A mapping method of a laser navigation robot and a laser navigation robot

Abstract

The application discloses a kind of mapping method and laser navigation robot of laser navigation robot, specifically includes: laser navigation robot is continuously acquired and records laser point cloud data in the process of constructing map, when the frame number of recorded laser point cloud data reaches preset frame number, straight line fitting is carried out based on recorded all laser point cloud data and records straight line;Repeat the above steps, until laser navigation robot map construction is completed, based on the optimal straight line that recorded all straight lines are obtained, the map is optimized and handled based on the optimal straight line, obtains the map after leveling optimization processing, the map of laser navigation robot is updated as the map after leveling optimization processing.The application is used for straight line fitting by real-time acquisition and record laser point cloud, greatly improve the reliability based on fitting straight line leveling, improve map leveling effect.

Description

Technical Field

[0001] This invention relates to the field of map building, specifically to a mapping method for a laser-guided robot and the laser-guided robot itself. Background Technology

[0002] With the maturity of robotics technology, robot mapping methods are becoming increasingly diverse. Currently, robot mapping methods mainly utilize LiDAR, visual sensors, and / or inertial navigation systems. Among these, robots using LiDAR for mapping typically display long, straight obstacles in the environment as straight lines on the map to enhance its visual appeal. This process is called mapping leveling. Currently, there are two main methods for mapping and leveling laser navigation robots. The first method involves acquiring a certain number of laser point cloud frames and fitting them with straight lines during the initial mapping phase. One line is selected from the fitted lines as the leveling direction, and the robot is controlled to map according to this direction. This method suffers from the drawback of insufficient laser point cloud data, resulting in unreliable fitted lines. This can lead to the robot's inability to accurately level itself if the environment becomes more complex later in the mapping process. Furthermore, even if the environment remains relatively stable during mapping, vibrations or other issues during the mapping process can cause the constructed map to become skewed. The second method involves extracting an obstacle fitting line from the constructed map after mapping is complete, and then controlling the map to level itself according to this direction. This method suffers from the problem of an unsightly, skewed map due to the lack of a leveling direction, and the extracted obstacle fitting line has poor accuracy, resulting in poor map leveling performance. Summary of the Invention

[0003] To address the aforementioned problems, this invention provides a mapping method for a laser-guided robot and a laser-guided robot in general. By real-time acquisition and recording of laser point clouds for straight line fitting, the reliability of leveling based on fitted straight lines is significantly improved, thus enhancing the map leveling effect. The specific technical solution of this invention is as follows:

[0004] A mapping method for a laser-guided robot specifically includes: the laser-guided robot continuously acquiring and recording laser point cloud data during the map construction process; when the number of frames of recorded laser point cloud data reaches a preset number of frames, straight line fitting is performed based on all recorded laser point cloud data and the straight line is recorded; the above steps are repeated until the laser-guided robot completes map construction; the optimal straight line is obtained based on all recorded straight lines; the map is leveled and optimized based on the optimal straight line; the leveled and optimized map is obtained; and the map of the laser-guided robot is updated to the leveled and optimized map.

[0005] Furthermore, the map construction method for the laser navigation robot also includes: after performing straight line fitting based on all recorded laser point cloud data and recording the straight line, deleting all recorded laser point cloud data and starting to record laser point cloud data again.

[0006] Furthermore, the map construction method of the laser navigation robot also includes: when the number of frames of the recorded laser point cloud data reaches a preset number of frames, firstly, downsampling is performed on all the recorded laser point cloud data, and then straight line fitting is performed based on the downsampled laser point cloud data and the straight line is recorded.

[0007] Furthermore, the method for fitting a straight line based on all recorded laser point cloud data and recording the straight line specifically includes: averaging all recorded laser point cloud data to obtain average laser point cloud data; and substituting the average laser point cloud data into the least squares formula in the form of parameters to obtain the fitted straight line.

[0008] Furthermore, the method for obtaining the optimal line based on all recorded lines specifically includes: calculating the total length of each line corresponding to all recorded lines; and selecting a line from all recorded lines that meets the total length requirement as the optimal line.

