[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"item-5362":3,"related-5362":79},{"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":64,"slug":67,"view_count":68,"doi":8,"paper":69,"created_at":78},5362,"深度语义分割模型的时序泛化评估:无人机小麦倒伏监测 5 年跨年测试","https:\u002F\u002Fbioengineer.org\u002Fai-models-trained-on-one-year-of-wheat-fail-the-next-five-year-drone-study-shows\u002F","Zhang G., Wang X., Wang D., Sun D., Zhang Y., Zhang D. (2026) 利用 5 年田间试验评估深度语义分割模型在无人机(UAV)小麦倒伏监测中的时序泛化能力。研究发现:同年内高精度不能保证跨年可靠性,模型从训练年份迁移到其他年份时性能显著下降,基于注意力的模型通常比全卷积网络更稳健,可能因为层级窗口自注意力能整合更长范围上下文,而卷积强调局部模式;即便最优架构在大时序漂移下也出现严重退化。强调时序泛化测试应成为农业 AI 模型部署前的必备环节。DOI: 10.1016\u002Fj.atech.2026.102631。",null,"Smart Agricultural Technology Vol.15 Article 102631 2026-10-16","2026-10-16T00:00:00Z","论文",10,false,79,{"impact":16,"substance":17,"depth":16,"authority":18,"freshness":19,"relevant":20,"comment":21},18,22,13,8,1,"基于5年跨年田间试验揭示农业AI模型时序泛化瓶颈，结论扎实且具部署指导价值，值得入选。",[23],{"name":9,"url":6},[25,26,27,28,29],"智慧农业","农业人工智能","无人机遥感","小麦倒伏","模型泛化",{"code":31,"slug":32,"name":33,"path":34,"parent_code":35},"A2","wheat","小麦","\u002Ftopic\u002Ffield-crops\u002Fwheat","A",[],{"status":38,"locked":13,"version":39,"method":40,"evidence":41},"classified","2026-10-22.2","rules",{"candidates":42,"agricultural_evidence":45},[43,49,54,59],{"code":31,"score":44,"supported":45,"title":46,"abstract":47,"tags":48},11,true,[33],[33],[],{"code":50,"score":20,"supported":13,"title":51,"abstract":52,"tags":53},"I1",[],[],[25],{"code":55,"score":20,"supported":13,"title":56,"abstract":57,"tags":58},"I2",[],[],[26],{"code":60,"score":20,"supported":13,"title":61,"abstract":62,"tags":63},"I3",[],[],[27],[65,66],"无人机 小麦倒伏 监测","深度语义分割 时序泛化","无人机小麦倒伏监测-5362",0,{"doi":8,"openalex_id":8,"authors":70,"venue":8,"cited_by_count":68,"oa_url":8,"card":71,"direction":75,"ingested_from":77},[],{"tldr":72,"method":73,"finding":74,"direction":75,"opportunity":76},"用5年田间试验评估深度语义分割模型在无人机小麦倒伏监测中的跨年泛化能力。","5年无人机小麦倒伏影像，对比注意力模型与全卷积网络的跨年迁移表现。","同年内高精度不能保证跨年可靠，注意力模型比全卷积网络更稳健，但大时序漂移下仍严重退化。","农业遥感与作物表型","可研究时序域自适应与跨年增量学习，建立农业AI模型部署前的时序泛化评测基准。","agent","2026-10-20T02:47:54.289085Z",{"total":80,"page":20,"page_size":80,"items":81},6,[82,153,221,270,315,389],{"id":83,"title":84,"url":85,"summary":86,"summary_zh":8,"content":8,"source_name":87,"source_url":8,"published_at":88,"category":11,"cover_url":8,"hotness":89,"is_selected":13,"score":90,"score_detail":91,"sources":94,"tags":99,"primary_topic":102,"secondary_topics":103,"classification":115,"search_phrases":140,"slug":143,"view_count":144,"doi":8,"paper":145,"created_at":152},2997,"基于无人机多光谱图像和VGG21模型的小麦渍害调控效果识别方法","https:\u002F\u002Fwww.toutiao.com\u002Farticle\u002F7685741381196825088","江苏省农业科学院农业信息研究所梁万杰等联合中国农科院农业环境与可持续发展研究所、湖北粮作所、扬州大学等团队，针对小麦渍害防控提出基于无人机多光谱图像和VGG21模型的快速无损识别方法。