Feature point position determination method and device, equipment and readable storage medium

Abstract

The invention discloses a feature point position determination method and device, equipment and a readable storage medium and belongs to the technical field of computers. The method comprises steps of obtaining a target feature point of a to-be-determined position; in response to the fact that the type of the target feature point is initialized, determining coordinates of the target feature point obtained in the last initialization process as initial coordinates of the target feature point; obtaining the prior information of the target feature point, wherein the prior information comprises a coefficient matrix of the incremental equation corresponding to the initial coordinates and a column vector of the incremental equation corresponding to the initial coordinates; and based on the coefficient matrix of the incremental equation corresponding to the initial coordinate and the column vector of the incremental equation corresponding to the initial coordinate, updating the initial coordinates to obtain target coordinates of the target feature point, and determining the target coordinates of the target feature point as the position of the target feature point. According to the method, the determined position of the target feature point is more accurate.

Description

technical field [0001] The embodiments of the present application relate to the field of computer technology, and in particular to a method, device, device, and readable storage medium for determining the position of a feature point. Background technique [0002] Visual-Inertial Odometry (Visual-Inertial Odometry, VIO), also known as Visual-inertial Navigation System (Visual-inertial System, VINS), can only use one or more cameras plus one or more inertial navigation units to complete positioning. A navigation and positioning method that can replace GPS (Global Positioning System, Global Positioning System) and laser positioning to a certain extent to achieve high-precision state estimation. [0003] In the related art, the target feature point at the position to be determined is obtained, and in response to the target feature point being initialized, the initialization result is determined as the target coordinate of the target feature point, and the target coordinate of th...

AI Technical Summary

Problems solved by technology

[0004] However, the above-mentioned method of directly determining the initialization result as the target coordinates of the target feature point makes the determined target coordinates not accurate enough, that is, the determined position of the target feature point is not accurate enough

Since the position of the target feature point is used to update the parameters in the VIO, when the position of the target feature point is not accurate enough, the accuracy of the parameters obtained after updating the parameters of the VIO with the position of the target feature point is also low, so that the VIO Positioning accuracy is also lower

Images