视觉跟随导航参数获取算法与验证系统的设计

Design of Visual Following Navigation Parameter Acquisition Algorithm and Verification System

视觉跟随导航参数空间坐标转换环节是连结视觉环境感知与移动机器人运动控制的关键步骤。为了兼顾移动机器人视觉跟随导航参数转换的实时性和准确性,本文结合摄像机针孔模型,并加入一定约束条件与假设条件,设计了一种适用于单目视觉感知局部区域近似平面的导航参数空间坐标转换算法。同时,研究基于Visual Studio 2013结合OpenCV库开发了导航参数转换算法验证软件并进行实验。试验结果表明,本文设计的算法测量值与理论值平均偏差为23.20mm,平均标准差为11.48,最小偏差值为5.02mm,能满足一般地面移动机器人单目视觉导航的要求。

The spatial coordinate conversion process of visual following navigation parameters is a key step in connecting visual environment perception with mobile robot motion control.In order to balance the real-time performance and accuracy of visual tracking navigation parameter conversion for mobile robots,this paper combines the camera pinhole model and adds certain constraints and assumptions to design a navigation parameter space coordinate conversion algorithm suitable for monocular visual perception of local area approximation planes.At the same time,a navigation parameter conversion algorithm validation software was developed based on Visual Studio 2013 and OpenCV library,and experiments were conducted.The experimental results show that the average deviation between the measured values and theoretical values of the algorithm designed in this article is 23.20 mm,the average standard deviation is 11.48,and the minimum deviation value is 5.02 mm,which can meet the requirements of monocular visual navigation for general ground mobile robots.