Spraying trajectory optimization method, device and system of spraying robot

Abstract

The present application relates to the technical field of manipulator, and particularly relates to a spraying trajectory optimization method, device and system of a spraying robot, the method comprising: scanning a to-be-sprayed object to obtain a target three-dimensional model and determining basic control points thereon; determining corresponding new control points by using curvature parameters of point clouds between adjacent basic control points; determining a target spraying path composed of the basic control points and a corresponding preliminary target function value by using surface distances between the basic control points; generating a plurality of Bezier curves by using a total sequence of control points formed by the target spraying path and the new control points; determining a final target function value of the target spraying path by using curvature parameters and the preliminary target function value at a connection position of adjacent Bezier curves; and optimizing the final target function value by using a preset path optimization algorithm to obtain an optimal target spraying path. The present application can realize accurate spraying of the to-be-sprayed object and achieve a better spraying effect.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of mechanical hands, in particular to a spraying trajectory optimization method, device and system of a spraying robot. BACKGROUND

[0002] A spraying robot is a robotic device used for automated spraying operations, widely used in industries such as automobile manufacturing, home appliance production, aerospace, etc. Its main function is to uniformly spray paint onto the surface of target objects through precise control of the spraying tool. In order to further improve spraying efficiency, reduce paint waste and ensure spraying quality, the optimization of spraying trajectory becomes particularly important. By optimizing the spraying path, the surface quality of the product can be improved, ensuring the stability and consistency of the spraying process.

[0003] Generally, Bezier curves are used to optimize the spraying trajectory of a spraying robot. Existing technologies usually need to first determine the general outline of the path, and then rely on experience to select the number of control points. At this time, if the number of control points is too small, the path may not accurately fit the curved surface of the object to be sprayed; if the number of control points is too large, it will lead to an overall path that is not coordinated enough, and some unnecessary subtle turns or complex curve shapes may appear, which will increase the instability of the spraying process. In addition, the optimization of Bezier curves usually relies on local adjustment (such as gradient descent method) to adjust the position of control points, which is easy to fall into local optimum, resulting in the path curvature or spraying uniformity not reaching the best, and poor adaptability to sudden changes, often requiring manual adjustment of control points. In addition, the repeated simulation verification and manual adjustment process is time-consuming and has low automation. SUMMARY

[0004] In order to solve the technical problems of inaccurate spraying path and poor spraying effect of existing Bezier curve automated spraying optimization technology, the purpose of the present application is to provide a spraying trajectory optimization method, device and system of a spraying robot, the technical solution adopted is as follows:

[0005] The present application provides a spraying trajectory optimization method of a spraying robot, the method comprising:

[0006] scanning the object to be sprayed to obtain a target three-dimensional model, and determining the basic control points on the target three-dimensional model;

[0007] determining the newly added control points between adjacent basic control points using the curvature parameters of the point cloud data between adjacent basic control points;

[0008] determining the target spraying path composed of the basic control points and the preliminary target function value corresponding to the target spraying path using the surface distance between the basic control points;

[0009] A multi-segment Bezier curve is generated using a total sequence of control points formed by the target spraying path and the new control points;

[0010] A final target function value corresponding to the target spraying path is determined using the curvature parameter at the connection between adjacent Bezier curves and the preliminary target function value;

[0011] The final target function value is optimized using a preset path optimization algorithm to obtain an optimal target spraying path.

[0012] Further, the determination of the base control points on the target three-dimensional model comprises:

[0013] A preset spraying width of the target three-dimensional model and a spraying origin in the base control points are determined.

[0014] From the spraying origin, an extended control point in the base control points is determined along a preset direction at a preset spraying width.

[0015] The extended control point is taken as a new spraying origin, and the step of determining an extended control point in the base control points from the spraying origin along a preset direction at a preset spraying width is executed cyclically until the base control points are repeated.

[0016] Further, the determination of the new control points between adjacent base control points using the curvature parameter of the point cloud data between the adjacent base control points comprises:

[0017] The curvature value of the point cloud data between the adjacent base control points is determined, and a corresponding curvature standard deviation is calculated;

[0018] The number of new control points between the adjacent base control points is calculated using the curvature standard deviation.

[0019] Further, the determination of the new control points between adjacent base control points using the curvature parameter of the point cloud data between the adjacent base control points comprises:

[0020] The total curvature sum value of all point cloud data between the adjacent base control points is determined.

[0021] The curvature average value of each pair of adjacent control points between the adjacent base control points is determined using the total curvature sum value and the number of new control points.

[0022] The first curvature sum value from one end of the adjacent base control points to a target control point is determined using the curvature average value; the target control point is between the adjacent base control points.

[0023] The target difference value between the first curvature sum value and the second curvature sum value of the currently traversed point cloud data is determined by traversing all point cloud data between the adjacent base control points.

