Search Results (56)

Accurate 3D object detection is crucial for autonomous vehicles (AVs) to navigate safely in complex environments. This paper introduces a novel fusion framework that integrates Camera image-based 2D object detection using YOLOv8 and LiDAR data-based 3D object detection using PointPillars, hence named C2L3-Fusion. Unlike conventional fusion approaches, which often struggle with feature misalignment, C2L3-Fusion enhances spatial consistency and multi-level feature aggregation, significantly improving detection accuracy. Our method outperforms state-of-the-art approaches such as YoPi-CLOCs Fusion Network, standalone YOLOv8, and standalone PointPillars, achieving mean Average Precision (mAP) scores of 89.91% (easy), 79.26% (moderate), and 78.01% (hard) on the KITTI dataset. Successfully implemented on the Nvidia Jetson AGX Xavier embedded platform, C2L3-Fusion maintains real-time performance while enhancing robustness, making it highly suitable for self-driving vehicles. This paper details the methodology, mathematical formulations, algorithmic advancements, and real-world testing of C2L3-Fusion, offering a comprehensive solution for 3D object detection in autonomous navigation. Full article

Purpose: Because of shared mechanisms such as decreased access to health care, rurality and poverty may act synergistically to decrease colorectal cancer (CRC) survival. Methods: We conducted a retrospective cohort analysis of SEER data (22 registries) with census tract-level measures of poverty/rurality for the period 2006–2015. Multivariable Cox proportional hazard regressions were applied to examine the independent and intersectional associations of persistent poverty and rurality on 5-year cause-specific CRC survival across five racial/ethnic groups. Results: Among 532,868 CRC patients, non-Hispanic Blacks (NHB) demonstrated lower 5-year survival probability (64.2% vs. 68.3% in non-Hispanic Whites [NHW], 66.5% in American Indian/Alaska Natives [AI/AN], 72.1% in Asian/Pacific Islanders, and 68.7% in Hispanic groups) (p-value < 0.001). In adjusted analysis, CRC patients living in rural areas with poverty were at a 1.2–1.6-fold increased risk of CRC death than those who did not live in these areas in five racial/ethnic groups. In particular, AI/AN patients living in rural areas with poverty were 66% more likely to die from CRC (95% CI, 1.32, 2.08). Conclusions: CRC patients who live in rural or poverty areas in SEER areas in the U.S. have a poorer survival compared with those who do not live in such areas regardless of race/ethnicity. Significantly greater risk of CRC death was observed in AI/ANs. Impact: Patient navigators, community education or screening, and other health care system interventions may be helpful to address these disparities by socioeconomic status, race, and geographic residence. Multi-level interventions aimed at institutional racism and medical mistrust may also be helpful. Full article

Background: Lung cancer is the leading cause of cancer-related mortality in the United States, which disproportionately affect racial and ethnic minorities. Disparities in lung cancer screening, diagnosis, treatment, and survival outcomes are due to a complex interplay of socioeconomic factors, structural racism, and limited access to high-quality care. This review aims to examine the underlying causes of these disparities and explore potential mitigation strategies to improve lung cancer care equity. Methods: A review of the literature was conducted, evaluating racial and ethnic disparities in lung cancer care. Disparities in lung cancer screening, genomic testing, surgical and systemic treatment, and survival were explored. Additionally, interventional strategies such as risk-based screening, patient navigation programs, and policy reforms were examined. Results: Racial and ethnic minority patients are diagnosed at younger ages with fewer pack-years yet are less likely to qualify for screening under current guidelines. They receive lower rates of guideline-concordant treatment, including surgery, radiation, chemotherapy, and biomarker testing, and have reduced access to specialty care. Socioeconomic barriers, medical mistrust, and geographic disparities further contribute to these inequities. Targeted interventions, including mobile screening programs, financial assistance initiatives, and culturally competent care, have shown promise in improving lung cancer outcomes. Conclusion: A multi-level approach, incorporating healthcare policy changes, improved screening criteria, and an enhanced community engagement strategy, is essential for achieving equitable lung cancer care, ultimately improving outcomes for racial minority populations. Full article

