KR-102962006-B1 - System and method for determining pose

Abstract

A system (100) and a method for determining a pose. The system (100) can obtain an initial pose of a moving device (110) (510). The system (100) can also determine a plurality of candidate poses of the moving device (110) based on the initial pose of the moving device (110) according to at least one map associated with the moving device (110) (520). The system (100) can determine a target initial pose of the moving device (110) based on a plurality of candidate poses and laser data obtained by the moving device (110) (530).

Inventors

• 후 리즈
• 린 후이
• 커 지엔화
• 루 웨이
• 인 쥔

Assignees

• 저장 다후아 테크놀러지 컴퍼니 리미티드

Dates

Publication Date

20260507

Application Date

20201230

Priority Date

20200709

Claims (20)

1. As a system, At least one storage device including an instruction set; and It includes at least one processor that communicates with the above-mentioned at least one storage device, and When executing the above instruction set, the at least one processor is the system To acquire the initial pose of the moving device; Determining a plurality of candidate poses of the moving device based on an initial pose of the moving device according to at least one map associated with the moving device; Determining the target initial pose of the moving device based on the above plurality of candidate poses and laser data acquired by the moving device, and To obtain the initial pose of the above-mentioned moving device, the at least one processor additionally the system Determining feature information of an image acquired by the above-mentioned moving device; A matching result is generated by matching the feature information of the above image with the reference feature information of multiple reference images stored in the feature database, and It is configured to obtain an initial pose of the moving device based on the above matching result, and In order to obtain an initial pose of the moving device based on the above matching result, the at least one processor is the system Determining a plurality of similarities between the feature information of the above image and the reference feature information of the plurality of reference images, and In response to a determination that all multiple similarities are below a similarity threshold, the moving device is moved to obtain a second image, and moving the moving device includes rotating the moving device at a certain angle or translating it by a distance smaller than a distance threshold without colliding with other objects. Determining the second feature information of the second image above, and Determining a plurality of second similarities between the second feature information of the second image and the reference feature information of the plurality of reference images, and Identifying a second similarity that exceeds the similarity threshold from the plurality of second similarities, and A system configured to determine an initial pose of the moving device based on a second reference pose or second reference feature information corresponding to a second reference image in which the second similarity exceeds the similarity threshold.
2. In paragraph 1 In order to obtain an initial pose of the moving device based on the above matching result, the at least one processor is the system A system configured to determine an initial pose of the moving device based on a reference pose or reference feature information corresponding to a reference image having a similarity exceeding the similarity threshold, in response to a decision in which any of the multiple similarities exceeds a similarity threshold.
3. In Article 1, In order to obtain an initial pose of the moving device based on the above matching result, the at least one processor, the system, A system configured to determine an initial pose of a moving device based on one or more of the previous poses of the moving device and odometry data of the moving device in response to a determination that all of the plurality of second similarities are below a similarity threshold.
4. In any one of paragraphs 1 to 3, The above feature database acquires a reference map, and Based on the above reference map, a plurality of reference poses of a reference movement device are obtained, and two adjacent reference poses of the plurality of reference poses satisfy a preset condition, and A plurality of images obtained by the reference movement device from the plurality of reference poses are determined as the plurality of reference images, and For each of the above plurality of reference images, reference feature information of the reference image is extracted and stored, occurs, A system in which the above reference feature information includes at least one of a reference feature point, a reference representation of the above reference feature point, or a reference coordinate of the above reference feature point.
5. In Article 4, A system in which the above-mentioned preset condition includes the time difference between points corresponding to each of the two adjacent poses exceeding a time threshold or the difference between two adjacent poses of the reference moving device exceeding a difference threshold.
6. In any one of paragraphs 1 to 3, The above at least one processor is the system Determining the matching result between the laser data acquired by the moving device and the map associated with the moving device; In response to a result where the above matching result satisfies a preset condition, it is determined whether the maximum similarity between the above feature information and the corresponding reference feature information of the above feature database is smaller than a similarity threshold, and A system configured to update the feature database by replacing reference feature information corresponding to the reference image with feature information of the image in response to a result in which the similarity is smaller than the similarity threshold.
7. In any one of paragraphs 1 to 3, The above at least one map includes at least two maps with different resolutions related to the moving device, and A system configured such that, in order to determine a plurality of candidate poses of a moving device based on an initial pose of the moving device according to at least one map associated with the moving device, at least one processor causes the system to determine a plurality of candidate poses of the moving device by rotating or moving the initial pose within a predetermined range on at least two maps.
