NAO Robot Indoor Navigation Based On Artificial Landmark

In recent decades,with the rapid development of robot technology and increasing intelligent of robots,robots are developing from the field of industrial production to our daily life.Robots have a wide possibility of application in daily life and their navigation abilities are the basis to finish the tasks.Of all the navigation modes for robots,visual navigation has a lot of advantages compared to the others.Natural landmarks' characteristics are difficult to extract,while artificial landmarks have the the advantages of detection and identification.This article studys a robot navigation method based on monocular vision and artificial landmarks.There are mainly parts of the thesis,we introduce the selection principle of artificial landmarks and naomark is selected to be the basis of the robot self-localization.We describe the model of the camera,the transformation relationship between the coordinate systems,study the methods for monocular camera calibration and realizes the Zhang zhengyou method for camera calibration.Finally,combined with the movement of the robot,the thesis gives the coversion of the image coordinate system to the camera coordinate system and then to the the world coordinate system,realizes the robot self-localization based on a single landmark.Inview of indoor environments,this thesis adopts a way of the topological map construction based on artificial landmarks,and researchs the shortest path algorithm for toplogical map which is improved combined with the actual situation.In order to reduce the error in navigation process casused by only depend on topological map.We study the model of the robot and introduces the Kalman filter theory.At last,we reasizes indoor navigation of NAO robot based on artificial landmarks.