CASIA OpenIR  > 毕业生  > 硕士学位论文
Thesis Advisor赵晓光
Degree Grantor中国科学院研究生院
Place of Conferral北京
Keyword采摘机器人 目标检测 目标跟踪 Kcf 三维测量 结构光 机械臂
(1) 介绍了水果采摘机器人的研究背景以及意义,阐述了目前国内外水果采摘机器人的研究现状,对目标检测和目标跟踪算法进行了一个综述,并对文章的内容和结构进行了介绍。
(2) 搭建了水果采摘机器人的视觉感知系统,开发了水果的识别和跟踪的图像处理软件。机器人能够实现水果的实时性目标检测与KCF目标跟踪,能够通过颜色分割以及形态学处理生成高质量的推荐域,文中设计了一种轻量级的‘半’全连接神经网络作为分类器,能够实现快速的分类。
(3) 搭建了水果采摘机器人的水果三维测量系统,机器人能够利用‘一’字型结构光以及单目摄像机完成水果的三维空间测量,此方法不需要对相机的内参数和畸变参数进行标定,直接利用Ridge回归从图像中恢复像素的深度信息,测量的速度与精度均达到了项目的需求。
(4) 研究了采摘机械臂的正向运动学分析与反向运动学分析,在Matlab下进行了仿真实验,验证了机械臂正向与反向运动学分析的正确性。
(5) 在对鲜桃种植园进行实地考察调研的基础上,确定了鲜桃采摘机器人的基本设计构架并搭建了相应的硬件平台,为机器人设计了轻便的履带式载体,并在载体上集成了采摘机械臂及其控制板、末端执行器、升降台、电源模块、视觉传感器模块等设备,并以一台工控机为核心控制器搭建了整个采摘机器人运动控制系统。最终独立完成整个采摘机器人的研制。
Other AbstractWith the rapid development of China's agricultural technology and people's consumption of fruits,The yield and planting area of peach, apple, pear and citrus has reached the top in the world.Fruit harvesting robot has very important significance to reduce labour intensity, to ensure harvesting in time and in quantities, and also help China complete agricultural modernization.This article focuses on the key technologies of vision-based mobile fruit harvesting robots. The main research content is as follows:
(1) The research background and significance of the fruit picking robot are introduced. The research status of robots at home and abroad is introduced. A detailed overview of target detection and target tracking algorithm is given. Finally, the content and structure of the article are introduced.
(2) Built a visual perception system for robots and developed image processing software for target recognition and target tracking. The robot can achieve rapid fruit target detection and KCF target tracking, and can generate high-quality proposed region through color segmentation and morphological processing. In this paper, a “semi” fully connected neural network is designed as a classifier, which can achieve rapid classification.
(3) A three-dimensional fruit measuring system for the robot is built. The robot can use the structured light and the monocular camera to complete the three-dimensional measurement of the fruit. This method does not need to calibrate the camera's internal parameters and distortion parameters, but uses the Ridge regression method. The depth information of the target pixel is directly calculated, and the speed and accuracy of the measurement meet the requirements of the project.
(4) The forward kinematics analysis and inverse kinematics analysis of the picking manipulator are studied. The simulation experiment is carried out under Matlab, and the correctness of the forward and reverse kinematic analysis of the manipulator arm is proved.
(5) After investigating the peach plantation, the basic design framework of the peach picking robot was determined and the corresponding hardware platform was set up. We designed a lightweight crawler carrier for the robot, and integrated the picking robot arm and its control board, end effector, lifting platform, power supply module, vision sensor module and other equipment on the carrier, and controlled it with an industrial computer as the core. The robot builds the entire picking robot motion control system. The development of the entire picking robot is ultimately completed independently.
Document Type学位论文
Recommended Citation
GB/T 7714
余永佳. 智慧农业水果自动采摘机器人的研究[D]. 北京. 中国科学院研究生院,2018.
Files in This Item:
File Name/Size DocType Version Access License
余永佳大论文.pdf(4563KB)学位论文 暂不开放CC BY-NC-SAApplication Full Text
Related Services
Recommend this item
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[余永佳]'s Articles
Baidu academic
Similar articles in Baidu academic
[余永佳]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[余永佳]'s Articles
Terms of Use
No data!
Social Bookmark/Share
All comments (0)
No comment.

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.