Stripe-like space target detection (SSTD) plays a crucial role in advancing space situational awareness, enabling missions like satellite navigation and debris monitoring. Existing unsupervised methods often falter in low signal-to-noise ratio (SNR) conditions, while fully supervised approaches require extensive and labor-intensive pixel-level annotations. To address these limitations, this paper introduces MRSA-Net, a novel encoder-decoder network specifically designed for SSTD. MRSA-Net incorporates multi-receptive field processing and multi-level feature fusion to effectively extract features of variable and low-SNR stripe-like targets. Building upon this, we propose the Collaborative Static-Dynamic Teaching (CSDT) architecture, a semi-supervised learning architecture that reduces reliance on labeled data by leveraging both static and dynamic teacher models. The framework uses the straight-line prior of stripe-like targets to customize linearity and presents an innovative Adaptive Pseudo-Labeling (APL) strategy, dynamically selecting high-quality pseudo-labels to enhance the student model’s learning process. Extensive experiments on AstroStripeSet and other real-world datasets demonstrate that the CSDT framework achieves state-of-the-art performance in SSTD. Using just 1/16 of the labeled data, CSDT outperforms the second-best Interactive Self-Training Mean Teacher (ISMT) method by 2.64% in mean Intersection over Union (mIoU) and 4.5% in detection rate ($P_d$), while exhibiting strong generalization in unseen scenarios. This work marks the first application of semi-supervised learning techniques to SSTD, offering a flexible and scalable solution for challenging space imaging tasks. Full article

Monitoring the stability and safety of climbing scaffolds in super-high-rise construction is critical to ensuring construction quality and worker safety. This study proposes a Global Navigation Satellite System (GNSS)-based real-time monitoring method to track scaffold displacement and assess structural performance. A multi-level data optimization framework integrating gross error elimination, data interpolation, robust Kalman filtering, and a Cumulative Sum Control Chart (CUSUM)-based early warning system is developed to enhance monitoring accuracy. The key objectives of this research are to improve real-time displacement tracking, suppress measurement noise, and establish an automated anomaly detection mechanism for climbing scaffolds under complex construction conditions. The proposed method was validated in a super-high-rise construction project in Tianjin, China. Experimental results demonstrated that the system effectively reduced high-frequency noise and gross errors, achieving root mean square error (RMSE) reductions of 51.4% in the E direction, 45.5% in the N direction, and 49.6% in the U direction. The system successfully tracked vertical climbing displacements of 4.4 m per ascent and horizontal deviations of 4 cm (E direction) and 2 cm (N direction). Additionally, the multi-level warning mechanism identified displacement anomalies based on predefined thresholds, providing an early warning function to enhance scaffold safety management. Compared to conventional monitoring methods, the proposed BeiDou/GNSS-based system provides higher precision, real-time adaptability, and enhanced automation, offering a scalable solution for intelligent construction safety management. The findings contribute to structural health monitoring (SHM) applications and can serve as a reference for future high-rise construction safety assessments. Full article

LiDAR-based place recognition (LPR) is crucial for the navigation and localization of autonomous vehicles and mobile robots in large-scale outdoor environments and plays a critical role in loop closure detection for simultaneous localization and mapping (SLAM). Existing LPR methods, which utilize 2D bird’s-eye view (BEV) projections of 3D point clouds, achieve competitive performance in efficiency and recognition accuracy. However, these methods often struggle with capturing global contextual information and maintaining robustness to viewpoint variations. To address these challenges, we propose R2SCAT-LPR, a novel, transformer-based model that leverages self-attention and cross-attention mechanisms to extract rotation-robust place feature descriptors from BEV images. R2SCAT-LPR consists of three core modules: (1) R2MPFE, which employs weight-shared cascaded multi-head self-attention (MHSA) to extract multi-level spatial contextual patch features from both the original BEV image and its randomly rotated counterpart; (2) DSCA, which integrates dual-branch self-attention and multi-head cross-attention (MHCA) to capture intrinsic correspondences between multi-level patch features before and after rotation, enhancing the extraction of rotation-robust local features; and (3) a combined NetVLAD module, which aggregates patch features from both the original feature space and the rotated interaction space into a compact and viewpoint-robust global descriptor. Extensive experiments conducted on the KITTI and NCLT datasets validate the effectiveness of the proposed model, demonstrating its robustness to rotation variations and its generalization ability across diverse scenes and LiDAR sensors types. Furthermore, we evaluate the generalization performance and computational efficiency of R2SCAT-LPR on our self-constructed OffRoad-LPR dataset for off-road autonomous driving, verifying its deployability on resource-constrained platforms. Full article

