[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"item-4320":3,"related-4320":72},{"id":4,"title":5,"url":6,"summary":7,"summary_zh":8,"content":9,"source_name":10,"source_url":6,"published_at":11,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":15,"score_detail":16,"sources":18,"tags":20,"primary_topic":24,"secondary_topics":30,"classification":31,"search_phrases":43,"slug":46,"view_count":15,"doi":47,"paper":48,"created_at":71},4320,"Design and stability analysis of an underactuated hand with passively rotating fingers","https:\u002F\u002Fdoi.org\u002F10.1115\u002F1.4072741","Abstract This paper presents an innovative design and stability analysis of an underactuated robotic finger with spatial mobility, designed to enhance gripping dexterity in robotic hands. The finger architecture incorporates a revolute joint at its base, enabling passive spatial rotation that facilitates both cylindrical and spherical grasping. With only two phalanges per finger, the design simplifies kinematic complexity while supporting precision and enveloping grasps. Stability criteria, based on the moment at the finger base joint induced by contact forces, are introduced to ensure reliable object gripping and prevent ejection during manipulation. The study also examines a differential mechanism that distributes a single actuation torque across multiple fingers, allowing adaptive and coordinated motion. This mechanism enhances the hand's ability to grasp diverse object shapes with minimal pre-grasp adjustments, leveraging passivity for autonomous adaptation. Theoretical findings are experimentally validated using a fully mechanical prototype, demonstrating versatility in performing cylindrical, spherical, parallel, and enveloping grasps. The integration of underactuation—both within individual fingers and among multiple fingers—reduces mechanical complexity, cost, and control effort while preserving functional adaptability. This work advances the development of compliant robotic hands suitable for applications requiring dexterity and robustness, such as agricultural robotics, logistics, assistive technologies, and waste sorting. Future research will focus on automating actuation and refining control strategies to further improve grasp stability and precision, paving the way for autonomous manipulation in unstructured environments.","本文提出了一种具有空间运动能力的欠驱动机械手指的创新设计与稳定性分析，旨在提升机器人手的抓取灵巧性。该手指结构在其基部集成了一个转动关节，能够实现被动空间旋转，从而支持圆柱形和球形抓取。每根手指仅有两个指节，该设计在简化运动学复杂度的同时支持精确抓取和包络抓取。本文引入了基于接触力在手指基部关节处产生的力矩的稳定性判据，以确保可靠的物体抓取并防止操作过程中的弹出。研究还考察了一种差动机构，该机构将单一驱动力矩分配至多根手指，实现自适应且协调的运动。该机构增强了手部以最少预抓取调整抓取多种物体形状的能力，利用被动性实现自主适应。理论研究成果通过全机械原型进行了实验验证，展示了执行圆柱形、球形、平行和包络抓取的多功能性。欠驱动——既包括单根手指内部的欠驱动，也包括多根手指之间的欠驱动——的集成降低了机械复杂度、成本和控制难度，同时保持了功能适应性。这项工作推进了顺应性机器人手的发展，适用于需要灵巧性和鲁棒性的应用场景，如农业机器人、物流、辅助技术和废物分拣。