The research in robot navigation and localization is to enable mobile robots to determine their position in environment real-time and reliably, which is necessary for mobile robots to navigate autonomously. Vision based localization has many advantages that other methods are unsurpassable and has become an important research field. Supported by the 863 National High-tech Project and National Natural Scientific Foundation, some key techniques of vision based localization and navigation for indoor mobile robots are studied in this thesis. The main contents of this paper are listed as follows: Firstly, the background and development status of mobile robots localization are briefly reviewed; a survey of mobile robot navigation based on vision is presented. Secondly,to remedy the disadvatage of popular lens,we design a landmark based on the unique character of the fisheye lens,which makes the robot get its position in the wide range easily. Thirdly, to improve the accuracy of the localization . we integrate the Cam shift tracking with multi-cus fusion in object recognition,then make use of the disparity of the binocular to gain the position of the object. The research of the localization based on the artificial landmark MR2 is also designed and the more accurate result is obtained. Fourthly, a real-time visual odometry algorithm with monocular is presented，furthermore, the validity and error of the algorithm are also analyzed. The method is based on two assumptions: indoor robots are moving on the planar ground and the camera view is in the planar structure. Experimental results show that the proposed method can achieve good precision. Fifthly,integrating the visual odometry and the artificial landmark MR2, we design a map building method, which gains the high accuracy in the map construction of mobile robot navigation. Finally, the achieved research results are summarized and future work is addressed.
修改评论