[0024] Determine the target position of the target point cloud data corresponding to the minimum target difference value, and take the target position as the position of the new control point between the adjacent basic control points.

[0025] Further, the surface distance between the basic control points is used to determine the target spraying path composed of the basic control points and the preliminary target function value corresponding to the target spraying path.

[0026] The surface distance between the basic control points and the number of basic control points are used to sum and establish an exponential function to determine the target spraying path composed of the basic control points and the preliminary target function value corresponding to the target spraying path.

[0027] Further, the total sequence of control points formed by the target spraying path and the new control points is used to generate a multi-segment Bezier curve, which comprises:

[0028] The total sequence of control points formed by the target spraying path and the new control points is combined and distributed according to the sequence order of the total sequence of control points and the preset control point combination number to obtain a multi-segment Bezier curve.

[0029] Further, the curvature parameter at the connection of adjacent Bezier curves and the preliminary target function value are used to determine the final target function value corresponding to the target spraying path, which comprises:

[0030] The curvature difference at the connection of adjacent Bezier curves and the preliminary target function value are used to calculate the final target function value corresponding to the target spraying path.

[0031] Further, the preset path optimization algorithm is used to optimize the final target function value to obtain an optimal target spraying path, which comprises:

[0032] A genetic algorithm is used to randomly generate a plurality of basic control point combinations corresponding to the final target function value as a primary population;

[0033] Based on elite selection, each final target function value is compared to determine a preferred parent population in the primary population;

[0034] Based on the preferred parent population, cross, mutation and repeated elite selection are performed to iteratively optimize to obtain an optimal target spraying path.

[0035] The present application provides a spraying trajectory optimization device of a spraying robot, which is used to realize the spraying trajectory optimization method of the spraying robot as claimed in any one of the above; the device comprises:

[0036] A point cloud generation module is used to scan a target three-dimensional model of a to-be-sprayed object to determine basic control points on the target three-dimensional model; and the curvature parameter of the point cloud data between adjacent basic control points is used to determine new control points between the adjacent basic control points.

[0037] The control point combination module is configured to determine a target spraying path composed of the basic control points and a preliminary target function value corresponding to the target spraying path by using surface distances between the basic control points; generate a plurality of Bezier curves by using a total sequence of control points formed by the target spraying path and the new control points; and determine a final target function value corresponding to the target spraying path by using curvature parameters at connection positions of adjacent Bezier curves and the preliminary target function value.

[0038] The path optimization module is configured to optimize the final target function value by using a preset path optimization algorithm to obtain an optimal target spraying path.

[0039] The present application provides a spraying trajectory optimization system of a spraying robot, which comprises a processor, a memory, and a spraying trajectory optimization program of a spraying robot stored in the memory and executable by the processor.

[0040] The present application has the following advantages:

[0041] Compared with the prior art, the number of control points is selected depending on experience, which may result in too few or too many control points, and adjusting the positions of the control points may easily fall into a local optimum, and the repeated simulation verification and manual adjustment process is time-consuming. The present application first determines all basic control points that can basically cover the object to be sprayed, and then determines the number and positions of new control points according to the surface changes of the object, and then determines the target function value by using all control points on each spraying path composed of the basic control points using a preset optimization algorithm, thereby determining the optimal control path, and finally generating each Bezier curve using the control points on the control path to determine the overall optimal spraying trajectory.

[0042] Specifically, since the basic goal of spraying is to achieve full coverage of the sprayed object, all basic control points that can basically cover the object to be sprayed are first determined. Next, since the final trajectory needs to be determined by a Bezier curve, the number and positions of new control points between all adjacent basic control points are determined according to the properties of the Bezier curve between each two basic control points. Then, the basic control points serve as the overall framework of the path, and the new control points draw the trajectory on this framework. Further, a preset optimization algorithm can be used to optimize the target function value by using all control points on each spraying path composed of the basic control points, thereby determining the optimal control path. Finally, a Bezier curve is generated using a preset number of control points, and the overall optimal spraying trajectory is determined, thereby achieving accurate spraying of the object to be sprayed and achieving better spraying effect. BRIEF DESCRIPTION OF DRAWINGS

[0043] In order to more clearly illustrate the technical solutions and advantages of the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or prior art description. Obviously, the drawings in the following description only constitute some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained without creative labor.

