Global Model Research Articles

Integrated Analysis of Utilization Efficiency of Intelligent Sports Venues Based on Data Mining Algorithms

With the rapid development of information technology and the increasing prosperity of the sports industry, the construction and management of intelligent sports venues have become an important way to improve the efficiency of sports resource utilization and meet diverse fitness needs. This paper focuses on the integrated analysis of the utilization efficiency of intelligent sports venues based on data mining algorithms. Aiming to reveal the inherent laws and potential value in the operation of sports venues through advanced data processing and analysis techniques, and provide a scientific basis for optimizing resource allocation and improving service quality. Intelligent sports venues utilize modern information technologies such as the Internet of Things, big data, and cloud computing to achieve comprehensive and intelligent monitoring and management of venue facilities, personnel activities, environmental conditions, and more. This new management model not only improves the operational efficiency of the venue but also enhances the user experience and meets personalized and diverse fitness needs. This paper focuses on the current situation of the opening of university sports venues to the outside world, aiming to reveal the pain points and bottlenecks in their operation process through detailed investigation and analysis. This study introduces advanced concepts of middleware architecture and constructs a multi-level data integration framework that includes local models, global models, user models, and mapping rules. This framework can effectively integrate distributed heterogeneous data sources, achieving seamless integration and efficient interaction of data. At the same time, with the help of distributed programming technology in cloud computing, it is possible to conduct in-depth mining and analysis of integrated massive data, discover potential value, and provide a scientific basis for decision support.

Predicting the habitat suitability and species richness of plants of Great Himalayan National Park under different climate change scenarios.

This study elucidates the distribution of plants in Great Himalayan National Park (GHNP), India, in current and different future climate change scenarios. The distribution of plants and habitat suitability in GHNP due to climate change was analyzed by MaxEnt, species distribution model (SDM) algorithm. In this study, species presence records were retrieved through field survey and published literature. We have projected the distribution of 44 plant species using MaxEnt and tested whether GHNP is performing well in conserving the plant species. Initially, we have constructed a model for each species and created the habitat suitability map from average of ascii files and later we added the maps of all species in order to make binary map to show the species richness in the selected climate scenarios. The model was created using the HADGEM-2 global circulation model in 2050 and 2070years by using climate change situations of RCP 2.6 and RCP 8.5. The area under curve (AUC) values in the final models of 44 plant species were in the range 0.70-0.97 that indicates statistically significant results. The model identified precipitation of driest month followed by altitude and annual mean temperature as most determining variables in the distribution of plants of GHNP in selected climate scenarios. In the present study, we found that overall suitable habitat increased for nine species, decreased for thirty-four, and unchanged for one plant species in terms of percent area change from current to future scenarios. So these nine species were found to be more adaptable towards changing climate than the other plant species in this study. The species richness was high in western and southwestern parts of GHNP in the current scenario, however under future climatic scenarios, species richness shows a decreasing trend. Based on our findings, it can be concluded that GHNP should be prioritized for conserving the plant species.

The contribution of transport emissions to ozone mixing ratios and methane lifetime in 2015 and 2050 in the Shared Socioeconomic Pathways (SSPs)