[0009] Further, the method for calculating the total length of all recorded lines specifically includes: selecting a line from all recorded lines whose total length has not been calculated as the current reference line, and obtaining the length of the current reference line; projecting all recorded lines except the current reference line onto the direction of the current reference line, obtaining the first projection value corresponding to each of the recorded lines except the current reference line; projecting all recorded lines except the current reference line onto a direction at 90° to the current reference line, obtaining the second projection value corresponding to each of the recorded lines except the current reference line; taking the larger of the first and second projection values ​​corresponding to the same line and multiplying it by a preset parameter as the third projection value of the line, obtaining the third projection value corresponding to each of the recorded lines except the current reference line; calculating the sum of the length of the current reference line and the third projection value corresponding to each of the recorded lines except the current reference line as the total length of the current reference line; repeating the above steps until the total length of all recorded lines is calculated.

[0010] Furthermore, if the total length requirement of the straight line is configured to require the total length of the straight line to be the maximum value, then the method of selecting a straight line that satisfies the total length requirement from all recorded straight lines as the optimal straight line specifically includes: selecting the straight line with the largest total length from all recorded straight lines as the optimal straight line.

[0011] Furthermore, the method for performing leveling optimization processing on the map based on the optimal straight line to obtain the leveled and optimized map specifically includes: obtaining the angle between the optimal straight line and the map constructed by the laser navigation robot; using the angle between the optimal straight line and the map constructed by the laser navigation robot as the rotation angle of the map, and obtaining the map rotated according to the rotation angle as the leveled and optimized map.

[0012] Furthermore, the mapping method for the laser navigation robot also includes: updating the positioning coordinates of the laser navigation robot based on the map after leveling and optimization, and obtaining the updated positioning coordinates of the laser navigation robot.

[0013] The present invention also discloses a laser navigation robot, which performs the mapping method of the laser navigation robot as described above.

[0014] This invention continuously acquires laser point cloud data and fits straight lines during the map-building process of a laser-guided robot. It continuously fits straight lines based on the real-time environment of the laser-guided robot. By comprehensively considering and comparing all recorded straight lines, it selects the optimal straight line as the straight direction, ensuring the reliability of leveling based on the optimal straight line. This results in a beautiful and straight map after leveling and optimization. Attached Figure Description

[0015] Figure 1 This is a schematic flowchart of a mapping method for a laser navigation robot according to an embodiment of the present invention. Detailed Implementation

[0016] To make the objectives, technical solutions, and advantages of this application clearer, the application is described and illustrated below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application. All other embodiments obtained by those skilled in the art based on the embodiments provided in this application without inventive effort are within the scope of protection of this application.

[0017] Unless otherwise defined, the technical or scientific terms used in this application shall have the ordinary meaning understood by one of ordinary skill in the art to which this application pertains. The terms “a,” “an,” “an,” “the,” and similar words used in this application do not indicate quantity limitation and may represent singular or plural. The terms “comprising,” “including,” “having,” and any variations thereof used in this application are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or apparatus that includes a series of steps or modules is not limited to the listed steps or units but may also include steps or units not listed, or may include other steps or units inherent to such processes, methods, products, or apparatus. The terms “first,” “second,” and “third” used in this application are merely to distinguish similar objects and do not represent a specific ordering of objects.

[0018] As a preferred embodiment of the present invention, the first embodiment provides a mapping method for a laser-guided robot, such as... Figure 1 As shown, the mapping method for the laser-guided robot specifically includes the following steps:

[0019] The laser navigation robot continuously acquires and records laser point cloud data during the map building process; specifically, the laser navigation robot refers to an intelligent robot that uses laser radar for positioning and navigation; the laser point cloud data is acquired by the laser radar mounted on the laser navigation robot.

[0020] The system determines whether the number of frames of the currently recorded laser point cloud data has reached the preset number of frames. Specifically, the preset number of frames is set by the laser navigation robot before it leaves the factory based on the user's requirements for mapping and leveling optimization accuracy. Within a reasonable range, the number of preset frames is negatively correlated with the mapping and leveling optimization accuracy. Within a reasonable range, the fewer the preset frames, the more straight lines the laser navigation robot fits during the mapping process, the better the straightness of the selected optimal straight line, and the higher the leveling optimization accuracy. At the same time, when the number of preset frames is less than the reasonable range, the amount of recorded laser point cloud data is insufficient to fit a straight line that meets the accuracy requirements.

