Indoor robot vision positioning system

Abstract

The invention relates to the field of robot navigation, particularly provides an indoor robot vision positioning system and aims to solve the technical problem about how to position an indoor robot accurately. In order to achieve the aim, the indoor robot vision positioning system provided by the invention comprises a road sign device, an image acquisition device and a server; the road sign devicecomprises a plurality of passive road signs; the passive road signs comprise a plurality of marker positions for setting marker points; the image acquisition device is used for acquiring an originalimage and transmitting the original image to the server; the server comprises a first judgment module, a second judgment module, an optimal passive road signal acquisition module and a robot coordinate acquisition module; the first judgment module is used for judging whether the passive road sign exists in the original image or not; the second judgment module is used for acquiring an ID code of the optimal passive road sign, judging whether the ID code exists in a preset database or not and acquiring the pixel coordinates of the center of each marker point when the optimal passive road sign isin a horizontal state; the optimal passive road sign acquisition module is used for acquiring an optimal passive road sign; and the robot coordinate acquisition module is used for acquiring the coordinates of the robot under a world coordinate system.

Description

technical field [0001] The invention relates to the field of robot navigation, in particular to an indoor robot vision positioning system. Background technique [0002] Indoor robot positioning is an important guarantee for completing navigation and other tasks. It can obtain accurate robot pose information based on current observation data and combined with prior map information. [0003] Currently, robot vision positioning can be divided into natural landmark positioning and artificial landmark positioning according to the types of landmarks. Natural landmark positioning refers to the use of existing features in the environment for positioning, without the need to arrange the environment, but the positioning algorithm is complex and poor in robustness. Artificial landmark positioning refers to installing artificially designed feature signs in the environment for positioning. This method has a small amount of calculation, high precision and strong robustness. [0004] Spe...

AI Technical Summary

Problems solved by technology

[0005] In order to solve the above problems in the prior art, that is, to solve the technical problem of how to accurately and robustly position an indoor robot, the present invention provides an indoor robot visual positioning system

Images