首页 | 本学科首页   官方微博 | 高级检索  
     检索      

株间除草装置横向偏移量识别与作物行跟踪控制
引用本文:胡 炼,罗锡文,张智刚,陈雄飞,林潮兴.株间除草装置横向偏移量识别与作物行跟踪控制[J].农业工程学报,2013,29(14):8-14.
作者姓名:胡 炼  罗锡文  张智刚  陈雄飞  林潮兴
作者单位:1. 华南农业大学南方农业机械与装备关键技术教育部重点实验室,广州 5106422. 华南农业大学工程学院,广州 510642;1. 华南农业大学南方农业机械与装备关键技术教育部重点实验室,广州 5106422. 华南农业大学工程学院,广州 510642;1. 华南农业大学南方农业机械与装备关键技术教育部重点实验室,广州 5106422. 华南农业大学工程学院,广州 510642;1. 华南农业大学南方农业机械与装备关键技术教育部重点实验室,广州 5106422. 华南农业大学工程学院,广州 510642;1. 华南农业大学南方农业机械与装备关键技术教育部重点实验室,广州 5106422. 华南农业大学工程学院,广州 510642
基金项目:国家科技支撑项目(2011BAD20B06);国家自然科学基金项目(31171864;61175081);948项目"精准农业智能关键技术引进与创新子课题"(2011-G32)
摘    要:株间机械除草技术与装置能有效摆脱田间除草的繁重体力劳动并消除化学除草方法所带来的危害,株间机械除草装置的牵引拖拉机在跟踪作物行时总会产生航向偏差,导致除草装置出现横向偏移,甚至无法进入除草的株间区域,同时还会增加伤苗率。为增大株间机械除草的作用区域和降低伤苗率,该文提出了通过作物行信息识别出株间机械除草装置与作物行横向偏移量的方法,并设计了株间机械除草作物行跟踪机构和控制器,实现了株间机械除草跟随作物行。采用正弦波和三角波2种标准信号作为横向偏移补偿量信号,对作物行跟踪控制器的性能进行了测试,试验结果表明:作物行跟踪控制器能较好地控制除草装置跟随横向偏移补偿信号,前进速度为0.5 m/s时正弦波信号跟踪最大误差10 mm,平均误差0.8 mm,三角波信号跟踪最大误差11 mm,平均误差1.2 mm。除草试验表明,作物行跟踪控制系统能较好地控制株间除草装置跟踪作物行,在0.5 m/s前进速度下跟踪最大误差为20.8 mm,平均误差2.5 mm;作物行跟踪控制明显减少了除草爪齿未进入株间区域的比例,在300 mm株距下,可保证93.3%的株间区域有除草爪齿进行除草作业,在200 mm株距下为85.9%;作物行跟踪控制降低了除草爪齿对作物的损伤,伤苗率从20%以上降到了12%以内,提高了株间机械除草的作业效果。

关 键 词:农业机械,除草,识别,跟踪,横向偏移,作物行
收稿时间:2013/4/29 0:00:00
修稿时间:6/4/2013 12:00:00 AM

Side-shift offset identification and control of crop row tracking for intra-row mechanical weeding
Hu Lian,Luo Xiwen,Zhang Zhigang,Chen Xiongfei and Lin Chaoxing.Side-shift offset identification and control of crop row tracking for intra-row mechanical weeding[J].Transactions of the Chinese Society of Agricultural Engineering,2013,29(14):8-14.
Authors:Hu Lian  Luo Xiwen  Zhang Zhigang  Chen Xiongfei and Lin Chaoxing
Abstract:Abstract: Intra-row mechanical weeding technique and device can free strenuous labor and eliminate the detriment of using chemical weed. The yaw error is barely avoidable when running a tracked crop-row tractor. It can result in an inconstant side-shift between the intra-row mechanical weed device and crop row and the increasing of crop damage and untreated weeds in intra-row area. The objective of this research is to develop a crop row tracking control algorithm for the optimized measurement of side-shift offset. The crop row line is obtained by using least square fit from 20 crops in several consecutive images, and the side-shift offset is estimated based on weeding device lateral position to keep a constant distance between the weeding device and crop row. Then, the PD control algorithm with bi-threshold dead band for transverse controller is developed to reduce the transverse error when the estimated offset exceeds the dead band. It needs to be ensured that the origin of the weeding device follows the desired route and parallels the crop row, by controlling velocity and direction of DC motor. The test results prove the good performance of standardized signal tracking using sine wave and triangle wave. The maximum and average sine wave tracking error is 10 mm and 0.8 mm, respectively, with a forward velocity of 0.2 m/s. The maximum and average triangle wave tracking error is 11 mm and 1.2 mm, respectively, with a forward velocity of 0.5 m/s. The results from weeding experiment in soil bin indicate that the side-shift enabled control of the transverse position of the weeding device and is able to follow the crop row line with an accuracy of ±13.4 mm at 0.2 m/s and ±20.8 mm at 0.5 m/s forward velocity. The untreated weeds in the intra-row area decreased significantly. The treated intra-row area achieve up to 93.3 % and 85.9 % of field surface for a mean plant spacing of 300 mm and 200 mm, respectively. The danger of crop damage is significantly reduced by using side-shift control. Compared to the rate of crop damage up to 20% without using the transverse controller, the rate of crop damage is down to less than 12% by using the side-shift control. The accuracy of intra-row weeding device tracking is high and acceptable.
Keywords:agricultural machinery  weed control  identification  tracking  side-shift  crop row
点击此处可从《农业工程学报》浏览原始摘要信息
点击此处可从《农业工程学报》下载免费的PDF全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号