A method of vision-based navigation for rescue robots using motion information

An effective and accurate navigation for rescue robots in some human inaccessible sites is very necessary. This paper proposes a method of vision-based navigation with high accuracy and real-time capability. Firstly, the system overview and the experimental platform are presented. Then an image processing algorithm using classic theories, which resulted in less calculated amount, is introduced to detect the guidance line. Lastly, the control model based on the results of image processing and motion information of the robot which makes the control more precise is developed. The results of the experiments show the feasibility of this method.