[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"item-3253":3,"related-3253":77},{"id":4,"title":5,"url":6,"summary":7,"summary_zh":8,"content":8,"source_name":9,"source_url":8,"published_at":10,"category":11,"cover_url":8,"hotness":12,"is_selected":13,"score":14,"score_detail":15,"sources":22,"tags":24,"primary_topic":30,"secondary_topics":36,"classification":37,"search_phrases":61,"slug":64,"view_count":65,"doi":8,"paper":66,"created_at":76},3253,"Visuomotor Robotic Pruning in Planar Orchills Using Hybrid Reinforcement Learning（基于混合强化学习的平面果园视觉运动机器人修剪）","https:\u002F\u002Farxivtldr.org\u002Fabs\u002F2609.24906","arXiv发表研究：休眠期树木修剪是劳动密集型但对维持现代高产果园至关重要的工作。本文针对现代平面树形训练系统（V型棚架苹果、UFO樱桃）的修剪任务，提出端到端管线学习闭环视觉运动控制器。控制器完全使用仿真和合成生成数据进行训练，以零样本方式部署到真实果园。管线包括平面果园树网格的合成生成、基于物理的果园仿真器构建、通过运动规划自动收集成功修剪轨迹，以及结合离线演示与在线仿真推出相结合的新型混合强化学习算法。控制器使用腕部相机的光流输入，避免完整三维重建需求，连续引导切割器穿越杂乱分支环境到指定切割点并在真实果园38次物理试验中展示零样本仿真到真实迁移。在V-Trellis苹果上达到49.9%成功率，UFO樱桃上46.0%。",null,"arXiv (Abhinav Jain, Cindy Grimm, Stefan Lee)","2026-09-20T00:00:00Z","论文",10,false,80,{"impact":16,"substance":17,"depth":16,"authority":18,"freshness":19,"relevant":20,"comment":21},18,22,13,9,1,"仿真到真实零样本迁移的果园修剪机器人研究，方法新颖、数据扎实，对智慧果园机械化有参考价值。",[23],{"name":9,"url":6},[25,26,27,28,29],"智慧农业","农业人工智能","农业机器人","强化学习","果园管理",{"code":31,"slug":32,"name":33,"path":34,"parent_code":35},"B2","fruit","果树","\u002Ftopic\u002Fhorticulture\u002Ffruit","B",[],{"status":38,"locked":13,"version":39,"method":40,"evidence":41},"classified","2026-10-22.2","rules",{"candidates":42,"agricultural_evidence":45},[43,50,56],{"code":31,"score":44,"supported":45,"title":46,"abstract":47,"tags":49},3,true,[],[48],"苹果",[],{"code":51,"score":52,"supported":13,"title":53,"abstract":54,"tags":55},"I2",2,[],[],[26,28],{"code":57,"score":20,"supported":13,"title":58,"abstract":59,"tags":60},"I1",[],[],[25],[62,63],"V-Trellis 苹果 机器人修剪","UFO 樱桃 零样本迁移","V-Trellis苹果机器人修剪-3253",5,{"doi":8,"openalex_id":8,"authors":67,"venue":8,"cited_by_count":68,"oa_url":8,"card":69,"direction":73,"ingested_from":75},[],0,{"tldr":70,"method":71,"finding":72,"direction":73,"opportunity":74},"提出端到端视觉运动控制器，用仿真合成数据训练，零样本迁移到真实果园完成修剪。","合成数据生成、物理仿真器、运动规划收集轨迹、混合强化学习、腕部相机光流。","真实果园38次试验零样本迁移成功，苹果49.9%、樱桃46.0%成功率。","农业人工智能与决策模型","可探索多季节、多树种泛化及真实数据微调，提升复杂冠层下的鲁棒性。","agent","2026-09-23T00:04:33.854148Z",{"total":78,"page":20,"page_size":78,"items":79},6,[80,140,203,270,340,389],{"id":81,"title":82,"url":83,"summary":84,"summary_zh":8,"content":8,"source_name":85,"source_url":83,"published_at":86,"category":11,"cover_url":8,"hotness":12,"is_selected":13,"score":87,"score_detail":88,"sources":93,"tags":95,"primary_topic":98,"secondary_topics":99,"classification":100,"search_phrases":117,"slug":120,"view_count":44,"doi":121,"paper":122,"created_at":139},3808,"Orchard worker detection using spatial-context attention on 2D LiDAR data","https:\u002F\u002Fdoi.org\u002F10.1016\u002Fj.biosystemseng.2026.104604","Orchard worker detection using spatial-context attention on 2D LiDAR data。