Indoor navigation and positioning method and device for robot

Abstract

The embodiments of the invention disclose an indoor navigation and positioning method and device for a robot. According to the indoor navigation and positioning method and device for the robot, a series of operations such as preprocessing, edge detection, camera distortion correction and feature information extraction are performed on an image captured by a camera, so that a feature pattern is obtained by means of fitting on the basis of the image; IPM transformation is performed on the feature pattern, so that the parameter equation of the feature pattern in a world coordinate system can be obtained; and whether the feature pattern is a valid pattern is judged according to pre-stored map information; if the feature pattern is a valid pattern, the theoretical pose information of the robotcan be determined according to the parameter equation of the valid pattern; the theoretical pose information and pose parameters collected by a wheel encoder are fused through using a Kalman filter, so that the position, speed and attitude angle of the robot can be calculated, so that the attitude of the robot can be adjusted; and therefore, the cumulative error of the wheel encoder can be eliminated in real time, and the accuracy of the positioning of the robot can be improved.

Description

technical field [0001] The invention relates to the technical field of computer vision detection, in particular to a method and device for indoor navigation and positioning of a robot. Background technique [0002] In recent years, intelligent robot technology has developed rapidly. Robot companies and scientific research institutions around the world are stepping up the development of various intelligent autonomous mobile service robots to be used in various environments to replace human labor, so as to reduce the pressure of human life. Service robots are gradually becoming practical and entering the public's field of vision. They can mainly replace human maintenance, transportation, cleaning, guardianship and other tasks. Service robots can not only get rid of annoying and boring repetitive labor work, but also are a high-tech technology with good application prospects. [0003] Robot navigation means that the robot makes path planning according to the known information ...

AI Technical Summary

Problems solved by technology

However, the path of this method can only go in the front, back, left, and right directions, and cannot go in any angle or arc to reduce the system efficiency. Moreover, when the vehicle is in various states such as skidding or power failure, the vehicle cannot find the QR code, and the problem of inability to locate occurs.

[0006] Due to the problem of various errors in a single sensor, multi-sensor fusion is an effective way to solve this problem. At present, most of the Kalman filters are used to fuse the data collected by multiple sensors for processing to improve the positioning accuracy, but these methods are not available. Use the visual sensor to collect and process the regular features of the ground, and then perform map matching according to the regular features of the ground, eliminate the cumulative error of the encoder, determine its own position and attitude angle, and realize its own navigation and positioning

Images