未来研究将聚焦于驱动自动化与控制策略优化，以进一步提升抓取稳定性和精度，为非结构化环境中的自主操作铺平道路。",null,"ASME Letters in Translational Robotics","2026-10-05T00:00:00Z","论文",10,false,0,{"impact":15,"substance":15,"depth":15,"authority":15,"freshness":15,"relevant":15,"comment":17},"该论文为机器人学基础研究，仅摘要末尾提及农业机器人应用场景，与三农、农业信息化主题无实质关联，不建议进入每日精选。",[19],{"name":10,"url":6},[21,22,23],"农业机器人","机器人灵巧手","欠驱动抓取",{"code":25,"slug":26,"name":27,"path":28,"parent_code":29},"F6","agricultural-robots","机器人与无人农场","\u002Ftopic\u002Finputs-equipment\u002Fagricultural-robots","F",[],{"status":32,"locked":14,"version":33,"method":34,"evidence":35},"classified","2026-10-22.2","rules",{"candidates":36,"agricultural_evidence":39},[37],{"code":25,"score":38,"supported":39,"title":40,"abstract":41,"tags":42},4,true,[],[21],[21],[44,45],"机器人灵巧手 农业机器人 欠驱动抓取","机器人灵巧手 农业机器人","机器人灵巧手农业机器人欠驱动抓取-4320","10.1115\u002F1.4072741",{"doi":47,"openalex_id":49,"authors":50,"venue":10,"cited_by_count":15,"oa_url":62,"card":63,"direction":69,"ingested_from":70},"W7170070531",[51,53,56,59],{"name":52,"orcid":9},"Léonie Plancoulaine",{"name":54,"orcid":55},"Sylvain Guégan","https:\u002F\u002Forcid.org\u002F0000-0002-8797-6078",{"name":57,"orcid":58},"Franck Plestan","https:\u002F\u002Forcid.org\u002F0000-0001-8971-5106",{"name":60,"orcid":61},"Damien Chablat","https:\u002F\u002Forcid.org\u002F0000-0001-7847-6162","https:\u002F\u002Farxiv.org\u002Fpdf\u002F2607.18950",{"tldr":64,"method":65,"finding":66,"direction":67,"opportunity":68},"设计并分析了一种欠驱动空间旋转手指的机械手，实现多种抓取并验证稳定性。","采用被动旋转关节与差动机构，仅两指节，理论建模加机械原型实验验证。","欠驱动设计降低复杂度与成本，同时保持对多种形状的自适应抓取能力。","其他","可探索将此类欠驱动手用于农业非结构化环境，结合视觉与自适应控制实现自主采摘。","农业人工智能与决策模型","openalex","2026-10-06T23:30:33.618628Z",{"total":73,"page":74,"page_size":73,"items":75},6,1,[76,130,196,267,326,379],{"id":77,"title":78,"url":79,"summary":80,"summary_zh":9,"content":81,"source_name":82,"source_url":9,"published_at":83,"category":84,"cover_url":9,"hotness":13,"is_selected":14,"score":85,"score_detail":86,"sources":92,"tags":94,"primary_topic":99,"secondary_topics":100,"classification":101,"search_phrases":125,"slug":128,"view_count":15,"doi":9,"paper":9,"created_at":129},5420,"广西印发具身智能产业创新发展五年行动计划 布局具身智能加现代农业场景","https:\u002F\u002Fwww.nanning.gov.cn\u002Fzt\u002Frdzt\u002Fyhyshj\u002Fyszc\u002Fbrphcx\u002Ft6740281.html","广西壮族自治区党委科技委员会办公室印发《具身智能产业创新发展五年行动计划（2026—2030年）》（桂科委办发〔2026〕12号），在农业领域明确要求凝练具身智能在农产品采摘、优势作物种植、养殖巡检等真实场景的技术需求，面向矿山开采、农产品采摘、海洋渔业、物流运输等场景开展特种机器人技术攻关。行动提出到2030年组织实施30项以上重点科技攻关项目、培育2家以上具身智能龙头企业、10家以上科技型中小企业、打造20个以上应用场景标杆，其中包括「具身智能+现代农业」场景。农业机器人与采摘装备由此获得明确的地方产业政策支持路径。","（十）挖掘特色应用场景需求。建立健全具身智能应用场景挖掘机制，面向企业、高校、科研院所、医疗机构等，在工业制造、教育培训、科研应用、民生服务等领域常态化征集具身智能应用场景需求，挖掘一批技术适配性强、应用价值高、示范效应好、可复制推广的具身智能应用场景。（责任单位：自治区数据局、发展改革委、教育厅、科技厅、工业和信息化厅、卫生健康委等）\n\n（十一）大力培育标杆应用场景。聚焦广西有色金属及关键金属材料、先进钢铁材料、现代绿色化工、新一代信息技术、机械及高端装备制造、新能源汽车及电池、林产加工及纸业、食品加工、生物医药及健康、绿色建材等十大支柱产业发展需求，围绕矿山无人化、高危环境巡检、材料分拣、精密装配、物料搬运、智能检测、远程运维、智能仓储等重要环节，打造“具身智能+制造业”应用场景；在优势作物种植、采摘、养殖巡检等现代农业领域开展创新应用与示范推广，打造“具身智能+现代农业”应用场景；在港口物流、能源、文旅等领域拓展具身智能应用，打造一批融合应用场景。到2030年，打造20个以上应用场景标杆。