目的/意义探索医学信息学跨学科主题演化路径识别方法,为该领域跨学科研究布局与科研管理提供参考。方法/过程首先利用BERTopic和大语言模型(large language model,LLM)在获取的医学信息学文献中识别全局主题和阶段主题;然后基于主题影...目的/意义探索医学信息学跨学科主题演化路径识别方法,为该领域跨学科研究布局与科研管理提供参考。方法/过程首先利用BERTopic和大语言模型(large language model,LLM)在获取的医学信息学文献中识别全局主题和阶段主题;然后基于主题影响力与学科多样性二维分析框架,确定跨学科主题;最后分析跨学科主题的演化路径。结果/结论基于Topic-LLM的分析方法,发现医学信息学跨学科主题的学科多样性和主题影响力呈现稳步增长趋势。展开更多
目的/意义构建双层分析框架,全面把握学科结构,识别新兴前沿领域,追踪主题演化。方法/过程检索2016—2025年PubMed、Scopus和Web of Science数据库医学信息学文献,采用BERTopic识别主题,并划分为新兴、稳定、衰退3种演化模式。基于Chrom...目的/意义构建双层分析框架,全面把握学科结构,识别新兴前沿领域,追踪主题演化。方法/过程检索2016—2025年PubMed、Scopus和Web of Science数据库医学信息学文献,采用BERTopic识别主题,并划分为新兴、稳定、衰退3种演化模式。基于ChromaDB构建检索增强生成系统,通过文档-主题映射实现微观验证与知识关联挖掘。结果/结论医学信息学主题演化呈现研究重心转移、技术融合深化、学科交叉增强3个特征。BERTopic-RAG框架为知识发现提供了新方法。展开更多
生成式人工智能正深刻变革教育,相关研究文献不断涌现。文章通过收集知网、Scopus和Web of Science中与研究主题相关的中英文文献,利用BERTopic主题建模技术对文献进行主题聚类,挖掘国内外高等教育中生成式人工智能的研究主题,分析研究...生成式人工智能正深刻变革教育,相关研究文献不断涌现。文章通过收集知网、Scopus和Web of Science中与研究主题相关的中英文文献,利用BERTopic主题建模技术对文献进行主题聚类,挖掘国内外高等教育中生成式人工智能的研究主题,分析研究现状,为教育领域的教学实践和学术研究提供参考。研究结果表明:在个性化学习与人机协同方面,生成式人工智能通过制定个性化学习路径显著提升教学效果;师生对生成式人工智能的接受程度受技术认知和使用体验等因素制约,并直接影响应用成效;教育创新需平衡技术赋能与过度依赖问题;教学设计与课程实践的革新更多体现在教学模式和资源生成方式上,但必须同步构建学术伦理防护与治理机制;人工智能时代亟须人才培养结构的优化升级,要求教育者重新定义核心能力目标。展开更多
Q-learning is a classical reinforcement learning method with broad applicability.It can respond effectively to environmental changes and provide flexible strategies,making it suitable for solving robot path-planning p...Q-learning is a classical reinforcement learning method with broad applicability.It can respond effectively to environmental changes and provide flexible strategies,making it suitable for solving robot path-planning problems.However,Q-learning faces challenges in search and update efficiency.To address these issues,we propose an improved Q-learning(IQL)algorithm.We use an enhanced Ant Colony Optimization(ACO)algorithmto optimizeQtable initialization.We also introduce the UCH mechanism to refine the reward function and overcome the exploration dilemma.The IQL algorithm is extensively tested in three grid environments of different scales.The results validate the accuracy of themethod and demonstrate superior path-planning performance compared to traditional approaches.The algorithm reduces the number of trials required for convergence,improves learning efficiency,and enables faster adaptation to environmental changes.It also enhances stability and accuracy by reducing the standard deviation of trials to zero.On grid maps of different sizes,IQL achieves higher expected returns.Compared with the original Q-learning algorithm,IQL improves performance by 12.95%,18.28%,and 7.98% on 10*10,20*20,and 30*30 maps,respectively.The proposed algorithm has promising applications in robotics,path planning,intelligent transportation,aerospace,and game development.展开更多
With the expanding applications of unmanned aerial vehicles(UAVs),precise flight evaluation has emerged as a critical enabler for efficient path planning,directly impacting operational performance and safety.Tradition...With the expanding applications of unmanned aerial vehicles(UAVs),precise flight evaluation has emerged as a critical enabler for efficient path planning,directly impacting operational performance and safety.Traditional path planning algorithms typically combine Dubins curves with local optimization to minimize trajectory length under 3D spatial constraints.However,these methods often overlook the correlation between pilot control quality and UAV flight dynamics,limiting their adaptability in complex scenarios.In this paper,we propose an intelligent flight evaluation model specifically designed to enhancemulti-waypoint trajectory optimization algorithms.Our model leverages a decision tree to integrate attitude parameters and trajectory matching metrics,establishing a quantitative link between pilot control quality and UAV flight states.Experimental results demonstrate that the proposed model not only accurately assesses pilot performance across diverse skill levels but also improves the optimality of generated trajectories.When integrated with our path planning algorithm,it efficiently produces optimal trajectories while strictly adhering to UAV flight constraints.This integrated framework highlights significant potential for real-time UAV training,performance assessment,and adaptive mission planning applications.展开更多
Rapidly-exploring Random Tree(RRT)and its variants have become foundational in path-planning research,yet in complex three-dimensional off-road environments their uniform blind sampling and limited safety guarantees l...Rapidly-exploring Random Tree(RRT)and its variants have become foundational in path-planning research,yet in complex three-dimensional off-road environments their uniform blind sampling and limited safety guarantees lead to slow convergence and force an unfavorable trade-off between path quality and traversal safety.To address these challenges,we introduce HS-APF-RRT*,a novel algorithm that fuses layered sampling,an enhanced Artificial Potential Field(APF),and a dynamic neighborhood-expansion mechanism.First,the workspace is hierarchically partitioned into macro,meso,and micro sampling layers,progressively biasing random samples toward safer,lower-energy regions.Second,we augment the traditional APF by incorporating a slope-dependent repulsive term,enabling stronger avoidance of steep obstacles.Third,a dynamic expansion strategy adaptively switches between 8 and 16 connected neighborhoods based on local obstacle density,striking an effective balance between search efficiency and collision-avoidance precision.In simulated off-road scenarios,HS-APF-RRT*is benchmarked against RRT*,GoalBiased RRT*,and APF-RRT*,and demonstrates significantly faster convergence,lower path-energy consumption,and enhanced safety margins.展开更多