[0044] Figure 1 A step flow chart of a spraying trajectory optimization method of a spraying robot provided by an embodiment of the present application;

[0045] Figure 2 A refinement flow chart of step S1 in a spraying trajectory optimization method of a spraying robot provided by an embodiment of the present application;

[0046] Figure 3 A refinement flow chart of step S2 in a spraying trajectory optimization method of a spraying robot provided by an embodiment of the present application;

[0047] Figure 4 A refinement flow chart of step S2 in a spraying trajectory optimization method of a spraying robot provided by another embodiment of the present application;

[0048] Figure 5 A structural schematic diagram of a hardware running environment of a spraying trajectory optimization system of a spraying robot related to the embodiment scheme of the present application;

[0049] Figure 6 A frame structural schematic diagram of a spraying trajectory optimization device of a spraying robot related to the embodiment scheme of the present application. DETAILED DESCRIPTION

[0050] In order to further illustrate the technical means and effects adopted by the present application to achieve the predetermined invention purpose, the following will combine the drawings and the preferred embodiments to specifically describe the spraying trajectory optimization method of a spraying robot according to the present application, the specific implementation, structure, features and effects of which are described in detail as follows. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. In addition, the specific features, structures or characteristics in one or more embodiments can be combined in any suitable form.

[0051] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the present application belongs.

[0052] The specific scheme of the spraying trajectory optimization method of a spraying robot provided by the present application will be specifically described below with reference to the drawings.

[0053] Embodiment one:

[0054] For the spraying trajectory optimization method of the spraying robot provided by the present application, please refer to Figure 1 , which shows the step flow chart of the spraying trajectory optimization method of the spraying robot provided by an embodiment of the present application.

[0055] The method comprises:

[0056] Step S1, scanning the object to be sprayed to obtain a target three-dimensional model, and determining a base control point on the target three-dimensional model;

[0057] For the object to be sprayed, first, use laser radar scanning technology to accurately scan the surface of the object. The laser radar captures the fine structure and contour of the object surface by emitting a laser beam and measuring its reflection time, thereby providing high-precision point cloud data. Next, for these point cloud data, process and optimize them through computer software to generate an accurate 3D model, i.e., a target three-dimensional model, which can accurately reflect the geometric shape and surface features of the object to be sprayed.

[0058] Specifically, please refer to Figure 2 , the step S1 comprises:

[0059] Step S11, determining a preset spraying width of the target three-dimensional model and a spraying origin in the base control points;

[0060] Step S12, determining an extended control point in the base control points from the spraying origin along a preset direction at a preset spraying width;

[0061] Step S13, taking the extended control point as a new spraying origin, and cyclically executing the step of determining an extended control point in the base control points from the spraying origin along a preset direction at a preset spraying width until the base control point appears repeatedly.

[0062] For the spraying robot, the spraying width is also fixed within a fixed distance from the object to be sprayed. Therefore, in order for the spraying to completely cover the object to be sprayed, the spraying interval must be at least the spraying width. Now, the following steps are used to determine all the base control points that can cover the object to be sprayed:

[0063] First, according to the set spraying distance of the spraying robot to the object to be sprayed, test the spraying width of the spraying robot (the average value of ten experiments can be used) as the preset spraying width of the target three-dimensional model.

[0064] Then, a spray origin is selected as a first base control point on the object to be sprayed. From the spray origin, two extension control points are selected to the right and upward of the object to be sprayed at a preset spray width. Through the extension control points, a spraying path can be gradually generated to ensure that the spraying operation can cover the entire object, and the extension control points are also base control points.

[0065] Next, the base control points extended from the previous step continue to extend to the right and upward at the spray width to form new base control points. In this way, the arrangement of the base control points is like a grid, gradually filling the surface of the object to be sprayed. This process is repeated until the base control points are repeated, that is, all the base control points are obtained.

[0066] When the grid structure of the control points completely covers the object to be sprayed, all the base control points needed are determined. At this time, it is assumed that there are N base control points in total.

[0067] In step S2, the curvature parameters of the point cloud data between adjacent base control points are used to determine the new control points between the adjacent base control points.

[0068] Specifically, in an embodiment, referring to Figure 3 , the step S2 comprises:

[0069] In step S21, the curvature value of the point cloud data between adjacent base control points is determined, and the corresponding curvature standard deviation is calculated.

[0070] In step S22, the number of new control points between adjacent base control points is calculated using the curvature standard deviation.

[0071] For the base control points obtained by the above process, although the basic function of covering the object is achieved, it cannot ensure the optimal coverage effect. The goal of this embodiment is to achieve optimal coverage through Bezier curves. For Bezier curves, when the curvature between two control points changes greatly, it means that the path changes sharply in that section, and there may be a large turn or bend. The shape of the Bezier curve is determined by the control points, and its shape is affected by the relative position and number of control points.

[0072] If the curvature varies significantly between control points, a single Bézier curve may not accurately fit sharp turns or changes in the path because Bézier curves have limited fitting ability at path bends. Therefore, to better capture these complex curvature variations, multiple Bézier curves are needed. In this case, increasing the number of base control points allows for more accurate control of each part of the curve, especially between base control points with significant curvature changes, thus avoiding unnatural curve shapes.