（责任单位：自治区科技厅、发展改革委、工业和信息化厅、交通运输厅、农业农村厅、文化和旅游厅、数据局等）","南宁市人民政府门户网站 2026-10-15","2026-10-15T00:00:00Z","政策",71,{"impact":87,"substance":88,"depth":89,"authority":90,"freshness":73,"relevant":74,"comment":91},18,20,16,11,"省级具身智能五年行动计划明确布局“具身智能+现代农业”场景，含2030年20个标杆等量化目标，政策条款具体，对农业机器人方向有实质参考价值。",[93],{"name":82,"url":79},[95,21,96,97,98],"智慧农业","具身智能","现代农业","产业规划",{"code":25,"slug":26,"name":27,"path":28,"parent_code":29},[],{"status":32,"locked":14,"version":33,"method":34,"evidence":102},{"candidates":103,"agricultural_evidence":39},[104,108,115,120],{"code":25,"score":38,"supported":39,"title":105,"abstract":106,"tags":107},[],[21],[21],{"code":109,"score":110,"supported":39,"title":111,"abstract":112,"tags":114},"D",3,[],[113],"渔业",[],{"code":116,"score":74,"supported":14,"title":117,"abstract":118,"tags":119},"I1",[],[],[95],{"code":121,"score":74,"supported":14,"title":122,"abstract":123,"tags":124},"J2",[],[],[97],[126,127],"广西 具身智能 现代农业","具身智能 农业机器人 应用场景","广西具身智能现代农业-5420","2026-10-21T00:09:54.769055Z",{"id":131,"title":132,"url":133,"summary":134,"summary_zh":135,"content":9,"source_name":136,"source_url":133,"published_at":137,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":138,"score_detail":139,"sources":144,"tags":146,"primary_topic":150,"secondary_topics":151,"classification":152,"search_phrases":170,"slug":173,"view_count":74,"doi":174,"paper":175,"created_at":195},5240,"Low-Cost Car-Like Autonomous Mobile Robot Under-Canopy Row Following, and Detection of Patterns Associated with Diseases in Plantain Crops","https:\u002F\u002Fdoi.org\u002F10.3991\u002Fijoe.v22i10.61223","Labor constraints and the need for frequent visual inspection of lower plantain (Musa paradisiaca) leaves in Colombia’s Orinoquia region motivate the development of low-cost robotic tools capable of autonomous under-canopy navigation and systematic image capture. This study reports the current stage of an ongoing research effort on a low-cost, small-scale, car-like agricultural robot for autonomous under-canopy row-following and systematic image capture in plantain fields. The platform uses a modular ROS 2 workflow, a low-cost 2D LiDAR, an inertial measurement unit, and an ESP32-based actuation layer. LiDAR data were processed through an adaptive measurement cone guided by the vehicle heading error to estimate the lateral distance to the crop row during forward row-following motion, and a fuzzy controller generated steering references. Initial field tests across five trials provided preliminary evidence of consistent under-canopy navigation under the tested conditions. In the first rowfollowing segment, the median MAE and RMSE of the lateral error were 0.0258 m and 0.0342 m, respectively; after the row-to-row maneuver, they were 0.0356 m and 0.0602 m, respectively. Additionally, an upward-facing RGB camera captured under-canopy images that were processed offline to create probability-based visual summaries of leaf-condition patterns. The results show that low-cost robotic navigation can help structure under-canopy data collection for scalable crop health monitoring in plantain plantations.","哥伦比亚奥里诺基亚地区劳动力受限，且需要对下部车前草（Musa paradisiaca）叶片进行频繁目视检查，这促使人们开发能够自主冠层下导航并系统采集图像的低成本机器人工具。本研究报道了针对一种低成本、小型、类车式农业机器人开展的在研工作的当前阶段，该机器人用于车前草田间的自主冠层下行跟随与系统图像采集。该平台采用模块化ROS 2工作流、低成本二维激光雷达、惯性测量单元以及基于ESP32的驱动层。