Biosystems Engineering","Biosystems Engineering","2026-09-30T00:00:00Z",79,{"impact":89,"substance":90,"depth":16,"authority":91,"freshness":12,"relevant":20,"comment":92},16,21,14,"提出基于二维激光雷达与空间上下文注意力的果园工人检测方法，属农业人工智能与智能农机细分领域的方法创新，具备一定专业参考价值。",[94],{"name":85,"url":83},[25,26,96,29,97],"目标检测","激光雷达",{"code":31,"slug":32,"name":33,"path":34,"parent_code":35},[],{"status":38,"locked":13,"version":39,"method":40,"evidence":101},{"candidates":102,"agricultural_evidence":45},[103,109,113],{"code":31,"score":104,"supported":45,"title":105,"abstract":107,"tags":108},11,[106],"orchard",[106],[],{"code":51,"score":52,"supported":13,"title":110,"abstract":111,"tags":112},[],[],[26,96],{"code":57,"score":20,"supported":13,"title":114,"abstract":115,"tags":116},[],[],[25],[118,119],"2D LiDAR 果园 工人检测","农业人工智能 智慧农业 果园管理 激光雷达","2DLiDAR果园工人检测-3808","10.1016\u002Fj.biosystemseng.2026.104604",{"doi":121,"openalex_id":123,"authors":124,"venue":85,"cited_by_count":68,"oa_url":83,"card":8,"direction":8,"ingested_from":138},"W7214801786",[125,127,129,132,135],{"name":126,"orcid":8},"Jintack Jeon",{"name":128,"orcid":8},"Huijong Chung",{"name":130,"orcid":131},"Gookhwan Kim","https:\u002F\u002Forcid.org\u002F0000-0002-7278-3476",{"name":133,"orcid":134},"Yong–Joo Kim","https:\u002F\u002Forcid.org\u002F0000-0002-1212-9018",{"name":136,"orcid":137},"Dae-Hyun Lee","https:\u002F\u002Forcid.org\u002F0000-0001-9544-5974","openalex","2026-09-30T23:30:05.310491Z",{"id":141,"title":142,"url":143,"summary":144,"summary_zh":145,"content":8,"source_name":146,"source_url":143,"published_at":147,"category":11,"cover_url":8,"hotness":12,"is_selected":13,"score":14,"score_detail":148,"sources":151,"tags":153,"primary_topic":156,"secondary_topics":157,"classification":158,"search_phrases":179,"slug":182,"view_count":183,"doi":184,"paper":185,"created_at":202},3780,"Epistemological Friction: Embodied Expertise and the ‘Last Mile’ of Agricultural Automation in the US Tree‐Fruit Industry","https:\u002F\u002Fdoi.org\u002F10.1111\u002Fntwe.70047","ABSTRACT Autonomous and AI‐driven agricultural technology is often framed as a panacea for labour dependency. While proponents envision autonomous technologies seamlessly optimising farming, this rationalised model misunderstands the reality of on‐farm integration. Based on an ethnographic and comparative study of 16 operations in the US West Coast Tree Fruit Industry (California and Washington), this article examines the ‘last mile’ of agricultural automation. The study reveals that integrating AI and robotics is a contested negotiation shaped by epistemological friction—the tension between the algorithmic logic of developers and the embodied, site‐specific expertise of farmers and farm workers. Rather than resulting in seamless labour replacement, we find that agricultural automation reconfigures work, creating an invisible reliance on ‘human middleware’ to navigate the biological recalcitrance of the orchard. Demonstrating how regional contexts and participatory implementation shape technological trajectories, this study extends Labour Process Theory. We argue for the need for a relational approach recognising the essential cognitive labour of farmers and farm workers that is demanded by digital agriculture. Ultimately, this study shows the ‘farm of the future’ depends on reconciling rigid systems with the situated expertise of the farm workforce.","自主与人工智能驱动的农业技术常被描绘为解决劳动力依赖问题的万能药。