Autonomous positioning method for robot

An autonomous positioning and robot technology, which is applied in the directions of instruments, image analysis, image enhancement, etc., can solve problems affecting the positioning accuracy of robots, and achieve the effects of improving image alignment accuracy, enhancing constraint relationships, and improving accuracy

CN112884838AActive Publication Date: 2021-06-01CHONGQING UNIV
15 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Publication Date
2021-06-01

Smart Images

  • Figure 1
    Figure 1
  • Figure 2
    Figure 2
  • Figure 3
    Figure 3
Patent Text Reader

Abstract

The autonomous positioning method for a robot comprises the following steps that: 1) enabling a robot to acquire an image of a current environment through a camera; 2) converting the current frame image acquired by the camera and the selected reference image serving as the positioning reference into an HSI color space; (3) extracting a projection point of a point Pj in a real environment space in the reference image from the reference image; (4) calculating the projection point of the point Pj in the current frame image; (5) calculating a projection error rj between the projection points; and 6) continuously iteratively performing solving by minimizing the objective function E (xi) to obtain the optimal solution of the camera pose. According to the autonomous positioning method for the robot, the precision of pose estimation of the robot can be improved by using more image information and constraint relationships through the image registration result with higher precision, namely, the precision of autonomous positioning of the robot is improved.
Need to check novelty before this filing date? Find Prior Art

Description

technical field

[0001] The invention relates to the fields of computer vision and robot technology, in particular to a method for autonomous positioning of a robot. Background technique

[0002] In the existing robot autonomous positioning method, the visual SLAM direct method adopted uses the assumption of photometric invariance. After the input image is directly converted into a grayscale image, the camera motion and point projection are simultaneously estimated according to the pixel grayscale information of the image. . However, according to the visual characteristics of the human eye, the human eye is more sensitive to color than grayscale; and the grayscale assumptions in the actual camera imaging system will be affected by the camera's automatic exposure and the specular reflection of the object surface. Therefore, using grayscale information alone may lead to failure of image alignment.

[0003] Simply put, image alignment, which aims to find the best image transfo...

Examples

Embodiment Construction

[0037] The present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0038] In the present embodiment, the autonomous positioning method of the robot comprises the following steps:

[0039] 1) The robot collects images of the current environment through the camera.

[0040] 2) Convert the current frame image collected by the camera and the reference image selected as the positioning reference into the HSI color space to obtain three components of H, S, and I.

[0041] 3) Extract the point P in the real environment space from the reference image j The projected point p in the reference image 1 j :

[0042]

[0043] pixel The grayscale value of The color components are represents the projected point the number of rows in the image array, represents the projected point the number of columns in the image array, projected point The image coordinates of .

[0044] In the above formula (1), K is the in...