Granted Patent B2

US 12,498,224 · App. 18/085,013 · Granted Dec 16, 2025

Inventors: Yi-Lun Wen (Taoyuan, TW); Chung-Hsien Wu (Hsinchu, TW)

Assignee: INDUSTRIAL TECHNOLOGY RESEARCH INSTITUTE

G01C21/005G06T7/73B64C39/024B64U2201/104G06T2207/20221G06T2207/30204

Abstract

Claims (49)

1 . A positioning system, for providing an environmental positioning information of a field using monocular vision, wherein the field has a plurality of virtual floors corresponding to different heights, each of the virtual floors has a plurality of marking positions, the positioning system comprising:

a computing device, configured to generate the environmental positioning information according to a first environmental image of the field, wherein the environmental positioning information has a positioning accuracy for the marking positions of each of the virtual floors, and the positioning accuracy has various levels; and

a first camera, configured to generate a marking image of a marking object, wherein the marking object is placed at a corresponding one the marking positions of the virtual floors, and the first camera is further configured to capture an accuracy positioning pattern and a plurality of positioning points of the marking object based on the marking image,

wherein the computing device is further configured to analyze the accuracy positioning pattern and the positioning points of the marking object, analyze a mapping relationship between the marking image and the marking object, evaluate the positioning accuracy based on the accuracy positioning pattern, the positioning points and the mapping relationship, and divide each of the virtual floors into a plurality of work zones according to the positioning accuracy; and

wherein the computing device is further configured to map the positioning points of the marking object to a plurality of pixels of the marking image, calculate a mean error of a plurality of coordinate positions of the pixels of the marking image, and evaluate the positioning accuracy according to the mean error.

2 . The positioning system according to claim 1 , wherein the first camera is further, configured to generate the first environmental image of the field which includes the marking image corresponding to each of the marking positions.

3 . The positioning system according to claim 1 , wherein the computing device is further configured to perform a lens distortion correction to correct the coordinate positions of the pixels mapped from the positioning points and calculate the mean error according to the coordinate positions which are corrected.

4 . The positioning system according to claim 1 , wherein the marking object has a predefined pattern, the predefined pattern includes a plurality of grids, and the positioning points of the marking object include junctions of the grids.

5 . The positioning system according to claim 1 , wherein one of the marking positions has a global navigation satellite system (GNSS) information, the computing device is further configured to calculate the coordinate positions of the pixels mapped from the positioning points according to the GNSS information.

6 . The positioning system according to claim 1 , wherein the computing device is further configured to divide the work zones according to the mean error;

when the mean error is greater than a first error threshold, the positioning accuracy corresponding to the work zones is at a “failed positioning” level;

when the mean error is less than the first error threshold and is greater than the second error threshold, the positioning accuracy corresponding to the work zones is at a “low positioning accuracy” level; and

when the mean error is less than the second error threshold, the positioning accuracy corresponding to the work zones is at a “high positioning accuracy” level.

7 . The positioning system according to claim 6 , wherein the marking object has a predefined pattern, the predefined pattern has a color information, the computing device is further configured to evaluate an environmental illuminance of the field according to the color information and divide the work zones according to the environmental illuminance.

8 . The positioning system according to claim 1 , wherein the marking object has a predefined pattern, the predefined pattern has a boundary information, the computing device is further configured to mark a plurality of boundary lines of the work zones according to the boundary information.

9 . The positioning system according to claim 1 , further comprises:

a second camera, configured to generate a second environmental image of the field, the second environmental image and the first environmental image have an overlapped area,

wherein, the computing device is further configured to stitch the first environmental image and the second environmental image to cover the field.

10 . A positioning method, for providing an environmental positioning information of a field using monocular vision, wherein the field has a plurality of virtual floors corresponding to different heights, each of the virtual floor has a plurality of marking positions, the positioning method comprising the following steps:

generating the environmental positioning information according to a first environmental image of the field by a computing device, wherein the environmental positioning information has a positioning accuracy for the marking positions of each of the virtual floors, and the positioning accuracy has various levels;

generating a marking image of a marking object by a first camera, wherein the marking object is placed at a corresponding one the marking positions of the virtual floors;

capturing an accuracy positioning pattern and a plurality of positioning points of the marking object based on the marking image, by the first camera;

analyzing the accuracy positioning pattern and the positioning points of the marking object, by the computing device;

analyzing a mapping relationship between the marking image and the marking object, by the computing device;

evaluating the positioning accuracy based on the accuracy positioning pattern, the positioning points and the mapping relationship, by the computing device;

dividing each of the virtual floors into a plurality of work zones according to the positioning accuracy, by the computing device;

mapping the positioning points of the marking object to a plurality of pixels of the marking image, by the computing device; and

calculating a mean error of a plurality of coordinate positions of the pixels of the marking image, and evaluating the positioning accuracy according to the mean error, by the computing device.

11 . The positioning method according to claim 10 , further comprises the following steps:

generating the first environmental image of the field which includes the marking image corresponding to each of the marking positions.

12 . The positioning method according to claim 10 , wherein, the step of calculating the mean error of the coordinate positions comprises:

performing a lens distortion correction to correct the coordinate positions of the pixels mapped from the positioning points; and

calculating the mean error according to the coordinate positions which are corrected.

13 . The positioning method according to claim 10 , wherein the marking object has a predefined pattern, the predefined pattern includes a plurality of grids, the positioning points of the marking object include junctions of the grids.

14 . The positioning method according to claim 10 , wherein one of the marking positions has a GNSS information, the positioning method further comprises the following steps:

calculating the coordinate positions of the pixels mapped from the positioning points according to the GNSS information.

15 . The positioning method according to claim 10 , further comprises the following steps:

dividing the work zones according to the mean error;

wherein, when the mean error is greater than a first error threshold, the positioning accuracy corresponding to the work zones is at a “failed positioning” level;

when the mean error is less than the first error threshold and is greater than the second error threshold, the positioning accuracy corresponding to the work zones is at a “low positioning accuracy” level; and

when the mean error is less than the second error threshold, the positioning accuracy corresponding to the work zones is at a “high positioning accuracy” level.

16 . The positioning method according to claim 15 , wherein the marking object has a predefined pattern, the predefined pattern has a color information, the positioning method further comprises the following steps:

evaluating an environmental illuminance of the field according to the color information; and

dividing the work zones according to the environmental illuminance.

17 . The positioning method according to claim 10 , wherein the marking object has a predefined pattern, the predefined pattern has a boundary information, the positioning method further comprises the following steps:

marking a plurality of boundary lines of the work zones according to the boundary information.

18 . The positioning method according to claim 10 , further comprises the following steps:

generating a second environmental image of the field, the second environmental image and the first environmental image have an overlapped area; and

stitching the first environmental image and the second environmental image to cover the field.

Assignments (1)

Priority Claims (1)

Continuity (1)

References Cited (60)