In the age of digital twins, the digitalization of the urban environment is one of the key aspects in the optimization of urban management. The goal of urban digitalization is to provide a digital representation of physical infrastructure, data, information, and procedures for the management of complex anthropogenic systems. To meet this new goal, one must be able to understand the urban system through the integrated use of different methods in a multi-level approach. In this context, mobile surveying is a consolidated method for data collection in urban environments. A recent innovation, the mobile mapping system (MMS), is a versatile tool used to collect geospatial data efficiently, accurately, and quickly, with reduced time and costs compared to traditional survey methods. This system combines various technologies such as GNSS (global navigation satellite system), IMU (inertial measurement unit), LiDAR (light detection and ranging), and high-resolution cameras to map and create three-dimensional models of the surrounding environment. The aim of this study was to analyze the limitations, possible implementations, and the state of the art of MMSs for road infrastructure monitoring in order to create a DT (digital twin) for road infrastructure management, with a specific focus on extracting value-added information from a survey dataset. The case study presented here was part of the Turin Digital Twin project. In this context, an MMS was tested in a specific area to evaluate its potential and integration with other data sources, adhering to the multi-level and multi-sensor approach of the DT project. A key outcome of this work was the integration of the extracted information into a comprehensive geodatabase, transforming raw geospatial data into a structured tool that supports predictive maintenance and strategic road asset management toward DT implementation. Full article

Low-Earth-orbit (LEO) satellites have unique advantages in communication, navigation, and remote sensing due to their low orbit, strong landing signal strength, and low launch cost. However, the optimization of the design of LEO constellations to obtain the optimal configuration to meet different missions faces great challenges. Traditional multi-objective optimization algorithms often struggle with designing constellations involving composite functions due to various constraints, which can result in premature termination and local optimality issues. This paper introduces a dynamic parameter-based non-dominated sorting differential evolution (D-NSDE) algorithm to obtian better solutions, which is capable of dynamically adjusting the boundary of feasible solutions and modifying operators according to the iteration process to mitigate these constraints. Additionally, we model a composite LEO constellation with multiple layers, constructing 2-/3-/4-layer configurations, and we include constraints from the third-generation BeiDou Navigation Satellite System (BDS-3) navigation constellations. Subsequently, we employ the D-NSDE algorithm to solve the corresponding multi-objective optimization problems and derive the optimal solution set. The results demonstrate that D-NSDE can generate complete and multi-level solution sets under diverse constraint conditions, with 75% of D-NSDE algorithm optimization solutions being able to achieve seamless positioning for 95% of global coverage. Furthermore, the PDOP median values are 5.12/4.23/2.97 without BDS-3 navigation constraints and 1.38/1.44/1.51 with BDS-3 navigation constraints. Additionally, simulation experiments conducted on standard function test sets confirm that the solution sets produced by the D-NSDE algorithm exhibit favorable distribution and convergence performance better than the Non-dominated Sorting Genetic Algorithm (NSGA)-III. Full article

The focus of this study is to investigate the underexplored operational effects of disengagements on the speed of an automated shuttle, providing novel insights into their disruptive impact on performance metrics. For this purpose, global positioning system data, disengagement records, weather reports, and roadway geometry data from an automated shuttle pilot program, from July to December 2023, at the University of North Carolina in Charlotte, were collected. The automated shuttle uses sensors for localization, navigation, and obstacle detection. A multi-level mixed-effects Gaussian regression model with a log-link function was employed to analyze the effect of disengagement events on the automated shuttle speed, while accounting for control variables such as roadway geometry, weather conditions, time-of-the-day, day-of-the-week, and number of intermediate stops. When these variables are controlled, disengagements significantly reduce the automated shuttle speed, with the expected log of speed decreasing by 0.803 units during such events. This reduction underscores the disruptive impact of disengagements on the automated shuttle’s performance. The analysis revealed substantial variability in the effect of disengagements across different route segments, suggesting that certain segments, likely due to varying traffic conditions, road geometries, and traffic control characteristics, pose greater challenges for autonomous navigation. By employing a multi-level mixed-effects model, this study provides a robust framework for quantifying the operational impact of disengagements. The findings serve as vital insights for advancing the reliability and safety of autonomous systems through targeted improvements in technology and infrastructure. Full article

