Path planning method and system for building coating robot

Abstract

The application relates to the field of automatic coating technology, in particular to a path planning method and system of a building coating robot, which comprises the following steps: generating a coating path according to coating task information; adopting a constructed path optimization model to perform path optimization on the generated coating path, including: expanding the generated coating path in different ways to generate individual schemes of a plurality of coating paths, and combining the individual schemes into a plurality of initial populations; constructing a plurality of evaluation functions; performing iterative optimization on the plurality of initial populations according to the constructed evaluation functions, selecting a plurality of individual schemes in each population after optimization, combining the individual schemes into a new population, and performing iterative optimization on the new population according to the evaluation functions to select an optimal coating path as the optimized coating path. The scheme can optimize the coating path, generate a coating path most meeting the user's requirements, guarantee the quality of coating, improve the coating efficiency, and reduce the coating cost.

Description

Technical Field

[0001] The present invention relates to the field of automated coating technology, and in particular to a path planning method and system for a building coating robot. Background Art

[0002] A coating robot is an automated device used to evenly apply a layer of paint, glue, or other materials to the surface of a workpiece. It is widely used in various industries, such as electronics manufacturing, automotive manufacturing, printed circuit board (PCB) manufacturing, and architectural decoration. The main function of a coating robot is to ensure uniform coating and consistent thickness, thereby improving production efficiency and product quality.

[0003] With the development of science and technology, the path planning of coating robots is becoming more and more intelligent. They can collect the coating information of the plate through images and coat according to preset rules, such as coating along the edge of the area to be coated, or coating from the center to the outside or from the outside to the inside along a spiral line. However, the preset rules have certain limitations. For complex plates, the coating path may not be the optimal path. For large coating areas and more coating times, the efficiency is low and it will cause waste of coating robot resources.

[0004] The quality of coating path planning directly affects the quality, efficiency and cost of coating. Therefore, there is an urgent need for a path planning method and system for building coating robots that can optimize the coating path, generate a coating path that best meets user needs, ensure coating quality, improve coating efficiency and reduce coating costs. Summary of the Invention

[0005] One of the purposes of the present invention is to provide a path planning method for a building coating robot, which can optimize the coating path, generate a coating path that best meets user needs, ensure coating quality, improve coating efficiency, and reduce coating costs.

[0006] The present invention provides a basic solution 1: a path planning method for a building coating robot, including the following contents:

[0007] Task acquisition step: obtain coating task information;

[0008] Path generation step: generating a coating path according to coating task information;

[0009] Path optimization step: using the constructed path optimization model to optimize the generated coating path;

[0010] Coating execution steps: coating according to the optimized coating path;

[0011] The path optimization step includes: S301, based on the generated coating path, coating task information, and the initial position of the coating robot, expanding the generated coating path in different ways to generate several individual coating path plans, and combining them into several initial groups;

[0012] S302, constructing several evaluation functions;

[0013] S303. According to the constructed evaluation function, several initial groups are iteratively optimized, and several individual solutions are selected from each optimized group to form a new group. According to the evaluation function, the new group is iteratively optimized to select the optimal coating path as the optimized coating path.

[0014] Furthermore, the coating task information includes: coating area information and coating parameters;

[0015] Coating area information, including: shape of the coating area, size of the coating area, and center coordinates of the minimum coating area;

[0016] Coating parameters include: coating movement speed and coating material drying speed.

[0017] Furthermore, the task acquisition step includes:

[0018] An image acquisition device is used to capture an image of the plate to be coated;

[0019] Use image recognition technology to extract the coated area on the plate in the image and identify the shape and size of the coated area;

[0020] Splitting the identified coating area into a plurality of cells according to a minimum coating area of ​​a preset size;

[0021] Set the coating thickness for each cell;

[0022] Establish a spatial coordinate system and mark the center coordinates of each cell as the node of the cell;

[0023] Set coating parameters.

[0024] Furthermore, the path generation step includes:

[0025] S201, taking the initial position of the coating robot as the starting point of the coating path;

[0026] S202, dividing the coating area into n coating sub-areas according to the size of the coating area, and sorting the coating sub-areas according to the distance between the center point of the coating sub-area and the starting point of the coating path; if there are multiple coating sub-areas whose center points are at the same distance from the starting point of the coating path, randomly sorting the coating sub-areas with the same distance;

[0027] S203: For each coating sub-region, any point is used as the starting point of the sub-region coating path, and the nodes other than the starting point are sorted according to their distance from the starting point of the sub-region coating path; if there are multiple nodes with the same distance from the starting point of the sub-region coating path, the nodes with the same distance are randomly encoded in sequence;

[0028] S204: Obtain the node closest to the current node as the next node. If there are multiple nodes with the same and closest distance to the current node, randomly select one node as the next node using a roulette wheel method and add it to the sub-region coating path.

[0029] S206, determining whether the sub-region coating path includes all nodes of the coating sub-region, if so, executing S207; if not, executing S208;

[0030] S207, output the sub-area coating path, and execute S209;

[0031] S208, update the next node to the current node, and execute S204;

[0032] S209 , according to the sorting result of the coating sub-areas, sequentially connect the starting point and end point of the sub-area coating path generated by each coating sub-area, and use the initial position of the coating robot as the starting point of the coating path to generate a coating path.

[0033] Furthermore, the step S301 includes:

[0034] According to the generated coating path and the set coating parameters, it is judged whether the node to be coated is located below a coated coating point. If so, it is judged whether the coating material on the coated coating point at the current time has dried according to the difference between the coating time of the coated coating point and the current time, as well as the drying speed of the coating material. If not, it is determined that the node currently to be coated is an unreasonable path node.