在小麦拔节-抽穗和抽穗-灌浆两个阶段开展对照、渍水胁迫、硅肥调控和氨基酸调控4个类别数据集，大疆精灵4多光谱无人机采集小麦冠层多光谱图像，测产评估调控效果。","智慧农业(中英文)2026,8(4):60-69","2026-09-15T12:42:00Z",25,78,{"impact":16,"substance":17,"depth":16,"authority":92,"freshness":80,"relevant":20,"comment":93},14,"多机构协作提出无人机多光谱结合VGG21的小麦渍害无损识别方法，方法新颖、数据扎实，对智慧农业植保监测有参考价值。",[95,96],{"name":87,"url":85},{"name":97,"url":98},"智慧农业(中英文)8(4) 60-69(2026)","https:\u002F\u002Ffinance.sina.com.cn\u002Froll\u002F2026-10-16\u002Fdoc-inivmprm8363597.shtml",[25,26,27,100,101],"小麦渍害","多光谱成像",{"code":31,"slug":32,"name":33,"path":34,"parent_code":35},[104,109],{"code":60,"slug":105,"name":106,"path":107,"parent_code":108},"remote-sensing","遥感与农情监测","\u002Ftopic\u002Fdigital-agriculture\u002Fremote-sensing","I",{"code":110,"slug":111,"name":112,"path":113,"parent_code":114},"H7","climate","气候适应与防灾减灾","\u002Ftopic\u002Fenvironment\u002Fclimate","H",{"status":38,"locked":13,"version":39,"method":40,"evidence":116},{"candidates":117,"agricultural_evidence":45},[118,123,127,132,136],{"code":60,"score":18,"supported":45,"title":119,"abstract":121,"tags":122},[120],"无人机多光谱",[120],[101,27],{"code":31,"score":44,"supported":45,"title":124,"abstract":125,"tags":126},[33],[33],[],{"code":110,"score":44,"supported":45,"title":128,"abstract":130,"tags":131},[129],"渍害",[129],[],{"code":50,"score":20,"supported":13,"title":133,"abstract":134,"tags":135},[],[],[25],{"code":55,"score":20,"supported":13,"title":137,"abstract":138,"tags":139},[],[],[26],[141,142],"江苏省农科院 小麦渍害 无人机多光谱","VGG21 小麦 渍害识别","江苏省农科院小麦渍害无人机多光谱-2997",3,{"doi":8,"openalex_id":8,"authors":146,"venue":8,"cited_by_count":68,"oa_url":8,"card":147,"direction":75,"ingested_from":77},[],{"tldr":148,"method":149,"finding":150,"direction":75,"opportunity":151},"用无人机多光谱图像和VGG21模型识别小麦渍害调控效果。","大疆精灵4多光谱无人机采集冠层图像，构建VGG21分类模型。","该方法可快速无损识别渍害及硅肥、氨基酸调控效果。","可探索多光谱与深度学习结合评估其他逆境调控措施，并迁移至多作物场景。","2026-09-20T00:03:07.858899Z",{"id":154,"title":155,"url":156,"summary":157,"summary_zh":8,"content":8,"source_name":158,"source_url":8,"published_at":159,"category":11,"cover_url":8,"hotness":12,"is_selected":13,"score":160,"score_detail":161,"sources":166,"tags":168,"primary_topic":171,"secondary_topics":172,"classification":178,"search_phrases":208,"slug":211,"view_count":212,"doi":8,"paper":213,"created_at":220},2854,"面向冬小麦水分含量的无人机遥感自动机器学习预测——MDPI Remote Sensing","https:\u002F\u002Fwww.mdpi.com\u002F2072-4292\u002F18\u002F18\u002F3161","本研究探索了无人机遥感快速准确评估冬小麦水分含量的潜力。