Abstract. We quantify the contributions of emissions from the transport sector to tropospheric ozone and the hydroxyl radical (OH) by means of model simulations with a global chemistry-climate model equipped with a source attribution method. For the first time we applied a method which also allows for quantifying contributions to OH which is invariant upon disaggregation or recombination and additive. Based on these quantified contributions, we analyse the ozone radiative forcing (RF) and methane lifetime reductions attributable to emissions from the transport sectors. The contributions were analysed for each transport sector separately and for 2015 as well as for 2050 under the Shared Socioeconomic Pathways (SSPs) SSP1-1.9, SSP2-4.5, and SSP3-7.0. In line with previous publications using the source attribution approach, we quantify ozone RF attributable to emissions from land transport, shipping, and aviation for the year 2015 of 121, 60, and 31 mW m−2, respectively. At the same time, we diagnose a relative reduction in methane lifetime due to transport emissions of 14.3 % (land transport), 8.5 % (shipping), and 3.8 % (aviation). These reductions are significantly larger than reported by previous studies due to the application of the source attribution method. Compared to 2015, only SSP1-1.9 shows a strong decrease in ozone RF and methane lifetime reduction attributable to the entire transport sector in 2050. For the projections of SSP2-4.5, we find similar effects of the total transport sector as for 2015, while the effects in SSP3-7.0 increase compared to 2015. This small change in the effects for the two projections compared to 2015 is caused by two main factors. Firstly, aviation emissions are projected to increase in SSP2-4.5 (increase of 107 %) and SSP3-7.0 (+86 %) compared to 2015, resulting in projected ozone RF of 55 mW m−2 (+78 %) and 50 mW m−2 (+61 %) for the year 2050 from aviation emissions. Secondly, the non-linear effects of atmospheric chemistry in polluted regions such as Europe and North America lead to rather small reductions in ozone and OH in response to emission reductions, especially from land transport emissions. In addition, the increase in emissions from land transport in other parts of the world, particularly in South Asia, leads to an increased contribution of ozone and OH. In particular, ozone formed by land transport emissions from South Asia causes strong RF that partially offsets the reductions in Europe and North America. Moreover, our results show that besides the non-linear response, lack of international cooperation, as in the SSP3-7.0 projection, hinders mitigation of ground-level ozone.

Future shifts in climatic conditions promoting northward expansion of the Mediterranean climate in the circum-Mediterranean region.

In the upcoming decades, precipitation and temperature patterns are expected to shift in the Mediterranean basin due to global warming, potentially having an influence on the environment and the economy in the area. Using monthly precipitation and temperature data from 15 global climate models (GCMs) developed as part of the Coupled Model Intercomparison Project Phase 6 (CMIP6), the Mediterranean Climate Envelop (MCE), as defined by Daget's (1977) criteria, is projected under two climate change scenarios: SSP2-4.5 and SSP5-8.5, and for two future periods: 2050s and 2070s. According to the findings, the MCE is expected to expand by 3.51 and 4.93% in the 2050s under the SSP2-4.5 and SSP5-8.5 scenarios, respectively, in comparison to the current state. This expansion is expected to reach 5.28 and 9.87% for SSP2-4.5 and SSP5-8.5, respectively, in 2070s. For both situations and durations, MCE contraction would be minor, however, at less than 1%. More than 99% of the present MCE would stay stable proportionately. The northern Mediterranean region is mostly concerned by the MCE's expansion. The SSP2-4.5 scenario predicts that by the 2070s, expansion zones will occupy 674,183 km2, with 64% of the area located in Southern Europe and 36% in Western Asia. In SSP5-8.5 scenario, this area is expected to be significantly larger, estimated to be approximately 1,256,881 km2; 67% in Southern Europe and 33% in Western Asia.