[0073] Based on the above description, the following formula can be constructed to represent the number of new control points required between adjacent basic control points A and B, i.e., the number of new control points: Formula explanation: Where, This indicates the number of new control points needed between adjacent basic control points A and B. This indicates rounding down, and k represents a custom control coefficient (e.g., 0.1). This represents the curvature value of the first newly added control point. This represents the curvature value of the first newly added control point, and so on. This represents the standard deviation of curvature of all point cloud data between adjacent basic control points A and B. Similarly, the number of new control points required between each pair of adjacent basic control points can be calculated.

[0074] It should be noted that each control point is itself point cloud data.

[0075] Specifically, in another embodiment, please refer to Figure 4 Step S2 includes:

[0076] Step S201: Determine the sum of curvature values ​​of all point cloud data between adjacent basic control points;

[0077] Step S202: Using the sum of curvature and the number of new control points, determine the average curvature of each pair of adjacent control points between adjacent basic control points;

[0078] Step S203: Using the average curvature, determine the first curvature and value from one end of an adjacent basic control point to the target control point; the target control point is between adjacent basic control points.

[0079] Step S204: Traverse all point cloud data between adjacent basic control points and determine the target difference between the first curvature sum and the second curvature sum of the currently traversed point cloud data;

[0080] Step S205: Determine the target location of the target point cloud data corresponding to the minimum target difference, and use the target location as the location of the new control point between adjacent basic control points.

[0081] The steps above have determined the number of additional control points needed between each pair of basic control points for better subsequent Bézier curve acquisition. The goal of this step is to precisely determine the locations of these additional control points. A common method is uniform distribution, where control points are evenly distributed between the two endpoints. However, while simple, this method may not accurately capture changes in path curvature, especially when the curve changes drastically. Therefore, this step employs curvature averaging, distributing the additional control points based on the sum of curvatures of all point cloud data along the path, ensuring that the locations of the additional control points better match the curvature characteristics of the curve.

[0082] As described above, to find the position of the i-th new control point (target control point) between adjacent basic control points A and B, the following operations are required:

[0083] First, calculate the sum of curvature of all point cloud data between adjacent basic control points A and B (including both basic control points and newly added control points). The subscript of the summation formula is t=1, and the superscript is N. A-B t represents the sequence number of the point cloud data between adjacent basic control points A and B, N A-B This indicates the number of point cloud data points between adjacent basic control points A and B.

[0084] Furthermore, based on the sum of curvature values and the number of newly added control points Calculate the average curvature of each pair of adjacent control points A and B (including both existing and newly added control points). The average curvature reflects the change in point cloud curvature between each pair of adjacent control points.

[0085] Then, starting from one end (here referring to A or B), for the position of the i-th newly added control point (target control point) between adjacent basic control points A and B, the first curvature sum (value) from one end to the position of the i-th newly added control point needs to be calculated as follows: .

[0086] Finally, starting from one end, traverse all point cloud data and calculate... Obtain the target difference and acquire The h value at that time, where h represents the index of the point cloud data being traversed. This represents the second curvature sum of the point cloud data that has been traversed. When this target difference reaches its minimum, the target position of the corresponding target point cloud data is the position of the newly added control point.

[0087] At this point, the location of the new control point among all adjacent basic control points can be determined.

[0088] Step S3: Using the surface distance between the basic control points, determine the target spraying path composed of the basic control points and the preliminary objective function value corresponding to the target spraying path;

[0089] Specifically, step S3 includes:

[0090] By using the surface distances between basic control points and the number of basic control points, an exponential function is summed and established to determine the target spraying path composed of basic control points and the preliminary objective function value corresponding to the target spraying path.

[0091] Basic control points are fundamental to determining the spraying path (here, the spraying path refers to path planning from ABC, not the specific spraying method). The length of the spraying path directly affects the efficiency of the spraying process. A shorter path means less time for the spraying equipment to perform the task. Spraying time is closely related to overall efficiency; a shorter path reduces the travel time of the spraying equipment, thus improving work efficiency. Furthermore, a shorter path results in a simpler travel path for the spraying equipment, avoiding unnecessary backtracking and waiting, ensuring the equipment operates efficiently. Therefore, optimizing the length of the spraying path not only helps save time but also reduces the burden on the equipment and improves workflow.

[0092] Based on the above description, the following formula can be used to construct the target spraying path composed of basic control points. Preliminary objective function value: Formula explanation: Where, The target spraying path is represented by the basic control points. The initial objective function value is given by exp, where exp represents the exponential function with base e, and N represents the number of basic control points. Indicates the basic control point The surface distance between them (along the object's surface). Here, i represents the index of the base control point.