This study examines the methods to plan the development of offshore oilfields over the years,which are used to support the decision-making on the development of offshore oilfields.About 100 papers are analysed and cat...This study examines the methods to plan the development of offshore oilfields over the years,which are used to support the decision-making on the development of offshore oilfields.About 100 papers are analysed and categorised into different groups of main early-stage decisions.The present study stands in contrast to the contributions of the operations research and system engineering review articles,on the one hand,and the petroleum engineering review articles,on the other.This is because it does not focus on one methodological approach,nor does it limit the literature analysis by offshore oilfield characteristics.Consequently,the present analysis may offer valuable insights,for instance,by identifying environmental planning decisions as a recent yet highly significant concern that is currently being imposed on decision-making process.Thus,it is evident that the incorporation of safety criteria within the technical-economic decision-making process for the design of production systems would be a crucial requirement at development phase.展开更多
Objective To systematically characterize the developmental trajectory and interdisciplinary integration of intelligent diagnosis in traditional Chinese medicine(TCM)through quantitative topic evolution analysis,we add...Objective To systematically characterize the developmental trajectory and interdisciplinary integration of intelligent diagnosis in traditional Chinese medicine(TCM)through quantitative topic evolution analysis,we addressed the fragmentation of existing research and clarified the long-term research structure and evolutionary patterns of the field.Methods A topic evolution analysis was performed on Chinese-language literature pertaining to intelligent diagnosis in TCM.Publications were retrieved from the China National Knowledge Infrastructure(CNKI),Wanfang Data,and China Science and Technology Journal Database(VIP),covering the period from database inception to July 3,2025.A hybrid segmentation approach,based on cumulative publication growth trends and inflection point detection,was applied to divide the research timeline into distinct stages.Subsequently,the latent Dirichlet allocation(LDA)model was used to extract research topics,followed by alignment and evolutionary analysis of topics across different stages.Results A total of 3919 publications published between 2003 and 2025 were included,and the research trajectory was divided into five stages based on data-driven breakpoint detection.The field exhibited a clear evolutionary shift from early rule-based systems and tonguepulse image and signal analysis(2006–2010),to machine-learning-based syndrome and prescription modeling(2011–2015),followed by deep-learning-driven pattern recognition and formula association(2016–2020).Since 2021,research has increasingly emphasized knowledge-graph construction,multimodal integration,and intelligent clinical decision-support systems,with recent studies(2024–2025)showing the emergence of large language models and agent-based diagnostic frameworks.Topic evolution analysis further revealed sustained cross-stage continuity in syndrome modeling and prescription association analysis,alongside the progressive consolidation of integrated intelligent diagnostic platforms.Conclusion By identifying key technological transitions and persistent core research themes,our findings offer a structured reference framework for the design of intelligent diagnostic systems,the construction of knowledge-driven clinical decision-support tools,and the alignment of AI models with TCM diagnostic logic.Importantly,the stage-based evolutionary insights derived from this analysis can inform future methodological choices,improve model interpretability and clinical applicability,and support the translation of intelligent TCM diagnosis from experimental research to real-world clinical practice.展开更多
Driven by the global energy transition and the urgent“dual carbon”goals,regional integrated energy system(RIES)planning is undergoing a paradigm shift from carbon reduction to negative carbon emissions.This paper pr...Driven by the global energy transition and the urgent“dual carbon”goals,regional integrated energy system(RIES)planning is undergoing a paradigm shift from carbon reduction to negative carbon emissions.This paper provides a comprehensive review of the theoretical frameworks and technical pathways for RIES planning from a carbon-centric perspective.A key contribution is the proposed Carbon-Energy-Economy(CEE)triple-dimensional governance framework,which endogenizes carbon factors into planning decisions through emission constraints,trading mechanisms,and capture technologies.We first analyze the fundamental characteristics of RIES and their critical role in achieving carbon neutrality,detailing advancements in multi-energy coupling models,energy router concepts,and standardized energy hub modeling.The paper further explores multi-energy flow analysis methods,and systematically compares the applicability and limitations of various planning algorithms,with emphasis on addressing uncertainties from renewable integration.Finally,we highlight the integration of artificial intelligence with traditional optimization methods,offering new pathways for intelligent,adaptive,and low-carbon RIES planning.This review underscores the transition towards data-physical fusion models,cooperative uncertainty optimization,multi-market planning,and innovative zero/negative-carbon technological routes.展开更多