An Impact-Based Flood Forecasting System for Citizen Empowerment

This work addresses the critical issue of flooding, a significant natural hazard, consistently ranked highest in the 2023 World Risk Index. The annual onslaught of tropical cyclones and the associated abnormal rainfall threaten lives, and destroy crops and property, thereby causing great economic damage, demanding urgent and science-based decision-making. We introduce an impact-based flood forecasting system as a proactive anticipatory action (i.e., AA) measure, linking climate services and disaster risk management to mitigate extreme weather impacts. Developed by the University of the Philippines Resilience Institute, the National Operational Assessment of Hazards (NOAH) Center, and Gerry Bagtasa of the Institute of Environmental Science and Meteorology, this system advances disaster resilience efforts, leveraging science-based forecasts to prioritize vulnerable barangays (villages). Unlike traditional early warning systems, the proposed automated system predicts flooding one day in advance and assesses exposure levels based on population distribution. By utilizing global weather forecast models locally calibrated for Philippine conditions and 100-year rain return flood hazard maps for the Philippines, the system forecasts river inundated areas, enabling local government units (LGUs) and humanitarian organizations to prioritize preparations for communities and allocate resources during flooding events effectively. This system enhances the efficiency of disaster response planning, fostering a proactive and impactful strategy to address the recurring threat of flooding in the country. The system also underscores the role of open data in disaster resilience, exemplified by NOAH’s system to disseminating big data, which aligns with open governance principles. By encouraging transparency and stakeholder collaboration, data exchange is promoted, providing actionable information that fosters collaboration between the LGUs, humanitarian organizations, and other stakeholders. Cognizant of the importance of community engagement in disaster resilience, the newly developed impact-based flood forecasting system encourages community involvement by providing easily accessible information, which can be used to validate the forecasts based on local knowledge and experience. This research contributes to the urgent call for anticipatory action in the face of escalating extreme weather events globally.

Global prescriptions and local adaptations: the role of political culture in shaping governance strategies for university funding

ABSTRACT This article explores the influence of New Public Management (NPM) reforms on local higher education (HE) governance in the context of globalisation, focussing on the funding mechanisms of universities in Hong Kong and Taiwan. Through the development of an analytical framework for governance strategies, the study examines the principles underlying funding allocation and distribution decisions in these regions. The findings show that, although both regions adopt performance-based funding aligned with NPM principles, significant differences emerge due to their distinct political cultures and societal expectations. Taiwan’s strategy is characterised by regulatory oversight driven by societal shifts and electoral pressures, while Hong Kong’s approach is market-oriented, reflecting its British colonial heritage and alignment with industry demands. This study highlights the role of path dependency in shaping localised adaptations of global governance models, demonstrating the diversity and adaptability of HE governance strategies. The paper concludes by proposing directions for further research into the interplay between political culture and HE governance, advocating for the development of context-sensitive frameworks that address evolving governance strategies.

Enhancing 3D planetary atmosphere simulations with a surrogate radiative transfer model

Abstract This work introduces an approach to enhancing the computational efficiency of 3D atmospheric simulations by integrating a machine-learned surrogate model into the OASIS global circulation model (GCM). Traditional GCMs, which are based on repeatedly numerically integrating physical equations governing atmospheric processes across a series of time-steps, are time-intensive, leading to compromises in spatial and temporal resolution of simulations. This research improves upon this limitation, enabling higher resolution simulations within practical timeframes. Speeding up 3D simulations holds significant implications in multiple domains. Firstly, it facilitates the integration of 3D models into exoplanet inference pipelines, allowing for robust characterisation of exoplanets from a previously unseen wealth of data anticipated from JWST and post-JWST instruments. Secondly, acceleration of 3D models will enable higher resolution atmospheric simulations of Earth and Solar System planets, enabling more detailed insights into their atmospheric physics and chemistry. Our method replaces the radiative transfer module in OASIS with a recurrent neural network-based model trained on simulation inputs and outputs. Radiative transfer is typically one of the slowest components of a GCM, thus providing the largest scope for overall model speed-up. The surrogate model was trained and tested on the specific test case of the Venusian atmosphere, to benchmark the utility of this approach in the case of non-terrestrial atmospheres. This approach yields promising results, with the surrogate-integrated GCM demonstrating above 99.0% accuracy and 147 factor GPU speed-up of the entire simulation compared to using the matched original GCM under Venus-like conditions.