Background/Objectives: EnBloc resections of bone tumors of the spine are very demanding as the target to achieve a tumor-free margin specimen (sometimes impossible due to the extracompartimental tumor extension) is sometimes conflicting with the integrity of neurological functions and spine stability. Methods: The surgical treatment of a huge multi-level chordoma of the thoracic spine with unusual extension is reported. Anteriorly, the tumor widely invaded the mediastinum and displaced the aorta; on the left side, it expanded in the subpleuric region; posteriorly, it was uncommonly distant 13 mm from the posterior wall. Results: EnBloc resection is largely performed for primary bone tumors of the spine and many reports have been published concerning brilliant solutions to difficult issues of surgical anatomy. One of the major challenges is still the compatibility between oncological and functional requirements. Conclusions: Oncological staging, careful imaging analysis, a multidisciplinary surgical team, and utilization of the most recent technologies like navigation and robotics have made an oncologically appropriate EnBloc resection of a multi-level chordoma of the thoracic spine possible without affecting the continuity of the spinal canal and without any involvement of its content by an original “L”-shaped osteotomy. Full article

The integration of artificial intelligence into daily life significantly enhances the autonomy and quality of life of visually impaired individuals. This paper introduces the Visual Impairment Spatial Awareness (VISA) system, designed to holistically assist visually impaired users in indoor activities through a structured, multi-level approach. At the foundational level, the system employs augmented reality (AR) markers for indoor positioning, neural networks for advanced object detection and tracking, and depth information for precise object localization. At the intermediate level, it integrates data from these technologies to aid in complex navigational tasks such as obstacle avoidance and pathfinding. The advanced level synthesizes these capabilities to enhance spatial awareness, enabling users to navigate complex environments and locate specific items. The VISA system exhibits an efficient human–machine interface (HMI), incorporating text-to-speech and speech-to-text technologies for natural and intuitive communication. Evaluations in simulated real-world environments demonstrate that the system allows users to interact naturally and with minimal effort. Our experimental results confirm that the VISA system efficiently assists visually impaired users in indoor navigation, object detection and localization, and label and text recognition, thereby significantly enhancing their spatial awareness and independence. Full article

Legged climbing unmanned ground vehicles (LC-UGVs) possess obstacle avoidance and wall transition capabilities, allowing them to move in 3D environments. Existing navigation methods for legged UGVs are only suitable for ground locomotion rather than 3D space. Although some wall transition methods have been proposed, they are specific to certain legged structures and have not been integrated into the navigation framework in full 3D environments. The planning of collision-free and accessible paths for legged climbing UGVs with any configuration in a 3D environment remains an open problem. This paper proposes a map representation suitable for the navigation planning of LC-UGVs in 3D space, named the Multi-Level Elevation Map (MLEM). Based on this map representation, we propose a universal hierarchical planning architecture. A global planner is applied to rapidly find cross-plane topological paths, and then a local planner and a motion generator based on motion primitives produces accessible paths and continuous motion trajectories. The hierarchical planning architecture equips the LC-UGVs with the ability to transition between different walls, thereby allowing them to navigate through challenging 3D environments. Full article