激光雷达数据通过由车辆航向误差引导的自适应测量锥进行处理，以在前进行跟随运动中估计与作物行的横向距离，并由模糊控制器生成转向参考。在五种试验条件下的初步田间测试提供了在所测条件下冠层下导航具有一致性的初步证据。在第一个行跟随段中，横向误差的中位MAE和RMSE分别为0.0258 m和0.0342 m；在行间机动之后，分别为0.0356 m和0.0602 m。此外，一台朝上的RGB相机采集了冠层下图像，并对其进行离线处理，以生成基于概率的叶片状况模式视觉摘要。结果表明，低成本机器人导航有助于构建冠层下数据采集流程，从而支持车前草种植园中可扩展的作物健康监测。","International Journal of Online and Biomedical Engineering (iJOE)","2026-10-16T00:00:00Z",70,{"impact":140,"substance":88,"depth":141,"authority":140,"freshness":142,"relevant":74,"comment":143},12,17,9,"低成本ROS 2农业机器人实现冠层下自主行跟随与叶片病害图像采集，方法新颖、田间数据初步可靠，但属细分领域阶段性成果，公共影响有限。",[145],{"name":136,"url":133},[95,21,147,148,149],"自主导航","作物病害检测","香蕉种植",{"code":25,"slug":26,"name":27,"path":28,"parent_code":29},[],{"status":32,"locked":14,"version":33,"method":34,"evidence":153},{"candidates":154,"agricultural_evidence":39},[155,161,166],{"code":25,"score":156,"supported":39,"title":157,"abstract":158,"tags":160},7,[],[21,159],"agricultural robot",[21],{"code":162,"score":74,"supported":14,"title":163,"abstract":164,"tags":165},"F5",[],[],[147],{"code":116,"score":74,"supported":14,"title":167,"abstract":168,"tags":169},[],[],[95],[171,172],"低成本农业机器人 大蕉 冠层导航","plantain 病害 图像检测","低成本农业机器人大蕉冠层导航-5240","10.3991\u002Fijoe.v22i10.61223",{"doi":174,"openalex_id":176,"authors":177,"venue":136,"cited_by_count":15,"oa_url":188,"card":189,"direction":193,"ingested_from":70},"W7222629670",[178,181,183,185],{"name":179,"orcid":180},"Danna Celena Montenegro","https:\u002F\u002Forcid.org\u002F0009-0003-0909-7806",{"name":182,"orcid":9},"Iván D. Duarte",{"name":184,"orcid":9},"Andres Leonardo Jutinico",{"name":186,"orcid":187},"Henry Borrero Guerrero","https:\u002F\u002Forcid.org\u002F0000-0003-4243-4205","https:\u002F\u002Fonline-journals.org\u002Findex.php\u002Fi-joe\u002Farticle\u002Fdownload\u002F61223\u002F17460",{"tldr":190,"method":191,"finding":192,"direction":193,"opportunity":194},"开发低成本类车机器人实现车前草冠层下自主行跟随与图像采集。","ROS 2、2D LiDAR、IMU、ESP32及模糊控制器，离线处理RGB图像","横向误差中位数MAE约0.026-0.036米，验证了低成本导航可行性。","智慧农业 \u002F 农业物联网","可研究多行自主切换、病害图像在线识别及复杂冠层下鲁棒导航。","2026-10-18T23:30:09.565705Z",{"id":197,"title":198,"url":199,"summary":200,"summary_zh":201,"content":9,"source_name":202,"source_url":199,"published_at":203,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":204,"score_detail":205,"sources":209,"tags":211,"primary_topic":215,"secondary_topics":216,"classification":217,"search_phrases":228,"slug":231,"view_count":74,"doi":232,"paper":233,"created_at":266},4991,"Direct‐Printed Textile Circuits for Smart Gloves With Enhanced Integration","https:\u002F\u002Fdoi.org\u002F10.1002\u002Fadem.71290","Integrating deformable sensors, conductive routing, embedded electronics, and local data processing on a textile carrier remains challenging for wearable gesture interfaces. Here, a direct‐printed textile circuit smart glove was constructed using a hybrid textile–electronic architecture. Low‐resistance carbon paste and Ag paste were sequentially screen‐printed to form the strain‐sensitive layer, electrodes, interconnects, and