Tiling pose calculation method and device based on grid fitting and tiling robot

Abstract

The application discloses a kind of based on grid fitting's brick laying pose calculation method, device and brick laying robot, the method includes: the first corner point coordinate information of multiple laid bricks of target brick laying area is determined;According to the preset brick grid optimization algorithm, and the first corner point coordinate information of multiple laid bricks, the corresponding brick grid relationship model of the target brick laying area is calculated;According to the brick grid relationship model, and the brick information of target to be laid brick, the second corner point coordinate information of the target to be laid brick is determined;According to the second corner point coordinate information of the target to be laid brick, the laying pose information of the target to be laid brick is determined;The laying pose information is used to indicate that the target to be laid brick is laid by brick laying equipment.It can be seen that the application can accurately calculate the coordinate information of to-be-laid brick according to the corner point coordinates of laid brick, to improve the accuracy and efficiency of subsequent brick laying.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data processing, and particularly relates to a brick laying pose calculation method and device based on grid fitting and a brick laying robot. BACKGROUND

[0002] With the development of construction technology and the explosion of engineering construction demand, more and more engineering construction begins to take intelligent means, for example, the work of laying bricks has begun to rely on brick laying equipment. However, the existing brick laying technology still generally adopts manual operation, and adopts methods or devices such as drawing a line on the ground with ink, customizing a metal grid, using a laser level or manual visual inspection to ensure the alignment of the laid bricks. The brick laying effect mainly depends on the experience and carefulness of workers, and the efficiency is low and the cost is high. Some brick laying technologies using brick laying equipment do not consider using the coordinates of the laid bricks to automatically calculate the coordinates of the to-be-laid bricks, so they cannot achieve more accurate brick laying effect. It can be seen that the existing technology has defects and needs to be solved. SUMMARY

[0003] The technical problem to be solved by the present application is to provide a brick laying pose calculation method and device based on grid fitting and a brick laying robot, which can accurately calculate the coordinate information of the to-be-laid bricks according to the corner point coordinates of the laid bricks, so as to improve the accuracy and efficiency of subsequent brick laying.

[0004] To solve the above technical problem, the present application discloses a brick laying pose calculation method based on grid fitting in the first aspect, which comprises:

[0005] determining first corner point coordinate information of a plurality of laid bricks in a target brick laying area;

[0006] calculating a brick grid relationship model corresponding to the target brick laying area according to a preset brick grid optimization algorithm and the first corner point coordinate information of the plurality of laid bricks;

[0007] determining second corner point coordinate information of a target to-be-laid brick according to the brick grid relationship model and brick information of the target to-be-laid brick;

[0008] determining laying pose information of the target to-be-laid brick according to the second corner point coordinate information of the target to-be-laid brick; the laying pose information is used to instruct a brick laying device to lay the target to-be-laid brick.

[0009] As an optional implementation manner, in the first aspect of the present application, the calculating of the brick grid relationship model corresponding to the target brick laying area according to the preset brick grid optimization algorithm and the first corner point coordinate information of the plurality of laid bricks comprises:

[0010] According to the first corner point coordinate information of the plurality of laid bricks and a preset brick grid optimization model, model parameters corresponding to the brick grid optimization model are obtained through fitting optimization calculation; the brick grid optimization model is used to limit the corner point parameter relationship between a plurality of virtual bricks of the same size arranged in a grid.

[0011] According to the model parameters, a brick grid relationship model corresponding to the target tiling area is determined.

[0012] As an optional implementation, in the first aspect of the present application, the brick grid optimization model includes a first brick grid optimization model and a second brick grid optimization model; and according to the first corner point coordinate information of the plurality of laid bricks and a preset brick grid optimization model, the model parameters corresponding to the brick grid optimization model are obtained through fitting optimization calculation, which includes:

[0013] According to the first corner point coordinate information of the plurality of laid bricks and a preset first brick grid optimization model, optimal grid slope parameters corresponding to the first brick grid optimization model are obtained through fitting optimization calculation; the first brick grid optimization model is used to limit the corner point parameter relationship between a plurality of virtual bricks of the same size but different intervals arranged in a grid; the optimal grid slope parameters are the optimization calculation values of the slope parameters in the linear relationship equations of different edges of the virtual bricks defined by the first brick grid optimization model;

[0014] The first corner point coordinate information of the plurality of laid bricks and the optimal grid slope parameters are input into a preset second brick grid optimization model, and model parameters corresponding to the second brick grid optimization model are obtained through fitting optimization calculation; the second brick grid optimization model is used to limit the corner point parameter relationship between a plurality of virtual bricks of the same size and same interval arranged in a grid; the slope parameters in the linear relationship equations of different edges of the virtual bricks defined in the second brick grid optimization model are the same as those in the first brick grid optimization model.

[0015] As an optional implementation, in the first aspect of the present application, the fitting optimization calculation is an optimization calculation based on least squares fitting; and / or, in the process of fitting optimization, the corner point coordinates with residual values greater than a preset standard deviation threshold value are deleted; the standard deviation threshold value is a preset multiple of the standard deviation of the residual values corresponding to all corner point coordinates; and the preset multiple is less than 1.

[0016] As an optional implementation, in the first aspect of the present application, the fitting optimization calculation is a fitting optimization calculation with a weight factor; the weight factor includes a center axis distance weight factor and / or a squareness weight factor; the center axis distance weight factor is proportional to the distance between each corner point coordinate information and the center of the target tiling area; the squareness weight factor is inversely proportional to the fitting residual between each corner point coordinate information and the corresponding corner point of the standard square in which it is located.

[0017] As an optional implementation, in the first aspect of the present application, the determination of the first corner point coordinate information of the plurality of laid tiles of the target tiling area includes:

[0018] Obtaining three-dimensional image information of the plurality of laid tiles of the target tiling area acquired by a three-dimensional camera;

[0019] According to the three-dimensional image information, calculating first three-dimensional corner point coordinates of the plurality of laid tiles in a camera coordinate system of the three-dimensional camera;

[0020] According to the first three-dimensional corner point coordinates of the plurality of laid tiles, fitting and optimizing calculation is performed to obtain a corner point plane equation corresponding to the three-dimensional corner point coordinates;

[0021] Calculating a coordinate system conversion relationship between a plane coordinate system of the corner point plane equation and the camera coordinate system;

[0022] According to the coordinate system conversion relationship, converting and calculating the first three-dimensional corner point coordinates of the plurality of laid tiles to obtain first corner point coordinate information of the plurality of laid tiles in the plane coordinate system;

[0023] And, the determination of the laying pose information of the target tile to be laid according to the second corner point coordinate information of the target tile to be laid includes:

[0024] According to the second corner point coordinate information of the target tile to be laid and the coordinate system conversion relationship, converting and calculating to obtain second three-dimensional corner point coordinates of the target tile to be laid in the camera coordinate system;

[0025] According to the second three-dimensional corner point coordinates of the target tile to be laid, calculating to obtain the laying pose information of the target tile to be laid.

[0026] As an optional implementation, in the first aspect of the present application, the fitting and optimizing calculation according to the first three-dimensional corner point coordinates of the plurality of laid tiles to obtain a corner point plane corresponding to the three-dimensional corner point coordinates includes:

[0027] Determining a corner point plane expression equation;

[0028] The first three-dimensional corner point coordinates of the plurality of laid bricks are substituted into the corner point plane expression equation, and a weighted fitting calculation is performed based on an optical axis weight to obtain a corner point plane equation corresponding to the three-dimensional corner point coordinates; the optical axis weight is inversely proportional to a distance between each first three-dimensional corner point coordinate and a camera optical axis of the three-dimensional camera.

[0029] The second aspect of the embodiment of the present application discloses a brick laying position calculation device based on grid fitting, and the device comprises:

[0030] A first coordinate determination module is configured to determine first corner point coordinate information of a plurality of laid bricks in a target brick laying area.

[0031] A grid model optimization module is configured to calculate a brick grid relationship model corresponding to the target brick laying area according to a preset brick grid optimization algorithm and the first corner point coordinate information of the plurality of laid bricks.

[0032] A second coordinate determination module is configured to determine second corner point coordinate information of a target to-be-laid brick according to the brick grid relationship model and brick information of the target to-be-laid brick.

[0033] A laying position calculation module is configured to determine laying position information of the target to-be-laid brick according to the second corner point coordinate information of the target to-be-laid brick; the laying position information is used to instruct a brick laying device to lay the target to-be-laid brick.

[0034] As an optional implementation, in the second aspect of the present application, the grid model optimization module comprises:

[0035] A parameter calculation unit is configured to calculate model parameters corresponding to the brick grid optimization model by fitting and optimization according to the first corner point coordinate information of the plurality of laid bricks and a preset brick grid optimization model; the brick grid optimization model is used to limit a corner point parameter relationship between a plurality of virtual bricks of the same size arranged in a grid.

[0036] A model determination unit is configured to determine the brick grid relationship model corresponding to the target brick laying area according to the model parameters.

[0037] As an optional implementation, in the second aspect of the present application, the brick grid optimization model comprises a first brick grid optimization model and a second brick grid optimization model; and the specific manner in which the parameter calculation unit calculates the model parameters corresponding to the brick grid optimization model according to the first corner point coordinate information of the plurality of laid bricks and the preset brick grid optimization model comprises:

[0038] According to the first corner point coordinate information of the plurality of laid bricks and a preset first brick grid optimization model, an optimal grid slope parameter corresponding to the first brick grid optimization model is obtained through fitting optimization calculation; the first brick grid optimization model is used to define the corner point parameter relationship between a plurality of virtual bricks arranged in a grid and having the same size but different intervals; the optimal grid slope parameter is an optimized calculation value of a slope parameter in a straight line relationship equation of different edges of the virtual bricks defined by the first brick grid optimization model.

