单 位：
中国农业大学现代精细农业系统集成研究教育部重点实验室农业部农业信息获取技术重点实验室;中国农业大学现代精细农业系统集成研究教育部重点实验室
关键词：
玉米;中耕除草;复合导航;GNSS;机器视觉
摘 要：
为了实现玉米中耕除草过程中的自动导航作业,提高自动除草的效率和准确性,设计了一种玉米中耕除草复合导航系统。系统由基于GNSS的农机自动导航部分和基于机器视觉的农具自动导航部分组成,可通过GNSS位置信息进行农机自动导航,同时根据摄像头获取的玉米作物行信息控制农具铲刀进行行间除草。对农机的转向控制部分和前轮转角检测部分进行了机械改装,以PLC和步进电机驱动器为基础设计了农机转向控制电路和农具液压控制电路;以横向偏差和横向偏差变化率作为模糊控制的输入变量设计了自适应模糊控制方法;采用摄像头获取玉米作物行,通过扫描滤波方法进行作物行检测。农机独立导航除草试验和农机具复合导航试验结果表明:在车速为0.6 m/s时,农机自动导航最大横向偏差为10.04 cm,平均偏差为4.62 cm;农机具复合导航时的最大偏差为6.35 cm,平均偏差为2.73 cm;农机具复合导航系统能较好地满足玉米中耕除草的要求。
译 名：
Design and Implementation of a Corn Weeding-cultivating Integrated Navigation System Based on GNSS and MV
作 者：
Zhang Man;Xiang Ming;Wei Shuang;Ji Yuhan;Qiu Ruicheng;Meng Qingkuan;Key Laboratory for Modern Precision Agriculture System Integration Research,Ministry of Education,China Agricultural University;Key Laboratory of Agricultural Information Acquisition Technology,Ministry of Agriculture;
关键词：
Corn;;Weeding-cultivating;;Integrated navigation;;GNSS;;Machine vision
摘 要：
In order to achieve the automatic navigation of corn weeding-cultivating operations and improve the efficiency and accuracy of automatic weeding,a corn weeding-cultivating integrated navigation system based on GNSS and MV was designed and developed. The system,which consisted of two parts,the agricultural vehicle automatic navigation based on GNSS and the weeding implement navigation based on machine vision,could achieve automatic navigation in the process of corn weeding-cultivating. In order to avoid the damage corn,the proper safety distance between implement moldboard and crop rows is necessary when the agricultural machinery is working, which is controlled by tracing the GNSS information of corn planting navigation and the image information of crop rows acquired by industrial camera. During the hardware part,the steering wheel controlling and the front wheel angle testing parts were designed and refitted. The steering control circuit and the implement moldboard hydraulic control circuit were designed based on PLC and stepper motor drive. To the vehicle navigation,the steering wheel controlling and the front wheel angle testing institutions were designed. The vehicle operated through tracing the corn row by the GNSS information of corn planting navigation. In the integrated navigation intelligent decision module,the fuzzy control system was taken as the main control algorithmfor navigation control decisions,and the lateral deviation of agricultural machinery and lateral deviation error rate were used as input variables of fuzzy control. In the implement visual navigation module,the crop row detection algorithm based on scanning filter was adopted,which could improve the precision of navigation line detection and processing efficiency. Corresponding experiments were designed to test the feasibility of the system,the accuracy of automatic navigation were calculated respectively. The results show that,in the speed of 0. 6 m / s,the maximum lateral deviation of GNSS navigation is 10. 04 cm,the average deviation is 4. 62 cm. The maximum lateral deviation of integrated navigation is 6. 35 cm,the average deviation is 2. 73 cm. The corn weeding-cultivating integrated navigation system can effectively satisfy the requirement of the corn weeding.