[0021] If the number of frames of laser point cloud data currently recorded has not reached the preset number of frames, the acquisition and recording of laser point cloud data will continue. Specifically, if the number of frames of laser point cloud data currently recorded has not reached the preset number of frames, it means that the amount of laser point cloud data currently recorded is insufficient to fit a straight line that meets the accuracy requirements. Therefore, the laser navigation robot is controlled to continue to acquire and record laser point cloud data until the number of frames of laser point cloud data recorded reaches the preset number of frames.

[0022] If the number of frames of the currently recorded laser point cloud data reaches the preset number of frames, then all the currently recorded laser point cloud data will be fitted with a straight line to obtain and record the straight line. Specifically, if the number of frames of the currently recorded laser point cloud data reaches the preset number of frames, it means that the amount of data of the currently recorded laser point cloud is sufficient to fit a straight line that meets the accuracy requirements. Therefore, the laser navigation robot is controlled to fit a straight line based on all the currently recorded laser point cloud data to obtain and record the fitted straight line.

[0023] Repeat the above steps until the laser navigation robot's map construction is complete. Integrate and calculate all recorded straight lines to obtain the optimal straight line. Based on the optimal straight line, perform leveling and optimization processing on the map to obtain the leveled and optimized map. Update the laser navigation robot's map with the leveled and optimized map. Specifically, by continuously acquiring laser point cloud data and fitting straight lines during the laser navigation robot's map construction process, the system continuously fits straight lines according to the real-time environment of the laser navigation robot. By comprehensively considering and comparing all recorded straight lines, the optimal straight line is selected as the straight direction, ensuring the reliability of leveling based on the optimal straight line and making the leveled and optimized map aesthetically pleasing and straight.

[0024] Based on the first embodiment described above, as a preferred embodiment of the present invention, the map construction method for the laser navigation robot in the second embodiment further includes: after performing line fitting based on all recorded laser point cloud data and recording the line, deleting all recorded laser point cloud data and restarting the recording of laser point cloud data. Specifically, when all recorded laser point cloud data is used for line fitting and the fitted line is recorded, it is considered that all currently recorded laser point cloud data has been used. In order to reduce the memory space occupied by the laser navigation robot, all laser point cloud data that has been fitted into a line is deleted, releasing the memory space occupied by the recorded laser data and improving the memory space utilization rate of the laser navigation robot.

[0025] Based on the above embodiments, as a preferred embodiment of the present invention, the map construction method of the laser navigation robot in the fourth embodiment of the present invention further includes: when the number of frames of recorded laser point cloud data reaches a preset number of frames, firstly, downsampling processing is performed on all recorded laser point cloud data, and then straight line fitting is performed based on the downsampled laser point cloud data and the straight line is recorded. Specifically, the downsampling processing refers to selecting a portion of the laser point cloud data from all recorded laser point cloud data; it should be noted that the downsampling processing method may be, but is not limited to, using a voxel filter for downsampling; the voxel filter is used to filter out noise data and outlier data in all recorded laser point cloud data without destroying the laser point cloud data. The purpose of downsampling all recorded laser point cloud data in this embodiment is to select a portion of the laser point cloud data from all recorded laser point cloud data to improve the accuracy of the fitted straight line, thereby improving the reliability of the fitted straight line.

[0026] Based on the above embodiments, as a preferred embodiment of the present invention, the method for fitting a straight line based on all recorded laser point cloud data and recording the straight line in the fourth embodiment of the present invention specifically includes: averaging all recorded laser point cloud data to obtain average laser point cloud data; and substituting the average laser point cloud data into the least squares formula in the form of parameters to obtain the fitted straight line. Specifically, averaging all recorded laser point cloud data means dividing the sum of all laser point cloud data by the number of frames of all laser point cloud data. This embodiment reduces the impact of sudden events such as environmental changes, occlusion, and positioning deviations on laser point cloud data by averaging all recorded laser point cloud data.

[0027] Based on the above embodiments, as a preferred embodiment of the present invention, in the fifth embodiment, the method for obtaining the optimal line based on all recorded straight lines specifically includes: calculating the total length of each recorded straight line; and selecting a straight line from all recorded straight lines that meets the total length requirement as the optimal line. Specifically, each straight line has a one-to-one corresponding total length, which is not the length of the line itself, but rather the length based on the relationship between the line's length and the remaining recorded straight lines. In this embodiment, the total length is used as a measure of the smoothness and straightness of each straight line. It should be noted that the total length requirement can be, but is not limited to, a limiting requirement for selecting a smoother and straighter fitted line. The total length requirement can be adjusted according to the complexity of the actual application scenario or the actual robot model. This embodiment compares the total lengths of all recorded straight lines to select a straight line that meets the total length requirement as the optimal line, thereby ensuring the leveling effect of the map leveling process based on this optimal line.

