A relocation test method, chip and robot based on mapping file

Abstract

The present invention discloses a repositioning test method, chip, and robot based on a mapping file. The method comprises: S1: obtaining a mapping file, then processing the point cloud file of the mapping file to obtain processed point cloud data and record the robot pose corresponding to the point cloud data; S2: reading the processed point cloud data and the robot pose corresponding to the point cloud data, and then performing a repositioning test calculation based on the mapping file; S3: determining the result of the repositioning test calculation based on the matching score calculated by the repositioning test, the robot pose corresponding to the point cloud data of the mapping file, and the repositioning method. This application uses a time-series laser file recorded by the mapping file to perform a full-map repositioning test retrospectively, greatly improving the efficiency of using the test results and reducing labor costs.

Description

Technical Field

[0001] The present invention relates to the field of intelligent robots, and in particular to a repositioning test method, a chip and a robot based on a mapping file. Background Art

[0002] Currently, most indoor intelligent mobile robots have the ability to autonomously locate and navigate, that is, they locate their own position and posture based on existing map data during the execution of tasks, thereby achieving autonomous navigation. When a robot has a map but does not know where it is located on the map, it needs to be relocated. Specifically, during the navigation process, the robot will perceive its surrounding environment information accordingly, and combine the existing map data to confirm whether there is a navigation error. Then, if a navigation error is confirmed, the current navigation position will be adjusted based on the perceived environmental information and the existing map data. Relocation is often the first step for a robot to start working, and the success or failure of relocation is a prerequisite for its subsequent normal operation.

[0003] Relocation testing is time-consuming and labor-intensive in robotic testing. Given limited testers and testing resources, maximizing the utilization of each test record is an urgent issue that needs to be addressed. Summary of the Invention

[0004] To address the above issues, the present invention provides a mapping-based relocation test method, chip, and robot. This application uses time-sequenced laser files recorded in the mapping to perform full-map relocation testing retrospectively, greatly improving the efficiency of test result utilization and reducing labor costs. The specific technical solutions of the present invention are as follows:

[0005] A repositioning test method based on a mapping file, the method comprising the following steps: S1: obtaining a mapping file, then processing a point cloud file of the mapping file to obtain processed point cloud data and record a robot pose corresponding to the point cloud data; S2: reading the processed point cloud data and the robot pose corresponding to the point cloud data, and then performing a repositioning test calculation through the mapping file; S3: determining a result of the repositioning test calculation based on a matching score calculated by the repositioning test, the robot pose corresponding to the point cloud data of the mapping file, and a repositioning method; wherein the mapping file comprises a point cloud file and a map file, the point cloud file comprises point cloud data and a robot pose corresponding to the point cloud data, and the map file is used for performing a repositioning test calculation.

[0006] Furthermore, in step S1, obtaining the mapping file includes the following steps: during the cleaning process, the robot records the point cloud data collected by the lidar each time, and records the robot posture corresponding to the point cloud data; after the robot completes cleaning, it records the walking map, and then saves the walking map, point cloud data and the robot posture corresponding to the point cloud data as a mapping file.

[0007] Furthermore, the point cloud file of the mapping file is processed to obtain processed point cloud data and record the robot pose corresponding to the point cloud data. The point cloud file of the mapping file is used to perform a re-run mapping test. During the re-run mapping test, the obtained point cloud data is dedistorted to obtain processed point cloud data and record the robot pose corresponding to the point cloud data.

[0008] Furthermore, in step S2, the acquired point cloud data is dedistorted, including the following steps: during the re-running mapping test, the robot acquires a frame of laser data and the moving speed when acquiring the frame of laser data, and then obtains the distortion value between the first laser point and the last laser point of the frame of laser data by multiplying the moving speed and the acquisition time of the frame of laser data; and supplements the frame of laser data according to the distortion value between the first laser point and the last laser point of the laser data to obtain the point cloud data with the distortion removed.

[0009] Furthermore, in step S2, the processed point cloud data and the robot pose corresponding to the point cloud data are read, and then a repositioning test calculation is performed through the mapping file, including the following steps: the robot uses the distorted point cloud data to traverse the map file of the mapping file to implement the repositioning test calculation and obtain the matching score in the repositioning test calculation.

[0010] Furthermore, in step S3, the result of the repositioning test calculation is determined based on the repositioning result, the robot pose corresponding to the point cloud data of the mapping file, and the repositioning method, including the following steps: if the robot performs forward repositioning, the matching score in the repositioning test calculation is compared with the score threshold; if the matching score in the repositioning test calculation is greater than the score threshold, a pose calibration is performed; if the matching score in the repositioning test calculation is less than or equal to the score threshold, the forward repositioning is judged to have failed; when performing the pose calibration, the robot pose corresponding to the point cloud data in the repositioning test calculation process is compared with the robot pose corresponding to the point cloud data in the mapping file; if the comparison value is less than the set threshold, the forward repositioning is judged to have succeeded; if the comparison value is greater than or equal to the set threshold, the forward repositioning is judged to have failed.