支持者设想自主技术能够无缝优化农业生产，然而这种理性化模型误解了农场整合的现实。本文基于对美国西海岸果树产业（加利福尼亚州和华盛顿州）16家经营主体的民族志比较研究，考察农业自动化的“最后一英里”。研究发现，人工智能与机器人技术的整合是一场充满争议的协商过程，其受到认识论摩擦的塑造——即开发者的算法逻辑与农民及农场工人具身的、场所特定的专业知识之间的张力。我们发现，农业自动化并未实现无缝的劳动力替代，而是重构了工作形态，制造了对“人类中间件”的隐性依赖，以应对果园的生物性阻力。本研究通过展示区域情境与参与式实施如何塑造技术轨迹，拓展了劳动过程理论。我们主张需要一种关系性方法，承认数字农业所要求的农民和农场工人不可或缺的认知劳动。最终，本研究表明“未来农场”取决于刚性系统与农场劳动力情境化专业知识之间的调和。","New Technology Work and Employment","2026-09-27T00:00:00Z",{"impact":16,"substance":17,"depth":16,"authority":91,"freshness":149,"relevant":20,"comment":150},8,"基于美西16家果园民族志研究，揭示AI与机器人落地的“最后一公里”认识论摩擦与“人类中间件”隐性依赖，对智慧农业人机协同具参考价值。",[152],{"name":146,"url":143},[154,25,26,27,155],"数字农业","果园自动化",{"code":31,"slug":32,"name":33,"path":34,"parent_code":35},[],{"status":38,"locked":13,"version":39,"method":40,"evidence":159},{"candidates":160,"agricultural_evidence":45},[161,166,170,175],{"code":57,"score":149,"supported":45,"title":162,"abstract":163,"tags":165},[],[154,164],"digital agriculture",[154,25],{"code":31,"score":78,"supported":45,"title":167,"abstract":168,"tags":169},[],[33,106],[],{"code":171,"score":20,"supported":13,"title":172,"abstract":173,"tags":174},"F6",[],[],[27],{"code":51,"score":20,"supported":13,"title":176,"abstract":177,"tags":178},[],[],[26],[180,181],"美国 果树 自动化","果园 机器人 劳动力","美国果树自动化-3780",4,"10.1111\u002Fntwe.70047",{"doi":184,"openalex_id":186,"authors":187,"venue":146,"cited_by_count":68,"oa_url":194,"card":195,"direction":201,"ingested_from":138},"W7214618881",[188,191],{"name":189,"orcid":190},"M. Anne Visser","https:\u002F\u002Forcid.org\u002F0000-0001-5047-7702",{"name":192,"orcid":193},"Alex Bladimir Hernandez","https:\u002F\u002Forcid.org\u002F0009-0001-5320-7203","https:\u002F\u002Fonlinelibrary.wiley.com\u002Fdoi\u002Fpdfdirect\u002F10.1111\u002Fntwe.70047",{"tldr":196,"method":197,"finding":198,"direction":199,"opportunity":200},"通过美西果树业民族志研究，揭示农业自动化落地中算法逻辑与农户经验知识的冲突。","对加州和华盛顿16个果园开展民族志与比较研究。","自动化未无缝替代劳动，反而依赖‘人类中间件’协调生物不确定性。","数字乡村与农业信息化","可探索参与式设计如何将农户隐性知识嵌入AI系统，弥合‘最后一英里’鸿沟。","智慧农业 \u002F 农业物联网","2026-09-29T23:30:12.281947Z",{"id":204,"title":205,"url":206,"summary":207,"summary_zh":208,"content":8,"source_name":209,"source_url":206,"published_at":210,"category":11,"cover_url":8,"hotness":12,"is_selected":13,"score":211,"score_detail":212,"sources":217,"tags":219,"primary_topic":220,"secondary_topics":221,"classification":222,"search_phrases":238,"slug":241,"view_count":44,"doi":242,"paper":243,"created_at":269},1590,"Semantic risk-aware motion planning for apple young-fruit thinning manipulators","https:\u002F\u002Fdoi.org\u002F10.1016\u002Fj.compag.2026.112374","Semantic risk-aware motion planning for apple young-fruit thinning manipulators。