To address low learning efficiency and inadequate path safety in spraying robot navigation within complex obstacle-rich environments—with dense,dynamic,unpredictable obstacles challenging conventional methods—this p...To address low learning efficiency and inadequate path safety in spraying robot navigation within complex obstacle-rich environments—with dense,dynamic,unpredictable obstacles challenging conventional methods—this paper proposes a hybrid algorithm integrating Q-learning and improved A*-Artificial Potential Field(A-APF).Centered on theQ-learning framework,the algorithmleverages safety-oriented guidance generated byA-APF and employs a dynamic coordination mechanism that adaptively balances exploration and exploitation.The proposed system comprises four core modules:(1)an environment modeling module that constructs grid-based obstacle maps;(2)an A-APF module that combines heuristic search from A*algorithm with repulsive force strategies from APF to generate guidance;(3)a Q-learning module that learns optimal state-action values(Q-values)through spraying robot-environment interaction and a reward function emphasizing path optimality and safety;and(4)a dynamic optimization module that ensures adaptive cooperation between Q-learning and A-APF through exploration rate control and environment-aware constraints.Simulation results demonstrate that the proposed method significantly enhances path safety in complex underground mining environments.Quantitative results indicate that,compared to the traditional Q-learning algorithm,the proposed method shortens training time by 42.95% and achieves a reduction in training failures from 78 to just 3.Compared to the static fusion algorithm,it further reduces both training time(by 10.78%)and training failures(by 50%),thereby improving overall training efficiency.展开更多
As carrier aircraft sortie frequency and flight deck operational density increase,autonomous dispatch trajectory planning for carrier-based vehicles demands efficient,safe,and kinematically feasible solutions.This pap...As carrier aircraft sortie frequency and flight deck operational density increase,autonomous dispatch trajectory planning for carrier-based vehicles demands efficient,safe,and kinematically feasible solutions.This paper presents an Iterative Safe Dispatch Corridor(iSDC)framework,addressing the suboptimality of the traditional SDC method caused by static corridor construction and redundant obstacle exploration.First,a Kinodynamic-Informed-Bidirectional Rapidly-exploring Random Tree Star(KIBRRT^(*))algorithm is proposed for the front-end coarse planning.By integrating bidirectional tree expansion,goal-biased elliptical sampling,and artificial potential field guidance,it reduces unnecessary exploration near concave obstacles and generates kinematically admissible paths.Secondly,the traditional SDC is implemented in an iterative manner,and the obtained trajectory in the current iteration is fed into the next iteration for corridor generation,thus progressively improving the quality of withincorridor constraints.For tractors,a reverse-motion penalty function is incorporated into the back-end optimizer to prioritize forward driving,aligning with mechanical constraints and human operational preferences.Numerical validations on the data of Gerald R.Ford-class carrier demonstrate that the KIBRRT^(*)reduces average computational time by 75%and expansion nodes by 25%compared to conventional RRT^(*)algorithms.Meanwhile,the iSDC framework yields more time-efficient trajectories for both carrier aircraft and tractors,with the dispatch time reduced by 31.3%and tractor reverse motion proportion decreased by 23.4%relative to traditional SDC.The presented framework offers a scalable solution for autonomous dispatch in confined and safety-critical environment,and an illustrative animation is available at bilibili.com/video/BV1tZ7Zz6Eyz.Moreover,the framework can be easily extended to three-dimension scenarios,and thus applicable for trajectory planning of aerial and underwater vehicles.展开更多
With the rapid development of intelligent navigation technology,efficient and safe path planning for mobile robots has become a core requirement.To address the challenges of complex dynamic environments,this paper pro...With the rapid development of intelligent navigation technology,efficient and safe path planning for mobile robots has become a core requirement.To address the challenges of complex dynamic environments,this paper proposes an intelligent path planning framework based on grid map modeling.First,an improved Safe and Smooth A*(SSA*)algorithm is employed for global path planning.By incorporating obstacle expansion and cornerpoint optimization,the proposed SSA*enhances the safety and smoothness of the planned path.Then,a Partitioned Dynamic Window Approach(PDWA)is integrated for local planning,which is triggered when dynamic or sudden static obstacles appear,enabling real-time obstacle avoidance and path adjustment.A unified objective function is constructed,considering path length,safety,and smoothness comprehensively.Multiple simulation experiments are conducted on typical port grid maps.The results demonstrate that the improved SSA*significantly reduces the number of expanded nodes and computation time in static environmentswhile generating smoother and safer paths.Meanwhile,the PDWA exhibits strong real-time performance and robustness in dynamic scenarios,achieving shorter paths and lower planning times compared to other graph search algorithms.The proposedmethodmaintains stable performance across maps of different scales and various port scenarios,verifying its practicality and potential for wider application.展开更多
Planning in lexical-prior-free environments presents a fundamental challenge for evaluating whether large language models(LLMs)possess genuine structural reasoning capabilities beyond lexical memorization.When predica...Planning in lexical-prior-free environments presents a fundamental challenge for evaluating whether large language models(LLMs)possess genuine structural reasoning capabilities beyond lexical memorization.When predicates and action names are replaced with semantically irrelevant random symbols while preserving logical structures,existing direct generation approaches exhibit severe performance degradation.This paper proposes a symbol-agnostic closed-loop planning pipeline that enables models to construct executable plans through systematic validation and iterative refinement.The system implements a complete generate-verify-repair cycle through six core processing components:semantic comprehension extracts structural constraints,language planner generates text plans,symbol translator performs structure-preserving mapping,consistency checker conducts static screening,Stanford Research Institute Problem Solver(STRIPS)simulator executes step-by-step validation,and VAL(Validator)provides semantic verification.A repair controller orchestrates four targeted strategies addressing typical failure patterns including first-step precondition