8. In Article 7, A system configured such that, in order to determine a target initial pose of a moving device based on a plurality of candidate poses and laser data acquired by the moving device, at least one processor, the system determines a target initial pose of a moving device based on a plurality of candidate poses of the moving device and at least two maps using a branching and boundary algorithm.
9. In Article 8, To determine a target initial pose of the moving device based on at least two maps and a plurality of candidate poses of the moving device using the above branching and boundary algorithm, the at least one processor is a system For each of at least two maps, determine one or more modified maps by downsampling the map; A system configured to determine a target initial pose of a moving device based on at least two maps, at least a modified map corresponding to the maps, and a plurality of candidate poses of the moving device using a branching and boundary algorithm.
10. In Article 8, The above at least two maps include a first map having a first resolution and a second map having a second resolution; To determine a target initial pose of the moving device based on at least two maps and a plurality of candidate poses of the moving device using the above branching and boundary algorithm, at least one processor, the system Based on the first map having a first resolution and the initial pose of the moving device, a plurality of first candidate poses of the moving device are determined among a plurality of candidate poses of the moving device; Using the branching and boundary algorithm above, a first target initial pose of the moving device is determined based on the plurality of first candidate poses; Based on a second map having a second resolution and a first target initial pose of the moving device, a plurality of second candidate poses of the moving device are determined among a plurality of candidate poses of the moving device; Using a branching and boundary algorithm, a second target initial pose of the moving device is determined based on a plurality of second candidate poses of the moving device; A system configured to designate a second target initial pose of the above-mentioned moving device as the target initial pose of the above-mentioned moving device.
11. A method implemented in a computing device comprising at least one processor, at least one storage medium, and a networked communication platform, Step of acquiring an initial pose of the moving device; A step of determining a plurality of candidate poses of the moving device based on an initial pose of the moving device according to at least one map associated with the moving device; and The method includes the step of determining a target initial pose of the moving device based on the plurality of candidate poses and laser data obtained by the moving device, The step of acquiring the initial pose of the above-mentioned moving device is: Determining feature information of an image acquired by the above-mentioned moving device, Generating a matching result by matching the feature information of the above image with the reference feature information of a plurality of reference images stored in a feature database, and It includes obtaining an initial pose of the moving device based on the above matching result, and Obtaining the initial pose of the moving device based on the above matching result is: Determining multiple similarities between the feature information of the above image and the reference feature information of the above multiple reference images In response to a determination that all multiple similarities are below a similarity threshold, obtaining a second image by rotating the moving device at a certain angle or translating it by a distance smaller than a distance threshold without colliding with other objects. Determining the second feature information of the second image above, Determining a plurality of second similarities between the second feature information of the second image and the reference feature information of the plurality of reference images, Identifying a second similarity that exceeds a similarity threshold from the plurality of second similarities, and A method comprising determining an initial pose of the moving device based on a second reference pose or second reference feature information corresponding to a second reference image in which the second similarity exceeds the similarity threshold.
12. As a non-transient computer-readable medium, comprising an executable instruction that directs at least one processor to perform a method when executed by at least one processor, The above method is; Step of acquiring an initial pose of the moving device; A step of determining a plurality of candidate poses of the moving device based on an initial pose of the moving device according to at least one map associated with the moving device; and The method includes the step of determining a target initial pose of the moving device based on the plurality of candidate poses and laser data obtained by the moving device, The step of acquiring the initial pose of the above-mentioned moving device is: Determining feature information of an image acquired by the above-mentioned moving device, Generating a matching result by matching the feature information of the above image with the reference feature information of a plurality of reference images stored in a feature database, and It includes obtaining an initial pose of the moving device based on the above matching result, and Obtaining the initial pose of the moving device based on the above matching result is: Determining multiple similarities between the feature information of the above image and the reference feature information of the above multiple reference images In response to a determination that all multiple similarities are below a similarity threshold, obtaining a second image by rotating the moving device at a certain angle without colliding with another object or translating it by a distance smaller than a distance threshold. Determining the second feature information of the second image above, Determining a plurality of second similarities between the second feature information of the second image and the reference feature information of the plurality of reference images, Identifying a second similarity that exceeds a similarity threshold from the plurality of second similarities, and A non-transient computer-readable medium comprising determining an initial pose of the moving device based on a second reference pose or second reference feature information corresponding to a second reference image in which the second similarity exceeds the similarity threshold.
13. delete
14. delete
15. delete
16. delete
17. delete
18. delete
19. delete
20. delete

