Research On Visual Navigation Of Autonomous Mobile Robot

Mobile robot navigation research based on vision system is quite active recently. Because computer vision carry a tremendous amount of information about the scene the robot represent, and describes roundly the environment, vision system plays a crucial role in mobile robot technique. This dissertation researches the vision system of a two-wheeled autonomous mobile robot. It contains mainly five parts:We firstly give a brief overview to the development of autonomous mobile robot and the status of mobile robot vision system investigation.Framework and hardware structure of the mobile robot designed by ourselves are presented. Structure of the vision system is established and real-time image acquisition is achieved.A camera calibration method based on polynomial fit for the vision system is put forward.Guass filter is adopted for image preprocessing and canny operator for edge detection. The real time line detection is accomplished by Hough algorithm, and then displayed on the interface between mobile robot and surveillant. The coordinate of the line is calculated and transferred to the path-layout module.At last, we make a research about kinds of color model and segmentation of color image, and import the threshold vector in YUV color space on this base, which avoids the expensive computing when we differentiate the thresholds of many colors, real-time capability is improved greatly using this technology.