[0093] It should be noted that, for ease of description, the spraying path here is represented as follows: Alternatively, it can be expressed in the form mentioned above, that is, it can be expressed as .

[0094] Step S4: Using the total sequence of control points formed by the target spraying path and the newly added control points, generate multiple segments of Bézier curves;

[0095] Specifically, step S4 includes:

[0096] Using the total sequence of control points formed by the target spraying path and the newly added control points, multiple Bézier curves are obtained by combining and allocating them according to the sequence order of the total sequence of control points and the preset number of control point combinations.

[0097] For each new control point added along the basic control point path (only new control points on the path are used; new control points at other locations are not considered), the goal is to optimize the spraying quality by generating Bézier curves, thereby improving spraying quality while ensuring spraying efficiency. If the curvature change at the connection point between adjacent Bézier curves is small, i.e., the curves transition smoothly at the connection point, the smoothness of the entire spraying path will be greatly improved. In this way, the spraying equipment can move more smoothly along the path, avoiding vibrations or deviations caused by sharp turns or path changes, thus achieving a uniform and high-quality spraying effect.

[0098] Assuming the target spraying path consists of the basic control points is Furthermore, new control points were added along the paths of these basic control points, resulting in the final total sequence of control points as follows: To optimize coating quality control, these control points can be grouped into sets of four (the preset number of control point combinations can be adjusted according to actual conditions) to generate a Bézier curve, thus obtaining multiple Bézier curve segments.

[0099] Step S5: Using the curvature parameters at the connection points of adjacent Bézier curves and the preliminary objective function value, determine the final objective function value corresponding to the target spraying path;

[0100] Specifically, step S5 includes:

[0101] By using the curvature difference at the connection point of adjacent Bézier curves and the preliminary objective function value, the final objective function value corresponding to the target spraying path is calculated.

[0102] Next, the curvature changes at all Bézier curve junctions are calculated. If these curvature changes are small, it indicates that the junctions between adjacent Bézier curves are very smooth, and the path transition is good. Therefore, not only is the spraying efficiency improved, but the spraying quality is also improved accordingly, achieving the dual optimization goals of efficiency and quality.

[0103] Based on the above description, the following formula can be constructed to represent the target spraying path composed of basic control points. The final objective function value:

[0104] Formula explanation: Where, The target spraying path is represented by the basic control points. The final objective function value, where exp represents the exponential function with base e. This indicates the point on the i-th Bézier curve. curvature, This indicates the point on the (i+1)th Bézier curve. The curvature, where 4 refers to the aforementioned generation of a Bézier curve by grouping every four control points; This represents the curvature difference at the junction of the i-th and (i+1)-th adjacent Bézier curves. The target spraying path is represented by the basic control points. The initial objective function value.

[0105] Step S6: Optimize the final objective function value using a preset path optimization algorithm to obtain the optimal target spraying path.

[0106] Specifically, in one embodiment, step S6 includes:

[0107] A genetic algorithm is used to randomly generate multiple combinations of basic control points corresponding to the final objective function value as the first generation population.

[0108] Based on elite selection, the optimal parent population in the first generation population is determined by comparing the values ​​of each final objective function.

[0109] Based on the selection of the parent population, crossover, mutation and repeated elite selection are performed to iteratively optimize and obtain the optimal target spraying path.

[0110] Based on the final objective function value of the target spraying path composed of basic control points, the optimality of the spraying efficiency and quality of the spraying path can be evaluated. Next, this embodiment optimizes the target spraying path composed of basic control points using a genetic algorithm:

[0111] (1) Randomly generate m different combinations of basic control points (N basic control points are fixed, only the order is different) as the initial population.

[0112] (2) Using the elite selection method, select the 10% with higher objective function values ​​from the current population as the "preferred parent generation" to prepare for the generation of the next generation of optimal target spraying path.

[0113] (3) Based on the preferred parent population, a new combination of basic control points is generated through single-point crossover operation.

[0114] (4) Apply mutation operation to the newly generated spraying path population, set an appropriate mutation rate as needed, and randomly mutate some paths to explore more possible spraying paths.

[0115] (5) Repeat the selection, crossover and mutation steps to improve the spraying path generation by generation until the fitness improvement no longer exceeds 1%. At this point, it is the spraying path composed of the optimal basic control points under the current stage, that is, the optimal target spraying path.

[0116] After determining the optimal target spraying path composed of basic control points, the next step is to use all the basic control points and newly added control points on this spraying path to generate the optimal trajectory of the path (at this point, the path simply refers to, for example, route ABC, and does not involve the trajectory of ABC). It is important to note that only newly added control points on this path can be used; newly added control points on other paths cannot be used to generate Bézier curves.