This article delineates the enhancement of an autonomous navigation and obstacle avoidance system for a quadruped robot dog. Part one of this paper presents the integration of a sophisticated multi-level dynamic control framework, utilizing Model Predictive Control (MPC) and Whole-Body Control (WBC) from MIT Cheetah. The system employs an Intel RealSense D435i depth camera for depth vision-based navigation, which enables high-fidelity 3D environmental mapping and real-time path planning. A significant innovation is the customization of the EGO-Planner to optimize trajectory planning in dynamically changing terrains, coupled with the implementation of a multi-body dynamics model that significantly improves the robot’s stability and maneuverability across various surfaces. The experimental results show that the RGB-D system exhibits superior velocity stability and trajectory accuracy to the SLAM system, with a 20% reduction in the cumulative velocity error and a 10% improvement in path tracking precision. The experimental results also show that the RGB-D system achieves smoother navigation, requiring 15% fewer iterations for path planning, and a 30% faster success rate recovery in challenging environments. The successful application of these technologies in simulated urban disaster scenarios suggests promising future applications in emergency response and complex urban environments. Part two of this paper presents the development of a robust path planning algorithm for a robot dog on a rough terrain based on attached binocular vision navigation. We use a commercial-of-the-shelf (COTS) robot dog. An optical CCD binocular vision dynamic tracking system is used to provide environment information. Likewise, the pose and posture of the robot dog are obtained from the robot’s own sensors, and a kinematics model is established. Then, a binocular vision tracking method is developed to determine the optimal path, provide a proposal (commands to actuators) of the position and posture of the bionic robot, and achieve stable motion on tough terrains. The terrain is assumed to be a gentle uneven terrain to begin with and subsequently proceeds to a more rough surface. This work consists of four steps: (1) pose and position data are acquired from the robot dog’s own inertial sensors, (2) terrain and environment information is input from onboard cameras, (3) information is fused (integrated), and (4) path planning and motion control proposals are made. Ultimately, this work provides a robust framework for future developments in the vision-based navigation and control of quadruped robots, offering potential solutions for navigating complex and dynamic terrains. Full article

The complexity and multi-dimensionality of energy transitions are broadly recognised, and insights from transition research increasingly support policy decision making. Sustainability transition scholars have been developing mostly qualitative socio-technical transition (STT) frameworks, and modelling has been argued to be complementary to these frameworks, for example for policy testing. We systematically evaluate five system dynamics (SD) energy models on their representation of key STT characteristics. Our results demonstrate that (i) the evaluated models incorporate most of the core characteristics of STT, and (ii) the policies tested in the models address different levels and aspects of the multi-level perspective (MLP) framework. In light of the increasing emergence of energy (transition) models, we recommend to systematically map models and their tested policy interventions into the MLP framework or other sustainability transition frameworks, creating an overview of tested policies (a “policy navigator”). This navigator supports policy makers and modellers alike, facilitating them to find previously tested policy options and related models for particular policy objectives. Full article

Perinatal mental health is a growing public health concern. Refugee and asylum-seeking women are particularly susceptible to experiencing perinatal mental illness and may encounter a range of challenges in accessing healthcare. This scoping review sought to identify the enablers of and barriers to healthcare access and healthcare provision for refugee and asylum-seeking women experiencing perinatal mental illness in the WHO European Region. The Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews was applied. Nine databases and six grey literature sources were initially searched in April 2022, and an updated search was completed in July 2023. The search identified 16,130 records, and after the removal of duplicates and the screening process, 18 sources of evidence were included in this review. A data extraction table was used to extract significant information from each individual source of evidence, which was then mapped to the seven dimensions of the candidacy framework. Empirical (n = 14; 77.8%) and non-empirical (n = 4; 22.2%) sources of evidence were included. The literature originated from seven countries within the WHO European Region, including the United Kingdom (n = 9; 50%), Germany (n = 3; 16.7%), Denmark (n = 2; 11.2%), Norway (n = 1; 5.6%), Greece (n = 1; 5.6%), Sweden (n = 1; 5.6%), and Switzerland (n = 1; 5.6%). The results indicate that, although enablers and barriers were apparent throughout the seven dimensions of candidacy, barriers and impeding factors were more frequently reported. There was also a notable overall lack of reported enablers at the system level. Unaddressed language barriers and lack of attention to the diversity in culturally informed perceptions of perinatal mental illness were the main barriers at the individual level (micro-level) to identifying candidacy, navigating healthcare systems, and asserting the need for care. The lack of culturally appropriate alignment of healthcare services was the key organizational (meso-level) barrier identified. The wider structural and political contexts (macro-level factors), such as lack of funding for consultation time, focus on Western diagnostic and management criteria, and lack of services that adequately respond to the needs of refugee and asylum-seeking women, negatively influenced the operating conditions and wider production of candidacy. It can be concluded that there are multilevel and interconnected complexities influencing access to and provision of perinatal mental healthcare for refugee and asylum-seeking women. Full article