[0039] The first corner point coordinate information of the plurality of laid bricks and the optimal grid slope parameter are input into a preset second brick grid optimization model, and a model parameter corresponding to the second brick grid optimization model is obtained through fitting optimization calculation; the second brick grid optimization model is used to define the corner point parameter relationship between a plurality of virtual bricks arranged in a grid and having the same size and the same interval; the slope parameter in the straight line relationship equation of different edges of the virtual bricks defined in the second brick grid optimization model is the same as that in the first brick grid optimization model.

[0040] As an optional implementation, in the second aspect of the present application, the fitting optimization calculation is an optimization calculation based on least square fitting; and / or, in the process of fitting optimization, the corner point coordinates with a residual value greater than a preset standard deviation threshold value are deleted; the standard deviation threshold value is a preset multiple of the standard deviation of the residual values corresponding to all corner point coordinates; and the preset multiple is less than 1.

[0041] As an optional implementation, in the second aspect of the present application, the fitting optimization calculation is a fitting optimization calculation with a weight factor; the weight factor includes a central axis distance weight factor and / or a squareness weight factor; the central axis distance weight factor is proportional to the distance between each corner point coordinate information and the center of the target tiling area; and the squareness weight factor is inversely proportional to the fitting residual between each corner point coordinate information and the corresponding corner point of the standard square in which it is located.

[0042] As an optional implementation, in the second aspect of the present application, the specific manner in which the first coordinate determination module determines the first corner point coordinate information of the plurality of laid bricks of the target tiling area includes:

[0043] Obtaining three-dimensional image information of a plurality of laid bricks of a target tiling area acquired by a three-dimensional camera;

[0044] According to the three-dimensional image information, calculating first three-dimensional corner point coordinates of the plurality of laid bricks in a camera coordinate system of the three-dimensional camera;

[0045] According to the first three-dimensional corner point coordinates of the plurality of laid bricks, an optimized calculation is performed to fit an angle point plane equation corresponding to the three-dimensional corner point coordinates;

[0046] A coordinate system conversion relationship between a plane coordinate system of the angle point plane equation and the camera coordinate system is calculated;

[0047] According to the coordinate system conversion relationship, the first three-dimensional corner point coordinates of the plurality of laid bricks are converted to obtain first corner point coordinate information of the plurality of laid bricks in the plane coordinate system;

[0048] In addition, the laying pose calculation module determines the laying pose information of the target to-be-laid brick according to the second corner point coordinate information of the target to-be-laid brick, and the specific manner comprises:

[0049] According to the second corner point coordinate information of the target to-be-laid brick and the coordinate system conversion relationship, second three-dimensional corner point coordinates of the target to-be-laid brick in the camera coordinate system are converted and calculated;

[0050] According to the second three-dimensional corner point coordinates of the target to-be-laid brick, the laying pose information of the target to-be-laid brick is calculated.

[0051] As an optional implementation, in the second aspect of the present application, the first coordinate determination module, according to the first three-dimensional corner point coordinates of the plurality of laid bricks, fits and optimizes the calculation to obtain the angle point plane corresponding to the three-dimensional corner point coordinates, and the specific manner comprises:

[0052] An angle point plane expression equation is determined;

[0053] The first three-dimensional corner point coordinates of the plurality of laid bricks are substituted into the angle point plane expression equation, and a weighted fitting calculation is performed based on an optical axis weight to obtain the angle point plane equation corresponding to the three-dimensional corner point coordinates; the optical axis weight is inversely proportional to the distance between each first three-dimensional corner point coordinate and the camera optical axis of the three-dimensional camera.

[0054] The third aspect of the present application discloses another grid fitting-based brick laying pose calculation device, and the device comprises:

[0055] A memory storing executable program codes;

[0056] A processor coupled with the memory;

[0057] The processor invokes the executable program codes stored in the memory to execute part or all of the steps of the grid fitting-based brick laying pose calculation method disclosed in the first aspect of the present application.

[0058] The fourth aspect of the present application discloses a tiling robot, which comprises an image acquisition device, a tiling device and a control device, and the control device is used to execute part or all steps of the tiling pose calculation method based on grid fitting disclosed in the first aspect of the present application.

[0059] The fifth aspect of the present application discloses a computer storage medium, which stores computer instructions, and the computer instructions are used to execute part or all steps of the tiling pose calculation method based on grid fitting disclosed in the first aspect of the present application when called.

[0060] Compared with the prior art, the embodiments of the present application have the following beneficial effects:

[0061] In the embodiments of the present application, a tiling pose calculation method based on grid fitting, a tiling robot and a tiling device are disclosed, and the method comprises the following steps: determining first corner point coordinate information of a plurality of laid bricks in a target tiling area; calculating a brick grid relationship model corresponding to the target tiling area according to a preset brick grid optimization algorithm and the first corner point coordinate information of the plurality of laid bricks; determining second corner point coordinate information of a target to-be-laid brick according to the brick grid relationship model and brick information of the target to-be-laid brick; determining a laying pose information of the target to-be-laid brick according to the second corner point coordinate information of the target to-be-laid brick; and the laying pose information is used to instruct a brick laying device to lay the target to-be-laid brick. It can be seen that the embodiments of the present application can perform fitting calculation based on a grid optimization algorithm and corner point coordinates of laid bricks, and determine the coordinates of a to-be-laid brick according to the calculated grid model, so as to accurately calculate the coordinate information of the to-be-laid brick according to the corner point coordinates of the laid bricks, thereby improving the accuracy and efficiency of subsequent tiling. BRIEF DESCRIPTION OF DRAWINGS

[0062] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0063] Figure 1 is a flowchart of a tiling pose calculation method based on grid fitting disclosed by the embodiments of the present application.

[0064] Figure 2 is a flowchart of another tiling pose calculation method based on grid fitting disclosed by the embodiments of the present application.

[0065] Figure 3is a structural schematic view of a brick laying pose calculation device based on grid fitting disclosed by an embodiment of the present application.

[0066] Figure 4 is a structural schematic view of another brick laying pose calculation device based on grid fitting disclosed by an embodiment of the present application.

[0067] Figure 5 is a structural schematic view of still another brick laying pose calculation device based on grid fitting disclosed by an embodiment of the present application. DETAILED DESCRIPTION

[0068] In order to enable persons skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by persons skilled in the art without creative efforts fall within the scope of the present application.

[0069] The terms "first", "second", and the like in the specification and claims of the present application and the above-described drawings are used to distinguish different objects, rather than to describe a particular order. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, device, product, or equipment including a series of steps or units is not limited to the listed steps or units, but can optionally include steps or units not listed, or can optionally include other steps or units inherent to the process, method, product, or equipment.

[0070] Reference herein to "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment can be included in at least one embodiment of the present application. The phrase appears at various places in the specification does not necessarily all refer to the same embodiment, nor is it necessarily mutually exclusive of other embodiments, or alternative embodiments. It is explicitly and implicitly understood by those skilled in the art that embodiments described herein can be combined with other embodiments.

[0071] The present application discloses a brick laying pose calculation method and device based on grid fitting and a brick laying robot, which can perform fitting calculation based on a grid optimization algorithm and corner point coordinates of laid bricks, and determine coordinates of to-be-laid bricks according to a calculated grid model, so as to accurately calculate the coordinate information of the to-be-laid bricks according to the corner point coordinates of the laid bricks, thereby improving the accuracy and efficiency of subsequent brick laying. The following will be described in detail respectively.

[0072] Embodiment one

[0073] Please refer to Figure 1, Figure 1 is a flowchart of a brick laying pose calculation method based on grid fitting disclosed by an embodiment of the present application. Wherein, Figure 1 The brick laying pose calculation method based on grid fitting described can be applied in a data processing system, a processing device or a processing server (wherein the server includes a local processing server or a cloud processing server). As Figure 1 shown, the brick laying pose calculation method based on grid fitting can include the following operations:

[0074] 101. Determine the first corner point coordinate information of a plurality of laid bricks in a target brick laying area.

[0075] Optionally, the target brick laying area can be a construction area to be implemented with brick laying, which can be a ground area or a wall area or a ceiling area.

[0076] Optionally, the brick described by the present application can be a square brick, such as a garden through brick with a size of 108mm by 108mm, or other types of bricks such as ceramic tiles, which are not limited by the present application.

[0077] Optionally, the first corner point coordinate information can be the two-dimensional or three-dimensional coordinates of the corner points of the laid bricks, which can be coordinate values in a specific coordinate system, which are not limited by the present application. Preferably, since the fitting algorithm of the two-dimensional grid has lower cost and difficulty, the first corner point coordinate information can adopt two-dimensional coordinates.

[0078] 102. According to the preset brick grid optimization algorithm and the first corner point coordinate information of the plurality of laid bricks, a brick grid relationship model corresponding to the target brick laying area is calculated.

[0079] Optionally, a description equation of the brick grid can be established in advance, and the first corner point coordinate information of the plurality of laid bricks is fitted and calculated to obtain the brick grid relationship model corresponding to the target brick laying area. Optionally, the brick grid relationship model is used to limit the parameter relationship between the position of different laid bricks in the target brick laying area and the corresponding corner point coordinates.

[0080] 103. According to the brick grid relationship model and the brick information of the target to-be-laid brick, the second corner point coordinate information of the target to-be-laid brick is determined.

[0081] Optionally, the brick information can be the position information of the target to-be-laid brick in all laid bricks in the target laying area, such as row and column information or serial number information, which is used to indicate the laying position of the target to-be-laid brick in the target laying area.

[0082] 104. According to the second corner point coordinate information of the target to-be-laid brick, the laying pose information of the target to-be-laid brick is determined.