errors andmid-segment statemaintenance issues.Comprehensive evaluation on PlanBench Mystery Blocksworld demonstrates substantial improvements over baseline approaches across both language models and reasoning models.Ablation studies confirm that each architectural component contributes non-redundantly to overall effectiveness,with targeted repair providing the largest impact,followed by deep constraint extraction and stepwise validation,demonstrating that superior performance emerges from synergistic integration of these mechanisms rather than any single dominant factor.Analysis reveals distinct failure patterns betweenmodel types—languagemodels struggle with local precondition satisfaction while reasoning models face global goal achievement challenges—yet the validation-driven mechanism successfully addresses these diverse weaknesses.A particularly noteworthy finding is the convergence of final success rates across models with varying intrinsic capabilities,suggesting that systematic validation and repair mechanisms play a more decisive role than raw model capacity in lexical-prior-free scenarios.This work establishes a rigorous evaluation framework incorporating statistical significance testing and mechanistic failure analysis,providingmethodological contributions for fair assessment and practical insights into building reliable planning systems under extreme constraint conditions.展开更多
Theauthor proposes a dual layer source grid load storage collaborative planning model based on Benders decomposition to optimize the low-carbon and economic performance of the distribution network.The model plans the ...Theauthor proposes a dual layer source grid load storage collaborative planning model based on Benders decomposition to optimize the low-carbon and economic performance of the distribution network.The model plans the configuration of photovoltaic(3.8 MW),wind power(2.5 MW),energy storage(2.2 MWh),and SVC(1.2 Mvar)through interaction between upper and lower layers,and modifies lines 2–3,8–9,etc.to improve transmission capacity and voltage stability.The author uses normal distribution and Monte Carlo method to model load uncertainty,and combines Weibull distribution to describe wind speed characteristics.Compared to the traditional three-layer model(TLM),Benders decomposition-based two-layer model(BLBD)has a 58.1%reduction in convergence time(5.36 vs.12.78 h),a 51.1%reduction in iteration times(23 vs.47 times),a 8.07%reduction in total cost(12.436 vs.13.528 million yuan),and a 9.62%reduction in carbon emissions(12,456 vs.13,782 t).After optimization,the peak valley difference decreased from4.1 to 2.9MW,the renewable energy consumption rate reached 93.4%,and the energy storage efficiency was 87.6%.Themodel has been validated in the IEEE 33 node system,demonstrating its superiority in terms of economy,low-carbon,and reliability.展开更多
In Human–Robot Interaction(HRI),generating robot trajectories that accurately reflect user intentions while ensuring physical realism remains challenging,especially in unstructured environments.In this study,we devel...In Human–Robot Interaction(HRI),generating robot trajectories that accurately reflect user intentions while ensuring physical realism remains challenging,especially in unstructured environments.In this study,we develop a multimodal framework that integrates symbolic task reasoning with continuous trajectory generation.The approach employs transformer models and adversarial training to map high-level intent to robotic motion.Information from multiple data sources,such as voice traits,hand and body keypoints,visual observations,and recorded paths,is integrated simultaneously.These signals are mapped into a shared representation that supports interpretable reasoning while enabling smooth and realistic motion generation.Based on this design,two different learning strategies are investigated.In the first step,grammar-constrained Linear Temporal Logic(LTL)expressions are created from multimodal human inputs.These expressions are subsequently decoded into robot trajectories.The second method generates trajectories directly from symbolic intent and linguistic data,bypassing an intermediate logical representation.Transformer encoders combine multiple types of information,and autoregressive transformer decoders generate motion sequences.Adding smoothness and speed limits during training increases the likelihood of physical feasibility.To improve the realism and stability of the generated trajectories during training,an adversarial discriminator is also included to guide them toward the distribution of actual robot motion.Tests on the NATSGLD dataset indicate that the complete system exhibits stable training behaviour and performance.In normalised coordinates,the logic-based pipeline has an Average Displacement Error(ADE)of 0.040 and a Final Displacement Error(FDE)of 0.036.The adversarial generator makes substantially more progress,reducing ADE to 0.021 and FDE to 0.018.Visual examination confirms that the generated trajectories closely align with observed motion patterns while preserving smooth temporal dynamics.展开更多
Efficient multiple unmanned aerial vehicles(UAVs)path planning is crucial for improving mission completion efficiency in UAV operations.However,during the actual flight of UAVs,the flight time between nodes is always ...Efficient multiple unmanned aerial vehicles(UAVs)path planning is crucial for improving mission completion efficiency in UAV operations.However,during the actual flight of UAVs,the flight time between nodes is always influenced by external factors,making the original path planning solution ineffective.In this paper,the multi-depot multi-UAV path planning problem with uncertain flight time is modeled as a robust optimization model with a budget uncertainty set.Then,the robust optimization model is transformed into a mixed integer linear programming model by the strong duality theorem,which makes the problem easy to solve.To effectively solve large-scale instances,a simulated annealing algorithm with a robust feasibility check(SA-RFC)is developed.The numerical experiment shows that the SA-RFC can find high-quality solutions within a few seconds.Moreover,the effect of the task location distribution,depot counts,and variations in robustness parameters on the robust optimization solution is analyzed by using Monte Carlo experiments.The results demonstrate that the proposed robust model can effectively reduce the risk of the UAV failing to return to the depot without significantly compromising the profit.展开更多
