Research On Indoor Positioning Technology Based On The Fusion Of Monocular Camera And Accelerometer

With the advancement of computer technology, sensor technology and artificial intelligence, the intelligence of mobile robots keep increasing. Autonomous positioning and navigation is one of the core technologies of autonomous mobile robots. The accuracy of positioning & naviation system is critical to the finish of tasks of robot in the real environment.In the paper, the mobile robot vision-based indoor positioning technology with the integration of accelerometer is studied. In the research of mobile robot localization based on vision systems, the camera’s angle in robot localization, the light or other external factors all have influences on the feature extraction and matching. Therefore the processing of image information directly affects the positioning accuracy. This paper uses the ORB (Oriented-Rotated Brief) algorithm for feature detection and matching. Based on the principle of mouse positioning, robot’s preliminary pose information is caculated by comparing the difference between consecutive images. Then Kalman filter is used to fuse the data information of the accelerometer and the vision sensor. The pose and location of robot is then estimated based on the Kalman filter.The algorithm is test on a PC platform. The digital camera connects with the PC through the USB interface. The PC interface can figure out the real-time information of the robot, what’s more, it can map out the running track of the car. Experimental results show that the position method is effective in improving the positioning accuracy, with good usability and stability.