基于视觉的AGV路径跟踪技术研究

发布时间：2018-04-11 20:52

本文选题：AGV图像处理 + 数字稳像　；参考：《青岛科技大学》2015年硕士论文

【摘要】：自动引导小车(AGV)是一种智能运送设备,是工厂柔性物流体系中的重要组成单元,以其自动化程度高、可靠性高、成本低、所组成运送系统柔性高等诸多优势获得了非常广泛的应用。基于视觉导引的AGV利用摄像机获取路径图像信息,与其它方式相比路径信息丰富。视觉导引方式主要包括两个重要部分,路径信息提取和运动控制,所以本文针对这两个方面进行了分析研究。主要内容有：(1)首先主要针对摄像机获取的AGV路径图像的预处理技术进行了探讨研究,主要包括路径图像增强、路径图像滤波、路径图像分割和路径图像位置校正等。预处理的目的是在尽可能保留路径全局信息的前提下,去除其它信息的干扰,减少后续相关处理的运算量。同时采用数字稳像技术对路径图像进行位置校正,解决由于摄像机随机抖动引起图像中路径位置变化而导致的后续路径跟踪不稳定的问题。(2)利用预处理后的AGV路径图像提取路径信息,首先根据路径标志线的边缘信息获得路径中心点集合,然后采用基于曲率角估计的方法对路径模型进行分类,主要分为直线路径模型、圆弧转弯路径模型和折线转弯路径模型,并根据路径模型特征推导出提取AGV导航所需的位置偏移量和角度偏移量的计算方法。(3)设计基于视觉导引的AGV路径模糊跟踪器,搭建了AGV运动模型,推导出AGV运动控制的状态方程。分析了小车的实际运行情况及控制方法,‘选择了模糊控制方法,并根据AGV的运动模型和控制要求设计出模糊控制器。最后利用Matlab工具进行仿真测试,仿真结果表明,本文所设计的模糊控制器能很好地纠正小车运动过程中产生的偏差,使AGV循迹标志线平稳运行。
[Abstract]:AGV is a kind of intelligent transportation equipment, which is an important component unit in the flexible logistics system of factories, because of its high degree of automation, high reliability and low cost.Many advantages such as high flexibility and high flexibility have been widely used.The vision-guided AGV uses the camera to obtain path image information, which is rich in path information compared with other methods.Visual guidance includes two important parts: path information extraction and motion control.Firstly, the preprocessing technology of AGV path image obtained by camera is discussed, including path image enhancement, path image filtering, path image segmentation and path image location correction.The purpose of preprocessing is to remove the interference of other information and reduce the computation of subsequent correlation processing on the premise of preserving the global information of the path as much as possible.At the same time, the digital image stabilization technology is used to correct the position of the path image.In order to solve the problem of unstable path tracking caused by random camera jitter in the image, the path information is extracted from the pre-processed AGV path image.Firstly, according to the edge information of the path mark line, the set of path center points is obtained, and then the path model is classified based on curvature angle estimation, which is divided into straight line path model, arc turning path model and broken line turning path model.According to the characteristics of the path model, the calculation method of position offset and angle offset for extracting AGV navigation is deduced. The AGV path fuzzy tracker based on visual guidance is designed, and the AGV motion model is built.The state equation of AGV motion control is derived.This paper analyzes the actual operation and control method of the trolley and selects the fuzzy control method, and designs a fuzzy controller according to the motion model and control requirements of AGV.The simulation results show that the fuzzy controller designed in this paper can correct the deviation in the motion of the car and make the AGV track mark run smoothly.
【学位授予单位】：青岛科技大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：TP391.41;TP242

【相似文献】