| 基于传感器融合阵列的果树冠层信息采集方法 |
| |
| 引用本文： | 刘理民,王金宇,毛文华,石光智,张晓辉,姜红花.基于传感器融合阵列的果树冠层信息采集方法[J].农业机械学报,2018,49(S1):347-353，359. |
| |
| 作者姓名： | 刘理民 王金宇 毛文华 石光智 张晓辉 姜红花 |
| |
| 作者单位： | 山东农业大学,山东农业大学,中国农业机械化科学研究院,山东农业大学,山东农业大学,山东农业大学 |
| |
| 基金项目： | 山东省重点研发计划项目(2015GNC112004)、山东省农业重大应用技术提升项目、山东省自然科学基金项目（ZR2018MC017）和国家重点研发计划项目（2017YFD0700500） |
| |
| 摘 要： | 针对果园上、中、下冠层不同稀疏度，提出一种多传感器阵列的果树冠层信息融合方法（简称传感器融合阵列），并进行了相关试验及验证。首先设计了果园冠层宽度信息的无线采集系统，并对比分析了6种非接触式测距传感器的动态识别能力；其次采用筛选出的激光传感器及超声波传感器阵列，收集3种果园上、中、下果树冠层信息；最后选出适合3种果园的传感器融合阵列，依据Box-Benhnken 中心组合试验法设计试验，对采用同种传感器阵列与传感器融合阵列测距方案进行响应面试验，并对得出的试验结果进行统计分析。试验结果表明：影响果树整体测量精度显著性水平从大到小依次为测距方案、车体速度、果园类型。车速为0.3~0.5m/s时，与人工测量相比，采用超声波传感器阵列收集果园冠层信息，相对误差为14.70%~20.04%；采用激光传感器阵列时，相对误差为9.13%~16.02%，采用传感器融合阵列时，相对误差为4.2%~10.24%。采用传感器融合阵列比单种传感器阵列精度高，更适合果园变量喷雾作业。
|
| 关 键 词： | 果树冠层 信息融合 传感器融合阵列 响应面试验 |
| 收稿时间： | 2018/7/15 0:00:00 |
Canopy Information Acquisition Method of Fruit Trees Based on Fused Sensor Array |
| |
| LIU Limin,WANG Jinyu,MAO Wenhu,SHI Guangzhi,ZHANG Xiaohui and JIANG Honghua.Canopy Information Acquisition Method of Fruit Trees Based on Fused Sensor Array[J].Transactions of the Chinese Society of Agricultural Machinery,2018,49(S1):347-353，359. |
| |
| Authors: | LIU Limin WANG Jinyu MAO Wenhu SHI Guangzhi ZHANG Xiaohui and JIANG Honghua |
| |
| Institution: | Shandong Agricultural University,Shandong Agricultural University,Chinese Academy of Agricultural Mechanization Sciences,Shandong Agricultural University,Shandong Agricultural University and Shandong Agricultural University |
| |
| Abstract: | In order to distinguish different sparse degrees of top, middle and bottom canopy in orchard, canopy information acquisition method of fruit trees was proposed based on fused sensor array. First of all, the wireless acquired system was designed, which was used to collect canopy width information in orchard. And in order to verify sensor ability to recognize objects in motion, the dynamic recognition capability of six non-contact distance sensors was compared and analyzed. Secondly, laser sensors and ultrasonic sensors array which were filtrated to collect canopy information was used, which incorporated sparse degree and width of top, middle and bottom canopy in three kinds of orchard. Finally, fused sensor array that was suitable for three kinds of orchard was selected. According to Box-Benhnken center combination test method, the orthogonal experiment was designed, including three factors and three levels. And appropriate ranging scheme was selected by response surface experiments and the results of testing were statistically analyzed. The significance level of the overall measurement accuracy of fruit trees from large to small was ranging scheme,car-body speed and orchard type in turn.When vehicle speed was 0.3~0.5m/s, using ultrasonic sensor array to collect canopy information of orchard, the relative error was 14.70%~20.04%. When laser sensor array was used to collect canopy information of orchard, its relative error was 9.13%~16.02%. And when fused sensor array was used for three orchards, its relative error was 4.2%~10.24%. Therefore, the fused sensor array scheme was more accurate than the single sensor array, and it was more suitable for the orchard variable spray operation. |
| |
| Keywords: | fruit tree canopy information fusion fused sensor array response surface experiment |
|
| 点击此处可从《农业机械学报》浏览原始摘要信息 |
| 点击此处可从《农业机械学报》下载免费的PDF全文 |
|