[0117] First, starting from the origin, four control points are selected, and a Bézier trajectory curve is generated based on these four control points. Next, grouping the control points into sets of four, all basic and newly added control points along the optimal target spraying path are traversed, generating and connecting multiple Bézier curve segments to form a continuous spraying trajectory. The start and end control points of each curve segment are connected to ensure a smooth transition and avoid abrupt turns. Thus, the trajectory composed of multiple Bézier curve segments serves as the motion trajectory of the spraying robot, accurately covering the object to be sprayed and ensuring uniform and precise spraying.

[0118] After generating the Bézier curve for the object to be sprayed by the painting robot, the robot will then perform the spraying operation according to the trajectory of this curve. During this process, it is crucial to ensure that the nozzle maintains a constant distance from the object's surface and remains parallel to it to guarantee the uniformity and coverage of the sprayed coating.

[0119] This solution's precise control minimizes coating waste while avoiding over-coating or uneven spraying, ensuring optimal spraying results and improving spraying quality and efficiency.

[0120] Compared to existing technologies that rely on experience to select the number of control points, resulting in either too few or too many control points, and are prone to getting stuck in local optima when adjusting the position of control points, coupled with the time-consuming process of repeated simulation verification and manual adjustment, this invention first determines all basic control points that can basically cover the object to be sprayed. Then, based on the changes in the object's surface, it determines the number and position of additional control points. Next, it uses a preset optimization algorithm to determine the objective function value through all control points on the spraying path composed of each basic control point, thereby determining the optimal control path. Finally, it uses the control points on the control path to generate each segment of the Bézier curve to determine the overall optimal spraying trajectory.

[0121] Specifically, since the basic goal of spraying is to achieve full coverage of the object, all basic control points that can essentially cover the object to be sprayed are first determined. Next, because the final trajectory needs to be determined using Bézier curves, the number and location of new control points between all adjacent basic control points are determined based on the properties of the Bézier curves between every two basic control points. Subsequently, the basic control points serve as the overall framework of the path, and the new control points are used to draw their trajectories within this framework. Then, a preset optimization algorithm is used to optimize the objective function value using all control points along each spraying path composed of basic control points, thereby determining the optimal control path. Finally, a Bézier curve is generated using a preset number of control points, thus determining the overall optimal spraying trajectory, achieving precise spraying of the object and resulting in better spraying effects. Example

[0122] This invention also proposes a spraying trajectory optimization system for a spraying robot. The spraying trajectory optimization system can be a spraying robot, a programmable logic controller, a computer, a server, or a combination of multiple systems for data processing and computation.

[0123] like Figure 5 As shown, Figure 5 This is a schematic diagram of the hardware operating environment of the spraying trajectory optimization system for the spraying robot involved in the embodiments of the present invention.

[0124] like Figure 5 As shown, the painting trajectory optimization system for the painting robot may include: a processor 1001, such as a CPU; a network interface 1004; a user interface 1003; a memory 1005; and a communication bus 1002. The communication bus 1002 is used to establish communication between these components. The user interface 1003 may include a display or an input unit such as a control panel; optionally, the user interface 1003 may also include a standard wired interface or a wireless interface. The network interface 1004 may optionally include a standard wired interface or a wireless interface (such as a Wi-Fi interface). The memory 1005 may be a high-speed RAM or a stable, non-volatile memory, such as a disk drive. Optionally, the memory 1005 may also be a storage device independent of the aforementioned processor 1001. The memory 1005, as a computer storage medium, may include the painting trajectory optimization program for the painting robot.

[0125] Those skilled in the art in this field can understand. Figure 5 The hardware structure shown does not constitute a limitation on the system and may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0126] Continue to refer to Figure 5 , Figure 5 The memory 1005, which is a computer-readable storage medium, may include an operating device, a user interface module, a network communication module, and a spraying trajectory optimization program for the spraying robot.

[0127] exist Figure 5 In this embodiment, the network communication module is mainly used to connect to the server and can communicate with the server for data; while the processor 1001 can call the spraying trajectory optimization program of the spraying robot stored in the memory 1005 and execute the steps in the above embodiments.

[0128] Based on the hardware structure of the above-described spraying trajectory optimization system for the spraying robot, various embodiments of the spraying trajectory optimization method for the spraying robot of the present invention are implemented.

[0129] In addition, the present invention also provides a spraying trajectory optimization device for a spraying robot, please refer to... Figure 6 The spraying trajectory optimization device for the spraying robot includes:

[0130] The point cloud generation module A10 is used to scan the object to be sprayed to obtain the target 3D model and determine the basic control points on the target 3D model; using the curvature parameters of the point cloud data between adjacent basic control points, new control points between adjacent basic control points are determined.