Computers and Electronics in Agriculture","苹果疏果机械臂的语义风险感知运动规划。《农业计算机与电子》","Computers and Electronics in Agriculture","2026-09-03T00:00:00Z",61,{"impact":213,"substance":214,"depth":214,"authority":213,"freshness":215,"relevant":20,"comment":216},12,15,7,"针对苹果疏果机器人的语义风险感知运动规划研究，提升作业安全与效率，细分领域进展。",[218],{"name":209,"url":206},[25,27,29],{"code":31,"slug":32,"name":33,"path":34,"parent_code":35},[],{"status":38,"locked":13,"version":39,"method":40,"evidence":223},{"candidates":224,"agricultural_evidence":45},[225,230,234],{"code":31,"score":91,"supported":45,"title":226,"abstract":228,"tags":229},[227],"apple",[48,227],[],{"code":171,"score":20,"supported":13,"title":231,"abstract":232,"tags":233},[],[],[27],{"code":57,"score":20,"supported":13,"title":235,"abstract":236,"tags":237},[],[],[25],[239,240],"农业机器人 智慧农业 果园管理","农业机器人 智慧农业","农业机器人智慧农业果园管理-1590","10.1016\u002Fj.compag.2026.112374",{"doi":242,"openalex_id":244,"authors":245,"venue":209,"cited_by_count":68,"oa_url":263,"card":264,"direction":73,"ingested_from":138},"W7166343784",[246,248,250,252,254,256,258,261],{"name":247,"orcid":8},"谢胜旺",{"name":249,"orcid":8},"T B Zhang",{"name":251,"orcid":8},"Xiaoning He",{"name":253,"orcid":8},"Yanan Hu",{"name":255,"orcid":8},"Shuqi Shang",{"name":257,"orcid":8},"Z G Zhu",{"name":259,"orcid":260},"Xia Gao","https:\u002F\u002Forcid.org\u002F0000-0001-7139-7883",{"name":262,"orcid":8},"Shengda Li","https:\u002F\u002Fdoi.org\u002F10.2139\u002Fssrn.7009815",{"tldr":265,"method":266,"finding":267,"direction":73,"opportunity":268},"提出语义风险感知运动规划方法，用于苹果疏果机械臂，提高作业安全与效率。","结合语义信息与风险度量，优化机械臂运动规划算法。","该方法能有效降低机械臂作业风险，提升疏果精度。","可探索将语义风险感知扩展到其他精细农业操作，如采摘、修剪，或结合实时传感器数据。","2026-09-04T23:30:01.595949Z",{"id":271,"title":272,"url":273,"summary":274,"summary_zh":275,"content":8,"source_name":276,"source_url":273,"published_at":277,"category":11,"cover_url":8,"hotness":12,"is_selected":13,"score":278,"score_detail":279,"sources":282,"tags":284,"primary_topic":288,"secondary_topics":289,"classification":290,"search_phrases":314,"slug":317,"view_count":68,"doi":318,"paper":319,"created_at":339},5401,"A HIERARCHICAL DISTRIBUTED EDGE AI FRAMEWORK FOR REAL-TIME APPLE LEAF DISEASE CLASSIFICATION","https:\u002F\u002Fdoi.org\u002F10.22190\u002Ffuacr260506002g","The application of artificial intelligence (AI) in agriculture has transformed crop management, particularly in the detection of plant diseases. Apple trees, a key agricultural commodity, are vulnerable to various diseases, which can severely impact crop yields. Traditional disease detection methods, often reliant on manual inspections, are time-consuming and prone to error. In response, this study proposes a novel two-stage edge AI framework for apple leaf disease classification that integrates RGB imagery with distributed edge computing. The first stage performs binary classification to distinguish between healthy and diseased leaves, achieving an F1 score of 