Map fusion method and fusion platform based on ORB features for multiple mobile robots

Abstract

The invention discloses a map fusion method based on ORB features for multiple mobile robots. The method includes the following steps of constructing a network of the robots; utilizing environmental point cloud information obtained by the robots in real time to construct regional grid maps; separately extracting ORB features of the grid maps, conducting feature point matching, and conducting purification on matched feature points; utilizing purified two-dimensional feature point pairs to calculate deviated affine transformation to obtain an optimal affine transformation matrix; conducting calculation of optimal transformation on feature points where the optimal affine transformation matrix is input to obtain a homographic matrix, obtaining an optimal matching point through solution of thehomographic matrix, and setting a fusion proportion to obtain a global grid map. ORB features of fused maps are extracted, and the problem that the fusion efficiency is not high is solved; during fusion, optimal affine transformation parameters of optimal matching point sets of the two maps are calculated, and the problem that the fusion precision can be reduced since overlapped regions are reduced is solved.

Description

technical field [0001] The present invention relates to the technical field of robots, and more specifically relates to a multi-mobile robot map fusion method and a fusion platform based on ORB features. Background technique [0002] With the advent of the intelligent age, robots have widely entered the field of vision of researchers. In recent years, robotics has developed vigorously. As a comprehensive interdisciplinary subject, robotics covers the latest achievements in the fields of control, navigation, machinery and computer science, and is one of the most active fields of current technological development. Remarkable achievements in robotics make it easier for us to deal with some information collection problems, such as environmental monitoring, search and rescue, and dynamic simultaneous localization and map creation SLAM (Simultaneous localization and mapping) systems. SLAM technology includes two parts of simultaneous positioning and map construction. The sensor c...

AI Technical Summary

Problems solved by technology

[0005] In view of this, the present invention provides a multi-mobile robot map fusion method and fusion platform based on ORB features, extracts the ORB features of the fusion map, and solves the problem of low fusion efficiency; the optimal matching point of two maps is calculated during fusion The optimal affine transformation parameters of the set solve the problem that the fusion accuracy will be reduced if there are few overlapping areas

Images