable geographically constrained regions. Full article

Self-supervised monocular depth estimation (MDE) has emerged as a cost-effective alternative to traditional depth sensing approaches, which often require expensive equipment or complex configurations. However, these models often fail to maintain high performance on unseen domains, presenting a considerable issue for real-world applications, such as autonomous driving, where the ability to adapt consistently across various conditions is essential for ensuring safety. In this paper, we propose a novel lightweight network designed to separate domain-specific style and domain-invariant content within feature representation statistics, removing the style information that contributes to the domain shift. Furthermore, to address the convolutional neural network’s limited local scope, we inject state-space model blocks inside our encoder architecture in order to capture extensive contextual information, yet having a minimum number of trainable parameters. Experiments on popular outdoor depth estimation benchmarks, KITTI and Make3D, demonstrate superior performance to other recent networks with similar size. Full article

Transparent object segmentation plays a critical role in indoor and outdoor scene understanding, particularly driven by the rapid advancements in autonomous driving and robotics. However, this task presents significant challenges due to the lack of distinct texture and chromatic features in transparent objects, causing their appearance to blend into the background. Existing methods face inherent architectural limitations: CNNs are restricted by limited receptive fields, while Transformer-based methods may inadvertently suppress the weak feature details of transparent surfaces due to the inherent low-pass filtering property of self-attention mechanisms, treating them as background noise. Consequently, these approaches struggle to consistently segment transparent objects across diverse scales, failing to preserve both fine details and large-scale structures. To address these limitations, we propose the Global Context-Aware Transformer (GCA-Trans). Specifically, we design a Multi-scale Context Mining (MCM) module that leverages parallel dilated convolutions with varying receptive fields to simultaneously extract features at multiple scales. This design allows the model to capture and fuse fine-grained local details (e.g., edges and textures) with coarse-grained global spatial context (e.g., overall object shapes), ensuring robust segmentation performance for transparent objects of varying scales. Extensive experiments on four benchmark datasets demonstrate that GCA-Trans sets a new state of the art, achieving significant improvements of 2.53% mIoU on Trans10K-v2, 2.1% IoU on RGB-D GSD, 2.2% IoU on GDD, and 1.9% IoU on GSD, validating the effectiveness and robustness of our approach. Full article

Low-speed automated driving (LSAD) shuttles operate in complex urban environments where abrupt braking can affect safety, service quality, and operational interpretability. This study proposes a telemetry-based workflow for the detection and characterization of hard braking (HB) events in autonomous shuttle operations. The workflow includes preprocessing of autonomous in-service telemetry data, deterministic HB detection under irregular sampling, evidence-based attribution using diagnostic and obstacle-related signals, and driving-context characterization through K-means clustering, applied to a 62-day dataset from an autonomous on-demand shuttle operating on a fixed 2.8 km urban loop in Turin. After preprocessing, 71% of the 16,670,518 observations are retained. The analysis identified 734 HB events, of which 89% are linked to specific contextual conditions, highlighting environmental and operational influences on safety-critical situations. Driving-context analysis relies on 316,280 observations collected at 1 Hz and yields a nine-cluster solution. When projected back onto the route through waypoint-level modal regimes, HB events are found to be over-represented in high-speed segments. These results show that HB events can be interpreted not only as a threshold exceedance, but as an operational indicator linked to route-level driving regimes. The proposed framework supports data-driven safety assessment and operational decision-making in autonomous shuttle systems by researchers and practitioners. Full article

Agriculture plays a relevant role in the food supply chain but is also a major contributor in terms of emissions. A possible solution to reduce its impact is to replace traditional machinery with innovative systems, such as agricultural rovers. In the proposed research, a case study of an agricultural rover, specifically designed to operate in orchards, is presented. The powertrain features a Li-ion battery pack as the primary energy source and a fuel cell system operating as a range extender unit. Hydrogen is stored on board using a metal hydride tank to enhance compactness. Once the traction and range extender power output control strategies were defined, experimental tests in a closed warehouse were performed. During the tests, the rover was manually controlled using a joystick, since the main focus was to evaluate the powertrain behavior rather than to test the autonomous driving algorithm. During the tests, different maneuvers in narrow spaces were performed. The results showed that the rover successfully accomplished the tasks and the range extender unit can effectively extend the rover autonomy up to +150% compared to the pure battery solution. This result was obtained considering a 15 min test carried out in an indoor environment with a polished concrete floor. Full article