Accuracy assessment of global vertical displacement loading tide models for the equatorial and Indian Ocean

Accuracy assessment of global vertical displacement loading tide models for the equatorial and Indian Ocean

Atmospheric Nitrogen Deposition Controls Interannual Variability of Net Primary Production in the Bohai Sea

AbstractIncreasing human activities and climate change have resulted in significant changes in net primary production (NPP) of the marginal sea in the past several decades. The Bohai Sea (BHS) is a shallow semi‐enclosed marginal sea of North China, the knowledge of the relative impacts of human activities and climate change on the NPP in the BHS remains limited. This study primarily aimed to elucidate the interannual variability of NPP and its influencing factors in the BHS during the 2002–2021 period by synthesizing four widely‐used global NPP models, namely the vertically generalized production model (VGPM), the Eppley‐VGPM, the size‐fractioned phytoplankton pigment absorption‐based model (SABPM), and the carbon, absorption, and fluorescence euphotic‐resolving model (CAFE). The multi‐model average NPP showed significant interannual variation, with an increasing trend from 2002 to 2012 and a decreasing trend from 2012 to 2021. The concentration of DIN in the BHS was primarily affected by atmospheric nitrogen deposition. The structural equation model demonstrated that chlorophyll‐a (Chl‐a) concentration, influenced by DIN concentration, had an important effect on NPP. In brief, atmospheric nitrogen deposition impacted the DIN concentration and thus controlled the (Chl‐a) concentration and NPP change in the BHS. Furthermore, atmospheric nitrogen deposition in the BHS have potential ecological impacts on the red tide features. These findings hold valuable implications for future marine resource planning and marine ecological management in the BHS.

SwinInver: 3D data-driven seismic impedance inversion based on Swin Transformer and adversarial training

As deep learning becomes increasingly prevalent in seismic impedance inversion, 3D data-driven approaches have garnered substantial interest. However, two critical challenges persist. First, existing methodologies predominantly rely on Convolutional Neural Networks (CNNs), which, due to the inherent locality of convolutional operations, are inadequate in capturing the global context of seismic data. This limitation notably hinders their performance in inverting complex subsurface structures, such as salt bodies. Second, the current inversion frameworks are prone to overfitting, particularly when trained on limited seismic datasets. To address these challenges, we propose SwinInver, a novel backbone network that integrates the Swin Transformer as its fundamental unit, coupled with a high-resolution network design to facilitate comprehensive global modeling of intricate subsurface structures. Furthermore, we incorporate adversarial training to enhance the inversion process and effectively mitigate overfitting. Experimental evaluations demonstrate that SwinInver significantly surpasses conventional CNN-based approaches in both synthetic and field data scenarios, providing a more accurate and reliable framework for seismic impedance inversion.

Estimation of glacier-stored freshwater volume present in major tributaries of the Brahmaputra basin.

Estimation of the glacier-stored freshwater is important to understand the water security in the Himalayan region. While previous work has studied the western and central Himalayan glaciers, the eastern counterpart received less and more scattered attention. In this study, an attempt is made to quantify the total glacier-stored freshwater in the Brahmaputra basin and later compared with previous global models. Using open-source tools such as COSI-Corr and the Himalayan Glacier Thickness Mapper (HIGTHIM), the surface velocity and thickness of 1075 glaciers (> 1 km2) in the Brahmaputra basin were estimated, resulting in a current ice-volume estimate of 283 × 109 m3. Based on the laminar flow model, the mean ice volume ranges from 8284 to 230,186 m3, with an average of 36,570 m3. Sub-basin-wise evaluations of the total glacier-stored freshwater availability in the basin were also conducted, revealing that the Siang (89.998 × 109 m3, 31.45%) and Lohit (84.371 × 109 m3, 29.49%) sub-basins have significantly larger ice volumes than others. The average mean ice volume for each sub-basin are as follows: Teesta (45,233 m3), Sankosh (45,552 m3), Manas (39,581.7 m3), Subansiri (40,922.4 m3), Kameng (41,241.2 m3), Siang (36,120.5 m3), Dibang (31,792.2 m3), and Lohit (30,340.6 m3). Teesta, Sankosh, and Manas exhibit relatively higher average mean ice volumes than others. In comparison with the global studies, the present study's findings are acceptable with ensemble ice-volume estimates considering an uncertainty of ± 17.35%. Therefore, these results serve as a primary input for assessing the future changes in water resources and hazards related to water in the Brahmaputra basin.

