A new monocular vision simultaneous localization and mapping process for high-precision positioning in structured indoor environments

Abstract

Abstract High-precision positioning in areas with unobtrusive features is difficult to achieve with visual based indoor positioning solutions. Most current solutions offer low precision and are expensive and/or require a priori training. A ceiling distributed on the same plane is regularly arranged and widespread in structured indoor environments. We propose a new monocular vision simultaneous localization and mapping (NMV SLAM) method for high-precision indoor positioning. First, image morphology technology is adopted to extract the ceiling corner accurately. Second, the geometric solution based on the a priori ceiling map realizes global positioning and makes up for the deficiency of time-consuming nonlinear optimization. Experiments show that the root mean square error of position measured by the proposed NMV SLAM process is less than 5.43 mm in a room with an entire ceiling. Moreover, the proposed NMV SLAM processes an average of 30.61 frames in 1 s on an i5-9400F central processing unit.