Robot autonomous positioning method and device based on depth camera

Abstract

The invention discloses a robot autonomous positioning method and device based on a depth camera. Different from a general visual positioning scheme in which a color image or a gray image is used forextracting a feature, a depth image with depth information is converted into a gray image containing the depth information, a unit pixel is equal to a fixed depth value during converting, and then linear converting is performed; or the depth value is divided by the maximum depth in the scene, and then multiplied by a maximum pixel value; feature points are extracted from the converted gray image containing the depth information and then mapped to a three-dimensional world through the depth value of the corresponding pixel, the obtained feature points are matched with reconstructed map points,and the current accurate position and orientation of a mobile robot are calculated. The disadvantage that a traditional visual positioning method is too sensitive to the change of ambient light changeis overcome, and high universality is achieved.

Description

technical field [0001] The present invention relates to the field of robots, in particular to an autonomous positioning method and device for a robot, in particular to a depth camera-based autonomous positioning method and device for a mobile robot. Background technique [0002] With the rapid development of mobile robots, computer vision, SLAM, artificial intelligence, and 3D reconstruction technologies, the research level of mobile robots integrating environmental perception, dynamic decision-making and planning, behavior control and execution has also been rapidly improved. The performance of mobile robots has been continuously improved, and the application range of mobile robots has also been greatly expanded. It has also been well applied, and the application prospect is very broad. [0003] Autonomous positioning is the core technology of mobile robots. How to improve the autonomous positioning accuracy of mobile robots without additional cost has become an urgent pr...

AI Technical Summary

Problems solved by technology

The main principles of these methods all use visual sensors to collect grayscale images or color images, and one of the biggest problems faced by the methods is poor stability.

The collected images are greatly affected by changes in ambient light. Daily factors such as sunrise and sunset, turning on and off the lights will cause huge changes in the gray scale and color of the image, which will lead to the instability of the autonomous positioning algorithm.

Images