Research On Indoor Composite Navigation Of Mobile Robot Based On Environmental Feature Recognition

Computer vision plays an increasingly important role in modern science and technology.Visual navigation,which can obtain abundant environmental information,has the advantages of intuition and high accuracy,so it has become a new development direction of navigation technology.Due to binocular stereo vision based positioning and inertial measurement based positioning have their respective limitations,a mobile robot indoor composite navigation method based on environmental feature recognition is proposed in this paper.By effectively combining inertial measurement,binocular vision and fuzzy map,the indoor positioning and navigation of mobile robot in indoor environment are realized simply and quickly.In this paper,the research of indoor composite navigation system based on environment feature recognition is as follows:Firstly,the development process of robot navigation technology is expounded.Popular mobile robot navigation methods are summarized,and for each,we analyze the research status at home and abroad.Then,the positioning technology of inertial measurement system is introduced,and its principle and advantages are analyzed in detail.Secondly,the design of binocular vision system based on environmental feature recognition is proposed.Combining the advantages of SIFT feature description and Harris corner detection,the improved Harris-SIFT algorithm is used to detect and extract corner points in the environment,which matches with SIFT descriptor in the offline visual database,and then the absolute position of the robot is solved.Thirdly,a data fusion mechanism for binocular vision and inertial measurement is put forward.In this paper,the inertial and visual modules operate at different data rate,and therefore an asynchronous Kalman filter is used to fuse results from the two modules.Finally,the hardware module of a stereo camera inertial measurement unit(SC-IMU)is established.And the indoor positioning experiment of the mobile robot is carried out on MT-R robot,in addition,the results of experiments illustrate feasibility and effectiveness of the method.