基于视觉测量的激光雷达设计与实现

发布时间：2019-01-11 10:25

【摘要】：如今移动机器人正在慢慢地走入我们的生活，，让我们的生活变得更加便捷，工作变得更加舒适，引领着智能化时代的发展。然而在移动机器人技术中，激光雷达可以说是移动机器人的核心部件，其主要用于定位与导航。但是激光雷达成本是很昂贵的，这也是制约移动机器人被广泛快速应用的一个原因。本课题主要是设计一种低成本的容易实现的激光雷达，主要是应用于SLAM导航和三维成像领域。本课题激光雷达的基本原理是激光三角测量法结合视觉测量技术来扫描成像。其具体过程是通过激光器产生激光光束，光束照射在被测物体表面形成散射，其散射光通过摄像头进入CCD成像平面上。然后通过对CCD获取的激光光斑的图像数据进行处理，可以计算出激光光斑中心的像素坐标。接着利用三角测量法，建立激光光路的几何模型，通过几何关系，可以根据激光光斑中心像素坐标计算出激光光斑的实际空间坐标。激光雷达通过上述方法每测量一次像素坐标数据，单片机就控制舵机旋转一度，同时将当前的舵机旋转角和测量数据通过串口传入上位机。根据像素坐标和旋转角计算出相对于舵机旋转中心的空间坐标，利用上位机软件绘制成像。激光雷达是移动机器人定位与导航的核心传感器，本课题主要实现了一种低成本激光雷达，但是在稳定性方面还有所欠佳。最后对全文进行总结，并提出了对本课题激光雷达的机械结构和摄像头的改进，来弥补稳定性的不足。
[Abstract]:Now mobile robots are slowly walking into our life, making our life more convenient, working more comfortable, leading the development of intelligent age. However, in mobile robot technology, lidar is the core component of mobile robot, which is mainly used for location and navigation. However, the cost of lidar is very expensive, which is one of the reasons restricting the rapid application of mobile robots. In this paper, a low cost and easy to implement lidar is designed, which is mainly used in SLAM navigation and 3D imaging. The basic principle of lidar is laser triangulation combined with vision measurement to scan imaging. The laser beam is generated by the laser, and the light beam shines on the surface of the measured object to form a scattering, and the scattering light enters the CCD imaging plane through the camera. Then the pixel coordinates of laser spot center can be calculated by processing the image data of laser spot acquired by CCD. Then the geometric model of laser beam path is established by using triangle measurement method. According to the geometric relation, the actual space coordinates of laser spot can be calculated according to the pixel coordinates of the center of laser spot. By the method mentioned above, the single chip microcomputer controls the rotation degree of the steering gear and transmits the current rotation angle and measurement data of the steering gear to the upper computer through the serial port of each pixel coordinate data measured by the method mentioned above. According to the pixel coordinate and rotation angle, the space coordinate relative to the rotation center of the steering gear is calculated, and the image is drawn by the upper computer software. Lidar is the core sensor of mobile robot positioning and navigation. This paper mainly implements a low cost lidar, but its stability is still poor. Finally, the paper summarizes the whole paper, and puts forward the improvement of the mechanical structure and camera of lidar in order to make up for the shortage of stability.
【学位授予单位】：华中科技大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TP391.41;TN958.98

【相似文献】