Setting Up Methane Mitigation Measures for Indian Rice Fields: Representative Emissions and New Interpretations

AbstractRice cultivation produces methane (CH4) due to anaerobic conditions induced by flood irrigation, significantly contributing to global warming. While most studies use national emission factors (EFs), our study synthesized 726 published measurements across India (the second largest methane emitter after China) to develop district‐level water regime‐specific EFs for estimating district‐scale emissions and warming potential. CH4 emissions from Indian rice fields increased from 3.7 (3.4–4.1) Tg to 4.8 (4.4–5.3) Tg during 1966–2017, driven by rice area and water‐regime variations. Meanwhile, district‐level emissions increased by ∼930%, influenced by management practices such as animal manure, fertilizer application, and water input, accurately reflecting regional variations compared to previous estimates. Employing a novel muti‐output random forest mitigation model (R2 ∼ 0.9), we found that a 25% warming reduction at the district‐level requires curtailing animal manure, nitrogen fertilizer, and water input by 8.5%, 12.9%, and 10.9%, respectively. These curtailments nearly double for a 50% mitigation scenario. Comparing our emissions with previous bottom‐up studies (used as inputs in global climate models) revealed discrepancies in prior national figures. With top‐down estimates, our emissions correlated positively, suggesting higher reliability. Including our new regionally validated data in global climate models may provide more accurate climate projections at the Indian and global scales.

Federated learning-based wavelength demodulation system for multi-point distributed multi-peak FBG sensors

Machine learning-based demodulation of multi-peak fiber Bragg grating (FBG) sensors has been extensively studied, demonstrating superior performance compared to conventional algorithms because it can neglect potential physical constraints. As the number of real-world monitoring points increases, the volume of fiber-optic sensing data grows exponentially. This necessitates aggregating data from various regions (e.g., via Wi-Fi), unlike traditional single-point monitoring, which challenges server storage capacity and communication efficiency. To address these issues, this paper proposes a federated learning (FL)-based framework for efficient wavelength demodulation of multi-peak FBGs in multipoint monitoring. Specifically, an arrayed waveguide grating (AWG) with multiplexing capability is employed at different monitoring points to convert spectral features into multi-channel transmission intensities, serving as training data for local models. Subsequently, the local model parameters, trained independently at each point, are uploaded to a central server to derive the optimal global model for demodulation across different monitoring points. The proposed demodulation framework is validated through stress demodulation experiments on multi-peak FBG sensors. Experimental results indicate strong multi-peak extraction performance with a demodulation error of ±0.64 pm. Additionally, the method demonstrates excellent applicability for generating high-precision global demodulation models through multi-node cooperative work under various scenarios.

Oceanic geodiversity along back-arc spreading centers reveals analogies with mid-ocean ridges

Oceanic geodiversity provides essential information on the dynamics of the Earth. Here, we focus on the geodiversity of three oceanic back-arc spreading centers: the Mariana Spreading Center, the Central-Southern Lau Basin spreading centers, and the East Scotia Ridge. We defined a method to identify their axial zones, obtaining spreading center depths along the basins. Results improve global plate boundary models and morphology variations, revealing that the average depths along the Mariana, East Scotia, and Lau Basin spreading ridges are 4.5, 3.5, and 2 km, respectively. We also measured new spreading rates based on five magnetic profiles crossing the three back-arc spreading centers, contributing to plate kinematic models. Furthermore, we computed subduction rates, including hinge velocities along the Mariana, South Sandwich, and Tonga Subductions, to understand the existing interactions between the subducting slab hinge motion and the kinematics of their related back-arc spreading centers. Our bathymetric, magnetic, and kinematic data show several differences among the Mariana, the East Scotia, and the Lau spreading centers, stressing the oceanic geodiversity in a similar geodynamic context. Our results also suggest a strong correlation between axial depth and full spreading rates along the back-arc spreading centers, a geological correspondence that allows a similar description of these divergent plate boundaries within the mid-ocean ridge classification. Finally, we show how hinge kinematics affects the relationship between convergence along subduction zones and back-arc spreading rates. All our findings contribute to understand how the oceanic geodiversity is directly related to geodynamic processes, increasing the knowledge of global tectonics.