[0083] Specifically, the paving pose information is used to instruct the brick paving device to pave the target to-be-paved brick. Optionally, the paving pose information can be sent to the brick paving device to control the brick paving device to pave the brick. Optionally, the paving pose information includes six-dimensional pose information corresponding to the end of the mechanical arm, and the brick paving device can include a mechanical arm and a brick paver, so that the mechanical arm can move according to the six-dimensional pose information and control the brick paver to pave the target to-be-paved brick.

[0084] It can be seen that the above embodiment can perform fitting calculation based on the grid optimization algorithm and the corner point coordinates of the paved bricks, and determine the coordinates of the to-be-paved brick according to the calculated grid model, so that the coordinates of the to-be-paved brick can be accurately calculated according to the corner point coordinates of the paved bricks, to improve the accuracy and efficiency of subsequent brick paving.

[0085] As an optional implementation, in the step 102, the brick grid relationship model corresponding to the target brick paving area is calculated according to the preset brick grid optimization algorithm and the first corner point coordinate information of the plurality of paved bricks, including:

[0086] The model parameters corresponding to the brick grid optimization model are calculated by fitting and optimization according to the first corner point coordinate information of the plurality of paved bricks and the preset brick grid optimization model;

[0087] The brick grid relationship model corresponding to the target brick paving area is determined according to the model parameters.

[0088] Optionally, the brick grid optimization model is used to limit the corner point parameter relationship between a plurality of virtual bricks of the same size arranged in a grid. Optionally, the brick grid optimization model can be used to limit the straight line equation expressions of the upper edge, lower edge, left edge and right edge of a plurality of virtual bricks, and all the parameters in the straight line equation expressions are combined into a parameter matrix that can be used for calculation to perform calculation.

[0089] Optionally, the fitting and optimization calculation can be an optimization calculation based on the least square method, for example, the first corner point coordinate information of the plurality of paved bricks can be substituted into the brick grid optimization model or the above-mentioned parameter matrix to perform calculation and optimization based on the least square method, until the optimal model parameters corresponding to the brick grid optimization model are obtained.

[0090] Optionally, in the process of fitting and optimization, the corner point coordinates with a residual value greater than a preset standard deviation threshold value can be deleted, wherein the standard deviation threshold value is a preset multiple of the standard deviation of the residual values corresponding to all the corner point coordinates. Preferably, the preset multiple is less than 1. Preferably, the preset multiple can be 0.25, which has a better model fitting effect in some specific implementation processes.

[0091] Optionally, the fitting optimization calculation can be a fitting optimization calculation with a weight factor, that is, each first corner point coordinate information can be multiplied by a corresponding weight factor and then substituted into the model for fitting optimization calculation. Optionally, the weight factor includes a center axis distance weight factor and / or a squareness weight factor. Optionally, the weight factor can be the product of the center axis distance weight factor and the squareness weight factor.

[0092] Specifically, the center axis distance weight factor is proportional to the distance between each corner point coordinate information and the center of the target tiling area. Optionally, the center of the target tiling area can be a specific coordinate point of the coordinate system in which each corner point coordinate information is located, for example, the coordinate origin, so that the size of the corresponding center axis distance weight factor of each corner point coordinate information can be determined by calculating the distance between the corner point coordinate information and the coordinate point. Optionally, the center axis distance weight factor can be an exponential function with e (natural constant) as the base number and the index including at least the negative number of the distance between each corner point coordinate information and the center of the target tiling area. Preferably, the index of the exponential function can be the ratio of the negative number of the distance between each corner point coordinate information and the center of the target tiling area and twice the first adjustment threshold, so that the function shape is close to the normal distribution, so as to achieve a reasonable center axis distance weight adjustment effect. Optionally, when each tile is a square, the first adjustment threshold can be the square value of the side length of each tile.

[0093] Specifically, the squareness weight factor is inversely proportional to the fitting residual between each corner point coordinate information and the corresponding corner point of the standard square in which it is located. Optionally, the tile in which each corner point coordinate information is located can be subjected to a similarity transformation and fitting calculation with the standard square to obtain the fitting residual between each corner point of the tile and the standard square. Furthermore, the reciprocal of the standard deviation of the fitting residuals of all corner points of the tile can be calculated to obtain the squareness weight factor corresponding to each corner point of the tile. Preferably, the reciprocal of the sum of the fitting residual between each corner point coordinate information and the corresponding corner point of the standard square in which it is located and a specific minimum number is determined as the squareness weight factor corresponding to each corner point. For example, the reciprocal of the sum of the standard deviation of the fitting residuals of all corner points of the tile and a specific minimum number can be determined as the squareness weight factor corresponding to each corner point. Optionally, the minimum number can be 0.000001. By setting in this way, the squareness weight factor can be prevented from being infinite due to the fitting residual being zero, so as to affect the subsequent calculation results.

[0094] It can be seen that by implementing the optional embodiment, the model parameters corresponding to the brick grid optimization model can be calculated by fitting and optimizing the first corner point coordinate information of the plurality of laid bricks, so as to determine the brick grid relationship model corresponding to the target brick laying area, thereby the accurate brick grid relationship model can be calculated, so that the coordinate information of the to-be-laid brick can be accurately calculated in the subsequent process, thereby improving the accuracy and efficiency of the subsequent brick laying.

[0095] As an optional embodiment, the brick grid optimization model can include a first brick grid optimization model and a second brick grid optimization model.

[0096] Specifically, the first brick grid optimization model is used to define the corner point parameter relationship between a plurality of virtual bricks arranged in a grid with the same size but different intervals, for example, the first brick grid optimization model can be used to define the straight line equation expressions of the upper edge, lower edge, left edge and right edge of a plurality of virtual bricks, and the parameters in all straight line equation expressions are combined into a parameter matrix that can be used for operation for calculation. Wherein, the intervals between different adjacent upper edge straight line equations and lower edge straight line equations are different, and the intervals between different adjacent left edge straight line equations and right edge straight line equations are different. Wherein, each straight line equation expression includes a uniform grid slope parameter, and since the grid arrangement relationship is that different straight lines are only parallel or perpendicular, that is, the slope parameter is a same value of positive number (when parallel) or negative number (when perpendicular). This grid optimization model is relatively loose, which is helpful to calculate a relatively optimal grid slope parameter.

[0097] Specifically, the second brick grid optimization model is used to define the corner point parameter relationship between a plurality of virtual bricks arranged in a grid with the same size and the same interval. Similarly, it can be used to define the straight line equation expressions of the upper edge, lower edge, left edge and right edge of a plurality of virtual bricks, and the parameters in all straight line equation expressions are combined into a parameter matrix that can be used for operation for calculation. Wherein, the intervals between different adjacent upper edge straight line equations and lower edge straight line equations are the same, and the intervals between different adjacent left edge straight line equations and right edge straight line equations are the same. The characteristic is that the slope parameters in the straight line relationship equations of different edges of the virtual bricks defined in the second brick grid optimization model are the same as those in the first brick grid optimization model, that is, they share a same slope parameter.

[0098] Correspondingly, in the above step, the model parameters corresponding to the brick grid optimization model are calculated by optimizing the first corner point coordinate information of the plurality of laid bricks and the preset brick grid optimization model, including:

[0099] According to the first corner point coordinate information of the plurality of laid bricks and a preset first brick grid optimization model, an optimal grid slope parameter corresponding to the first brick grid optimization model is obtained through fitting optimization calculation; wherein the optimal grid slope parameter is an optimized calculation value of a slope parameter in a linear relationship equation of different edges of a virtual brick defined by the first brick grid optimization model;

[0100] The first corner point coordinate information of the plurality of laid bricks and the optimal grid slope parameter are input into a preset second brick grid optimization model, and a model parameter corresponding to the second brick grid optimization model is obtained through fitting optimization calculation.

[0101] Optionally, the fitting optimization calculation of the first brick grid optimization model or the second brick grid optimization model can be an optimization calculation based on least square fitting. For example, the first corner point coordinate information of the plurality of laid bricks can be substituted into the first brick grid optimization model or the second brick grid optimization model or a corresponding parameter matrix to perform arithmetic optimization based on least square fitting until an optimal fitting result is obtained.

[0102] Optionally, in the process of fitting optimization of the first brick grid optimization model or the second brick grid optimization model, corner point coordinates with a residual value greater than a preset standard deviation threshold value can be deleted, wherein the standard deviation threshold value is a preset multiple of the standard deviation of the residual values corresponding to all corner point coordinates. Preferably, the preset multiple is less than 1. Preferably, the preset multiple can be 0.25, which has a better model fitting effect in some specific implementation processes.

[0103] Optionally, the fitting optimization calculation of the first brick grid optimization model or the second brick grid optimization model can be fitting optimization calculation with a weight factor, that is, each first corner point coordinate information can be multiplied by a corresponding weight factor and then substituted into the model for fitting optimization calculation. Optionally, the weight factor includes a center axis distance weight factor and / or a squareness weight factor. Optionally, the weight factor can be the product of the center axis distance weight factor and the squareness weight factor.

[0104] Specifically, the center-axis distance weight factor is proportional to the distance between each corner point coordinate information and the center of the target tile region. Optionally, the center of the target tile region can be a specific coordinate point of the coordinate system in which each corner point coordinate information is located, such as the coordinate origin, so that the size of the corresponding center-axis distance weight factor of each corner point coordinate information can be determined by calculating the distance between each corner point coordinate information and the coordinate point. Optionally, the center-axis distance weight factor can be an exponential function with e (natural constant) as the base number and the index including at least the negative number of the distance between each corner point coordinate information and the center of the target tile region. Preferably, the index of the exponential function can be the ratio of the negative number of the distance between each corner point coordinate information and the center of the target tile region and twice the first adjustment threshold, so that the function shape is close to the normal distribution, so as to achieve a reasonable center-axis distance weight adjustment effect. Optionally, when each tile is square, the first adjustment threshold can be the square value of the side length of each tile.