Shenzhen,a major city in southern China,has experienced rapid advancements in Unmanned Aerial Vehicle(UAV)technology,resulting in extensive logistics networks with thousands of daily flights.However,frequent disruptio...Shenzhen,a major city in southern China,has experienced rapid advancements in Unmanned Aerial Vehicle(UAV)technology,resulting in extensive logistics networks with thousands of daily flights.However,frequent disruptions due to its subtropical monsoon climate,including typhoons and gusty winds,present ongoing challenges.Despite the growing focus on operational costs and third-party risks,research on low-altitude urban wind fields remains scarce.This study addresses this gap by integrating wind field analysis into UAV path planning,introducing key innovations to the classical model.First,UAV wind resistance and turbulence constraints are analyzed,mapping high-wind-speed and turbulence-prone zones in the airspace.Second,wind dynamics are incorporated into path planning by considering airspeed and groundspeed variation,optimizing waypoint selection and flight speed adjustments to improve overall energy efficiency.Additionally,a wind-aware Theta*algorithm is proposed,leveraging wind vectors to expedite search process,while Computational Fluid Dynamics(CFD)techniques are employed to calculate wind fields.A case study of Shenzhen,examining wind patterns over the past decade,demonstrates a 6.23%improvement in groundspeed and a 7.69%reduction in energy consumption compared to wind-agnostic models.This framework advances UAV logistics by enhancing route safety and energy efficiency,contributing to more cost-effective operations.展开更多
文摘目的/意义探索医学信息学跨学科主题演化路径识别方法,为该领域跨学科研究布局与科研管理提供参考。方法/过程首先利用BERTopic和大语言模型(large language model,LLM)在获取的医学信息学文献中识别全局主题和阶段主题;然后基于主题影响力与学科多样性二维分析框架,确定跨学科主题;最后分析跨学科主题的演化路径。结果/结论基于Topic-LLM的分析方法,发现医学信息学跨学科主题的学科多样性和主题影响力呈现稳步增长趋势。
文摘目的/意义构建双层分析框架,全面把握学科结构,识别新兴前沿领域,追踪主题演化。方法/过程检索2016—2025年PubMed、Scopus和Web of Science数据库医学信息学文献,采用BERTopic识别主题,并划分为新兴、稳定、衰退3种演化模式。基于ChromaDB构建检索增强生成系统,通过文档-主题映射实现微观验证与知识关联挖掘。结果/结论医学信息学主题演化呈现研究重心转移、技术融合深化、学科交叉增强3个特征。BERTopic-RAG框架为知识发现提供了新方法。
文摘生成式人工智能正深刻变革教育,相关研究文献不断涌现。文章通过收集知网、Scopus和Web of Science中与研究主题相关的中英文文献,利用BERTopic主题建模技术对文献进行主题聚类,挖掘国内外高等教育中生成式人工智能的研究主题,分析研究现状,为教育领域的教学实践和学术研究提供参考。研究结果表明:在个性化学习与人机协同方面,生成式人工智能通过制定个性化学习路径显著提升教学效果;师生对生成式人工智能的接受程度受技术认知和使用体验等因素制约,并直接影响应用成效;教育创新需平衡技术赋能与过度依赖问题;教学设计与课程实践的革新更多体现在教学模式和资源生成方式上,但必须同步构建学术伦理防护与治理机制;人工智能时代亟须人才培养结构的优化升级,要求教育者重新定义核心能力目标。
基金Financial supports from the National Natural Science Foundation of China(GrantNo.52374123&51974144)Project of Liaoning Provincial Department of Education(GrantNo.LJKZ0340)Liaoning Revitalization Talents Program(Grant No.XLYC2211085)are greatly acknowledged.
文摘Q-learning is a classical reinforcement learning method with broad applicability.It can respond effectively to environmental changes and provide flexible strategies,making it suitable for solving robot path-planning problems.However,Q-learning faces challenges in search and update efficiency.To address these issues,we propose an improved Q-learning(IQL)algorithm.We use an enhanced Ant Colony Optimization(ACO)algorithmto optimizeQtable initialization.We also introduce the UCH mechanism to refine the reward function and overcome the exploration dilemma.The IQL algorithm is extensively tested in three grid environments of different scales.The results validate the accuracy of themethod and demonstrate superior path-planning performance compared to traditional approaches.The algorithm reduces the number of trials required for convergence,improves learning efficiency,and enables faster adaptation to environmental changes.It also enhances stability and accuracy by reducing the standard deviation of trials to zero.On grid maps of different sizes,IQL achieves higher expected returns.Compared with the original Q-learning algorithm,IQL improves performance by 12.95%,18.28%,and 7.98% on 10*10,20*20,and 30*30 maps,respectively.The proposed algorithm has promising applications in robotics,path planning,intelligent transportation,aerospace,and game development.
基金funded in part by the Fundamental Research Funds for the Central Universities under Grant NS2023052in part by the Natural Science Foundation of Jiangsu Province of China under Grants No.BK20231439 and No.BK20222012.
文摘With the expanding applications of unmanned aerial vehicles(UAVs),precise flight evaluation has emerged as a critical enabler for efficient path planning,directly impacting operational performance and safety.Traditional path planning algorithms typically combine Dubins curves with local optimization to minimize trajectory length under 3D spatial constraints.However,these methods often overlook the correlation between pilot control quality and UAV flight dynamics,limiting their adaptability in complex scenarios.In this paper,we propose an intelligent flight evaluation model specifically designed to enhancemulti-waypoint trajectory optimization algorithms.Our model leverages a decision tree to integrate attitude parameters and trajectory matching metrics,establishing a quantitative link between pilot control quality and UAV flight states.Experimental results demonstrate that the proposed model not only accurately assesses pilot performance across diverse skill levels but also improves the optimality of generated trajectories.When integrated with our path planning algorithm,it efficiently produces optimal trajectories while strictly adhering to UAV flight constraints.This integrated framework highlights significant potential for real-time UAV training,performance assessment,and adaptive mission planning applications.
基金supported in part by 14th Five Year National Key R&D Program Project(Project Number:2023YFB3211001)the National Natural Science Foundation of China(62273339,U24A201397).
