为了实现室内移动机器人的自定位,考虑到室内的复杂程度和环境特色,设计一种基于单目视觉的移动机器人定位系统。机器人在室内带有天花板路标信息的环境下,以颜色和形状结合的路标作为特征,通过机器人头顶与天花板垂直的摄像机,运用彩色图像分割轮廓矩匹配的方法识别路标,解析路标携带的世界坐标信息,从而计算出机器人的全局位置和航向角。该方法大大提高了图像处理和特征提取的速度,通过实验验证,其满足定位系统的实时性和鲁棒性要求。%In order to realize the self⁃location of the indoor mobile robot,a mobile robot location system based on monocu⁃lar vision was designed in consideration of the indoor complexity and environment feature. The landmark combining color and shape is taken as the robot feature in the indoor environment with the ceiling landmark information. The color image segmenta⁃tion and contour matching method is used to recognize the landmark by means of the camera whose head is vertical to the cei⁃ling,so as to analyze the world coordinate information of the landmark,and calculate the global position and azimuth angle of the robot. This method can improve the speed of image processing and feature extraction greatly. The experimental results show this method can satisfy the requirements of system real⁃time performance and robustness.
展开▼