[0105] Specifically, the squareness weight factor is inversely proportional to the fitting residual between each corner point coordinate information and the corresponding corner point of the standard square in which it is located. Optionally, a similarity transformation and fitting calculation can be performed on each tile and the standard square to obtain the fitting residual between each corner point of the tile and the standard square. Further, the reciprocal of the standard deviation of the fitting residuals of all corner points of the tile can be calculated to obtain the squareness weight factor corresponding to each corner point of the tile. Preferably, the reciprocal of the sum of the fitting residual between each corner point coordinate information and the corresponding corner point of the standard square in which it is located and a specific minimum number is determined as the squareness weight factor corresponding to each corner point. For example, the reciprocal of the sum of the standard deviation of the fitting residuals of all corner points of the tile and a specific minimum number can be determined as the squareness weight factor corresponding to each corner point. Optionally, the minimum number can be 0.000001. In this way, the squareness weight factor can be prevented from being infinite due to the fitting residual being zero, so as to affect the subsequent calculation results.

[0106] In a specific embodiment, a mathematical model of a tile grid is constructed according to a specific tile laying scenario. First, in an ideal case, it is assumed that all tiles are square, and the horizontal lines and vertical lines of the grid composed of tiles are parallel to each other, and the horizontal lines and vertical lines are perpendicular to each other. In particular, two mathematical models of tile grid are constructed. In the first mathematical model of tile grid, it is assumed that all tiles in the grid are of equal size, and the gaps between the tiles are not equal. Therefore, in the first mathematical model of tile grid, the horizontal line equation of the upper edge of the rth row (r = 0, 1, 2,...) of tiles can be set as y = kx + rw + h r Similarly, the horizontal line equation of the lower edge of the rth row of tiles can be set as y = kx + (r + 1)w + h rSimilarly, the vertical line equation of the left edge of the brick in the cth column (c = 0, 1, 2,...) can be set as: x = -ky + cw + v c Similarly, the vertical line equation of the right edge of the brick in the cth column can be set as: x = -ky + (c + 1)w + v c .

[0107] where k is the slope, all the line equations share, w is the offset parameter of the brick size, h r is the offset parameter of the rth row of the brick joint, and v c is the offset parameter of the cth column of the brick joint.

[0108] In order to facilitate subsequent fitting calculation, the above first brick grid mathematical model is written in matrix form, and the following is obtained:

[0109]

[0110] In the second brick grid mathematical model, it is assumed that all bricks are of equal size and the gaps between the bricks are equal, so in this grid mathematical model, the horizontal line equation of the upper edge of the brick in the rth row can be set as: y = kx + rw + rg + a, similarly, the horizontal line equation of the lower edge of the brick in the rth row can be set as: y = kx + (r + 1)w + rg + a, similarly, the vertical line equation of the left edge of the brick in the cth column can be set as: x = -ky + cw + cg + b, and similarly, the vertical line equation of the right edge of the brick in the cth column can be set as: x = -ky + (c + 1)w + cg + b.

[0111] where k is the slope, all the line equations share, and are shared with the first brick grid mathematical model, w is the offset parameter of the brick size, g is the offset parameter of the brick joint size, a is the offset parameter of the 0th row of the brick joint, and b is the offset parameter of the 0th column of the brick joint.

[0112] In order to facilitate subsequent fitting calculation, the above second brick grid mathematical model is written in matrix form, and the following is obtained:

[0113]

[0114] Both the matrix forms of the two brick grid mathematical models are regarded as the form of AX = Y, and the fitting optimization solution of the two brick grid mathematical models is converted into a least squares problem of matrix for solving, and the objective function of the solution is: min arg (‖AX-Y‖ 2 ), and the least squares solution is: X = (A T A) -1 A T Y.

[0115] Further, in the above fitting optimization solution of the two brick grid mathematical models, considering that the camera is used to acquire the brick image to determine the corner point coordinates in the specific embodiment, the imaging of the brick in the center of the camera image is better, and the imaging of the edge is worse, at this time, the camera coordinate system or the plane coordinate system converted from the camera coordinate system has its coordinate origin located at the center of the region, therefore, a central axis distance weight factor λ is set to improve the fitting weight of the corner point coordinates of the central region r , and the formula is as follows:

[0116]

[0117] Wherein, x and y are the corner point coordinates, and σ is an adjustment threshold for adjusting the distance between the corner point coordinates and the center of the region (i.e. the origin) and the weight relationship, which is generally taken as the size of the edge length of a brick.

[0118] Further, in the above fitting optimization solution of the two brick grid mathematical models, considering the squareness requirement of the brick, the four corner points of any brick are fitted with a similar transformation and a unit square to obtain the standard deviation e of the residual s . Since the residual of good squareness is smaller, the fitting weight is larger, and here the squareness weight factor λ e is the reciprocal of the residual. In order to avoid a denominator of 0, a very small number is added to the denominator:

[0119]

[0120] Finally, in the fitting optimization solution of the two brick grid mathematical models, the total weight is the product of the central axis distance weight factor λ r and the squareness weight factor λ e : λ = λ r λ e .

[0121] In the specific weighted fitting optimization calculation, first, in the first step, the first corner point coordinate information of all laid bricks is multiplied by the corresponding weight, and then substituted into the above matrix of the first brick grid mathematical model to obtain:

[0122]

[0123] Then, according to the above least square solution formula, the eucalyptus matrix is fitted, in the fitting process, iteration is increased, the corner point coordinates with a residual greater than 0.25 times the standard deviation are removed, and only the k value of the fitting model result, i.e. the slope parameter, is taken after the fitting is completed.

[0124] In the second step, the first corner point coordinate information of all laid bricks is multiplied by the corresponding weight, and then substituted into the above matrix of the second brick grid mathematical model to obtain:

[0125]

[0126] Then, the eucalyptus matrix is ​​fitted according to the least squares solution formula. During the fitting process, iterations are added to remove the corner coordinates with residuals greater than 0.25 times the standard deviation. After the fitting is completed, the model parameters w, g, a, b are calculated, and the second brick grid mathematical model is used as the ideal calculation model.

[0127] Specifically, when the tiling robot performs tiling work based on the ideal calculation model obtained above, it can calculate the position and posture of the tiling bricks based on the row and column numbers of the bricks to be laid in the current tiling field of view, provided by the tiling task planning. Specifically, based on the row number of the brick to be laid and the ideal calculation model obtained above, the equations of the upper and lower edges of the brick to be laid can be calculated. Then, based on the column number of the brick to be laid and the ideal calculation model obtained above, the equations of the left and right edges of the brick to be laid can be calculated. Subsequently, based on these four line equations, four intersection points A, B, C, and D are obtained, and the coordinates of the four intersection points are transformed from the brick corner coordinate system to the camera coordinate system. Further, the coordinates of the four intersection points in the camera coordinate system are connected to the diagonal points to find the central intersection point E, thus establishing a coordinate system on the brick to be laid, with the central intersection point E as the origin. The direction is the X-axis direction, with The direction is the Y-axis, and the six-dimensional posture information of the brick to be laid is finally obtained so that the brick-laying robot can control the robotic arm on the brick-laying device to perform the brick-laying operation.

[0128] As can be seen, by implementing this optional implementation method, the optimal grid slope parameter can be obtained by fitting and optimizing the first brick grid optimization model, and then the model parameters can be obtained by fitting and optimizing the second brick grid optimization model based on the optimal grid slope parameter. In this way, an accurate brick grid relationship model can be calculated, so that the coordinate information of the bricks to be laid can be accurately calculated in the future, thereby improving the accuracy and efficiency of subsequent brick laying.

[0129] Example 2

[0130] Please see Figure 2 , Figure 2 This is a flowchart illustrating another method for calculating paving pose based on mesh fitting disclosed in an embodiment of the present invention. Figure 2 The described mesh-fitting-based method for calculating paving pose can be applied to data processing systems, processing devices, or processing servers (including local processing servers or cloud processing servers). Figure 2As shown, the brick laying pose calculation method based on grid fitting can include the following operations:

[0131] 201. Obtain three-dimensional image information of a plurality of laid bricks of a target brick laying area acquired by a three-dimensional camera.

[0132] Optionally, the three-dimensional image information can include two-dimensional images and point cloud information. Optionally, the three-dimensional camera can be an image acquisition device provided on a laying robot, and preferably it can be a Microsoft Azure Kinect DK three-dimensional camera, which can acquire two-dimensional images and point cloud information of a plurality of laid bricks of the target brick laying area in the current field of view, so as to facilitate subsequent calculation.

[0133] 202. Calculate first three-dimensional corner point coordinates of the plurality of laid bricks in a camera coordinate system of the three-dimensional camera according to the three-dimensional image information.

[0134] Specifically, all the bricks in the field of view can be identified and the four corner points of each brick can be located and calculated according to the three-dimensional image information, and then the three-dimensional coordinates (x, y, z) of the four corner points of each brick in the camera coordinate system (such as a color camera coordinate system) can be calculated by fusing the two-dimensional images and the point cloud information.

[0135] Further, the bricks can be numbered by rows and columns (r, c) according to their positions in the image coordinate system, and the number of the to-be-laid bricks can be calculated according to the brick laying plan to obtain the brick information of the laid bricks and the to-be-laid bricks, so as to facilitate subsequent model optimization calculation and laying pose calculation. These technical details can be referred to the description in Embodiment One, and will not be further described here.

