论文部分内容阅读
在月面巡视探测器漫游过程中,需要测量月面巡视探测器的位置和姿态信息,以使巡视探测器行驶在规划好的路径上,并确保其安全。定位系统可以采取多种方式。在文章中,利用着陆器立体视觉系统对月面巡视探测器定位方法进行了研究,它采用立体视觉测量技术以及彩色图像分割的方法,进行跟踪测量月面巡视探测器位姿信息,能够对惯导系统加里程计的定位方法进行修正,减小月面巡视探测器的计算工作量。在定位方法的研究中,采用彩色图像分割方法对月面巡视探测器进行识别,能够适应变光照条件,并且运算速度快,达到实时图像伺服控制的目的。通过月面巡视探测器上制作特征点及采用基于四边形约束的特征点匹配方法,提高了图像匹配概率。通过卡尔曼滤波技术,能够满足任意时刻的月面巡视探测器定位要求。
During the lunar rover rover rover, it is necessary to measure the position and attitude information of the lunar rover detector, so that the rover probe can be driven on the planned path and ensured its safety. Positioning systems can take a variety of ways. In this paper, the lander stereo vision system is used to study the lunar rover positioning method, which uses stereo vision measurement technology and color image segmentation method to track and measure lunar rover position and attitude information, Guide system plus the positioning method of the odometer to be amended to reduce the lunar rover detector calculation workload. In the research of positioning method, the method of color image segmentation is used to recognize the lunar surface inspection detector, which can adapt to the condition of variable light illumination and has the advantages of fast calculation speed and real-time image servo control. Through the production of feature points on lunar patrol detectors and the matching of feature points based on tetragonal constraint, the image matching probability is improved. By Kalman filter technology, to meet the arbitrary time of the lunar rover detector positioning requirements.