Description

System and method for determining pose related Cross-applications-reference This application claims priority to Chinese Patent Application No. 202010655087.1 filed on July 9, 2020 and Chinese Patent Application No. 202010963179.6 filed on September 14, 2020, the full contents of which are incorporated herein by reference. Technology field The present disclosure generally relates to automation technology, and in particular to a system and method for determining the pose of a moving device. With the advancement of automation and computer technologies, pose determination is becoming increasingly important. Generally, the pose of a mobile device (e.g., a robot) is determined based on a single type of data acquired by the device (e.g., image data, laser data, odometry data). For instance, the pose of a mobile device is determined based on laser data and a preset global map of the region where the device is located. However, in some cases, a single type of data cannot guarantee the accuracy of pose determination. For instance, since the actual environment surrounding the mobile device changes dynamically and a preset global map cannot reflect the real-time conditions of the actual environment, this can affect the accuracy of pose determination based on laser data and the preset global map. Therefore, it is desirable to provide a system and method for accurately and efficiently determining the pose of a mobile device. According to one aspect of the present disclosure, a system may be provided. The system may include at least one storage device comprising a set of instructions; and at least one processor communicating with the at least one storage device, and when executing the set of instructions, the at least one processor may be configured to cause the system to acquire an initial pose of a moving device; to determine a plurality of candidate poses of the moving device based on the initial pose of the moving device according to at least one map associated with the moving device; and to determine a target initial pose of the moving device based on the plurality of candidate poses and laser data acquired by the moving device. In some embodiments, to obtain an initial pose of the moving device, the at least one processor may be configured to determine feature information of an image obtained by the moving device of the system; generate a matching result by matching the feature information of the image with reference feature information of a plurality of reference images stored in a feature database, and obtain an initial pose of the moving device based on the matching result. In some embodiments, to obtain an initial pose of the moving device based on the matching result, the at least one processor may be configured such that the system determines a plurality of similarities between the feature information of the image and the reference feature information of the plurality of reference images, identifies a similarity that exceeds a similarity threshold from the plurality of similarities; and determines an initial pose of the moving device based on a reference pose or reference feature information corresponding to a reference image having a similarity that exceeds the similarity threshold. In some embodiments, to obtain an initial pose of the moving device based on the matching result, the at least one processor may be configured to move the moving device to obtain a second image in response to a determination that all of the plurality of similarities are below a similarity threshold, determine second feature information of the second image, determine a plurality of second similarities between the second feature information of the second image and the reference feature information of the plurality of reference images, identify a second similarity exceeding the similarity threshold from the plurality of second similarities; and determine an initial pose of the moving device based on a second reference pose or second reference feature information corresponding to a second reference image where the second similarity exceeds the similarity threshold. In some embodiments, to obtain an initial pose of the moving device based on the matching result, the at least one processor may be configured such that the system determines an initial pose of the moving device based on one or more previous poses of the moving device and odometry data of the moving device in response to a determination that all of the plurality of second similarities are below a similarity threshold. In some embodiments, the feature database may be generated by obtaining a reference map, obtaining a plurality of reference poses of a reference moving device based on the reference map, two adjacent reference poses of the plurality of reference poses satisfying a preset condition, determining a plurality of images obtained by the reference moving device from the plurality of reference poses as the plurality of reference images, and for each of the plurality o