component pads. Packaged modules were attached with silver‐filled epoxy electrically conductive adhesive and protected by localized silicone‐rubber encapsulation. The sensor produced a linear response over 2.5%–15% strain ( R 2 = 0.9995), with response and recovery times of approximately 200 and 100 ms, and stable electrical output during a 10 000s test corresponding to about 4000 cycles at 0.4 Hz. An ESP32‐based threshold classifier processed five finger channels locally and recognized eight static gestures with an average accuracy of 95.2% over 480 trials. The encoded gestures were then used to command a uHand 2.0 bionic hand and an agricultural robot. These results demonstrate the system‐level feasibility of the hybrid textile–electronic architecture for integrating textile sensing, on‐device gesture recognition, wireless communication, and robotic command execution.","将可变形传感器、导电布线、嵌入式电子器件和本地数据处理集成到纺织品载体上，对于可穿戴手势接口而言仍具挑战性。本文采用混合纺织-电子架构构建了一种直写印刷纺织电路智能手套。通过依次丝网印刷低阻碳浆和银浆，形成应变敏感层、电极、互连线和元件焊盘。封装模块采用银填充环氧导电胶粘接，并通过局部硅橡胶封装进行保护。该传感器在2.5%–15%应变范围内呈线性响应（R² = 0.9995），响应时间和恢复时间分别约为200 ms和100 ms，在10 000 s测试（对应0.4 Hz下约4000次循环）中电输出保持稳定。基于ESP32的阈值分类器在本地处理五个手指通道，在480次试验中以平均95.2%的准确率识别八种静态手势。随后，编码手势被用于控制uHand 2.0仿生手和一台农业机器人。这些结果证明了混合纺织-电子架构在集成纺织传感、设备端手势识别、无线通信和机器人指令执行方面的系统级可行性。","Advanced Engineering Materials","2026-10-14T00:00:00Z",81,{"impact":87,"substance":206,"depth":87,"authority":207,"freshness":142,"relevant":74,"comment":208},22,14,"直印纺织电路智能手套实现本地手势识别并驱动农业机器人，系统级验证扎实，对农业人机交互有参考价值。",[210],{"name":202,"url":199},[95,21,212,213,214],"可穿戴设备","人机交互","柔性传感",{"code":25,"slug":26,"name":27,"path":28,"parent_code":29},[],{"status":32,"locked":14,"version":33,"method":34,"evidence":218},{"candidates":219,"agricultural_evidence":39},[220,224],{"code":25,"score":156,"supported":39,"title":221,"abstract":222,"tags":223},[],[21,159],[21],{"code":116,"score":74,"supported":14,"title":225,"abstract":226,"tags":227},[],[],[95],[229,230],"直印纺织电路 智能手套","ESP32 手势识别 农业机器人","直印纺织电路智能手套-4991","10.1002\u002Fadem.71290",{"doi":232,"openalex_id":234,"authors":235,"venue":202,"cited_by_count":15,"oa_url":9,"card":261,"direction":193,"ingested_from":70},"W7221824460",[236,238,240,242,244,246,248,250,253,255,258],{"name":237,"orcid":9},"Hao Wu",{"name":239,"orcid":9},"Huimin Ren",{"name":241,"orcid":9},"Shubei Zhi",{"name":243,"orcid":9},"Xutao Huang",{"name":245,"orcid":9},"Junxian Guo",{"name":247,"orcid":9},"Siyu Li",{"name":249,"orcid":9},"Lingxiao Cao",{"name":251,"orcid":252},"Zhonghao Wang","https:\u002F\u002Forcid.org\u002F0009-0004-7227-3929",{"name":254,"orcid":9},"Yunlong Ni",{"name":256,"orcid":257},"Zhi‐Zhu He","https:\u002F\u002Forcid.org\u002F0000-0001-8861-9964",{"name":259,"orcid":260},"Zhen