[0136] 203. According to the first three-dimensional corner point coordinates of the plurality of laid bricks, an angle point plane equation corresponding to the three-dimensional corner point coordinates is calculated by fitting and optimization.

[0137] 204. Calculate the coordinate system conversion relationship between the plane coordinate system of the angle point plane equation and the camera coordinate system.

[0138] 205. According to the coordinate system conversion relationship, the first three-dimensional corner point coordinates of the plurality of laid bricks are converted to obtain the first corner point coordinate information of the plurality of laid bricks in the plane coordinate system.

[0139] 206. According to a preset brick grid optimization algorithm and the first corner point coordinate information of the plurality of laid bricks, a brick grid relationship model corresponding to the target brick laying area is calculated.

[0140] 207. According to the brick grid relationship model and the brick information of the target to-be-laid brick, the second corner point coordinate information of the target to-be-laid brick is determined.

[0141] 208. Based on the coordinate information of the second corner point of the target brick to be laid, and the coordinate system transformation relationship, the coordinates of the second three-dimensional corner point of the target brick to be laid in the camera coordinate system are transformed and calculated.

[0142] 209. Based on the coordinates of the second and third-dimensional corner points of the target brick to be laid, the laying posture information of the target brick to be laid is calculated.

[0143] In this embodiment of the invention, for the relevant descriptions and technical details of steps 206-207, please refer to the detailed description of steps 102-103 in Embodiment 1. The embodiments of the invention will not repeat the details.

[0144] As can be seen, the embodiments of the present invention can further calculate the transformation relationship between the camera coordinate system and the planar coordinate system, so as to calculate the coordinates of the corner point in the planar coordinate system according to the transformation relationship, so as to facilitate subsequent mesh optimization calculation, reduce the workload of optimization calculation, and at the same time, after the corner coordinates of the brick to be laid are calculated, the three-dimensional corner coordinates can be calculated through the transformation relationship to facilitate further calculation of laying pose information, thereby effectively improving the efficiency of brick corner calculation, and thus improving the efficiency and accuracy of brick laying work.

[0145] As an optional implementation, step 203 above, which involves fitting and optimizing the calculation to obtain the corner plane corresponding to the three-dimensional corner coordinates based on the first three-dimensional corner coordinates of multiple laid bricks, includes:

[0146] Determine the plane expression equation for the corner point;

[0147] Substitute the coordinates of the first three-dimensional corner points of multiple laid bricks into the corner point plane expression equation, and perform weighted fitting calculation based on the optical axis weight to obtain the corner point plane equation corresponding to the three-dimensional corner point coordinates.

[0148] Specifically, the optical axis weight is inversely proportional to the distance between the coordinates of each first 3D corner point and the optical axis of the 3D camera. Optionally, the optical axis of the 3D camera is generally the origin of the camera coordinate system where the coordinates of the first 3D corner point are located; therefore, the optical axis weight is inversely proportional to the distance between the coordinates of each first 3D corner point and the origin. Preferably, the optical axis weight can be an exponential function with base e (natural constant) and an exponent that includes at least the negative of the distance between the coordinates of each first 3D corner point and the optical axis of the 3D camera. Preferably, the exponent of this exponential function can be the ratio of the negative of the distance between the coordinates of each first 3D corner point and the optical axis of the 3D camera to twice the second adjustment threshold, so that its function shape approximates a normal distribution, thereby achieving a reasonable center axis distance weight adjustment effect. Optionally, when each brick is a square, the first adjustment threshold can be the square of the side length of each brick.

[0149] Optionally, the weighted fitting calculation can be a weighted fitting calculation based on least square optimization.

[0150] In one specific embodiment, a plane closest to all the corner points of the brick needs to be fitted, and a weighted fitting is used, with the weight factor decreasing with the distance of the corner point to the optical axis of the camera, i.e. the farther the corner point is to the optical axis of the camera, the smaller the weight is, and thus a plane fitting weight factor λ is used p , and the formula is as follows:

[0151]

[0152] where x and y are the coordinate values of the corner point in the camera coordinate system, and the value of σ is used to adjust the degree of attenuation with x and y, and is generally the length of the side of the brick.

[0153] Subsequently, the general form of the plane equation ax+by+cz+d=0 is determined, and the objective function of the least square optimization of the weighted fitting is determined: Through the SVD (Singular Value Decomposition) algorithm method, the corresponding corner point plane equation is fitted and calculated according to the above objective function and the coordinates of all the corner points of the brick.

[0154] Further, all the corner points can be vertically projected to the corner point plane according to the plane normal vector direction to obtain the projected corner point p c , and a coordinate system is constructed on the corner point plane, the XOY of the coordinate system is coincident with the plane, and the Z axis is perpendicular to the plane normal vector. Denoted as the corner point plane coordinate system.

[0155] For the general form of the plane equation ax+by+cz+d=0, (a 2 +b 2 +c 2 =1), it can be seen as being obtained by translation and rotation transformation from z=0, and the coordinate transformation from the plane coordinate system to the camera coordinate system is denoted as c H p , and the following transformation relationship is obtained:

[0156]

[0157] The following can be obtained by conversion:

[0158]

[0159] where α is the rotation angle around the x axis, β is the rotation angle around the y axis, and z0 is the translation value in the direction of the z axis, and α, β, and z0 can be solved as α=arcsin(a), z0=-d. The solved α, β, and z0 are substituted into the above formula to obtain c Hp , thereby calculating the coordinate system conversion relationship between the plane coordinate system of the corner point plane equation and the camera coordinate system.

[0160] Further, the projected corner point is converted from the camera coordinate system to the brick corner point plane coordinate system, denoted as p p :

[0161] p p = c H p -1 p c .

[0162] Subsequently, the X and Y values of the projected and transformed corner point p p can be used for grid fitting. The details of the grid fitting steps can be referred to the description in Embodiment One, which will not be repeated here.

[0163] As can be seen, by implementing the optional embodiment, the first three-dimensional corner point coordinates of the multiple laid bricks can be substituted into the corner point plane expression equation, and a weighted fitting calculation is performed based on the optical axis weight, to obtain the corner point plane equation corresponding to the three-dimensional corner point coordinates, so that the accurate plane equation can be calculated, and the plane coordinate information of the corner point of the to-be-laid brick can be accurately calculated subsequently, to improve the calculation efficiency and accuracy of subsequent grid fitting.

[0164] Embodiment Three

[0165] Please refer to Figure 3 , Figure 3 is a structural schematic diagram of a grid fitting-based brick laying pose calculation device disclosed by an embodiment of the present application. Wherein, Figure 3 The grid fitting-based brick laying pose calculation device described can be applied in a data processing system, a processing device or a processing server (wherein the server includes a local processing server or a cloud processing server). As Figure 3 indicated, the grid fitting-based brick laying pose calculation device can include:

[0166] A first coordinate determination module 301 is configured to determine the first corner point coordinate information of the multiple laid bricks in the target brick laying area.

[0167] Optionally, the target brick laying area can be a construction area to be implemented for brick laying, which can be a ground area or a wall area or a ceiling area.

[0168] Optionally, the brick described by the present application can be a square brick, such as a garden through brick with a size of 108mm by 108mm, or other types of bricks such as ceramic tiles, which are not limited by the present application.

[0169] Optionally, the first corner point coordinate information can be two-dimensional coordinates or three-dimensional coordinates of the corners of the laid bricks, which can be coordinate values in a specific coordinate system, and the application does not make any limitation. Preferably, the first corner point coordinate information can adopt two-dimensional coordinates, because the fitting algorithm of the two-dimensional grid has lower cost and difficulty.

[0170] The grid model optimization module 302 is configured to calculate a brick grid relationship model corresponding to the target tiling area according to a preset brick grid optimization algorithm and the first corner point coordinate information of the plurality of laid bricks.

[0171] Optionally, a description equation of the brick grid can be established in advance, and the brick grid relationship model corresponding to the target tiling area can be obtained by fitting calculation according to the first corner point coordinate information of the plurality of laid bricks. Optionally, the brick grid relationship model is used to limit the parameter relationship between the positions of different laid bricks in the target tiling area and the corresponding corner point coordinates.

[0172] The second coordinate determination module 303 is configured to determine second corner point coordinate information of the target to-be-laid brick according to the brick grid relationship model and brick information of the target to-be-laid brick.

[0173] Optionally, the brick information can be position information of the target to-be-laid brick in all laid bricks in the target tiling area, such as row and column information or serial number information, which is used to indicate the laying position of the target to-be-laid brick in the target tiling area.

[0174] The laying pose calculation module 304 is configured to determine laying pose information of the target to-be-laid brick according to the second corner point coordinate information of the target to-be-laid brick.

[0175] Specifically, the laying pose information is used to indicate the laying of the target to-be-laid brick by the brick laying device. Optionally, the laying pose information can be sent to the brick laying device to control the brick laying device to lay the bricks. Optionally, the laying pose information includes six-dimensional pose information of the corresponding end of the mechanical arm, and the brick laying device can include a mechanical arm and a brick laying device, so that the mechanical arm can move according to the six-dimensional pose information and control the brick laying device to lay the target to-be-laid brick.

[0176] It can be seen that the above embodiment of the application can perform fitting calculation based on the grid optimization algorithm and the corner point coordinates of the laid bricks, and determine the coordinates of the to-be-laid brick according to the calculated grid model, so as to accurately calculate the coordinate information of the to-be-laid brick according to the corner point coordinates of the laid bricks, thereby improving the accuracy and efficiency of subsequent tiling.

[0177] As an optional implementation, as shown in Figure 4 The grid model optimization module 302 includes:

[0178] The parameter calculation unit 3021 is configured to fit and optimize calculation to obtain model parameters corresponding to the brick grid optimization model according to the first corner point coordinate information of the plurality of laid bricks and the preset brick grid optimization model.