95.50%. This optimization is achieved by utilizing lightweight models on edge devices, thereby enhancing computational efficiency. The second stage employs a consensus-based multiclass classification approach, achieving a 96.44% F1 score in identifying specific diseases. This dual-stage system, supported by an edge AI architecture, ensures real-time disease detection while minimizing resource consumption and enhancing scalability in large-scale agricultural environments. The full system achieved a 95.50% accuracy and F1 score on the special dataset. The results demonstrate the system’s efficacy in controlled laboratory conditions, offering a promising solution for precise, real-time agricultural monitoring. This study underscores the potential of distributed edge AI systems for scalable and efficient disease detection in precision agriculture.","人工智能（AI）在农业中的应用已深刻改变了作物管理方式，尤其是在植物病害检测方面。苹果树作为重要的农业经济作物，易受多种病害侵袭，严重影响作物产量。传统的病害检测方法通常依赖人工检查，耗时且易出错。为此，本研究提出了一种新颖的两阶段边缘AI框架，用于苹果叶片病害分类，该框架将RGB图像与分布式边缘计算相结合。第一阶段执行二分类以区分健康叶片与患病叶片，F1分数达到95.50%。该优化通过在边缘设备上采用轻量级模型实现，从而提升了计算效率。第二阶段采用基于共识的多类分类方法，在识别具体病害方面F1分数达到96.44%。该双阶段系统在边缘AI架构的支持下，实现了实时病害检测，同时最大限度地降低了资源消耗，并增强了在大规模农业环境中的可扩展性。完整系统在专用数据集上达到了95.50%的准确率和F1分数。结果表明，该系统在受控实验室条件下具有有效性，为精准、实时的农业监测提供了一种有前景的解决方案。本研究凸显了分布式边缘AI系统在精准农业中实现可扩展且高效病害检测的潜力。","Facta Universitatis Series Automatic Control and Robotics","2026-10-18T00:00:00Z",76,{"impact":16,"substance":90,"depth":280,"authority":213,"freshness":149,"relevant":20,"comment":281},17,"该研究提出两级边缘AI框架，在苹果叶片病害识别上取得95%以上F1值，方法新颖、数据可靠，对精准农业实时监测有参考价值，但尚处实验室验证阶段，产业影响有限。",[283],{"name":276,"url":273},[25,26,285,286,287],"边缘计算","作物病害识别","苹果病害",{"code":31,"slug":32,"name":33,"path":34,"parent_code":35},[],{"status":38,"locked":13,"version":39,"method":40,"evidence":291},{"candidates":292,"agricultural_evidence":45},[293,297,303,309],{"code":31,"score":16,"supported":45,"title":294,"abstract":295,"tags":296},[227],[33,48,227],[287],{"code":57,"score":215,"supported":45,"title":298,"abstract":299,"tags":302},[],[300,301],"精准农业","precision agriculture",[25],{"code":304,"score":215,"supported":45,"title":305,"abstract":306,"tags":308},"I5",[],[285,307],"edge computing",[285],{"code":51,"score":65,"supported":45,"title":310,"abstract":311,"tags":313},[],[312],"artificial intelligence",[286,26],[315,316],"苹果叶片病害 边缘AI","分布式边缘计算 病害分类","苹果叶片病害边缘AI-5401","10.22190\u002Ffuacr260506002g",{"doi":318,"openalex_id":320,"authors":321,"venue":276,"cited_by_count":68,"oa_url":273,"card":334,"direction":73,"ingested_from":138},"W7222964752",[322,325,328,331],{"name":323,"orcid":324},"Milan Grujev","https:\u002F\u002Forcid.org\u002F0009-0002-3211-6844",{"name":326,"orcid":327},"Dragan H. Stojanovic","https:\u002F\u002Forcid.org\u002F0000-0003-4893-3814",{"name":329,"orcid":330},"Miloš Ilić","https:\u002F\u002Forcid.org\u002F0000-0002-9696-2874",{"name":332,"orcid":333},"Aleksandar