Data processing method and device, method and device for achieving positioning and intelligent equipment

Abstract

The invention discloses a data processing method and device, a method and device for achieving positioning and intelligent equipment. Loop detection is conducted through semantic line segments, and due to the fact that the semantic line segments have the advantages of being resistant to illumination and shielding, the problem that loopback detection cannot be effectively achieved due to the fact that the positions of the feature points are changed due to illumination changes is well avoided, the effectiveness of loopback detection is improved, the positioning accuracy is improved, and the invention is particularly suitable for complex scenes.

Description

technical field [0001] This application relates to but is not limited to the field of intelligent robots, especially a data processing method and device and a method for realizing positioning. Background technique [0002] In the use of intelligent robots, loop detection information is needed to eliminate the cumulative error of positioning. Among them, the loopback detection refers to detecting whether the robot has ever been to the current location during the operation of the intelligent robot. The function of the loopback detection information is to help the intelligent robot eliminate the cumulative error caused by its own positioning, which is of great significance for optimizing the positioning of the intelligent robot. [0003] In related technologies, one kind of loop closure detection is feature point-based loop closure detection, and this method needs to construct a descriptor-based Bag-of-Words (BoW, Bag-of-Words) model. The bag-of-words model maps the descripto...

AI Technical Summary

Problems solved by technology

However, changes in light will cause the image of the same location to interfere with the position of feature points and descriptors due to changes in light. Therefore, this method is limited by the stability of feature points and descriptors, and drastic changes in descriptors will It will lead to changes in the statistical characteristics of the vocabulary space, which will affect the effect of loop detection and reduce the effectiveness of loop detection.

[0004] Another loop detection method using the feature point position and the weak geometric consistency strategy of Hough (Hough) voting, if the position of the feature point changes due to illumination changes, the loop detection cannot be effectively realized.

[0005] There is also a loop detection method based on point descriptors and line descriptors. A bag-of-words model is specially constructed for line features, which is used in conjunction with the bag-of-words model of points. This method is improved compared to the scene where only feature points are used. , however, in scenes with large illumination changes, it still cannot meet the needs of loop detection

Images