[0179] The model determination unit 3022 is configured to determine the brick grid relationship model corresponding to the target paving area according to the model parameters.

[0180] Optionally, the brick grid optimization model is used to limit the corner point parameter relationship between a plurality of virtual bricks of the same size arranged in a grid. Optionally, the brick grid optimization model can be used to limit the straight line equation expressions of the straight lines where the upper edges, lower edges, left edges and right edges of the plurality of virtual bricks are located, and all the parameters in the straight line equation expressions are combined into a parameter matrix that can be used for calculation to perform calculation.

[0181] Optionally, the fitting and optimization calculation can be an optimization calculation based on the least square method. For example, based on the least square method, the first corner point coordinate information of the plurality of laid bricks can be substituted into the brick grid optimization model or the above-mentioned parameter matrix to perform calculation and optimization until the optimal model parameters corresponding to the brick grid optimization model are obtained.

[0182] Optionally, in the process of fitting and optimization, the corner point coordinates with residual values greater than a preset standard deviation threshold value can be deleted, wherein the standard deviation threshold value is a preset multiple of the standard deviation of the residual values corresponding to all the corner point coordinates. Preferably, the preset multiple is less than 1. Preferably, the preset multiple can be 0.25, which has a better model fitting effect in some specific implementation processes.

[0183] Optionally, the fitting and optimization calculation can be fitting and optimization calculation with a weight factor, that is, each first corner point coordinate information can be multiplied by a corresponding weight factor and then substituted into the model for fitting and optimization calculation. Optionally, the weight factor includes a center axis distance weight factor and / or a squareness weight factor. Optionally, the weight factor can be the product of the center axis distance weight factor and the squareness weight factor.

[0184] Specifically, the center-axis distance weight factor is proportional to the distance between each corner point coordinate information and the center of the target tiling area. Optionally, the center of the target tiling area can be a specific coordinate point of the coordinate system in which each corner point coordinate information is located, for example, the coordinate origin, so that the size of the corresponding center-axis distance weight factor of each corner point coordinate information can be determined by calculating the distance between each corner point coordinate information and the coordinate point. Optionally, the center-axis distance weight factor can be an exponential function with e (natural constant) as the base number and the index including at least the negative number of the distance between each corner point coordinate information and the center of the target tiling area. Preferably, the index of the exponential function can be the ratio of the negative number of the distance between each corner point coordinate information and the center of the target tiling area and twice the first adjustment threshold, so that the function shape is close to the normal distribution, so as to achieve a reasonable center-axis distance weight adjustment effect. Optionally, when each tile is a square, the first adjustment threshold can be the square value of the side length of each tile.

[0185] Specifically, the squareness weight factor is inversely proportional to the fitting residual between each corner point coordinate information and the corresponding corner point of the standard square in which it is located. Optionally, the tile in which each corner point coordinate information is located can be subjected to similarity transformation and fitting calculation with the standard square to obtain the fitting residual between each corner point of the tile and the standard square. Furthermore, the reciprocal of the standard deviation of the fitting residuals of all corner points of the tile can also be calculated to obtain the squareness weight factor corresponding to each corner point of the tile. Preferably, the reciprocal of the sum of the fitting residual between each corner point coordinate information and the corresponding corner point of the standard square in which it is located and a specific minimum number is determined as the squareness weight factor corresponding to each corner point. For example, the reciprocal of the sum of the standard deviation of the fitting residuals of all corner points of the tile and a specific minimum number can be determined as the squareness weight factor corresponding to each corner point. Optionally, the minimum number can be 0.000001. By setting in this way, the squareness weight factor can be prevented from being infinite due to the fitting residual being zero, so as to affect the subsequent calculation results.

[0186] It can be seen that by implementing the optional embodiment, the model parameters corresponding to the tile grid optimization model can be calculated and optimized according to the first corner point coordinate information of the plurality of laid tiles, so as to determine the tile grid relationship model corresponding to the target tiling area, thereby enabling the accurate tile grid relationship model to be calculated, so that the coordinate information of the tile to be laid can be accurately calculated subsequently, thereby improving the accuracy and efficiency of subsequent tiling.

[0187] As an optional embodiment, the tile grid optimization model can include a first tile grid optimization model and a second tile grid optimization model.

[0188] Specifically, the first brick grid optimization model is used to define the parameter relationship between the corner points of a plurality of virtual bricks arranged in a grid with the same size but different intervals. For example, the first brick grid optimization model can be used to define the linear equation expressions of the straight lines on which the upper edges, lower edges, left edges and right edges of the plurality of virtual bricks are located, and the parameters in all linear equation expressions are combined into a parameter matrix that can be used for calculation for calculation. Wherein, the intervals between different adjacent upper edge linear equations and lower edge linear equations are different, and the intervals between different adjacent left edge linear equations and right edge linear equations are different. Each linear equation expression includes a uniform grid slope parameter. Due to the grid arrangement relationship, different straight lines are only in parallel or perpendicular relationship, i.e. the slope parameter is a same value of positive number (when parallel) or negative number (when perpendicular). This grid optimization model is relatively loose, which is helpful to obtain a relatively optimal grid slope parameter.

[0189] Specifically, the second brick grid optimization model is used to define the parameter relationship between the corner points of a plurality of virtual bricks arranged in a grid with the same size and the same interval. Similarly, it can be used to define the linear equation expressions of the straight lines on which the upper edges, lower edges, left edges and right edges of the plurality of virtual bricks are located, and the parameters in all linear equation expressions are combined into a parameter matrix that can be used for calculation for calculation. Wherein, the intervals between different adjacent upper edge linear equations and lower edge linear equations are the same, and the intervals between different adjacent left edge linear equations and right edge linear equations are the same. The characteristic is that the slope parameters in the linear relationship equations of the different edges of the virtual bricks defined in the second brick grid optimization model are the same as those in the first brick grid optimization model, i.e. they share a same slope parameter.

[0190] Correspondingly, the specific manner of the parameter calculation unit 3021 for optimizing and calculating the model parameters corresponding to the brick grid optimization model according to the first corner point coordinate information of the plurality of laid bricks and the preset brick grid optimization model includes:

[0191] According to the first corner point coordinate information of the plurality of laid bricks and the preset first brick grid optimization model, the optimal grid slope parameter corresponding to the first brick grid optimization model is fitted and optimized to calculate; wherein, the optimal grid slope parameter is the optimized calculation value of the slope parameter in the linear relationship equation of the different edges of the virtual brick defined by the first brick grid optimization model;

[0192] The first corner point coordinate information of the plurality of laid bricks and the optimal grid slope parameter are input into the preset second brick grid optimization model, and the model parameters corresponding to the second brick grid optimization model are fitted and optimized to calculate.

[0193] Optionally, the fitting optimization calculation for the first brick grid optimization model or the second brick grid optimization model can be an optimization calculation based on least squares fitting. For example, the first corner point coordinate information of the plurality of laid bricks can be substituted into the first brick grid optimization model or the second brick grid optimization model or the corresponding parameter matrix based on least squares fitting to perform optimization calculation until the optimal fitting result is obtained.

[0194] Optionally, in the process of fitting optimization of the first brick grid optimization model or the second brick grid optimization model, the corner point coordinates with a residual value greater than a preset standard deviation threshold value can be deleted, wherein the standard deviation threshold value is a preset multiple of the standard deviation of the residual values corresponding to all corner point coordinates. Preferably, the preset multiple is less than 1. Preferably, the preset multiple can be 0.25, which has a better model fitting effect in some specific implementation processes.

[0195] Optionally, the fitting optimization calculation for the first brick grid optimization model or the second brick grid optimization model can be a fitting optimization calculation with a weight factor, that is, each first corner point coordinate information can be multiplied by a corresponding weight factor and then substituted into the model for fitting optimization calculation. Optionally, the weight factor includes a center axis distance weight factor and / or a squareness weight factor. Optionally, the weight factor can be the product of the center axis distance weight factor and the squareness weight factor.

[0196] Specifically, the center axis distance weight factor is proportional to the distance between each corner point coordinate information and the center of the target brick laying area. Optionally, the center of the target brick laying area can be a specific coordinate point of the coordinate system where each corner point coordinate information is located, for example, the coordinate origin, so that the distance between each corner point coordinate information and the coordinate point can be calculated to determine the size of the corresponding center axis distance weight factor. Optionally, the center axis distance weight factor can be an exponential function with e (natural constant) as the base number and the index including at least the negative number of the distance between each corner point coordinate information and the center of the target brick laying area. Preferably, the index of the exponential function can be the ratio of the negative number of the distance between each corner point coordinate information and the center of the target brick laying area and twice the first adjustment threshold value, so that the function shape is close to the normal distribution, so as to achieve a reasonable center axis distance weight adjustment effect. Optionally, the first adjustment threshold value can be the square value of the side length of each brick when each brick is a square.

[0197] Specifically, the squareness weight factor is inversely proportional to the fitting residual between each corner point coordinate information and the corresponding corner point of the standard square. Optionally, a similarity transformation and fitting calculation can be performed between the tile and the standard square to obtain the fitting residual between each corner point of the tile and the standard square. Furthermore, the reciprocal of the standard deviation of the fitting residual of all corner points of the tile can be calculated to obtain the squareness weight factor corresponding to each corner point of the tile. Preferably, the reciprocal of the sum of the specific minimum number and the fitting residual between each corner point coordinate information and the corresponding corner point of the standard square is determined as the squareness weight factor corresponding to each corner point. For example, the reciprocal of the sum of the specific minimum number and the standard deviation of the fitting residual of all corner points of the tile can be determined as the squareness weight factor corresponding to each corner point. Optionally, the minimum number can be 0.000001. By setting in this way, the squareness weight factor can be prevented from being infinite due to the fitting residual being zero, so as to affect the subsequent calculation results.