[0131] The control point combination module A20 is used to determine the target spraying path composed of the basic control points and the preliminary objective function value corresponding to the target spraying path by using the surface distance between the basic control points; to generate multiple Bézier curves by using the total sequence of control points formed by the target spraying path and the newly added control points; and to determine the final objective function value corresponding to the target spraying path by using the curvature parameter at the connection of adjacent Bézier curves and the preliminary objective function value.

[0132] The path optimization module A30 is used to optimize the final objective function value using a preset path optimization algorithm to obtain the optimal target spraying path.

[0133] Furthermore, the point cloud generation module A10 is also used for:

[0134] Determine the preset spraying width and the spraying origin in the basic control points of the target 3D model;

[0135] Starting from the spraying origin, determine the extended control points in the basic control points along the preset direction and with the preset spraying width;

[0136] Using the extended control point as the new spraying origin, the steps of determining the extended control point in the basic control point along a preset direction with a preset spraying width from the spraying origin are repeated until the basic control point appears repeatedly.

[0137] Furthermore, the point cloud generation module A10 is also used for:

[0138] Determine the curvature values ​​of point cloud data between adjacent basic control points and calculate the corresponding standard deviation of curvature;

[0139] The number of new control points between adjacent basic control points is calculated using the standard deviation of curvature.

[0140] Furthermore, the point cloud generation module A10 is also used for:

[0141] Determine the sum of curvature of all point cloud data between adjacent basic control points;

[0142] Using the sum of curvature and the number of new control points, determine the average curvature of each pair of adjacent control points between adjacent basic control points;

[0143] The first curvature and value from one end of an adjacent base control point to a target control point are determined using the average curvature; the target control point is located between the adjacent base control points.

[0144] Traverse all point cloud data between adjacent basic control points and determine the target difference between the first curvature sum and the second curvature sum of the currently traversed point cloud data;

[0145] Determine the target location of the target point cloud data corresponding to the minimum target difference, and use the target location as the location of the new control point between adjacent basic control points.

[0146] Furthermore, the control point combination module A20 is also used for:

[0147] By using the surface distances between basic control points and the number of basic control points, an exponential function is summed and established to determine the target spraying path composed of basic control points and the preliminary objective function value corresponding to the target spraying path.

[0148] Furthermore, the control point combination module A20 is also used for:

[0149] Using the total sequence of control points formed by the target spraying path and the newly added control points, multiple Bézier curves are obtained by combining and allocating them according to the sequence order of the total sequence of control points and the preset number of control point combinations.

[0150] Furthermore, the control point combination module A20 is also used for:

[0151] By using the curvature difference at the connection point of adjacent Bézier curves and the preliminary objective function value, the final objective function value corresponding to the target spraying path is calculated.

[0152] Furthermore, the path optimization module A30 is also used for:

[0153] A genetic algorithm is used to randomly generate multiple combinations of basic control points corresponding to the final objective function value as the first generation population.

[0154] Based on elite selection, the optimal parent population in the first generation population is determined by comparing the values ​​of each final objective function.

[0155] Based on the selection of the parent population, crossover, mutation and repeated elite selection are performed to iteratively optimize and obtain the optimal target spraying path.

[0156] The specific implementation of the spraying trajectory optimization device for the spraying robot of the present invention is basically the same as the various embodiments of the spraying trajectory optimization method for the spraying robot described above, and will not be repeated here.

[0157] Furthermore, the present invention also provides a computer-readable storage medium. The computer-readable storage medium stores a spraying trajectory optimization program for a spraying robot, wherein when the spraying trajectory optimization program for the spraying robot is executed by a processor, it implements the steps of the spraying trajectory optimization method for the spraying robot as described above.

[0158] The method implemented when the spraying trajectory optimization program of the spraying robot is executed can be referred to in various embodiments of the spraying trajectory optimization method of the spraying robot of the present invention, and will not be repeated here.

[0159] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0160] The various embodiments in this specification are described in a progressive manner. For the same or similar parts between the various embodiments, please refer to each other. Each embodiment focuses on describing the differences from the reference embodiments.

[0161] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, apparatus, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0162] The above description is merely a preferred embodiment of the present invention and does not limit the scope of protection of the present invention. Any equivalent structural / method transformations made based on the content of the present invention's specification and drawings under the inventive concept of the present invention, or direct / indirect applications in the related technical fields, are included within the scope of protection of the present invention.