Sexual Dimorphism in the Immunometabolic Role of Gpr183 in Mice

Abstract Aim Excessive eating and intake of a Western diet negatively affect the intestinal immune system, resulting in compromised glucose homeostasis and lower gut bacterial diversity. The G protein-coupled receptor GPR183 regulates immune cell migration and intestinal immune response and has been associated with tuberculosis, type 1 diabetes, and inflammatory bowel diseases. We hypothesized that with these implications, GPR183 has an important immunometabolic role and investigated this using a global Gpr183 knock-out mouse model. Material and methods Wild-type (WT) and Gpr183 deficient (Gpr183-/-) mice were fed a high-fat, high-sucrose diet (HFSD) for 15 weeks. We investigated changes in weight, body composition, fecal IgA levels, fecal microbiome, and glucose tolerance before and after the diet. Macrophage infiltration into visceral fat was determined by flow cytometry, and hepatic gene expression was measured. Results and conclusions A sexual dimorphism was discovered, where female Gpr183-/- mice showed adverse metabolic outcomes compared to WT counterparts with inferior glucose tolerance, lower fecal IgA levels, and increased macrophage infiltration in visceral fat. In contrast, male Gpr183-/- mice had significantly lower fasting blood glucose after diet than male WT mice. Liver gene expression showed reduced inflammation and macrophage markers in Gpr183-/- livers, regardless of sex, while the pancreatic islet area did not differ between the groups. No conclusive differences were found after microbiome sequencing. To conclude, Gpr183 maintains metabolic homeostasis in female but not in male mice independent of diet. If confirmed in humans, future therapy targeting GPR183 should consider this sexual dimorphism.

The Schatten current sheet

Space weather models endeavoring to connect remote observations to in-situ measurements at various locations in the heliosphere invariably require a coronal model to connect the photosphere magnetically to the inner heliosphere. The most famous and popular implementation of this connection is a potential field source surface (PFSS) model out to the source surface, typically located at 2.5 solar radii, combined with a Schatten current sheet (SCS) model. While the PFSS model is mostly understood, the SCS has been utilized in heliospheric physics for nearly 50 years with little understanding of it’s physical and mathematical underpinnings. In this overview article, I lay out the mathematical formalism of the SCS, describe how it differs from the PFSS, and summarize several techniques used to combine the PFSS and SCS to create a global coronal model from the photosphere to the inner heliosphere.

Reducing the Uncertainty in the Tropical Precipitation through a Multi‐Criteria Decision‐Making Approach

ABSTRACTThe inherent model uncertainty in precipitation projections is found to be more dominant over tropical regions thereby reducing the reliability of using them in climate change impact assessment studies. To address such issues, a subset of well performing global climate models (GCMs) can provide narrow range of possible future outcomes, which can be helpful in formulating mitigation and adaptation strategies that are more targeted and efficient. In this study, climate models are selected based on their performance in simulating relative humidity and vertical velocity since these variables play an important role in precipitation simulation and significantly contribute toward the intermodel spread. The models are evaluated by using various statistical performance measures and ranked using multi‐criteria decision‐making approaches. Finally, based on Jenks natural breaks optimization algorithm, subset of GCMs consisting of ACCESS1.0, ACCESS1.3 and INM‐CM4 models, are considered as the best possible subset for precipitation simulation over tropical land regions. Two observational precipitation datasets are further considered to investigate the effectiveness of the proposed framework. The proposed methodology is validated to be effective in identifying the best climate models since the resulting subset is capable of both capturing observed precipitation and minimizing the uncertainty in future projections. Hence, this methodology can be utilized further for performance evaluation of GCMs focusing different geography and climatic drivers.

