Robot indoor positioning method based on ceiling images

Abstract

The invention provides a robot indoor positioning method based on ceiling images. The method comprises the following steps of arranging a pattern obviously different from an image in an adjacent vision field on a ceiling; programming to make a robot traverse indoor positions, shooting pictures through a camera on the top, and splicing collected images to obtain a global image; comparing the actualindoor size with the pixel size of the global image, and obtaining coordinate conversion relationsbetween image coordinates and actual coordinates; calculating the pose of the robot in the global image through an optical flow method and image matching, and according to the coordinate conversion relations, determining the current actual pose of the robot. The robot indoor positioning method basedon the ceiling images calculates the pose of the robot in the global image through the optical flow method and image matching, and determines the current actual pose of the robot according to the coordinate conversion relations. The robot indoor positioning method based on the ceiling images is high in adoptability and good in stability.

Description

Technical field [0001] The invention belongs to the field of visual navigation, and particularly relates to a method for indoor positioning of a robot based on a ceiling image. Background technique [0002] With the widespread application of robots, smart cars and other equipment in life and production, people have higher and higher requirements for their intelligence. The realization of autonomous positioning and navigation is the basic requirement and core technology of mobile robots. Especially in indoor conditions with high accuracy requirements and complex sports environments, how to effectively locate and navigate is an important and meaningful issue. [0003] At present, relatively mature indoor positioning and navigation technologies mainly include magnetic track navigation, inertial navigation, and visual navigation. Among them, magnetic trajectory navigation needs to lay a magnetic trajectory, and the robot can only move based on the trajectory, which is poor in flexibil...

AI Technical Summary

Problems solved by technology

However, due to the complexity of indoor object movement, such as walking people, moving objects, etc., it is very likely to change the image characteristics of the object, resulting in the failure of the positioning and navigation algorithm.

[0004] Therefore, the problems existing in the prior art are: the indoor positioning and navigation of the robot has weak adaptability and poor stability.

Images