Claims

1. A method for optimizing the spraying trajectory of a spraying robot, characterized in that, The method includes: The object to be sprayed is scanned to obtain a target 3D model, and the basic control points on the target 3D model are determined. By using the curvature parameters of point cloud data between adjacent basic control points, new control points between adjacent basic control points can be determined. By utilizing the surface distances between basic control points, the target spraying path composed of the basic control points and the preliminary objective function value corresponding to the target spraying path are determined; Using the total sequence of control points formed by the target spraying path and the newly added control points, multiple Bézier curves are generated; By using the curvature parameters at the junction of adjacent Bézier curves and the preliminary objective function value, the final objective function value corresponding to the target spraying path is determined; The optimal target spraying path is obtained by optimizing the final objective function value using a preset path optimization algorithm. The determination of the basic control points on the target 3D model includes: Determine the preset spraying width and the spraying origin in the basic control points of the target 3D model; Starting from the spraying origin, determine the extended control points in the basic control points along the preset direction and with the preset spraying width; Using the extended control point as the new spraying origin, the step of determining the extended control point in the basic control point along the preset direction with a preset spraying width starting from the spraying origin is repeated until the basic control point appears repeatedly. The step of determining new control points between adjacent basic control points using the curvature parameters of point cloud data includes: Determine the curvature values ​​of point cloud data between adjacent basic control points and calculate the corresponding standard deviation of curvature; The number of new control points between adjacent basic control points is calculated using the standard deviation of curvature. The step of determining new control points between adjacent basic control points using the curvature parameters of point cloud data includes: Determine the sum of curvature of all point cloud data between adjacent basic control points; Using the sum of curvature and the number of new control points, determine the average curvature of each pair of adjacent control points between adjacent basic control points; The first curvature and value from one end of an adjacent base control point to a target control point are determined using the average curvature; the target control point is located between the adjacent base control points. Traverse all point cloud data between adjacent basic control points and determine the target difference between the first curvature sum and the second curvature sum of the currently traversed point cloud data; Determine the target location of the target point cloud data corresponding to the minimum target difference, and use the target location as the location of the new control point between adjacent basic control points.

2. The method for optimizing the spraying trajectory of a spraying robot according to claim 1, characterized in that, The process of determining the target spraying path composed of the basic control points and the preliminary objective function value corresponding to the target spraying path by utilizing the surface distance between the basic control points includes: By using the surface distances between basic control points and the number of basic control points, an exponential function is summed and established to determine the target spraying path composed of basic control points and the preliminary objective function value corresponding to the target spraying path.

3. The method for optimizing the spraying trajectory of a spraying robot according to claim 1, characterized in that, The process of generating multiple Bézier curves using the total sequence of control points formed by the target spraying path and newly added control points includes: Using the total sequence of control points formed by the target spraying path and the newly added control points, multiple Bézier curves are obtained by combining and allocating them according to the sequence order of the total sequence of control points and the preset number of control point combinations.

4. The method for optimizing the spraying trajectory of a spraying robot according to claim 1, characterized in that, The step of determining the final objective function value corresponding to the target spraying path by using the curvature parameter at the connection point of adjacent Bézier curves and the preliminary objective function value includes: By using the curvature difference at the connection point of adjacent Bézier curves and the preliminary objective function value, the final objective function value corresponding to the target spraying path is calculated.

5. The method for optimizing the spraying trajectory of a spraying robot according to claim 1, characterized in that, The step of optimizing the final objective function value using a preset path optimization algorithm to obtain the optimal target spraying path includes: A genetic algorithm is used to randomly generate multiple combinations of basic control points corresponding to the final objective function value as the first generation population. Based on elite selection, the optimal parent population in the first generation population is determined by comparing the values ​​of each final objective function. Based on the selection of the parent population, crossover, mutation and repeated elite selection are performed to iteratively optimize and obtain the optimal target spraying path.

6. A spraying trajectory optimization device for a spraying robot, characterized in that, The apparatus is used to implement the spraying trajectory optimization method for a spraying robot as described in any one of claims 1 to 5; the apparatus includes: The point cloud generation module is used to scan the object to be sprayed to obtain the target 3D model and determine the basic control points on the target 3D model; using the curvature parameters of the point cloud data between adjacent basic control points, new control points between adjacent basic control points are determined. The control point combination module is used to determine the target spraying path composed of the basic control points and the preliminary objective function value corresponding to the target spraying path by using the surface distance between the basic control points; to generate multiple Bézier curves by using the total sequence of control points formed by the target spraying path and the newly added control points; and to determine the final objective function value corresponding to the target spraying path by using the curvature parameter at the connection of adjacent Bézier curves and the preliminary objective function value. The path optimization module is used to optimize the final objective function value using a preset path optimization algorithm to obtain the optimal target spraying path.

7. A spraying trajectory optimization system for a spraying robot, characterized in that, The system includes a processor, a memory, and a spraying trajectory optimization program for a spraying robot stored in the memory and executable by the processor, wherein when the spraying trajectory optimization program for the spraying robot is executed by the processor, it implements the steps of the spraying trajectory optimization method for a spraying robot as described in any one of claims 1 to 5.

Images