基于被动式红外路标的移动机器人室内定位技术研究与嵌入式实现

发布时间：2018-01-05 04:35

本文关键词：基于被动式红外路标的移动机器人室内定位技术研究与嵌入式实现　出处：《北京工业大学》2012年硕士论文　论文类型：学位论文

【摘要】：在自主移动机器人研究中，机器人自定位具有非常重要的地位，是实现机器人导航和其它智能行为的前提。论文以机器人视觉为出发点，通过研究单应矩阵定位算法，对基于被动式红外路标的定位方法进行了研究。以嵌入式处理器平台为核心构建了一套完整的具有实用性的移动机器人室内定位系统。论文的主要研究内容如下： 1.设计了一种被动式红外点阵路标，适用于移动机器人在室内大范围移动的自定位。该路标具有易检测、易识别、利于定位计算、制作方便等特点。 2.通过分析定位误差产生的原因，对摄像机实际使用中倾斜安装的情况提出了垂直关系假设算法。该算法运算简单，并且能够对定位模块快速的进行初始标定。采用基于单应矩阵的优化定位算法和相机安装位置初始标定方法，，保证定位精度的同时，补偿了由于实际实用中出现的安装误差引起的定位偏差。该算法易于嵌入式系统实现。 3.通过实验分析其定位效果，表明机器人定位模块在嵌入式系统上实现与红外路标相结合，能够现机器人的定位功能。并且新的算法在保证定位精度的同时，能够很好的实现机器人定位。此定位模块在室内环境下使用，不容易受环境光的干扰，能够降低使用类似定位传感器的机器人成本。
[Abstract]:In the research on autonomous mobile robot, robot self localization is very important, is a prerequisite for robot navigation and other intelligent behaviors. Based on robot vision as a starting point, through the study of homography matrix localization algorithm, a set of methods based on passive infrared landmarks were studied. The embedded processor as the core of the platform a complete set of practical mobile robot indoor positioning system. The main research contents of this paper are as follows:
1., a passive infrared dot matrix roadmap is designed, which is suitable for mobile robot's self localization in large indoor mobile location. The road sign has the characteristics of easy detection, easy identification, convenient location calculation and convenient production.
2. through the analysis of the positioning errors of the camera, the actual use of inclined installation of the proposed algorithm. The vertical relation hypothesis algorithm is simple, fast and able to positioning module for the initial calibration. By using matrix optimization localization algorithm and camera installation position initial calibration method based on homography, ensure the precision at the same time compensate the positioning deviation caused by actual installation error in practice. This algorithm is easy to be implemented in the embedded system.
3. through the analysis of the experiments the locating results show that the robot positioning module combined with infrared landmark in the embedded system, can the robot localization function. And the new algorithm ensures the positioning accuracy at the same time, the robot can achieve a very good location. This location module used in the indoor environment, not easy to interference by the environment light, can reduce the use of similar positioning robot cost.

【学位授予单位】：北京工业大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：TP242;TP368.1

【参考文献】