Indoor Localization Method of Mobile Educational Robot Based on Visual Sensor

<p>This article aims to study the mobile positioning method of mobile educational robots indoors. In order for robots to be able to unblocked indoors, they can avoid obstacles well. Vision sensors are the direct source of information for the entire machine vision system, and are mainly composed of one or two graphics sensors, sometimes accompanied by light projectors and other auxiliary equipment. This paper presents an indoor positioning method for mobile educational robots based on visual sensors. Build some models to compare which algorithm is more in line with the positioning of indoor mobile educational robots. The experimental results in this paper show that the positioning accuracy of the optical flow meter and the odometer on the short-haired carpet is equivalent (both are less than the index 4.52%); the positioning error of the optical flowmeter on the long-haired carpet is the largest 7%, and the positioning error of the odometer is the largest it reached 83%. The error of the algorithm positioning method after the visual odometer fusion is obviously smaller than that of the optical flow method. This shows that the algorithm after visual process fusion is more suitable for indoor mobile educational robot positioning than this optical flow method.</p> <p>&nbsp;</p>