[0198] It can be seen that by implementing the optional embodiment, the optimal grid slope parameter can be obtained through fitting and optimization calculation according to the first tile grid optimization model, and the model parameter can be obtained through fitting and optimization calculation according to the optimal grid slope parameter and the second tile grid optimization model, so that an accurate tile grid relationship model can be calculated, so that the coordinate information of the to-be-laid tile can be accurately calculated subsequently, and the accuracy and efficiency of subsequent tiling can be improved.

[0199] As an optional embodiment, the specific manner in which the first coordinate determination module 301 determines the first corner point coordinate information of the plurality of laid tiles in the target tiling area includes:

[0200] Obtain three-dimensional image information of the plurality of laid tiles in the target tiling area obtained by the three-dimensional camera;

[0201] According to the three-dimensional image information, calculate first three-dimensional corner point coordinates of the plurality of laid tiles in the camera coordinate system of the three-dimensional camera;

[0202] According to the first three-dimensional corner point coordinates of the plurality of laid tiles, fitting and optimization calculation is performed to obtain a corner point plane equation corresponding to the three-dimensional corner point coordinates;

[0203] Calculate the coordinate system conversion relationship between the plane coordinate system of the corner point plane equation and the camera coordinate system;

[0204] According to the coordinate system conversion relationship, the first three-dimensional corner point coordinates of the plurality of laid tiles are converted to obtain first corner point coordinate information of the plurality of laid tiles in the plane coordinate system;

[0205] And the paving pose calculation module 304 determines the paving pose information of the target to-be-paved tile according to the second corner point coordinate information of the target to-be-paved tile. The specific manner of the paving pose calculation module 304 includes the following steps.

[0206] According to the second three-dimensional corner point coordinate of the target to-be-paved tile and the coordinate system conversion relationship, the second three-dimensional corner point coordinate of the target to-be-paved tile in the camera coordinate system is converted and calculated.

[0207] According to the second three-dimensional corner point coordinate of the target to-be-paved tile, the paving pose information of the target to-be-paved tile is calculated.

[0208] Optionally, the three-dimensional image information can include two-dimensional image and point cloud information. Optionally, the three-dimensional camera can be an image acquisition device arranged on the paving robot, and preferably, it can be a Microsoft Azure Kinect DK three-dimensional camera, which can acquire two-dimensional image and point cloud information of a plurality of paved tiles in the target paving area in the current field of view, so as to facilitate subsequent calculation.

[0209] Specifically, all the tiles in the field of view can be recognized according to the three-dimensional image information, and the four tile corner points of each tile are located and calculated, and then the three-dimensional coordinates (x, y, z) of the four corner points of each tile in the camera coordinate system (such as the color camera coordinate system) are calculated by fusing the two-dimensional image and the point cloud information.

[0210] Further, the tiles in the image coordinate system can be numbered by rows and columns (r, c) according to the positions of the tiles, and the numbers of the to-be-paved tiles are calculated according to the paving planning, so as to obtain the tile information of the paved tiles and the to-be-paved tiles, thereby facilitating subsequent model optimization calculation and paving pose calculation.

[0211] It can be seen that the optional embodiment can further calculate the conversion relationship between the camera coordinate system and the plane coordinate system, so as to calculate the coordinates of the corner points in the plane coordinate system according to the conversion relationship, so as to facilitate subsequent grid optimization calculation and reduce the workload of optimization calculation. In addition, the three-dimensional corner point coordinates of the to-be-paved tile can be calculated through the conversion relationship after the corner point coordinates of the to-be-paved tile are calculated, so as to further calculate the paving pose information, thereby effectively improving the efficiency of tile corner point calculation and further improving the efficiency and accuracy of the paving work.

[0212] As an optional embodiment, the first coordinate determination module 301 determines the specific manner of fitting and optimizing the three-dimensional corner point coordinate corresponding to the corner point plane according to the first three-dimensional corner point coordinate of the plurality of paved tiles, which includes the following steps.

[0213] Determine the corner point plane expression equation.

[0214] The first three-dimensional corner point coordinates of the plurality of laid bricks are substituted into the corner point plane expression equation, and a weighted fitting calculation is performed based on the optical axis weight to obtain a corner point plane equation corresponding to the three-dimensional corner point coordinates.

[0215] Specifically, the optical axis weight is inversely proportional to the distance between each first three-dimensional corner point coordinate and the camera optical axis of the three-dimensional camera. Optionally, the camera optical axis of the three-dimensional camera is generally the origin of the camera coordinate system in which the first three-dimensional corner point coordinates are located, and thus the optical axis weight is inversely proportional to the distance between each first three-dimensional corner point coordinate and the coordinate origin. Preferably, the optical axis weight can be an exponential function with e (natural constant) as the base number, and the exponent at least includes the negative number of the distance between each first three-dimensional corner point coordinate and the camera optical axis of the three-dimensional camera. Preferably, the exponent of the exponential function can be the ratio of the negative number of the distance between each first three-dimensional corner point coordinate and the camera optical axis of the three-dimensional camera and twice the second adjustment threshold, so that the function shape is close to the normal distribution, so as to achieve a reasonable central axis distance weight adjustment effect. Optionally, the first adjustment threshold can be the square value of the side length of each brick when each brick is square.

[0216] Optionally, the weighted fitting calculation can be a weighted fitting calculation based on least squares optimization.

[0217] As can be seen, by implementing the optional embodiment, the first three-dimensional corner point coordinates of the plurality of laid bricks can be substituted into the corner point plane expression equation, and a weighted fitting calculation is performed based on the optical axis weight to obtain a corner point plane equation corresponding to the three-dimensional corner point coordinates, so that an accurate plane equation can be calculated, so that the plane coordinate information of the corner point of the to-be-laid brick can be accurately calculated subsequently, so as to improve the calculation efficiency and accuracy of subsequent grid fitting.

[0218] Embodiment Four

[0219] Please refer to Figure 5 , Figure 5 It is another grid fitting-based brick laying pose calculation device disclosed in the embodiments of the present application. Figure 5 The grid fitting-based brick laying pose calculation device described can be applied in a data processing system, a processing device or a processing server (wherein the server includes a local processing server or a cloud processing server). As Figure 5 shown, the grid fitting-based brick laying pose calculation device can include:

[0220] a memory 401 storing executable program codes;

[0221] a processor 402 coupled with the memory 401;

[0222] The processor 402 invokes executable program code stored in the memory 401 to perform the steps of the grid fitting based tile pose calculation method described in Embodiment One or Embodiment Two.

[0223] Embodiment Five

[0224] The computer readable storage medium stores a computer program for electronic data exchange, wherein the computer program causes a computer to perform the steps of the grid fitting based tile pose calculation method described in Embodiment One or Embodiment Two.

[0225] Embodiment Six

[0226] The computer program product includes a non-transitory computer readable storage medium storing a computer program, and the computer program is operable to cause a computer to perform the steps of the grid fitting based tile pose calculation method described in Embodiment One or Embodiment Two.

[0227] Embodiment Seven

[0228] The tile robot can include an image acquisition device, a tile device, and a control device, wherein the control device is configured to perform the steps of the grid fitting based tile pose calculation method described in Embodiment One or Embodiment Two to control the tile device to perform the tile work in combination with the image obtained by the image acquisition device. The details of the devices in the tile robot can refer to the description in Embodiment One or Two, which will not be repeated here.

[0229] The above describes certain embodiments of the present specification, and other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims can be performed in an order different than the order in the embodiments and still achieve the desired result. In addition, the processes depicted in the figures do not necessarily have to be performed in the specific order shown or sequentially.

[0230] Each of the embodiments in the specification is described in a progressive manner, and the same or similar parts between each embodiment can be referred to each other, and each embodiment focuses on the difference from other embodiments. In particular, for the device, equipment, and non-volatile computer readable storage medium embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can refer to the part of the method embodiment.

[0231] The apparatus, device, nonvolatile computer readable storage medium and method provided by the embodiments of the present specification are corresponding, therefore, the apparatus, device, nonvolatile computer storage medium also has similar beneficial technical effects as the corresponding method, since the beneficial technical effects of the method have been described in detail above, therefore, the beneficial technical effects of the corresponding apparatus, device, nonvolatile computer storage medium will not be described here.

[0232] The controller can be implemented in any suitable manner, for example, the controller can take the form of, for example, a microprocessor or processor and a computer readable medium storing computer readable program code, such as software or firmware, executable by the (micro)processor, logic gates, switches, an application specific integrated circuit (ASIC), a programmable logic controller and an embedded microcontroller, examples of the controller include but are not limited to the following microcontrollers: ARC 625D, Atmel AT91SAM, Microchip PIC18F26K20 and Silicone Labs C8051F320, the memory controller can also be implemented as part of the control logic of the memory. Those skilled in the art will also know that, in addition to implementing the controller in the form of pure computer readable program code, the same function can also be achieved by logically programming the method steps in the form of logic gates, switches, application specific integrated circuits, programmable logic controllers and embedded microcontrollers. Therefore, such a controller can be considered as a hardware component, and the means included therein for implementing various functions can also be considered as structures within the hardware component. Alternatively, the means for implementing various functions can even be considered as both a software module implementing the method and a structure within the hardware component.

[0233] The system, apparatus, module or unit illustrated by the above embodiments can be specifically implemented by a computer chip or entity, or by a product with certain functions. A typical implementation device is a computer. Specifically, the computer may, for example, be a personal computer, a laptop computer, a cellular phone, a camera phone, a smartphone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.

[0234] For the convenience of description, the above apparatus is described as various units respectively in terms of functions. Of course, the functions of each unit can be implemented in the same or more software and / or hardware when implementing the present specification.