[0011] Furthermore, in step S3, the repositioning result is determined based on the matching score calculated by the repositioning test, the robot posture corresponding to the point cloud data of the mapping file, and the repositioning method, including the following steps: if the robot performs reverse repositioning, the matching score calculated in the repositioning test is compared with the score threshold; if the matching score calculated in the repositioning test is greater than the score threshold, the reverse repositioning is judged to have failed; if the matching score calculated in the repositioning test is less than or equal to the score threshold, the reverse repositioning is judged to have succeeded.

[0012] Furthermore, if it is determined that the relocalization fails, the robot records the point cloud file in the mapping file.

[0013] A chip is used to store a program, and the program is configured to execute the above-mentioned relocation test method based on a mapping file.

[0014] A robot is equipped with a main control chip, which is the above-mentioned chip. The robot is provided with a laser radar for acquiring point cloud data.

[0015] Compared with the existing technology, the technical solution of the present application performs repositioning tests by using the point cloud data obtained during the cleaning process and re-running the cleaned map to build the map, which greatly improves the efficiency of using the repositioning test results and reduces labor costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 This is a flow chart of a relocation test method based on a mapping file according to an embodiment of the present invention. DETAILED DESCRIPTION

[0017] The technical solutions in the embodiments of the present invention will be described in detail below in conjunction with the accompanying drawings in the embodiments of the present invention. It should be understood that the specific embodiments described below are only used to explain the present invention and are not used to limit the present invention.

[0018] like Figure 1 As shown, a relocalization test method based on mapping files is mainly centered around relocalization testing. The laser file with time sequence recorded by mapping is used to backtrack and perform relocalization testing on the entire map. This can improve the effect of each relocalization test, allow the relocalization algorithm to be tested more, and reduce labor costs to a certain extent. Among them, forward relocalization is when the robot uses the current environment map for relocalization. At this time, the robot needs to relocalize successfully to match the correct position; reverse relocalization is when the robot uses another environment map for relocalization. At this time, the robot needs to relocalize and re-build the map and clean it if the relocalization fails. The method includes the following steps:

[0019] Step S1: The robot obtains a mapping file, wherein the mapping file includes a point cloud file and a map file. The point cloud file includes point cloud data and the robot pose corresponding to the point cloud data. The map file is used for relocalization test calculations. The point cloud file of the mapping file is then processed to obtain the processed point cloud data and record the robot pose corresponding to the point cloud data. Obtaining the mapping file includes the following steps: during the cleaning process, the robot will record the point cloud data collected by the lidar each time and record the robot pose corresponding to the point cloud data; after completing the cleaning, the robot records the walking map and then saves the walking map, point cloud data, and the robot pose corresponding to the point cloud data as a mapping file. The mapping file includes a point cloud file and a map file. The point cloud file is used to rerun the mapping test, and the map file is used for relocalization test calculations, that is, used as a matching map for relocalization matching calculations. The mapping file also includes IMU data and other data used by the robot to construct the map. Re-running mapping means that the robot uses a processor or computer to run again to build a map based on the data in the mapping file, and records the point cloud data obtained during the run and the robot posture corresponding to the point cloud data. The data in the mapping file also includes laser data and IMU data.

[0020] As one of the embodiments, in step S1, the robot processes the point cloud file of the mapping file to obtain processed point cloud data and records the robot pose corresponding to the point cloud data, that is, the point cloud file of the mapping file is used to perform a re-run mapping test. During the re-run mapping test, the obtained point cloud data is de-distorted to obtain the processed point cloud data and record the robot pose corresponding to the point cloud data, that is, the real-time laser data in the robot mapping file is recorded in a file, and a simulation test is used to perform a re-run mapping test. Then, laser data with a format different from that of the real-time laser data is generated, and the generated laser data is de-distorted. Then, the de-distorted laser data and the robot pose corresponding to the laser data are output, and recording the robot pose corresponding to the de-distorted laser data is to save the robot pose corresponding to the real-time laser data in the mapping file and the de-distorted laser data output by the re-run mapping test in a one-to-one correspondence. The obtained point cloud data is dedistorted, including the following steps: during the re-running mapping test, the robot obtains a frame of laser data and the movement speed when obtaining the frame of laser data, and then obtains the distortion value between the first laser point and the last laser point of the frame of laser data by multiplying the movement speed and the acquisition time of the frame of laser data; the frame of laser data is supplemented according to the distortion value between the first laser point and the last laser point of the laser data to obtain the point cloud data with the distortion removed. The above method is to dedistort by speed, and dedistortion can also be performed by acceleration and other methods. When the robot obtains a frame of data while standing still, this frame of data will not be distorted. However, in actual cleaning, the robot is inevitably in motion, and the acquisition of laser data requires a certain amount of time, so the laser data obtained by the robot during movement is distorted. The main purpose of rerunning the map is to dedistort the real-time laser data. During map construction, distorted point clouds are typically recorded during cleaning to meet real-time requirements. If the robot has dedistorted laser data and robot pose, relocalization test calculations can be performed directly without rerunning the map. Although the robot can dedistort the laser data during mapping, it does not record the dedistorted laser data and robot pose. Therefore, data processing methods are required to dedistort the laser data to obtain the dedistorted laser data. Currently, the robot only writes undistorted laser data during mapping, and some tests and machine runs before mapping also only record undistorted laser data. The robot should not write out the dedistorted laser data in subsequent operations, as writing multiple files can cause other problems, such as delayed sensor data processing. To reuse previous tests, rerunning the map to dedistort the laser data is necessary.There are various methods for dedistorting laser data. The robot can also use other methods to remove distortion in laser data, or save the dedistorted laser data when the robot builds the map. The method or steps for dedistorting laser data can be selected according to actual needs and are not the only one.