在开花期和灌浆期使用配备多光谱、RGB和热红外相机的无人机获取高分辨率冠层遥感图像。集成地面真值采样数据与无人机遥感数据,使用自动机器学习(AutoML)框架建立回归模型预测冬小麦水分含量(MC)。结果表明,MC预测在灌浆期表现最佳,TIR传感器精度最高(R²=0.812,MAE=0.0204,RMSE=0.0274)。多传感器融合相比单传感器方法进一步提升预测性能,MC预测的R²达0.876、MAE 0.0191、RMSE 0.0259。来自中国农业科学院农田灌溉研究所。","MDPI Remote Sensing","2026-09-15T00:00:00Z",74,{"impact":162,"substance":163,"depth":164,"authority":18,"freshness":19,"relevant":20,"comment":165},15,21,17,"中国农科院团队用AutoML融合多光谱、RGB与热红外无人机数据预测冬小麦水分含量，多传感器融合R²达0.876，方法新颖、结论可靠，对精准灌溉有实用价值，值得进入每日精选。",[167],{"name":158,"url":156},[25,26,27,169,170],"冬小麦","多传感器融合",{"code":31,"slug":32,"name":33,"path":34,"parent_code":35},[173,174],{"code":60,"slug":105,"name":106,"path":107,"parent_code":108},{"code":55,"slug":175,"name":176,"path":177,"parent_code":108},"agricultural-ai","农业AI与大模型","\u002Ftopic\u002Fdigital-agriculture\u002Fagricultural-ai",{"status":38,"locked":13,"version":39,"method":40,"evidence":179},{"candidates":180,"agricultural_evidence":45},[181,188,193,198,204],{"code":60,"score":182,"supported":45,"title":183,"abstract":186,"tags":187},20,[184,185],"遥感","remote sensing",[184],[27],{"code":31,"score":189,"supported":45,"title":190,"abstract":191,"tags":192},12,[33],[33],[169],{"code":55,"score":189,"supported":45,"title":194,"abstract":196,"tags":197},[195],"机器学习",[195],[26],{"code":199,"score":144,"supported":45,"title":200,"abstract":201,"tags":203},"H3",[],[202],"灌溉",[],{"code":50,"score":20,"supported":13,"title":205,"abstract":206,"tags":207},[],[],[25],[209,210],"农业人工智能 多传感器融合 无人机遥感 智慧农业","农业人工智能 多传感器融合","农业人工智能多传感器融合无人机遥感智慧农业-2854",2,{"doi":8,"openalex_id":8,"authors":214,"venue":8,"cited_by_count":68,"oa_url":8,"card":215,"direction":75,"ingested_from":77},[],{"tldr":216,"method":217,"finding":218,"direction":75,"opportunity":219},"用无人机多传感器遥感结合AutoML预测冬小麦水分含量。","无人机多光谱、RGB、热红外图像+地面真值，AutoML回归建模。","灌浆期热红外精度最高，多传感器融合将R²提升至0.876。","可探索不同生育期与品种的泛化性，及将水分预测接入灌溉决策系统。","2026-09-18T00:03:30.758419Z",{"id":222,"title":223,"url":224,"summary":225,"summary_zh":8,"content":8,"source_name":226,"source_url":8,"published_at":227,"category":11,"cover_url":8,"hotness":12,"is_selected":13,"score":90,"score_detail":228,"sources":230,"tags":232,"primary_topic":235,"secondary_topics":236,"classification":237,"search_phrases":258,"slug":261,"view_count":68,"doi":8,"paper":262,"created_at":269},5365,"基于经济型多光谱成像与知识蒸馏模型压缩的小麦线条花叶病毒早期检测","https:\u002F\u002Fwww.ebiotrade.com\u002Fnewsf\u002F2026-10\u002F20261017000601916.htm","研究人员利用配备六个光谱滤光片的消费级数码单反(DSLR)相机在自然环境条件下获取了一个新的小麦线条花叶病毒(WhSMV)多光谱数据集。