Competency mapping among Rashtriya Bal Swasthya Karyakram nurses in Kerala

Background: In India, birth defects, nutritional deficiencies, disabilities, and developmental delays among children cause significant public health challenges. The Ministry of Health and Family Welfare, aims to address these issues through the Rashtriya Bal Swasthya Karyakram (RBSK) program. RBSK nurses are the crucial part of this program. They screen the children between zero to eighteen years as a part of early intervention. This study mainly aimed to map the competencies among RBSK nurses from randomly selected districts of Kerala. Methods: This qualitative study employed key informant interviews with the state nodal officer, district early intervention centre managers, and district coordinators; and in-depth interviews (IDIs) with RBSK nurses, as well as report and document review. Inductive coding was applied to two selected interview transcripts from each district, followed by deductive coding with the rest of the transcribed data to derive themes. Results: The findings are presented using the ‘core competencies of public health professionals’ as revised and adopted by the council on linkages between academia and public health practice in 2021. RBSK nurses in Kerala exhibited proficiency in multiple domains, including knowledge of child health issues, clinical skills, communication, patient education, care for vulnerable population, organizing activities, electronic health record usage, multitasking, teamwork, and problem-solving skills. Conclusions: This dedication and proficiency demonstrated by RBSK nurses in Kerala, serve as a global model for emphasising the importance of early intervention and comprehensive care for children.

Enhancing federated learning with dynamic weight adjustment based on particle swarm optimization

Federated learning (FL) stands as a promising distributed machine learning approach today, allowing model training on local clients without data sharing. However, varied contributions among clients often impact the global model’s performance, while FL confronts challenges like communication overhead, data heterogeneity, and privacy concerns. In this paper, we introduce a novel federated learning server aggregation algorithm: the Federated Learning Algorithm with Optimized Weight Aggregation via Particle Swarm Optimization Algorithm (AdpFedPSO). This approach dynamically adjusts client model contribution weights based on their performance and stability, aiming to enhance global model accuracy, convergence speed, and make model aggregation smarter and more adaptable. Through experimental validation on real datasets, we find that AdpFedPSO enhances accuracy by about 15% on the 0.6-Dirichlet MNIST dataset, 7.3% on the FashionMNIST dataset, and 13.4% on the CIFAR-10 dataset compared to the traditional FL aggregation method, FedAvg. These results indicate that AdpFedPSO not only enhances the accuracy of the global model but also expedites the model’s convergence speed. Additionally, it demonstrates resilience across various levels of client numbers, participation rates, and client heterogeneity, providing valuable reference and guidance for the further development of FL technology. Moreover, the concept of employing the particle swarm optimization algorithm for FL model aggregation also offers new insights and directions for future research.

Integrated Navigation Method for Orchard-Dosing Robot Based on LiDAR/IMU/GNSS

To enhance the localization reliability and obstacle avoidance performance of the dosing robot in complex orchards, this study proposed an integrated navigation method using LiDAR, IMU, and GNSS. Firstly, the tightly coupled LIO-SAM algorithm was used to construct an orchard grid map for path planning and obstacle avoidance. Then, a global localization model based on RTK-GNSS was developed to achieve accurate and efficient initial localization of the robot’s coordinates and heading, and a Kalman filter was applied to integrate GNSS and IMU to improve robustness. Next, an improved A* algorithm was introduced to ensure the global operational path maintained a safe distance from obstacles, while the DWA algorithm handled dynamic obstacle avoidance. Field tests showed that the global localization model achieved an accuracy of 2.215 cm, with a standard deviation of 1 cm, demonstrating stable positioning performance. Moreover, the global path maintained an average safe distance of 50.75 cm from the obstacle map. And the robot exhibited a maximum absolute lateral deviation of 9.82 cm, with an average of 4.16 cm, while maintaining a safe distance of 1 m from dynamic obstacles. Overall, the robot demonstrated smooth and reliable autonomous navigation, successfully completing its tasks.