基于全景视觉的网球服务机器人的设计与研究

发布时间：2018-04-26 02:11

本文选题：网球服务机器人 + 反射式全景视觉系统　；参考：《北方工业大学》2017年硕士论文

【摘要】：本论文提出了一个基于全景视觉的网球服务机器人的设计方案,并将该方案予以实现。该网球服务机器人以中型组足球机器人的机械为原型在原有的基础上进行机械上的改造和设计,并以STM32F407微控制器为核心设计了机器人下位机核心驱动板,完成了机器人硬件框架。机器人采用了笔记本电脑作为其上位机的核心控制单元,并在其上位机上研究和设计了上位机软件,该软件通过USB连接反射式全景视觉系统采集机器人四周的图像。同时机器人可以通过手机远程控制启动和关闭。课题设计和研究主要包含三方面:机械、硬件和软件。机械方面需要对原有的机械结构重新设计,设计机器人的吸球装置和网球的储存装置。硬件方面需要设计机器人的整体硬件架构,包括机器人的行走和吸球驱动器、核心板的设计和上位机与下位机的连接以及通信。软件方面是课题的一个难点,首先是图像的采集,单纯的依靠视觉作为机器人的感官原件,对光照的要求叫严格,在很多场景下需要改变摄像头参数才能更好的识别网球,所以课题的首先需要克服光照环境对识别效率的影响。课题的软件设计是先通过对采集图像的预处理工作,自动调节摄像头参数,使直接通过摄像头获得的图像适应课题后续的图像处理工作。然后对图像进行增强和颜色空间的转换,通过课题研究的网球识别算法进行网球的识别及其位置信息的确定,再由行为策略模块对网球位置信息统计并生成行为指令,控制机器人进行网球的拾取工作。首先,第一章介绍了网球服务机器人的研究现状及背景,叙述了可以研究内容的意义以及内容。第二章介绍机器人整体的设计方案。同时介绍了课题中所用到的各项关键技术。第三章展示了机器人的机械设计方案。第四章介绍了机器人的硬件框架的设计,包括行走轮机构、吸球器机构、机载电源、下位机控制单元和上位机控制单元的各个硬件设计。第五章介绍了机器人上位机的软件的设计。详细介绍了软件的框架和层次结构,并针对框架中设计的功能模块进行了详细设计介绍。最后是调试的结果分析以及本课题的结论与展望。
[Abstract]:In this paper, a design scheme of tennis service robot based on panoramic vision is proposed and implemented. The tennis service robot is designed on the basis of the mechanical prototype of the medium-sized soccer robot, and the core driving board of the robot is designed with the STM32F407 microcontroller as the core. The hardware frame of the robot is completed. The robot adopts the notebook computer as the core control unit of its upper computer, and studies and designs the upper computer software on its upper computer. The software collects the images around the robot through USB connected with the reflective panoramic vision system. At the same time, the robot can be remotely controlled by the mobile phone to start and close. The project design and research mainly include three aspects: mechanical, hardware and software. In the mechanical aspect, the original mechanical structure should be redesigned to design the ball suction device of the robot and the storage device of the tennis ball. In the aspect of hardware, we need to design the whole hardware architecture of the robot, including the walking and sucking drive of the robot, the design of the core board, the connection and communication between the upper computer and the lower computer. The software aspect is a difficult point of the subject, first is the image collection, relying solely on the vision as the robot's sensory original, the requirement of illumination is strict, in many scenes, we need to change the camera parameters to recognize tennis ball better. Therefore, the first need to overcome the impact of lighting environment on the efficiency of recognition. The software design of the subject is to adjust the camera parameters automatically through the preprocessing of the collected images, so that the images obtained directly through the camera can adapt to the subsequent image processing work of the subject. Then the image is enhanced and the color space is transformed. The tennis ball recognition and location information are identified by the tennis recognition algorithm studied in this paper. Then the behavior strategy module statistics the tennis position information and generates the behavior instructions. Control the robot to pick up tennis balls. The first chapter introduces the research status and background of tennis service robot, and describes the significance and content of the research content. The second chapter introduces the overall design of the robot. At the same time, the key technologies used in the subject are introduced. The third chapter shows the mechanical design of the robot. Chapter 4 introduces the design of the hardware frame of the robot, including the hardware design of the walking wheel mechanism, the ball absorber mechanism, the airborne power supply, the lower computer control unit and the upper computer control unit. The fifth chapter introduces the software design of robot upper computer. The framework and hierarchy of the software are introduced in detail, and the functional modules designed in the framework are introduced in detail. Finally, the result analysis of debugging and the conclusion and prospect of this subject are given.
【学位授予单位】：北方工业大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP242

【参考文献】