评估十二个深度学习模型识别 WhSMV 感染不同阶段的性能,并首次将模型压缩技术应用于多光谱图像用于 WhSMV 感染的早期检测。比较了深度学习算法与压缩模型的性能,揭示压缩模型达到与大型深度学习模型相当的性能。WhSMV 是小麦土传媒介传播病毒,巴西与南非均有报道。","Smart Agricultural Technology 2026-10-17","2026-10-17T00:00:00Z",{"impact":16,"substance":17,"depth":164,"authority":18,"freshness":19,"relevant":20,"comment":229},"该研究构建了新的多光谱数据集，并首次将模型压缩用于小麦病毒早期检测，方法新颖、结论可靠，对智慧农业植保具有参考价值。",[231],{"name":226,"url":224},[25,26,101,233,234],"小麦病害","模型压缩",{"code":31,"slug":32,"name":33,"path":34,"parent_code":35},[],{"status":38,"locked":13,"version":39,"method":40,"evidence":238},{"candidates":239,"agricultural_evidence":45},[240,244,250,254],{"code":31,"score":44,"supported":45,"title":241,"abstract":242,"tags":243},[33],[33],[],{"code":55,"score":245,"supported":45,"title":246,"abstract":247,"tags":249},4,[],[248],"深度学习",[26],{"code":50,"score":20,"supported":13,"title":251,"abstract":252,"tags":253},[],[],[25],{"code":60,"score":20,"supported":13,"title":255,"abstract":256,"tags":257},[],[],[101],[259,260],"小麦线条花叶病毒 多光谱 检测","知识蒸馏 模型压缩 小麦","小麦线条花叶病毒多光谱检测-5365",{"doi":8,"openalex_id":8,"authors":263,"venue":8,"cited_by_count":68,"oa_url":8,"card":264,"direction":75,"ingested_from":77},[],{"tldr":265,"method":266,"finding":267,"direction":75,"opportunity":268},"用经济型多光谱成像结合知识蒸馏压缩模型，实现小麦线条花叶病毒早期检测。","消费级单反加六波段滤光片采集多光谱数据，评估12个深度学习模型并做模型压缩。","压缩模型性能与大型深度学习模型相当，可用于WhSMV感染早期识别。","可探索低成本多光谱设备与轻量模型在田间病害早期预警中的部署与跨区域泛化。","2026-10-20T02:47:54.487590Z",{"id":271,"title":272,"url":273,"summary":274,"summary_zh":8,"content":8,"source_name":275,"source_url":8,"published_at":276,"category":277,"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":310,"slug":313,"view_count":68,"doi":8,"paper":8,"created_at":314},4859,"机器人变身小麦育种专家——山东农大小麦育种具身机器人原型机公开亮相","https:\u002F\u002Fnews.sciencenet.cn\u002Fsbhtmlnews\u002F2026\u002F10\u002F391582.shtm","近日在山东泰安举行的小麦智能育种研讨会上,小麦育种具身机器人原型机首次公开亮相。该机器人由山东农业大学教授王宏伟、张亮团队联合研发,采用实时动态差分定位与视觉融合导航技术,可在试验田小区间自主移动作业;搭载滑台高清相机,可对田间小麦进行连续影像采集,并利用视觉算法自动量化株高、穗部性状等表型参数。系统结合智能决策模型对田间单株进行高通量初筛,自主识别具有进一步考察价值的候选植株。