Milosavljević","https:\u002F\u002Forcid.org\u002F0000-0003-3057-7727",{"tldr":335,"method":336,"finding":337,"direction":201,"opportunity":338},"提出两级边缘AI框架，用RGB图像实时分类苹果叶片病害。","分布式边缘计算+轻量模型，先二分类再共识多分类。","两级系统准确率与F1均达95.50%，多分类F1为96.44%。","仅在受控实验室验证，可延伸至田间复杂光照与多作物泛化及边缘能耗优化。","2026-10-20T23:30:33.830748Z",{"id":341,"title":342,"url":343,"summary":344,"summary_zh":8,"content":8,"source_name":345,"source_url":8,"published_at":346,"category":11,"cover_url":8,"hotness":12,"is_selected":13,"score":347,"score_detail":348,"sources":350,"tags":352,"primary_topic":355,"secondary_topics":356,"classification":357,"search_phrases":377,"slug":380,"view_count":68,"doi":8,"paper":381,"created_at":388},5130,"DefoliationCAM:面向葡萄藤食叶严重度实时评估的边缘AI系统 [预印本]","https:\u002F\u002Farxiv.org\u002Fabs\u002F2610.20879","Fuchen Li等提出DefoliationCAM(D-CAM),一种太阳能供电的边缘AI系统,用于估算葡萄藤树冠食叶程度并通过LoRa将结果传输至Web仪表盘。流水线使用YOLOX-Nano在4K图像中定位葡萄藤主干,并使用基于MobileNet的轻量级通道注意力分类器对320×320像素的斑块进行分类。29,674个田间采集斑块被标注为背景或四种损伤等级之一;合并中度和重度类别将类别不平衡比从约40:1降至8.8:1。合成增强使macro-F1从0.768提升至0.838。分类器以更少的参数量(1.08M vs 23.52M)在macro-F1上优于ResNet-50。利用TensorRT FP16在NVIDIA Jetson Orin Nano上每张4K图像处理时间约1.5秒,完整运行周期57-60秒。两台设备为期五周的田间部署验证了自主太阳能运行、LoRa通信和远程食叶监测的可行性。","arXiv 2610.20879 2026 年 10 月 14 日","2026-10-14T00:00:00Z",75,{"impact":89,"substance":17,"depth":16,"authority":104,"freshness":149,"relevant":20,"comment":349},"预印本提出太阳能边缘AI系统，实现葡萄藤食叶严重度实时评估与LoRa远程监测，方法新颖、数据规模大、田间验证充分，具智慧农业应用价值。",[351],{"name":345,"url":343},[25,26,285,353,354],"葡萄藤","病害评估",{"code":31,"slug":32,"name":33,"path":34,"parent_code":35},[],{"status":38,"locked":13,"version":39,"method":40,"evidence":358},{"candidates":359,"agricultural_evidence":45},[360,365,369,373],{"code":31,"score":104,"supported":45,"title":361,"abstract":363,"tags":364},[362],"葡萄",[362],[],{"code":57,"score":20,"supported":13,"title":366,"abstract":367,"tags":368},[],[],[25],{"code":51,"score":20,"supported":13,"title":370,"abstract":371,"tags":372},[],[],[26],{"code":304,"score":20,"supported":13,"title":374,"abstract":375,"tags":376},[],[],[285],[378,379],"DefoliationCAM 葡萄藤 食叶","Jetson Orin Nano 边缘AI 葡萄","DefoliationCAM葡萄藤食叶-5130",{"doi":8,"openalex_id":8,"authors":382,"venue":8,"cited_by_count":68,"oa_url":8,"card":383,"direction":201,"ingested_from":75},[],{"tldr":384,"method":385,"finding":386,"direction":201,"opportunity":387},"提出太阳能边缘AI系统DefoliationCAM，实时评估葡萄藤食叶严重度并远程传输。","YOLOX-Nano定位主干，MobileNet通道注意力分类，TensorRT","轻量分类器macro-F1达0.838，优于ResNet-50，田间部署验证可行。","可探索多作物迁移、损伤等级细化及边缘模型自适应更新以提升泛化性。","2026-10-17T00:13:44.932340Z",{"id":390,"title":391,"url":392,"summary":393,"summary_zh":394,"content":8,"source_name":395,"source_url":392,"published_at":346,"category":11,"cover_url":8,"hotness":12,"is_selected":13,"score":396,"score_detail":397,"sources":399,"tags":401,"primary_topic":405,"secondary_topics":410,"classification":417,"search_phrases":468,"slug":471,"view_count":20,"doi":472,"paper":473,"created_at":494},4988,"Hierarchical cross-modal spatiotemporal attention network for crop yield prediction in complex Himalayan agroecosystems","https:\u002F\u002Fdoi.org\u002F10.3389\u002Ffpls.2026.1933033","Introduction Accurate crop yield