文摘Rapidly-exploring Random Tree(RRT)and its variants have become foundational in path-planning research,yet in complex three-dimensional off-road environments their uniform blind sampling and limited safety guarantees lead to slow convergence and force an unfavorable trade-off between path quality and traversal safety.To address these challenges,we introduce HS-APF-RRT*,a novel algorithm that fuses layered sampling,an enhanced Artificial Potential Field(APF),and a dynamic neighborhood-expansion mechanism.First,the workspace is hierarchically partitioned into macro,meso,and micro sampling layers,progressively biasing random samples toward safer,lower-energy regions.Second,we augment the traditional APF by incorporating a slope-dependent repulsive term,enabling stronger avoidance of steep obstacles.Third,a dynamic expansion strategy adaptively switches between 8 and 16 connected neighborhoods based on local obstacle density,striking an effective balance between search efficiency and collision-avoidance precision.In simulated off-road scenarios,HS-APF-RRT*is benchmarked against RRT*,GoalBiased RRT*,and APF-RRT*,and demonstrates significantly faster convergence,lower path-energy consumption,and enhanced safety margins.
基金the Strategic Research Plan of the Centre for Marine Technology and Ocean Engineering(CENTEC),which is financed by the Portuguese Foundation for Science and Technology(Fundação para a Ciência e a Tecnologia FCT)under contract UIDB/UIDP/00134/2020.
文摘This study examines the methods to plan the development of offshore oilfields over the years,which are used to support the decision-making on the development of offshore oilfields.About 100 papers are analysed and categorised into different groups of main early-stage decisions.The present study stands in contrast to the contributions of the operations research and system engineering review articles,on the one hand,and the petroleum engineering review articles,on the other.This is because it does not focus on one methodological approach,nor does it limit the literature analysis by offshore oilfield characteristics.Consequently,the present analysis may offer valuable insights,for instance,by identifying environmental planning decisions as a recent yet highly significant concern that is currently being imposed on decision-making process.Thus,it is evident that the incorporation of safety criteria within the technical-economic decision-making process for the design of production systems would be a crucial requirement at development phase.
基金Grants of National Natural Science Foundation of China(82274685).
文摘Objective To systematically characterize the developmental trajectory and interdisciplinary integration of intelligent diagnosis in traditional Chinese medicine(TCM)through quantitative topic evolution analysis,we addressed the fragmentation of existing research and clarified the long-term research structure and evolutionary patterns of the field.Methods A topic evolution analysis was performed on Chinese-language literature pertaining to intelligent diagnosis in TCM.Publications were retrieved from the China National Knowledge Infrastructure(CNKI),Wanfang Data,and China Science and Technology Journal Database(VIP),covering the period from database inception to July 3,2025.A hybrid segmentation approach,based on cumulative publication growth trends and inflection point detection,was applied to divide the research timeline into distinct stages.Subsequently,the latent Dirichlet allocation(LDA)model was used to extract research topics,followed by alignment and evolutionary analysis of topics across different stages.Results A total of 3919 publications published between 2003 and 2025 were included,and the research trajectory was divided into five stages based on data-driven breakpoint detection.The field exhibited a clear evolutionary shift from early rule-based systems and tonguepulse image and signal analysis(2006–2010),to machine-learning-based syndrome and prescription modeling(2011–2015),followed by deep-learning-driven pattern recognition and formula association(2016–2020).Since 2021,research has increasingly emphasized knowledge-graph construction,multimodal integration,and intelligent clinical decision-support systems,with recent studies(2024–2025)showing the emergence of large language models and agent-based diagnostic frameworks.Topic evolution analysis further revealed sustained cross-stage continuity in syndrome modeling and prescription association analysis,alongside the progressive consolidation of integrated intelligent diagnostic platforms.Conclusion By identifying key technological transitions and persistent core research themes,our findings offer a structured reference framework for the design of intelligent diagnostic systems,the construction of knowledge-driven clinical decision-support tools,and the alignment of AI models with TCM diagnostic logic.Importantly,the stage-based evolutionary insights derived from this analysis can inform future methodological choices,improve model interpretability and clinical applicability,and support the translation of intelligent TCM diagnosis from experimental research to real-world clinical practice.
基金supported by the Natural Science Foundation of China(Grants U2166211)Zhejiang Provincial Natural Science Foundation of China(Grants LY24E070006 and LMS25E070002).
文摘Driven by the global energy transition and the urgent“dual carbon”goals,regional integrated energy system(RIES)planning is undergoing a paradigm shift from carbon reduction to negative carbon emissions.This paper provides a comprehensive review of the theoretical frameworks and technical pathways for RIES planning from a carbon-centric perspective.A key contribution is the proposed Carbon-Energy-Economy(CEE)triple-dimensional governance framework,which endogenizes carbon factors into planning decisions through emission constraints,trading mechanisms,and capture technologies.We first analyze the fundamental characteristics of RIES and their critical role in achieving carbon neutrality,detailing advancements in multi-energy coupling models,energy router concepts,and standardized energy hub modeling.The paper further explores multi-energy flow analysis methods,and systematically compares the applicability and limitations of various planning algorithms,with emphasis on addressing uncertainties from renewable integration.Finally,we highlight the integration of artificial intelligence with traditional optimization methods,offering new pathways for intelligent,adaptive,and low-carbon RIES planning.This review underscores the transition towards data-physical fusion models,cooperative uncertainty optimization,multi-market planning,and innovative zero/negative-carbon technological routes.
基金supported by the National Natural Science Foundation of China(Grant No.52374156).
