A visual positioning method and an apparatus

Abstract

A visual positioning method and an apparatus suitable for positioning an electronic device having an image sensing element. The method comprises: utilizing a particle filter to spray a plurality of particles to multiple positions around an estimated position by the electronic device, then identifying a plurality of map feature points in a visible range corresponding to each individual particle, and combining to form a visual map feature point set; then, using the image sensing element to extract the image and analyzing a plurality of image feature points of the image, and then comparing the above map feature point in the visual map feature point set with the image feature points of the image to obtain a comparison result; and at last, selecting a particle most close to the comparison result, wherein the position of the particle is the position of the electronic device.

Description

technical field [0001] The disclosure relates to a visual positioning method and device. Background technique [0002] The purpose of research and development of positioning and navigation technology is to estimate the position of the mobile carrier during the actual environment movement, and then achieve the accurate and stable navigation requirements of the mobile carrier. This technology can be divided into two categories: one is to build an electronic map of the environment, and use this map as the reference information when positioning the mobile carrier; Simultaneous positioning and environmental electronic map construction are carried out in the system. [0003] If the technology of building an electronic map of the environment is adopted, the mobile carrier can collect environmental information through sensing elements during the movement, and further compare it with the pre-built electronic map of the environment to complete its position estimation. for example, ...

AI Technical Summary

Problems solved by technology

[0006] However, when the particle filter algorithm is generally used for the positioning of the visual image sensing element of the mobile carrier, each particle must be compared with the image and the map feature points, so when the number of particles increases, the calculation amount will follow. significantly increased

In addition, most of the positioning technologies of the above-mentioned visual image sensing components are still in the simulation or offline positioning calculation, and cannot complete the positioning of the mobile carrier in real time, making these methods unable to be used for the positioning and navigation needs of the mobile carrier in the actual environment

Images