[0028] Based on the above embodiments, as a preferred embodiment of the present invention, the method for calculating the total length of all recorded straight lines in the sixth embodiment of the present invention specifically includes:

[0029] Select a line from all recorded lines whose total length has not been calculated as the current reference line, and obtain the length of the current reference line; specifically, obtaining the length of the current reference line means obtaining the mathematical definition of the length of the current reference line, and the method of obtaining the length of the current reference line can be, but is not limited to, obtaining the coordinates of the two endpoints of the current reference line by calculation.

[0030] Project all recorded lines except the current reference line onto the direction of the current reference line to obtain the first projection value corresponding to each of the recorded lines except the current reference line; specifically, if there are N recorded lines, project the N-1 recorded lines except the current reference line onto the direction of the current reference line to obtain N-1 first projection values ​​corresponding to each of the N-1 lines.

[0031] Project all recorded lines except the current reference line in a direction 90° to the current reference line to obtain the second projection value corresponding to each of the recorded lines except the current reference line; specifically, if there are N recorded lines, then project the N-1 recorded lines except the current reference line in a direction to the current reference line to obtain the N-1 second projection values ​​corresponding to each of the N-1 lines.

[0032] The larger of the first and second projection values ​​corresponding to the same straight line is multiplied by a preset parameter to obtain the third projection value of that straight line. This process is repeated for all recorded straight lines except the current reference line. Specifically, this step selects the larger of the first and second projection values ​​for each straight line and multiplies it by a preset parameter, using the product as the third projection value of that line. The preset parameter is a value adjusted based on the actual environmental complexity and the actual laser navigation robot model to limit the relationship between the first, second, and third projection values.

[0033] The sum of the third projection values ​​corresponding to the current reference line length and all other recorded lines (excluding the current reference line) is calculated as the total length of the current reference line. This process is repeated until the total length of each recorded line is calculated. This embodiment compares all fitted lines in the calculation records and calculates the total length of each line to assess the reliability of leveling based on that line.

[0034] Based on the above embodiments, in a preferred embodiment of the present invention, the total length requirement of the straight lines is configured to require the total length of the straight lines to be the maximum value. Therefore, the method of selecting a straight line from all recorded straight lines that meets the total length requirement as the optimal straight line specifically includes: selecting the straight line with the largest total length from all recorded straight lines as the optimal straight line. Specifically, the total length requirement of the straight lines can be set according to the actual application environment of the laser navigation robot and the user's accuracy requirements for the optimal straight line.

[0035] Based on the above embodiments, in a preferred embodiment of the present invention, the method for performing leveling optimization processing on the map based on the optimal straight line to obtain the leveled and optimized map specifically includes: obtaining the angle between the optimal straight line and the map constructed by the laser navigation robot; using the angle between the optimal straight line and the map constructed by the laser navigation robot as the rotation angle of the map, and obtaining the map rotated according to the rotation angle as the leveled and optimized map. Specifically, the method of rotating the map according to the rotation angle can be, but is not limited to, using algorithms with image processing capabilities such as OpenCV. This embodiment achieves map leveling optimization by rotating the map based on the optimal straight line, solving the problem of map skew caused by vibration during the mapping process of the laser navigation robot, while ensuring map accuracy and improving the straightness and visibility of straight lines in the map.

[0036] Based on the above embodiments, as a preferred embodiment of the present invention, in the ninth embodiment, the mapping method for the laser navigation robot further includes: updating the positioning coordinates of the laser navigation robot based on the leveled and optimized map, and obtaining the updated positioning coordinates of the laser navigation robot. Specifically, the method for obtaining the updated positioning coordinates of the laser navigation robot can be, but is not limited to, using an algorithm with image processing capabilities such as OpenCV based on the leveled and optimized map. This embodiment ensures the accuracy and reliability of map-based positioning by updating the positioning coordinates of the laser navigation robot according to the leveled and optimized map.