prediction is essential for precision agriculture, climate-resilient farm management, and timely decision-making. However, yield prediction remains challenging because of complex, nonlinear, and multi-scale interactions among crop physiology, soil conditions, weather variability, and management practices. Conventional statistical, machine learning, and deep learning approaches, including convolutional neural networks (CNNs), recurrent neural networks (RNNs), and long short-term memory (LSTM) networks, often have limitations in capturing long-range temporal dependencies, hierarchical spatial relationships, heterogeneous data modalities, and predictive uncertainty. Methods This study proposes a Hierarchical Cross-Modal Spatio-Temporal Attention Network (HCM-STAN) for crop yield prediction using a unified attention-based framework for integrating heterogeneous agro-environmental data. The architecture incorporates four principal components: (i) hierarchical spatio-temporal attention operating across tree, block, orchard, and regional scales; (ii) a cross-modal attention module for adaptive fusion of IoT-based soil sensor observations, weather data, plant physiological measurements, and drone-based multispectral imagery; (iii) a soil-aware attention head that explicitly models soil memory, hysteresis effects, and root-zone dynamics; and (iv) an uncertainty-aware prediction module combining Monte Carlo dropout and quantile regression to generate calibrated prediction intervals. The framework was evaluated using a multi-source dataset collected from Himalayan apple orchards during 2023–2026, comprising 1,247 monitored trees across three commercial orchards. Results The proposed HCM-STAN achieved an R 2 of 0.89 for four-week-ahead crop yield prediction, representing a 15.6% improvement over the CNN-LSTM baseline. Furthermore, zero-shot transfer to mango orchards achieved an R 2 of 0.67, corresponding to approximately 75% of the performance obtained under full training. Discussion The results demonstrate that integrating hierarchical spatio-temporal representation learning, cross-modal attention, soil-aware modelling, and uncertainty estimation can effectively address the heterogeneous and dynamic nature of crop yield prediction in complex mountain agroecosystems. The observed zero-shot transfer performance further indicates the potential of HCM-STAN to generalize across crop systems without crop-specific retraining. Overall, the proposed framework provides a promising foundation for reliable, uncertainty-aware yield forecasting and can support precision agriculture, climate-smart farm management, and data-driven decision-making in heterogeneous agroecosystems.","引言 准确的作物产量预测对于精准农业、气候适应性农场管理以及及时决策至关重要。然而，由于作物生理、土壤条件、天气变异和管理措施之间复杂、非线性且多尺度的相互作用，产量预测仍然具有挑战性。传统的统计方法、机器学习以及深度学习方法，包括卷积神经网络（CNN）、循环神经网络（RNN）和长短期记忆（LSTM）网络，在捕捉长程时间依赖关系、层次化空间关系、异构数据模态以及预测不确定性方面往往存在局限。方法 本研究提出了一种用于作物产量预测的层次化跨模态时空注意力网络（HCM-STAN），采用统一的基于注意力的框架来整合异构农业环境数据。