The navigation method of wheeled mobile robot based on data fusion obtained from onboard sensors and camera

Abstract

In this paper, we propose a method for constructing a mobile robot navigation system based on data fusion obtained from on-board sensors and cameras using an extended Kalman filter. Moreover, the video device we use is a low cos web-camera, that leads to whole system price reduction. The approach can significantly improve the accuracy of mobile robot navigation where global positioning systems are not available. The method proposed consists of two stages. In the first stage video processing and providing additional information about mobile robot movement occurs. A second stage is the fusion of the navigation data obtained from the video stream and the data received from the onboard sensors of the mobile robot. The experimental study of the proposed approach proved its higher accuracy and efficiency comparing with methods based on onboard sensors or mobile robot video camera only.