研究团队与山东极智生物科技有限公司达成合作意向。","中国科学报 科学网 2026 年 10 月 13 日 第 1 版","2026-10-13T00:00:00Z","报道",81,{"impact":17,"substance":182,"depth":280,"authority":18,"freshness":12,"relevant":20,"comment":281},16,"国内小麦育种具身机器人原型机首次公开亮相，技术路线与产学研合作信息具体，具备行业示范价值，值得进入每日精选。",[283],{"name":275,"url":273},[25,26,285,286,287],"种业振兴","小麦育种","具身机器人",{"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},[293,297,302,306],{"code":31,"score":189,"supported":45,"title":294,"abstract":295,"tags":296},[33],[33],[286],{"code":298,"score":20,"supported":13,"title":299,"abstract":300,"tags":301},"F1",[],[],[285],{"code":50,"score":20,"supported":13,"title":303,"abstract":304,"tags":305},[],[],[25],{"code":55,"score":20,"supported":13,"title":307,"abstract":308,"tags":309},[],[],[26],[311,312],"山东农业大学 小麦育种 具身机器人","王宏伟 小麦 表型采集","山东农业大学小麦育种具身机器人-4859","2026-10-14T00:04:08.166418Z",{"id":316,"title":317,"url":318,"summary":319,"summary_zh":320,"content":8,"source_name":321,"source_url":318,"published_at":322,"category":11,"cover_url":8,"hotness":12,"is_selected":13,"score":278,"score_detail":323,"sources":326,"tags":328,"primary_topic":332,"secondary_topics":333,"classification":335,"search_phrases":353,"slug":356,"view_count":20,"doi":357,"paper":358,"created_at":388},4806,"A study on the identification of wheat powdery mildew using chlorophyll fluorescence imaging","https:\u002F\u002Fdoi.org\u002F10.3389\u002Ffpls.2026.1837843","Introduction Wheat powdery mildew poses a serious threat to global wheat production, and its accurate identification is key to smart agriculture. Methods This study combines chlorophyll fluorescence imaging with deep learning algorithms to investigate the temporal dynamics of key fluorescence parameters (such as maximum photochemical efficiency, QYmax) during the progression of powderymildew. A QYmax-based chlorophyll fluorescence image dataset for wheat powdery mildew was constructed. Building on this, the LCRNet model was proposed, which effectively processes fluorescence image data through a collaborative mechanism integrating the Large Selective Adaptive Fusion (LSAF) and Coordinated Attention (CA) modules. Results On an independent test set, the model achieved an accuracy of 94% and an F1 score of 94.1%, outperforming comparison models. Discussion This study proposes a method for classifying the severity of powdery mildew in wheat that reduces the computational cost of fluorescence data processing, providing important methodological support and technical insights for advancing the intelligent, precise, and green transformation of crop disease prevention and control systems.","引言 小麦白粉病对全球小麦生产构成严重威胁，其准确识别是实现智慧农业的关键。方法 本研究将叶绿素荧光成像与深度学习算法相结合，探究白粉病发展过程中关键荧光参数（如最大光化学效率QYmax）的时间动态变化，构建了基于QYmax的小麦白粉病叶绿素荧光图像数据集。