文摘To address low learning efficiency and inadequate path safety in spraying robot navigation within complex obstacle-rich environments—with dense,dynamic,unpredictable obstacles challenging conventional methods—this paper proposes a hybrid algorithm integrating Q-learning and improved A*-Artificial Potential Field(A-APF).Centered on theQ-learning framework,the algorithmleverages safety-oriented guidance generated byA-APF and employs a dynamic coordination mechanism that adaptively balances exploration and exploitation.The proposed system comprises four core modules:(1)an environment modeling module that constructs grid-based obstacle maps;(2)an A-APF module that combines heuristic search from A*algorithm with repulsive force strategies from APF to generate guidance;(3)a Q-learning module that learns optimal state-action values(Q-values)through spraying robot-environment interaction and a reward function emphasizing path optimality and safety;and(4)a dynamic optimization module that ensures adaptive cooperation between Q-learning and A-APF through exploration rate control and environment-aware constraints.Simulation results demonstrate that the proposed method significantly enhances path safety in complex underground mining environments.Quantitative results indicate that,compared to the traditional Q-learning algorithm,the proposed method shortens training time by 42.95% and achieves a reduction in training failures from 78 to just 3.Compared to the static fusion algorithm,it further reduces both training time(by 10.78%)and training failures(by 50%),thereby improving overall training efficiency.
基金support of the National Key Research and Development Plan(Grant No.2021YFB3302501)the financial support of the National Science Foundation of China(Grant No.12161076)the financial support of the Fundamental Research Funds for the Central Universities(Grant No.DUT24LAB129).
文摘As carrier aircraft sortie frequency and flight deck operational density increase,autonomous dispatch trajectory planning for carrier-based vehicles demands efficient,safe,and kinematically feasible solutions.This paper presents an Iterative Safe Dispatch Corridor(iSDC)framework,addressing the suboptimality of the traditional SDC method caused by static corridor construction and redundant obstacle exploration.First,a Kinodynamic-Informed-Bidirectional Rapidly-exploring Random Tree Star(KIBRRT^(*))algorithm is proposed for the front-end coarse planning.By integrating bidirectional tree expansion,goal-biased elliptical sampling,and artificial potential field guidance,it reduces unnecessary exploration near concave obstacles and generates kinematically admissible paths.Secondly,the traditional SDC is implemented in an iterative manner,and the obtained trajectory in the current iteration is fed into the next iteration for corridor generation,thus progressively improving the quality of withincorridor constraints.For tractors,a reverse-motion penalty function is incorporated into the back-end optimizer to prioritize forward driving,aligning with mechanical constraints and human operational preferences.Numerical validations on the data of Gerald R.Ford-class carrier demonstrate that the KIBRRT^(*)reduces average computational time by 75%and expansion nodes by 25%compared to conventional RRT^(*)algorithms.Meanwhile,the iSDC framework yields more time-efficient trajectories for both carrier aircraft and tractors,with the dispatch time reduced by 31.3%and tractor reverse motion proportion decreased by 23.4%relative to traditional SDC.The presented framework offers a scalable solution for autonomous dispatch in confined and safety-critical environment,and an illustrative animation is available at bilibili.com/video/BV1tZ7Zz6Eyz.Moreover,the framework can be easily extended to three-dimension scenarios,and thus applicable for trajectory planning of aerial and underwater vehicles.
文摘With the rapid development of intelligent navigation technology,efficient and safe path planning for mobile robots has become a core requirement.To address the challenges of complex dynamic environments,this paper proposes an intelligent path planning framework based on grid map modeling.First,an improved Safe and Smooth A*(SSA*)algorithm is employed for global path planning.By incorporating obstacle expansion and cornerpoint optimization,the proposed SSA*enhances the safety and smoothness of the planned path.Then,a Partitioned Dynamic Window Approach(PDWA)is integrated for local planning,which is triggered when dynamic or sudden static obstacles appear,enabling real-time obstacle avoidance and path adjustment.A unified objective function is constructed,considering path length,safety,and smoothness comprehensively.Multiple simulation experiments are conducted on typical port grid maps.The results demonstrate that the improved SSA*significantly reduces the number of expanded nodes and computation time in static environmentswhile generating smoother and safer paths.Meanwhile,the PDWA exhibits strong real-time performance and robustness in dynamic scenarios,achieving shorter paths and lower planning times compared to other graph search algorithms.The proposedmethodmaintains stable performance across maps of different scales and various port scenarios,verifying its practicality and potential for wider application.
基金supported by the Information,Production and Systems Research Center,Waseda University,and partly supported by the Future Robotics Organization,Waseda Universitythe Humanoid Robotics Institute,Waseda University,under the Humanoid Project+1 种基金the Waseda University Grant for Special Research Projects(grant numbers 2024C-518 and 2025E-027)was partly executed under the cooperation of organization between Kioxia Corporation andWaseda University.
文摘Planning in lexical-prior-free environments presents a fundamental challenge for evaluating whether large language models(LLMs)possess genuine structural reasoning capabilities beyond lexical memorization.When predicates and action names are replaced with semantically irrelevant random symbols while preserving logical structures,existing direct generation approaches exhibit severe performance degradation.This paper proposes a symbol-agnostic closed-loop planning pipeline that enables models to construct executable plans through systematic validation and iterative refinement.The system implements a complete generate-verify-repair cycle through six core processing components:semantic comprehension extracts structural constraints,language planner generates text plans,symbol translator performs structure-preserving mapping,consistency checker conducts static screening,Stanford Research Institute Problem Solver(STRIPS)simulator executes step-by-step validation,and VAL(Validator)provides semantic verification.A repair controller orchestrates four targeted strategies addressing typical failure patterns including first-step precondition errors andmid-segment statemaintenance issues.Comprehensive evaluation on PlanBench Mystery Blocksworld demonstrates substantial improvements over baseline approaches across both language models and reasoning models.Ablation studies confirm that each architectural component contributes non-redundantly to overall effectiveness,with targeted repair providing the largest impact,followed by deep constraint extraction and stepwise validation,demonstrating that superior performance emerges from synergistic integration of these mechanisms rather than any single dominant factor.Analysis reveals distinct failure patterns betweenmodel types—languagemodels struggle with local precondition satisfaction while reasoning models face global goal achievement challenges—yet the validation-driven mechanism successfully addresses these diverse weaknesses.A particularly noteworthy finding is the convergence of final success rates across models with varying intrinsic capabilities,suggesting that systematic validation and repair mechanisms play a more decisive role than raw model capacity in lexical-prior-free scenarios.This work establishes a rigorous evaluation framework incorporating statistical significance testing and mechanistic failure analysis,providingmethodological contributions for fair assessment and practical insights into building reliable planning systems under extreme constraint conditions.
