番茄采摘机器人的SLAM运动控制系统设计  

SLAM Motion Control System Design for Tomato Picking Robot

作　　者：戴月[1] 陈波[2] DAI Yue;CHEN Bo(Experimental Training Center,Chuzhou Vocational and Technical College,Chuzhou 239000,China;School of Computer Science and Technology,Southwest University of Science and Technology,Mianyang 621010,China)

机构地区：[1]滁州职业技术学院实验实训中心,安徽滁州239000 [2]西南科技大学计算机科学与技术学院,四川绵阳621010

出　　处：《太原学院学报（自然科学版）》2022年第1期37-43,共7页Journal of TaiYuan University:Natural Science Edition

基　　金：2020年度安徽高校自然科学研究项目(KJ2020A0995);安徽省重大线上教学改革研究项目(2020zdxsjg252)。

摘　　要：大规模季节性水果依赖人工采摘,需要花费大量的时间,效率比较低,番茄采摘的自动化对提升农业科技含量和水平发挥着至关重要的作用。基于同步定位与建图(simultaneous localization and mapping,SLAM)技术对番茄机器人进行定位导航,实现了番茄采摘机器人的运动控制。采用BP神经网络对番茄机器人末端执行器在运动学方程中的位置进行求解,开发采摘机器人运动控制系统,提出了基于SLAM的番茄采摘机器人的导航方法,阐述了激光雷达、自主建图、定位和地图匹配。对机器人智能采摘进行试验,结果表明,算法的最短用时为6.6 s,最长用时为7.0 s,平均用时为6.7 s,能够达到预期的效果,这对实现番茄的机器人采摘具有一定的参考价值。Large-scale seasonal fruits rely on manual picking,which takes a lot of time and is relatively inefficient.The automation of tomato picking plays a crucial role in improving the content and level of agricultural science and technology.Based on simultaneous localization and mapping(SLAM)technology,the motion control of tomato picking robot was realized.BP neural network was used to solve the kinematics equation to get the position of tomato robot end-effector and the motion control system of tomato picking robot was developed.The navigation method of tomato picking robot based on SLAM was proposed,and laser radar,autonomous mapping,localization and map matching were described.The results based on the experoment of the robort showed that the shortest time the algorithm used was 6.6 s,the longest was 7.0 s,and the average time was 6.7 s,achieving the expected effect,with certain reference value for the realization of robot tomato picking.

关 键 词：采摘机器人 番茄采摘 运动控制 SLAM 

分 类 号：TP242.6[自动化与计算机技术—检测技术与自动化装置]