With the development of autonomous driving and intelligent connected vehicle technologies, the vehicle cabin is shifting from a simple transportation space to an intelligent mobile space integrating infotainment, interaction, and rest, and passenger comfort has gradually become an important factor affecting user experience, system trust, and perceived safety. Focusing on three categories of cabin environmental factors, namely the acoustic, optical, and thermal environments, this study develops an experimental design and comprehensive modeling method for passenger comfort evaluation. First, controlled single-factor experiments were conducted to establish quantitative mapping relationships between physical environmental parameters and subjective comfort ratings. The analytic hierarchy process (AHP) was then used to determine the weights of each indicator, and a penalty-based aggregation mechanism was introduced to construct a comprehensive comfort evaluation model. Finally, external validation was performed on an independent vehicle platform to examine the model’s applicability and consistency. The results show that acoustic comfort decreases as the sound pressure level increases, whereas optical and thermal comfort exhibit nonlinear behavior with optimal intervals. AHP weight results show that the thermal environment has the highest weight (0.4280), followed by the acoustic environment (0.3305) and the optical environment (0.2415). The external validation results indicate that the proposed model exhibits good predictive consistency across three steady-state operating conditions, with a mean absolute error of 0.122, a root-mean-square error of 0.150, and a Pearson correlation coefficient of 0.960. The findings show that the penalty-based aggregation model can effectively characterize the limiting-factor effect under the joint action of multiple environmental factors, providing a computable and interpretable evaluation framework for intelligent cockpit environmental control and automotive engineering experimental teaching. The conclusions of this study are mainly applicable to the current experimental platform and steady-state operating conditions, and further validation is still required with more vehicle models, dynamic road scenarios, and complex multi-environment factor disturbances. Full article

Road freight transport (RFT) faces growing pressure from increasing freight demand, stricter environmental requirements, and persistent driver shortages. Automation technologies (ATes)—especially semi-autonomous driving—are increasingly viewed as a practical pathway toward improving the sustainability performance of freight operations; however, their effects depend strongly on infrastructure and operational conditions. This study evaluates the sustainability potential of autonomous and semi-autonomous trucks through an integrated framework combining (i) a structured review of technical and regulatory developments, (ii) surveys of transport enterprises (TEes) and road users (RUs), (iii) SWOT/TOWS analysis, and (iv) a cost minimization logistics model that links operational feasibility to infrastructure readiness (IR). The proposed model minimizes cost per tonne-kilometre and introduces an Infrastructure Readiness Score (IRS) to represent the share of a route that can be operated in automated mode; it also accounts for fuel savings from platooning and higher maintenance and capital costs of semi-autonomous vehicles (SAVs). Results indicate that, as IRS increases, semi-autonomous operations achieve higher daily mileage and lower unit costs, with a break-even point at approximately IRS ≈ 0.125. Beyond this threshold, unit costs decline from EUR 0.0433 to EUR 0.0348 per tonne-kilometre as IRS rises toward 0.6, after which further infrastructure improvements yield diminishing mileage gains. These cost and utilization improvements imply sustainability benefits via improved energy efficiency and reduced emissions intensity per tonne-kilometre. Nevertheless, survey evidence highlights major adoption barriers, including insufficient IR, regulatory uncertainty, technological reliability concerns, and limited public trust in fully autonomous systems. Overall, the findings support semi-autonomous trucking as the most feasible near-term stage of transition, while emphasizing that infrastructure upgrades and governance mechanisms are critical for scaling sustainability gains. Full article