文摘Theauthor proposes a dual layer source grid load storage collaborative planning model based on Benders decomposition to optimize the low-carbon and economic performance of the distribution network.The model plans the configuration of photovoltaic(3.8 MW),wind power(2.5 MW),energy storage(2.2 MWh),and SVC(1.2 Mvar)through interaction between upper and lower layers,and modifies lines 2–3,8–9,etc.to improve transmission capacity and voltage stability.The author uses normal distribution and Monte Carlo method to model load uncertainty,and combines Weibull distribution to describe wind speed characteristics.Compared to the traditional three-layer model(TLM),Benders decomposition-based two-layer model(BLBD)has a 58.1%reduction in convergence time(5.36 vs.12.78 h),a 51.1%reduction in iteration times(23 vs.47 times),a 8.07%reduction in total cost(12.436 vs.13.528 million yuan),and a 9.62%reduction in carbon emissions(12,456 vs.13,782 t).After optimization,the peak valley difference decreased from4.1 to 2.9MW,the renewable energy consumption rate reached 93.4%,and the energy storage efficiency was 87.6%.Themodel has been validated in the IEEE 33 node system,demonstrating its superiority in terms of economy,low-carbon,and reliability.
基金The authors extend their appreciation to Prince Sattam bin Abdulaziz University for funding this research work through the project number(PSAU/2024/01/32082).
文摘In Human–Robot Interaction(HRI),generating robot trajectories that accurately reflect user intentions while ensuring physical realism remains challenging,especially in unstructured environments.In this study,we develop a multimodal framework that integrates symbolic task reasoning with continuous trajectory generation.The approach employs transformer models and adversarial training to map high-level intent to robotic motion.Information from multiple data sources,such as voice traits,hand and body keypoints,visual observations,and recorded paths,is integrated simultaneously.These signals are mapped into a shared representation that supports interpretable reasoning while enabling smooth and realistic motion generation.Based on this design,two different learning strategies are investigated.In the first step,grammar-constrained Linear Temporal Logic(LTL)expressions are created from multimodal human inputs.These expressions are subsequently decoded into robot trajectories.The second method generates trajectories directly from symbolic intent and linguistic data,bypassing an intermediate logical representation.Transformer encoders combine multiple types of information,and autoregressive transformer decoders generate motion sequences.Adding smoothness and speed limits during training increases the likelihood of physical feasibility.To improve the realism and stability of the generated trajectories during training,an adversarial discriminator is also included to guide them toward the distribution of actual robot motion.Tests on the NATSGLD dataset indicate that the complete system exhibits stable training behaviour and performance.In normalised coordinates,the logic-based pipeline has an Average Displacement Error(ADE)of 0.040 and a Final Displacement Error(FDE)of 0.036.The adversarial generator makes substantially more progress,reducing ADE to 0.021 and FDE to 0.018.Visual examination confirms that the generated trajectories closely align with observed motion patterns while preserving smooth temporal dynamics.
基金supported by the National Natural Science Foundation of China(72571094,72271076,71871079)。
文摘Efficient multiple unmanned aerial vehicles(UAVs)path planning is crucial for improving mission completion efficiency in UAV operations.However,during the actual flight of UAVs,the flight time between nodes is always influenced by external factors,making the original path planning solution ineffective.In this paper,the multi-depot multi-UAV path planning problem with uncertain flight time is modeled as a robust optimization model with a budget uncertainty set.Then,the robust optimization model is transformed into a mixed integer linear programming model by the strong duality theorem,which makes the problem easy to solve.To effectively solve large-scale instances,a simulated annealing algorithm with a robust feasibility check(SA-RFC)is developed.The numerical experiment shows that the SA-RFC can find high-quality solutions within a few seconds.Moreover,the effect of the task location distribution,depot counts,and variations in robustness parameters on the robust optimization solution is analyzed by using Monte Carlo experiments.The results demonstrate that the proposed robust model can effectively reduce the risk of the UAV failing to return to the depot without significantly compromising the profit.
基金supported by the National Natural Science Foundation of China(No.U2433214)。
文摘Shenzhen,a major city in southern China,has experienced rapid advancements in Unmanned Aerial Vehicle(UAV)technology,resulting in extensive logistics networks with thousands of daily flights.However,frequent disruptions due to its subtropical monsoon climate,including typhoons and gusty winds,present ongoing challenges.Despite the growing focus on operational costs and third-party risks,research on low-altitude urban wind fields remains scarce.This study addresses this gap by integrating wind field analysis into UAV path planning,introducing key innovations to the classical model.First,UAV wind resistance and turbulence constraints are analyzed,mapping high-wind-speed and turbulence-prone zones in the airspace.Second,wind dynamics are incorporated into path planning by considering airspeed and groundspeed variation,optimizing waypoint selection and flight speed adjustments to improve overall energy efficiency.Additionally,a wind-aware Theta*algorithm is proposed,leveraging wind vectors to expedite search process,while Computational Fluid Dynamics(CFD)techniques are employed to calculate wind fields.A case study of Shenzhen,examining wind patterns over the past decade,demonstrates a 6.23%improvement in groundspeed and a 7.69%reduction in energy consumption compared to wind-agnostic models.This framework advances UAV logistics by enhancing route safety and energy efficiency,contributing to more cost-effective operations.