[0235] Those skilled in the art will appreciate that embodiments of the present description can be readily used as a method, an apparatus (system) or a computer program product. Accordingly, embodiments of the present description can take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, embodiments of the present description can take the form of a computer program product on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROMs, optical storage devices, and the like) embodying computer readable program code.

[0236] The present description is described in reference to flow diagrams and / or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the present description. It will be understood that each block of the flow diagrams and / or block diagrams, and combinations of blocks in the flow diagrams and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general purpose computer, special purpose computer, embedded processing device or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flow diagrams and / or block diagrams block or blocks. Figure 1 one or more functions specified in the flow diagram and / or block diagram block or blocks. Figure 1 one or more functions specified in the flow diagram and / or block diagram block or blocks.

[0237] These computer program instructions can also be stored in a computer- readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instructions which implement the flow diagrams and / or block diagrams block or blocks. Figure 1 one or more functions specified in the flow diagram and / or block diagram block or blocks. Figure 1 one or more functions specified in the flow diagram and / or block diagram block or blocks.

[0238] These computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the flow diagrams and / or block diagrams block or blocks. Figure 1 one or more functions specified in the flow diagram and / or block diagram block or blocks. Figure 1 one or more functions specified in the flow diagram and / or block diagram block or blocks.

[0239] In one typical configuration, a computing device includes one or more processors (CPUs), input / output interfaces, network interfaces, and memory.

[0240] The memory can include non-persistent memory and / or volatile memory, such as random access memory (RAM) and / or cache memory. The memory can also include non-volatile memory, such as read-only memory (ROM), electrically programmable read-only memory (EPROM), electrically erasable read-only memory (EEPROM), flash memory, or a combination of non-volatile memories in different forms. The memory is an example of computer-readable media.

[0241] Computer-readable media includes permanent and non-permanent, movable and non-movable media that can be implemented by any method or technology to store information. The information can be computer-readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disc read-only memory (CD-ROM), digital versatile disc (DVD), or other optical storage, magnetic cassette, magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information accessible to a computing device. According to the definition herein, computer-readable media does not include transitory media such as modulated data signals and carriers.

[0242] It should also be noted that the terms "comprising", "containing", or any other variant thereof are intended to cover non-exclusive inclusion, such that processes, methods, articles or devices that include a series of elements not only include those elements, but also include other elements not explicitly listed, or inherent to such processes, methods, articles or devices. Without more limitations, the element defined by the statement "comprising a" does not exclude the presence of additional identical elements in the process, method, article or device that includes the element.

[0243] The specification can be described in the general context of computer-executable instructions being executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, and the like that perform particular tasks or implement particular abstract data types. The specification can also be practiced in distributed computing environments where tasks are performed by remote processing devices that are connected through a communication network. In a distributed computing environment, program modules can be located in both local and remote computer storage media, including storage devices.

[0244] Finally, it should be noted that: the embodiment of the application discloses a kind of based on grid fitting's brick laying pose calculation method, device and brick laying robot disclosed only for the preferred embodiment of the application, only for describing the technical scheme of the application, not for its limitation;Although the application is described in detail with reference to the foregoing embodiments, those skilled in the art should understand;It can still modify the technical scheme recorded in the foregoing embodiments, or make equivalent replacement to part of the technical features;And these modifications or replacements do not make the essence of the corresponding technical scheme deviate from the spirit and scope of the technical scheme of the embodiments of the application.

Claims

1. A mesh fitting based tile pose computation method, characterized in that, The method comprises: determining first corner point coordinate information of a plurality of laid tiles of a target tiling area; calculating a tile grid relationship model corresponding to the target tiling area according to a preset tile grid optimization algorithm and the first corner point coordinate information of the plurality of laid tiles; determining second corner point coordinate information of a target tile to be laid according to the tile grid relationship model and tile information of the target tile to be laid; determining laying pose information of the target tile to be laid according to the second corner point coordinate information of the target tile to be laid; the laying pose information is used to instruct a tile laying device to lay the target tile to be laid.

2. The mesh fitting based tile pose computation method of claim 1, wherein, The calculation of the tile grid relationship model corresponding to the target tiling area according to the preset tile grid optimization algorithm and the first corner point coordinate information of the plurality of laid tiles comprises: fitting and optimizing the model parameters corresponding to the tile grid optimization model according to the first corner point coordinate information of the plurality of laid tiles and the preset tile grid optimization model; the tile grid optimization model is used to limit the corner point parameter relationship between a plurality of virtual tiles of the same size arranged in a grid; determining the tile grid relationship model corresponding to the target tiling area according to the model parameters.

3. The mesh fitting based tile pose computation method of claim 2, wherein, The tile grid optimization model comprises a first tile grid optimization model and a second tile grid optimization model; the fitting and optimization of the model parameters corresponding to the tile grid optimization model according to the first corner point coordinate information of the plurality of laid tiles and the preset tile grid optimization model comprises: fitting and optimizing optimal grid slope parameters corresponding to the first tile grid optimization model according to the first corner point coordinate information of the plurality of laid tiles and the preset first tile grid optimization model; the first tile grid optimization model is used to limit the corner point parameter relationship between a plurality of virtual tiles of the same size but different intervals arranged in a grid; the optimal grid slope parameters are optimized calculation values of slope parameters in linear relationship equations of different edges of the virtual tiles defined by the first tile grid optimization model; inputting the first corner point coordinate information of the plurality of laid tiles and the optimal grid slope parameters into the preset second tile grid optimization model to fit and optimize model parameters corresponding to the second tile grid optimization model; the second tile grid optimization model is used to limit the corner point parameter relationship between a plurality of virtual tiles of the same size and same intervals arranged in a grid; slope parameters in linear relationship equations of different edges of the virtual tiles defined in the second tile grid optimization model are the same as those in the first tile grid optimization model.

4. The mesh fitting based tile pose computation method of claim 2, wherein, The fitting and optimization calculation is an optimization calculation based on least square fitting; and / or, in the process of fitting and optimization, deleting corner point coordinates with residual values greater than a preset standard deviation threshold value; the standard deviation threshold value is a preset multiple of the standard deviation of residual values corresponding to all corner point coordinates; the preset multiple is less than 1.

5. The mesh fitting based tile pose computation method of claim 2, wherein, The fitting optimization calculation is a fitting optimization calculation with a weight factor; the weight factor includes a central axis distance weight factor and / or a squareness weight factor; the central axis distance weight factor is proportional to the distance between each corner point coordinate information and the center of the target tiling area; the squareness weight factor is inversely proportional to the fitting residual between each corner point coordinate information and the corresponding corner point of the standard square in which the corner point is located.

6. The mesh fitting based tile pose computation method of claim 1, wherein, The first corner point coordinate information of the plurality of laid tiles of the target tiling area is determined by: acquiring three-dimensional image information of the plurality of laid tiles of the target tiling area acquired by a three-dimensional camera; calculating first three-dimensional corner point coordinates of the plurality of laid tiles in a camera coordinate system of the three-dimensional camera according to the three-dimensional image information; fitting and optimizing to obtain a corner point plane equation corresponding to the three-dimensional corner point coordinates according to the first three-dimensional corner point coordinates of the plurality of laid tiles; calculating a coordinate system conversion relationship between a plane coordinate system of the corner point plane equation and the camera coordinate system; converting and calculating the first three-dimensional corner point coordinates of the plurality of laid tiles to obtain first corner point coordinate information of the plurality of laid tiles in the plane coordinate system according to the coordinate system conversion relationship; and, the second corner point coordinate information of the target tile to be laid is used to determine the laying pose information of the target tile to be laid, including: converting and calculating the second corner point coordinate information of the target tile to be laid and the coordinate system conversion relationship to obtain second three-dimensional corner point coordinates of the target tile to be laid in the camera coordinate system; calculating the laying pose information of the target tile to be laid according to the second three-dimensional corner point coordinates of the target tile to be laid.

7. The mesh fitting based tile pose computation method of claim 6, wherein, The fitting and optimization calculation of the first three-dimensional corner point coordinates of the plurality of laid tiles to obtain the corner point plane corresponding to the three-dimensional corner point coordinates includes: determining a corner point plane expression equation; substituting the first three-dimensional corner point coordinates of the plurality of laid tiles into the corner point plane expression equation, and performing a weighted fitting calculation based on an optical axis weight to obtain a corner point plane equation corresponding to the three-dimensional corner point coordinates; the optical axis weight is inversely proportional to the distance between each first three-dimensional corner point coordinate and the camera optical axis of the three-dimensional camera.

8. A mesh fitting based tile pose computation apparatus, characterized by, The device includes: a first coordinate determination module for determining first corner point coordinate information of a plurality of laid tiles of a target tiling area; a grid model optimization module for calculating a tile grid relationship model corresponding to the target tiling area according to a preset tile grid optimization algorithm and the first corner point coordinate information of the plurality of laid tiles; a second coordinate determination module for determining second corner point coordinate information of a target tile to be laid according to the tile grid relationship model and tile information of the target tile to be laid; a laying pose calculation module for determining laying pose information of the target tile to be laid according to the second corner point coordinate information of the target tile to be laid; the laying pose information is used to instruct a tile laying device to lay the target tile to be laid.

9. A mesh fitting based tile pose computation apparatus, characterized by, The device includes: a memory storing executable program code; a processor coupled to the memory; the processor invokes the executable program code stored in the memory to execute the grid fitting based tile pose calculation method according to any one of claims 1-7.

10. A tiling robot characterized by, the tile robot comprises an image acquisition device, a tile laying device, and a control device, and the control device is configured to execute the grid fitting based tile pose calculation method according to any one of claims 1-7.

Images