Deep reinforcement learning often suffers from inefficient exploration, which is commonly addressed by introducing an auxiliary model that assigns intrinsic rewards when the agent encounters novel scenarios. However, such approaches increase training complexity and computational overhead. This paper proposes an Attribution-Guided Reinforcement Learning (AGRL) framework that exploits real-time attribution analysis to guide exploration in autonomous driving decision-making. The proposed method is built upon the Kolmogorov–Arnold-Network-based Interpretable Deep Reinforcement Learning (KAN-IDRL) framework. Specifically, action-wise attribution patterns are computed online, and perturbations are applied to the state inputs to measure attribution sensitivity. The resulting attribution-sensitivity signal identifies actions whose decision rationales are more locally responsive to state changes, and these actions are therefore preferentially explored. In addition, local attribution results collected from a pretrained interpretable policy are aggregated into global feature-importance scores, which are then used to initialize a trainable prior attention gate in a Prior-Attention-Enhanced Kolmogorov–Arnold Network (PAE-KAN). This design allows the policy to incorporate attribution-derived prior knowledge while maintaining sufficient adaptability for task-specific learning. Experiments across multiple autonomous driving scenarios demonstrate that the proposed AGRL framework achieves faster convergence and competitive final performance compared with representative baseline methods. These findings indicate that attribution information can be transformed from a post hoc interpretability tool into an effective guidance signal for improving reinforcement learning. Full article

How non-dominant specialized firms sustain systemic innovation influence in modular service ecosystems without occupying architectural control positions remains theoretically underdeveloped. This study develops a dual-path Resource–Strategy–Capability (R–S–C) mechanism framework to explain how structurally distinct network positions generate divergent innovation trajectories among non-dominant firms. The empirical analysis draws on large-scale patent collaboration network data from China’s autonomous driving industry, covering 26 hidden champion firms and 14 global leading enterprises across 2009–2023. The framework identifies two divergent pathways: firms occupying structural hole positions adopt specialization-deepening strategies that build module-anchoring capabilities, while firms with high betweenness centrality adopt T-shaped strategies that build interface-bridging capabilities—both enabling systemic influence without architectural control. To make the resource construct theoretically precise, the framework distinguishes four categories of network-derived resources operative in the R–S–C mechanism—informational, coordination, reputational, and module-definition resources—and specifies three microfoundational processes through which strategic orientation translates into capability: experiential learning, codification of routines, and legitimation through external recognition. Institutional policy environments moderate these mechanisms by reshaping network structural heterogeneity rather than directly driving firm outcomes. The study challenges the canonical prediction of structural hole theory by demonstrating that brokerage positions generate specialization deepening rather than scope expansion when absorptive capacity constraints are binding, extends service ecosystem theory by introducing non-dominant firm pathways to systemic value co-creation, and reframes institutional policy as a network-structural moderator with transferable implications beyond the Chinese context. Full article

AI and Machine Learning (ML) have advanced rapidly, yet their theoretical underpinnings remain incomplete. We developed an integrated framework combining mathematical theory, uncertainty quantification, and dynamic validation across autonomous platforms such as unmanned aerial vehicles and self-driving cars. We address key challenges in generalization bounds, safety-guaranteed control, and multimodal sensor fusion by exploring the role of Large Language Models (LLMs) in experiment design and teaching material generation. Preliminary simulation and system-level results demonstrate the feasibility of bridging theoretical AI models with real-world engineering systems. The proposed framework aims to provide a reproducible research and teaching platform that fosters interpretable, robust, and certifiable AI applications. Full article

LiDAR is widely used in autonomous driving. Although LiDAR point cloud data can provide stable and reliable information about the environment, it also faces the problem of a huge amount of data. One of the reasons is that point cloud data contains a large amount of noise and outliers. Outlier removal of point clouds can reduce the impact of these disturbances and improve the quality of the point cloud, but it will inevitably eliminate some valid points, which affects subsequent perception tasks. To overcome this limitation, this paper proposes a fuzzy outlier removal (FOR) method based on fuzzy theory and informativeness. It uses fuzzy theory to model the uncertainty of the membership degree of each point in each dimension, calculates the informativeness sum of each point based on membership degree, and filters points according to the informativeness. FOR is characterized by filtering the point cloud in the edge region on the premise of retaining the point cloud in the center region, so as to preserve the environmental information in the center region and reduce the impact of outlier removal on subsequent perception tasks. The experiments focus on the contradictory relationship between outlier removal and perception accuracy, and verify the effectiveness of FOR with multiple object detection models on the autonomous driving datasets KITTI and nuScenes. The experimental results indicate that, compared with other point cloud outlier removal methods, FOR has the advantage of reducing inference time while retaining detection accuracy, demonstrating balanced high performance across different datasets and detection models. Full article