[0037] Based on the above embodiments, as a preferred embodiment of the present invention, the tenth embodiment of the present invention provides a laser navigation robot, wherein the laser navigation robot executes the mapping method of the laser navigation robot as described in the foregoing embodiments during the mapping process.

[0038] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. This computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. References to memory, storage, databases, or other media used in the embodiments provided in this application can all include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable memory (PROM), electrically programmable memory (EPROM), electrically erasable programmable memory (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory.

[0039] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0040] The above embodiments are merely illustrative of several implementations of the present invention, and their descriptions are relatively specific and detailed, but they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application.

Claims

1. A mapping method for a laser-guided robot, characterized in that, Specifically, it includes: During the map building process, the laser navigation robot continuously acquires and records laser point cloud data. When the number of frames of recorded laser point cloud data reaches the preset number of frames, it performs line fitting based on all the recorded laser point cloud data and records the line. Repeat the above steps until the laser navigation robot map is built. Based on all the recorded straight lines, obtain the optimal straight line, perform leveling and optimization processing on the map based on the optimal straight line, obtain the leveled and optimized map, and update the laser navigation robot map to the leveled and optimized map. The method for obtaining the optimal line based on all recorded lines specifically includes: calculating the total length of each line corresponding to all recorded lines; and selecting a line from all recorded lines that meets the total length requirement as the optimal line. The method for calculating the total length of all lines corresponding to the recorded lines specifically includes: Select one line from all recorded lines whose total length has not been calculated as the current reference line, and obtain the length of the current reference line; Project all the recorded lines except the current reference line onto the direction of the current reference line to obtain the first projection value corresponding to each of the recorded lines except the current reference line. Project all the recorded lines except the current reference line onto the direction at 90° with the current reference line to obtain the second projection value corresponding to each of the recorded lines except the current reference line. The larger of the first and second projection values ​​corresponding to the same straight line and the product of the preset parameter is taken as the third projection value of the straight line. The third projection values ​​corresponding to all the recorded straight lines except the current reference straight line are obtained one by one. The total length of the current reference line is calculated by summing the third projection values ​​corresponding one-to-one with the length of the current reference line and all the lines in the records other than the current reference line. Repeat the above steps until the total length of the line corresponding to all the recorded lines has been calculated.

2. The mapping method for a laser-guided robot according to claim 1, characterized in that, The map construction method for the laser navigation robot further includes: performing straight line fitting based on all recorded laser point cloud data and recording the straight line, then deleting all recorded laser point cloud data and starting to record laser point cloud data again.

3. The mapping method for a laser-guided robot according to claim 1, characterized in that, The map construction method of the laser navigation robot further includes: when the number of frames of the recorded laser point cloud data reaches a preset number of frames, first downsampling is performed on all the recorded laser point cloud data, and then straight line fitting is performed based on the downsampled laser point cloud data and the straight line is recorded.

4. The mapping method for a laser-guided robot according to claim 3, characterized in that, The method for fitting a straight line based on all recorded laser point cloud data and recording the straight line specifically includes: averaging all recorded laser point cloud data to obtain average laser point cloud data; and substituting the average laser point cloud data into the least squares formula as parameters to obtain the fitted straight line.

5. The mapping method for a laser-guided robot according to claim 1, characterized in that, The requirement for the total length of the line is configured to require the total length of the line to be the maximum value. The method of selecting a line that meets the requirement for the total length of the line from all the recorded lines as the optimal line specifically includes: selecting the line with the largest total length from all the recorded lines as the optimal line.

6. The mapping method for a laser-guided robot according to claim 1, characterized in that, The method for performing leveling and optimization processing on the map based on the optimal straight line to obtain the leveled and optimized map specifically includes: obtaining the angle between the optimal straight line and the map constructed by the laser navigation robot; using the angle between the optimal straight line and the map constructed by the laser navigation robot as the rotation angle of the map, and obtaining the map after rotating according to the rotation angle as the leveled and optimized map.

7. The mapping method for a laser-guided robot according to claim 6, characterized in that, The mapping method for the laser navigation robot further includes: updating the positioning coordinates of the laser navigation robot based on the map after leveling and optimization, and obtaining the updated positioning coordinates of the laser navigation robot.

8. A laser-guided robot, characterized in that, The laser navigation robot performs the mapping method for laser navigation robots as described in any one of claims 1 to 7.

Images