在此基础上，提出了LCRNet模型，该模型通过融合大选择性自适应融合（LSAF）模块与协调注意力（CA）模块的协同机制，有效处理荧光图像数据。结果 在独立测试集上，该模型达到了94%的准确率和94.1%的F1分数，优于对比模型。讨论 本研究提出了一种小麦白粉病严重程度分类方法，降低了荧光数据处理的计算成本，为推动作物病害防控体系向智能化、精准化和绿色化转型提供了重要的方法学支撑和技术启示。","Frontiers in Plant Science","2026-10-12T00:00:00Z",{"impact":16,"substance":17,"depth":16,"authority":92,"freshness":324,"relevant":20,"comment":325},9,"将叶绿素荧光成像与深度学习结合用于小麦白粉病严重度识别，方法新颖、数据可靠，对智慧植保有实质参考价值。",[327],{"name":321,"url":318},[25,26,329,330,331],"病害识别","叶绿素荧光","小麦白粉病",{"code":31,"slug":32,"name":33,"path":34,"parent_code":35},[334],{"code":55,"slug":175,"name":176,"path":177,"parent_code":108},{"status":38,"locked":13,"version":39,"method":40,"evidence":336},{"candidates":337,"agricultural_evidence":45},[338,342,347],{"code":31,"score":92,"supported":45,"title":339,"abstract":340,"tags":341},[32],[33,32],[],{"code":55,"score":19,"supported":45,"title":343,"abstract":344,"tags":346},[],[248,345],"deep learning",[26,329],{"code":50,"score":348,"supported":45,"title":349,"abstract":350,"tags":352},7,[],[25,351],"smart agriculture",[25],[354,355],"小麦白粉病 叶绿素荧光 深度学习","LCRNet 小麦病害识别","小麦白粉病叶绿素荧光深度学习-4806","10.3389\u002Ffpls.2026.1837843",{"doi":357,"openalex_id":359,"authors":360,"venue":321,"cited_by_count":68,"oa_url":379,"card":380,"direction":386,"ingested_from":387},"W7221549180",[361,363,366,369,371,373,375,377],{"name":362,"orcid":8},"Yijia Chen",{"name":364,"orcid":365},"Chengyuan Hao","https:\u002F\u002Forcid.org\u002F0000-0002-6002-3963",{"name":367,"orcid":368},"Zhiyu Li","https:\u002F\u002Forcid.org\u002F0000-0003-3771-0199",{"name":370,"orcid":8},"Xupeng An",{"name":372,"orcid":8},"Jingxiang Huang",{"name":374,"orcid":8},"Xinchang Chen",{"name":376,"orcid":8},"Shidong Wang",{"name":378,"orcid":8},"ZhiChao Chen","https:\u002F\u002Fwww.frontiersin.org\u002Fjournals\u002Fplant-science\u002Farticles\u002F10.3389\u002Ffpls.2026.1837843\u002Fpdf",{"tldr":381,"method":382,"finding":383,"direction":384,"opportunity":385},"该研究利用叶绿素荧光成像与深度学习识别小麦白粉病并分级。","构建QYmax荧光图像数据集，提出融合LSAF与CA模块的LCRNet模型。","模型在独立测试集上准确率达94%、F1为94.1%，优于对比模型。","农业人工智能与决策模型","可探索多病害、多作物荧光图像通用模型，并融合田间时序数据实现早期预警。","智慧农业 \u002F 农业物联网","openalex","2026-10-13T23:30:09.060156Z",{"id":390,"title":391,"url":392,"summary":393,"summary_zh":8,"content":394,"source_name":395,"source_url":8,"published_at":396,"category":277,"cover_url":8,"hotness":12,"is_selected":13,"score":397,"score_detail":398,"sources":400,"tags":402,"primary_topic":403,"secondary_topics":404,"classification":405,"search_phrases":425,"slug":428,"view_count":80,"doi":8,"paper":8,"created_at":429},4382,"山东农业大学研发小麦育种具身机器人 