Li","https:\u002F\u002Forcid.org\u002F0000-0001-6010-0298",{"tldr":262,"method":263,"finding":264,"direction":193,"opportunity":265},"开发直接印刷纺织电路智能手套，集成传感、本地手势识别与机器人控制。","丝网印刷碳\u002F银浆料制备柔性电路，ESP32本地阈值分类，控制仿生手与农业机器人。","传感器线性度R²=0.9995，8种手势识别平均准确率95.2%，可稳定控制机器人。","可探索将智能手套用于农业机器人遥操作、采摘辅助或农户动作数据采集与技能数字化。","2026-10-15T23:30:09.135126Z",{"id":268,"title":269,"url":270,"summary":271,"summary_zh":9,"content":9,"source_name":272,"source_url":9,"published_at":273,"category":84,"cover_url":9,"hotness":13,"is_selected":14,"score":274,"score_detail":275,"sources":279,"tags":281,"primary_topic":285,"secondary_topics":286,"classification":287,"search_phrases":321,"slug":324,"view_count":15,"doi":9,"paper":9,"created_at":325},4837,"山东出台数字强省建设'十五五'规划 建设特色农业'大脑'支持30个智慧农业应用场景","https:\u002F\u002Fnew.qq.com\u002Frain\u002Fa\u002F20261012A08ZIW00","山东发布数字强省建设'十五五'规划,在农业数智化方面,实施'人工智能+农业'行动,开展农业传感器与专用芯片、模型算法等研发攻关,大力发展智能农机与农业机器人,支持'无人农场'建设。建设特色农业'大脑',省级每年支持建设30个智慧农业应用场景,支持德州、聊城'吨半粮'等引领区数智化建设。建好用好'齐鲁农云''齐鲁农超''好粮有网'等平台,打造农业农村'数字大脑',建设80个数字农业发展县。","腾讯新闻 2026 年 10 月 12 日","2026-10-12T00:00:00Z",77,{"impact":206,"substance":276,"depth":87,"authority":73,"freshness":277,"relevant":74,"comment":278},23,8,"省级数字强省规划含农业数智化具体条款与量化目标，信息增量足，值得进入每日精选。",[280],{"name":272,"url":270},[282,95,283,284,21],"数字乡村","无人农场","农业人工智能",{"code":25,"slug":26,"name":27,"path":28,"parent_code":29},[],{"status":32,"locked":14,"version":33,"method":34,"evidence":288},{"candidates":289,"agricultural_evidence":39},[290,296,300,305,311,316],{"code":116,"score":291,"supported":39,"title":292,"abstract":293,"tags":295},15,[95],[95,294],"数字农业",[95],{"code":25,"score":277,"supported":39,"title":297,"abstract":298,"tags":299},[],[21,283],[21,283],{"code":162,"score":110,"supported":39,"title":301,"abstract":302,"tags":304},[],[303],"农机",[],{"code":306,"score":110,"supported":39,"title":307,"abstract":308,"tags":310},"I5",[],[309],"农业传感",[],{"code":312,"score":74,"supported":14,"title":313,"abstract":314,"tags":315},"I2",[],[],[284],{"code":317,"score":74,"supported":14,"title":318,"abstract":319,"tags":320},"I6",[],[],[282],[322,323],"山东 数字强省 十五五规划","齐鲁农云 智慧农业应用场景","山东数字强省十五五规划-4837","2026-10-14T00:04:01.888905Z",{"id":327,"title":328,"url":329,"summary":330,"summary_zh":9,"content":9,"source_name":331,"source_url":9,"published_at":332,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":333,"score_detail":334,"sources":336,"tags":338,"primary_topic":341,"secondary_topics":342,"classification":343,"search_phrases":366,"slug":369,"view_count":15,"doi":9,"paper":370,"created_at":378},4734,"面向农业机器人作物行分割的边缘可部署轻量感知网络","https:\u002F\u002Farxiv.org\u002Fabs\u002F2610.11764","提出仅1.31万参数、逐层渐进学习的作物行分割网络,推理能耗25.1毫焦,较主流网络参数量显著降低且保持检测精度。适配资源受限田间机器人部署,定位低成本农机边缘智能新路径。未经实地验证未经同行评议","arXiv 2026-10-08 [预印本]","2026-10-08T00:00:00Z",66,{"impact":207,"substance":88,"depth":141,"authority":277,"freshness":156,"relevant":74,"comment":335},"预印本提出超轻量作物行分割网络，参数与能耗指标亮眼，但未经实地验证与同行评议，属细分技术进展，可作精选备选。",[337],{"name":331,"url":329},[95,284,21,339,340],"边缘计算","作物行识别",{"code":25,"slug":26,"name":27,"path":28,"parent_code":29},[],{"status":32,"locked":14,"version":33,"method":34,"evidence":344},{"candidates":345,"agricultural_evidence":39},[346,350,354,358,362],{"code":25,"score":142,"supported":39,"title":347,"abstract":348,"tags":349},[21],[],[21],{"code":162,"score":110,"supported":39,"title":351,"abstract":352,"tags":353},[],[303],[],{"code":116,"score":74,"supported":14,"title":355,"abstract":356,"tags":357},[],[],[95],{"code":312,"score":74,"supported":14,"title":359,"abstract":360,"tags":361},[],[],[284],{"code":306,"score":74,"supported":14,"title":363,"abstract":364,"tags":365},[],[],[339],[367,368],"作物行分割 