[0021] Step S2: Read the processed point cloud data and the robot pose corresponding to the point cloud data, and then perform a relocalization test calculation using the mapping file. In step S2, the dedistorted point cloud data and the robot pose corresponding to the point cloud data are read to perform a relocalization test calculation using the mapping file. This process includes the following steps: the robot uses the dedistorted point cloud data to traverse the map file in the mapping file to perform the relocalization test calculation and obtain a matching score for the relocalization test calculation. During the relocalization process, the robot uses the dedistorted point cloud data to traverse the map file in the pre-stored mapping file. This traversal process is a map matching process. Each match produces a matching result containing three parameters: matching score, matching area, and matching area ratio. The matching score is the percentage of overlap between a point on the point cloud map and an obstacle at the corresponding matching location on the global laser grid map. For example, if a point on the point cloud map completely overlaps with the obstacle at the corresponding matching location, the matching score is 1; if there is 60% overlap, the matching score is 0.6. During cleaning, distorted point cloud data is recorded for real-time performance, and the point cloud files obtained in the saved mapping files are also distorted point cloud data. In actual applications, since the point cloud files are dedistorted during mapping, relocalization matching with the distorted point cloud may not produce a match. Therefore, the dedistorted point cloud data is used for relocalization matching. The robot pose corresponding to the point cloud is obtained to verify the relocalization results.

[0022] Step S3: The robot determines the relocalization result based on the matching score calculated by the relocalization test, the robot pose corresponding to the point cloud data in the mapping file, and the relocalization method. In step S3, the relocalization result is determined based on the relocalization result, the robot pose corresponding to the point cloud data in the point cloud file of the mapping file, and the relocalization method, including the following steps: if the robot is performing forward relocalization, the matching score calculated in the relocalization test is compared with the score threshold; if the matching score calculated in the relocalization test is greater than the score threshold, a pose calibration is performed; if the matching score calculated in the relocalization test is less than or equal to the score threshold, the forward relocalization is judged to have failed; when performing the pose calibration, the robot pose corresponding to the point cloud data in the relocalization test calculation process is compared with the robot pose corresponding to the point cloud data in the mapping file; if the comparison value is less than the set threshold, the forward relocalization is judged to have succeeded; if the comparison value is greater than or equal to the set threshold, the forward relocalization is judged to have failed. In step S3, the relocalization result is determined based on the relocalization result, the robot pose corresponding to the point cloud data during the rerun of the mapping test, and the relocalization method. The steps include the following: if the robot performs reverse relocalization, the matching score calculated in the relocalization test is compared with the score threshold; if the matching score calculated in the relocalization test is greater than the score threshold, the reverse relocalization is judged to have failed; if the matching score calculated in the relocalization test is less than or equal to the score threshold, the reverse relocalization is judged to have succeeded. If the relocalization is judged to have failed, the robot records the point cloud file in the mapping file.

[0023] A chip is used to store a program, and the program is configured to execute the above-mentioned relocation test method based on a mapping file.

[0024] A robot is equipped with a main control chip, which is the above-mentioned chip. The robot is provided with a laser radar for acquiring point cloud data.

[0025] Compared with the existing technology, the technical solution of the present application performs repositioning tests by using the point cloud data obtained during the cleaning process and re-running the cleaned map to build the map, which greatly improves the efficiency of using the repositioning test results and reduces labor costs.