WheatX 首次亮相","http:\u002F\u002Fwww.shandong.gov.cn\u002Fart\u002F2026\u002F10\u002F6\u002Fart_97564_765520.html","山东农业大学张亮团队联合山东省农业科学院、中国农业大学等单位研发小麦育种具身机器人 WheatX,集感知、决策与选育功能于一体,可完成大田麦穗快速筛选、对量化达标麦穗喷涂颜料原位标记,实现「感知—决策—执行」全流程闭环;团队已与极智生物达成合作,将首款小麦育种大模型 iwheat 嵌入机器人,推动从原型机向产品级升级。","小麦试验田里，一台长宽高均超过两米的“大家伙”在麦株前稳稳停住。机械臂移动滑台探出，前端圆环装置精准套住一株麦穗缓缓上滑，环内八枚微型摄像头从多个角度对其精细成像。随后，穗粒数、穗长等性状的量化数据全部生成。\n\n这是由山东农业大学、小麦育种全国重点实验室王宏伟、张亮教授团队联合研发的小麦育种具身机器人原型机，近日在该校举办的小麦智能育种研讨会上首次公开亮相。\n\n王宏伟介绍，传统小麦育种中，育种家经年积累了大量宝贵育种经验，能从成千上万的杂交后代里选育出性状优良的品系。但在大田环境下，人工筛选不仅劳动强度大，筛选结果还容易受主观经验差异影响。同时，现有田间装备多以被动扫描和影像采集为主，能够完成“看”和“测”，但在自主判断、主动干预和选育执行等方面仍存在不足。该研究主要聚焦传统小麦育种主要依赖人工、选育效率低等行业共性问题，将为大田育种智能化提供新的装备与技术路径。\n\n“在模型训练时加入大量育种家先验经验是该款机器人特色之一。”王宏伟说，团队的目标，是让机器人不仅能够“看得见”，还能够“作判断、去操作”，在田间完成类似育种家的选育动作。\n\n为让机器人“学会”育种，团队种植了2000份不同的小麦种质资源，逐一拍摄影像后，请育种家按标准打分标注，再将标注结果输入系统，育种家凭多年经验作出的判断被转化为机器可学习的数据。机器人结合育种家经验构建智能决策模型，具备了初步的自主选育判断能力。\n\n同时，该机器人采用RTK与视觉融合导航技术，可在试验田小区间自主移动作业；搭载滑台高清相机，对田间小麦进行连续影像采集，并利用视觉算法自动量化株高、穗部性状等表型参数。在此基础上，系统结合智能决策模型，可对田间单株进行高通量初筛，自主识别具有进一步考察价值的候选植株。\n\n“对于初筛获得的目标植株，机器人还可通过自主研发的多功能机械臂开展进一步精细作业，包括单株抓取、环形多角度精细成像等操作。”张亮解释，对于量化后确认符合标准的麦穗，机器人会像育种家在田间“绑绳子做标记”一样，喷涂颜料原位标记，完成从大范围快速筛选到目标植株精细评价和标记的连续作业，实现“感知—决策—执行”的全流程闭环。\n\n该研究获山东省重点研发计划项目支持，并联合山东省农业科学院、中国农业大学等单位协同攻关。国家小麦产业技术体系副首席曹新有表示，该机器人集感知、决策与选育功能于一体，为智能育种由算法模型走向田间自主作业提供了关键技术支撑。\n\n“用机器人辅助人工选择，这是很难但很有意义的方向。”张亮坦言，大田场景麦穗密度高、类型多，场景极其复杂，加之天气影响，对机器人精准作业都是极大考验，系统在抓取精准度、学习经验深度等指标上还需进一步升级改良。目前，团队已与极智生物科技有限公司达成合作意向，将我国首款小麦育种大模型iwheat嵌入WheatX育种机器人，联合研发小麦智能育种系统，推动原型机迭代升级向产品级迈进。（记者 刘姝彤 通讯员 屈越）","山东省人民政府网 2026-10-06","2026-10-06T00:00:00Z",82,{"impact":17,"substance":17,"depth":164,"authority":18,"freshness":19,"relevant":20,"comment":399},"国内首款小麦育种具身机器人原型机亮相，实现感知—决策—执行闭环，属智慧育种领域实质性技术突破，值得进入每日精选。",[401],{"name":395,"url":392},[25,26,285,286,287],{"code":31,"slug":32,"name":33,"path":34,"parent_code":35},[],{"status":38,"locked":13,"version":39,"method":40,"evidence":406},{"candidates":407,"agricultural_evidence":45},[408,412,417,421],{"code":31,"score":189,"supported":45,"title":409,"abstract":410,"tags":411},[33],[33],[286],{"code":55,"score":245,"supported":45,"title":413,"abstract":414,"tags":416},[],[415],"大模型",[26],{"code":298,"score":20,"supported":13,"title":418,"abstract":419,"tags":420},[],[],[285],{"code":50,"score":20,"supported":13,"title":422,"abstract":423,"tags":424},[],[],[25],[426,427],"山东农业大学 小麦育种机器人 WheatX","小麦智能育种 iwheat 大模型","山东农业大学小麦育种机器人WheatX-4382","2026-10-08T00:06:07.051280Z"]