轻量网络 农业机器人","边缘部署 作物行 田间机器人","作物行分割轻量网络农业机器人-4734",{"doi":9,"openalex_id":9,"authors":371,"venue":9,"cited_by_count":15,"oa_url":9,"card":372,"direction":193,"ingested_from":377},[],{"tldr":373,"method":374,"finding":375,"direction":193,"opportunity":376},"提出仅1.31万参数的轻量作物行分割网络，适配田间机器人边缘部署。","逐层渐进学习，参数量1.31万，推理能耗25.1毫焦。","参数量显著降低且保持检测精度，定位低成本农机边缘智能路径。","该网络未经实地验证，可开展田间试验并探索多作物、多光照条件下的泛化能力。","agent","2026-10-12T00:11:14.925915Z",{"id":380,"title":381,"url":382,"summary":383,"summary_zh":9,"content":9,"source_name":384,"source_url":9,"published_at":385,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":386,"score_detail":387,"sources":389,"tags":391,"primary_topic":395,"secondary_topics":400,"classification":402,"search_phrases":424,"slug":427,"view_count":428,"doi":9,"paper":429,"created_at":436},4648,"[预印本]面向农业机器人在多变天气与地形下基于视觉-语言模型的上下文自适应农药喷洒(Context-Aware Adaptive Pesticide Spraying for Agricultural Robots under Changing Weather and Terrain Using Vision-Language Models)","https:\u002F\u002Farxiv.org\u002Fabs\u002F2610.08807","arXiv论文(2610.08807,2026年10月6日提交)提出基于视觉-语言模型(VLM)的上下文自适应喷洒框架,通过空间推理整合作物、农药、气象数据驱动喷洒决策,并结合MPPI轨迹跟踪控制。作物行检测精度至少提升30%,机器人可在两种环境中灵活调节喷洒量与行驶速度并减少农药漂移。[预印本未经同行评议,尚未实地部署验证]。","arXiv 2026-10-06 [预印本]","2026-10-06T00:00:00Z",72,{"impact":87,"substance":88,"depth":87,"authority":13,"freshness":73,"relevant":74,"comment":388},"预印本提出VLM驱动的自适应喷洒框架，方法新颖且减药减排指向明确，但未经同行评议与实地验证，宜作前沿动态收录而非重点精选。",[390],{"name":384,"url":382},[95,21,392,393,394],"精准施药","视觉语言模型","农药减量",{"code":396,"slug":397,"name":398,"path":399,"parent_code":29},"F4","crop-protection","农药与绿色防控","\u002Ftopic\u002Finputs-equipment\u002Fcrop-protection",[401],{"code":25,"slug":26,"name":27,"path":28,"parent_code":29},{"status":32,"locked":14,"version":33,"method":34,"evidence":403},{"candidates":404,"agricultural_evidence":39},[405,412,416,420],{"code":396,"score":406,"supported":39,"title":407,"abstract":410,"tags":411},21,[408,409],"农药","pesticide",[408],[394,392],{"code":25,"score":142,"supported":39,"title":413,"abstract":414,"tags":415},[21],[],[21],{"code":116,"score":74,"supported":14,"title":417,"abstract":418,"tags":419},[],[],[95],{"code":312,"score":74,"supported":14,"title":421,"abstract":422,"tags":423},[],[],[393],[425,426],"农业机器人 视觉语言模型 农药喷洒","MPPI 轨迹跟踪 作物行检测","农业机器人视觉语言模型农药喷洒-4648",2,{"doi":9,"openalex_id":9,"authors":430,"venue":9,"cited_by_count":15,"oa_url":9,"card":431,"direction":193,"ingested_from":377},[],{"tldr":432,"method":433,"finding":434,"direction":193,"opportunity":435},"提出基于视觉-语言模型的上下文自适应农药喷洒框架，结合MPPI控制实现精准喷洒。","视觉-语言模型空间推理，融合作物、农药与气象数据，MPPI轨迹跟踪控制。","作物行检测精度至少提升30%，可灵活调节喷洒量与速度并减少农药漂移。","预印本未实地验证，可探索真实多变天气地形下的VLM鲁棒性与多模态数据融合优化。","2026-10-11T00:09:47.810289Z"]