该架构包含四个主要组成部分：（i）在树、地块、果园和区域尺度上运行的层次化时空注意力；（ii）用于自适应融合基于物联网的土壤传感器观测、天气数据、植物生理测量和无人机多光谱影像的跨模态注意力模块；（iii）显式建模土壤记忆、滞后效应和根区动态的土壤感知注意力头；（iv）结合蒙特卡洛 dropout 和分位数回归以生成校准预测区间的不确定性感知预测模块。该框架使用2023—2026年在喜马拉雅苹果园采集的多源数据集进行了评估，该数据集包含三个商业果园中1，247棵监测树木。结果 所提出的 HCM-STAN 在提前四周作物产量预测中达到了0.89的 R²，相较于 CNN-LSTM 基线提升了15.6%。此外，零样本迁移至芒果园达到了0.67的 R²，相当于完整训练下所获性能的约75%。讨论 结果表明，整合层次化时空表征学习、跨模态注意力、土壤感知建模和不确定性估计，能够有效应对复杂山地农业生态系统中作物产量预测的异构性和动态性。观察到的零样本迁移性能进一步表明，HCM-STAN 具有在无需作物特定再训练的情况下跨作物系统泛化的潜力。总体而言，该 p","Frontiers in Plant Science",81,{"impact":16,"substance":17,"depth":16,"authority":91,"freshness":19,"relevant":20,"comment":398},"提出分层跨模态时空注意力网络，融合物联网土壤传感、气象、生理与无人机多光谱数据，苹果园四周产量预测R²达0.89并实现芒果园零样本迁移，方法新颖、数据扎实，对智慧农业产量预报有参考价值。",[400],{"name":395,"url":392},[25,26,402,403,404],"产量预测","多模态融合","遥感",{"code":51,"slug":406,"name":407,"path":408,"parent_code":409},"agricultural-ai","农业AI与大模型","\u002Ftopic\u002Fdigital-agriculture\u002Fagricultural-ai","I",[411,412],{"code":31,"slug":32,"name":33,"path":34,"parent_code":35},{"code":413,"slug":414,"name":415,"path":416,"parent_code":409},"I3","remote-sensing","遥感与农情监测","\u002Ftopic\u002Fdigital-agriculture\u002Fremote-sensing",{"status":38,"locked":13,"version":39,"method":40,"evidence":418},{"candidates":419,"agricultural_evidence":45},[420,428,433,439,443,450,456,462],{"code":51,"score":104,"supported":45,"title":421,"abstract":422,"tags":427},[],[423,424,425,426],"深度学习","机器学习","machine learning","deep learning",[26,403],{"code":31,"score":19,"supported":45,"title":429,"abstract":430,"tags":432},[],[48,431,106,227],"芒果",[],{"code":413,"score":149,"supported":45,"title":434,"abstract":435,"tags":438},[],[436,437],"无人机多光谱","multispectral",[402,404],{"code":57,"score":215,"supported":45,"title":440,"abstract":441,"tags":442},[],[300,301],[25],{"code":444,"score":78,"supported":45,"title":445,"abstract":446,"tags":449},"H2",[],[447,448],"土壤","soil",[],{"code":304,"score":78,"supported":45,"title":451,"abstract":452,"tags":455},[],[453,454],"物联网","IoT",[],{"code":457,"score":44,"supported":45,"title":458,"abstract":459,"tags":461},"H6",[],[460],"农业生态",[],{"code":463,"score":44,"supported":45,"title":464,"abstract":465,"tags":467},"H7",[],[466],"气候适应",[],[469,470],"喜马拉雅 苹果园 产量预测","HCM-STAN 作物产量预测","喜马拉雅苹果园产量预测-4988","10.3389\u002Ffpls.2026.1933033",{"doi":472,"openalex_id":474,"authors":475,"venue":395,"cited_by_count":68,"oa_url":392,"card":489,"direction":201,"ingested_from":138},"W7221795850",[476,479,481,483,485,487],{"name":477,"orcid":478},"Shabir Ahmad Mir","https:\u002F\u002Forcid.org\u002F0000-0003-2311-218X",{"name":480,"orcid":8},"Umarfarooque Momin",{"name":482,"orcid":8},"Sudeep Kumar E",{"name":484,"orcid":8},"Itfaq Ahmad Mir",{"name":486,"orcid":8},"Ab Rouf Khan",{"name":488,"orcid":8},"Ghulam Mohammad Mir",{"tldr":490,"method":491,"finding":492,"direction":73,"opportunity":493},"提出分层跨模态时空注意力网络HCM-STAN，融合多源农业数据预测喜马拉雅山区苹果产量。","用IoT土壤、气象、生理和无人机多光谱数据，结合分层时空与跨模态注意力及不确定性","四周前产量预测R²达0.89，比CNN-LSTM提升15.6%，零样本迁移芒果R²为0.67。","可探索跨作物零样本迁移的领域自适应机制，并验证模型在更多山地农业生态系统的泛化能力。","2026-10-15T23:30:08.750140Z"]