[0026] Obviously, the above-mentioned embodiments are only some embodiments of the present invention, rather than all embodiments, and the technical solutions between the various embodiments can be combined with each other. In addition, if the terms "center", "up", "down", "left", "right", "vertical", "horizontal", "inside", "outside" and the like appear in the embodiments, the orientation or position relationship indicated is based on the orientation or position relationship shown in the accompanying drawings, which is only for the convenience of describing the present invention and simplifying the description, and does not indicate or imply that the device or element referred to must have a specific orientation or be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present invention. If the terms "first", "second", "third" and the like appear in the embodiments, it is to facilitate the distinction between related features and cannot be understood as indicating or implying their relative importance, order or number of technical features.

[0027] Those skilled in the art will appreciate that all or part of the steps in the above-described method embodiments can be implemented using hardware associated with program instructions. These programs can be stored in a computer-readable storage medium (e.g., ROM, RAM, magnetic disk, optical disk, or other medium capable of storing program code). When executed, the program performs the steps of the above-described method embodiments.

[0028] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the above embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A relocation test method based on a mapping file, characterized in that: The method comprises the following steps: S1: Obtain a mapping file, then process the point cloud file of the mapping file to obtain processed point cloud data and record the robot pose corresponding to the point cloud data; S2: Read the processed point cloud data and the robot pose corresponding to the point cloud data, and then perform relocalization test calculations using the mapping file; S3: Determine the result of the relocalization test calculation based on the matching score calculated by the relocalization test, the robot pose corresponding to the point cloud data of the mapping file, and the relocalization method; The mapping file includes a point cloud file and a map file. The point cloud file includes point cloud data and the robot pose corresponding to the point cloud data. The map file is used for relocation test calculation. In step S1, the point cloud file of the mapping file is processed to obtain processed point cloud data and record the robot pose corresponding to the point cloud data. The point cloud file of the mapping file is used to rerun the mapping test. During the rerun mapping test, the obtained point cloud data is dedistorted to obtain processed point cloud data and record the robot pose corresponding to the point cloud data. The obtained point cloud data is subjected to dedistortion processing, including the following steps: During the rerun mapping test, the robot acquires a frame of laser data and its movement speed when acquiring the frame. The distortion value between the first and last laser points of the frame is calculated by multiplying the movement speed by the acquisition time. The frame of laser data is supplemented according to the distortion value between the first laser point and the last laser point of the laser data to obtain point cloud data with the distortion removed.

2. The relocation test method based on the mapping file according to claim 1, characterized in that: In step S1, obtaining the mapping file includes the following steps: During the cleaning process, the robot records the point cloud data collected by the lidar each time, and records the robot posture corresponding to the point cloud data; After the robot completes cleaning, it records the walking map and then saves the walking map, point cloud data, and the robot pose corresponding to the point cloud data as a mapping file.

3. The relocation test method based on the mapping file according to claim 1, characterized in that: In step S2, the processed point cloud data and the robot pose corresponding to the point cloud data are read, and then a repositioning test calculation is performed using the mapping file, including the following steps: The robot uses the distorted point cloud data to traverse the map file of the mapping file to implement the relocalization test calculation and obtain the matching score in the relocalization test calculation.

4. The relocation test method based on the mapping file according to claim 3, characterized in that: In step S3, the result of the relocalization test calculation is determined based on the matching score calculated by the relocalization test, the robot pose corresponding to the point cloud data of the mapping file, and the relocalization method, including the following steps: If the robot is performing forward relocalization, the matching score calculated in the relocalization test is compared with the score threshold; If the matching score calculated in the relocalization test is greater than the score threshold, a posture check is performed. If the matching score calculated in the relocalization test is less than or equal to the score threshold, the forward relocalization is judged to have failed. During pose calibration, the robot pose corresponding to the point cloud data in the relocalization test calculation process is compared with the robot pose corresponding to the point cloud data in the mapping file. If the comparison value is less than the set threshold, the forward relocalization is judged to be successful. If the comparison value is greater than or equal to the set threshold, the forward relocalization is judged to have failed.

5. The relocation test method based on the mapping file according to claim 4 is characterized in that: In step S3, the relocation result is determined based on the relocation result, the robot pose corresponding to the point cloud data of the mapping file, and the relocation method, including the following steps: If the robot performs reverse relocation, the matching score calculated in the relocation test is compared with the score threshold; If the matching score in the relocation test calculation is greater than the score threshold, the reverse relocation is judged to have failed. If the matching score in the relocation test calculation is less than or equal to the score threshold, the reverse relocation is judged to have succeeded.

6. The relocation test method based on the mapping file according to claim 5, characterized in that: If relocalization fails, the robot records the point cloud file in the mapping file.

7. A chip for storing a program, characterized in that: The program is configured to execute the mapping file-based relocation test method according to any one of claims 1 to 6.

8. A robot equipped with a main control chip, characterized in that: The main control chip is the chip according to claim 7, and the robot is provided with a laser radar for acquiring point cloud data.

Images