[0035] Eliminate unreasonable path nodes, and randomly generate several unreasonable path node coating paths for unreasonable path nodes. Each unreasonable path node coating path is combined with the coating path after excluding the unreasonable path nodes to form a new coating path. The coating path is used as an individual solution to form the first initial group.

[0036] Extract unreasonable path nodes, and for the previous node of the unreasonable path node, reselect other nodes that are not unreasonable path nodes and are closest to it as the next node to form a new coating path. Then, cross the new coating paths to form several new coating paths. The coating paths are used as individual solutions to form the second initial group.

[0037] Extract unreasonable path nodes, and randomly insert them into the coating path from which the unreasonable path nodes have been removed for multiple times to form several new coating paths. The coating paths are used as individual solutions to form the third initial group.

[0038] Furthermore, the step S302 includes:

[0039] Time evaluation function, used to evaluate the time required to execute the coating path:

[0040] ;

[0041] in is the time at which the function is evaluated, To preset standard time, is the number of nodes, including the node with the initial position of the coating robot as the starting point, is the coating time from node i to node i+1, is the path length from node i to node i+1, is the coating speed, Time for coating equipment setup and other non-coating operations;

[0042] Length evaluation function, used to evaluate the total length of the coating path:

[0043] ;

[0044] in Evaluate the function value for the length, It is the preset standard length;

[0045] Comprehensive evaluation function:

[0046] ;

[0047] in is the comprehensive evaluation value, is the weight constant, is the normalized time function evaluation value, Evaluate the function for the normalized length.

[0048] Furthermore, the step S303 includes:

[0049] S30301, initialize the number of iterations k=1;

[0050] S30302. Calculate the time evaluation function values ​​and length evaluation function values ​​of the individual solutions in the first initial group, the second initial group, and the third initial group according to the constructed evaluation function;

[0051] S30303: Sort the individual solutions in the first initial group, the second initial group, and the third initial group according to the time evaluation function value; and sort the individual solutions in the first initial group, the second initial group, and the third initial group according to the length evaluation function value;

[0052] S30304. Sort the individual solutions in each population according to the time evaluation function value and the length evaluation function value of each individual solution, and select x individual solutions from the initial population using a roulette wheel method; the selection probability is proportional to the magnitude of the time evaluation function value and the length evaluation function value;

[0053] S30305. Perform mutation and crossover on the selected individual solutions, and eliminate unfeasible individual solutions to form two new first groups, two new second groups, and two new third groups, including: a new first group selected based on the time evaluation function value, a new first group selected based on the length evaluation function value, a new second group selected based on the time evaluation function value, a new second group selected based on the length evaluation function value, a new third group selected based on the time evaluation function value, and a new third group selected based on the length evaluation function value;

[0054] S30306. Sort the individual solutions of each group respectively, and sort them from large to small according to the evaluation function value;

[0055] S30307. Obtain the first r individual solutions from each group to form a new group;

[0056] S30308. Perform mutation and crossover on the new comprehensive population, eliminate unfeasible individual solutions, form a new population, and calculate the comprehensive evaluation value of each individual solution based on the comprehensive evaluation function;

[0057] S30308. Determine whether the current number of iterations is equal to the preset number. If so, obtain the individual solution with the largest comprehensive evaluation value as the optimal coating path. If not, merge two new first groups, two new second groups, and two new third groups respectively, update the first initial group, the second initial group, and the third initial group, update the number of iterations k=k+1, and then execute S30302.

[0058] Furthermore, the r is set accordingly according to the ethnic group.

[0059] Furthermore, the step S302 includes:

[0060] Time evaluation function, used to evaluate the time required to execute the coating path:

[0061] ;

[0062] in is the time at which the function is evaluated, is the coating time from node i to node i+1, is the path length from node i to node i+1, is the coating speed, Time for coating equipment setup and other non-coating operations, The time required for the end point of the coating path to return to the starting point;

[0063] Length evaluation function, used to evaluate the total length of the coating path:

[0064] ;

[0065] in The length of the path from the end point of the coating path back to the starting point;

[0066] Comprehensive evaluation function:

[0067] ;

[0068] in is the comprehensive evaluation value, is the weight constant, is the normalized time function evaluation value, Evaluate the function for the normalized length.

[0069] The second purpose of the present invention is to provide a path planning system for a building coating robot, which can optimize the coating path, generate a coating path that best meets user needs, ensure coating quality, improve coating efficiency, and reduce coating costs.

[0070] The present invention provides a second basic solution: a path planning system for a building coating robot, which is used to execute the above-mentioned path planning method for the building coating robot.

[0071] Beneficial effect: After obtaining the coating task information, this solution generates a coating path according to the coating task information, and then uses the constructed path optimization model to optimize the generated coating path, and finally performs coating according to the optimized coating path; compared with coating directly after generating the coating path, the optimized coating path can better meet user needs. Specifically, user needs can be represented by an evaluation function. For example, if the path is optimal or the time is optimal, the corresponding weights can be set. The path optimization in this solution is based on the generated coating path, the coating task information, and the initial position of the coating robot. It is expanded in different ways based on the generated coating path to generate several individual solutions for the coating path and combine them into several initial groups; construct several evaluation functions, which can be constructed specifically according to user needs. According to the constructed evaluation functions, several initial groups are iteratively optimized, and several individual solutions are selected from each optimized group to combine into a new group. According to the evaluation function, the new group is iteratively optimized to select the optimal coating path as the optimized coating path.

[0072] The coating path consists of several nodes, and the number of nodes is very large. Compared with directly randomly generating several individual schemes to form a group, optimizing and combining a large number of nodes to form the optimal coating path, there is no established optimization direction and the optimization efficiency is low. This scheme takes into account the distance between nodes in the initial planning of the coating path, and plans a relatively better coating path as much as possible. Its gene itself has certain advantages, so the generated coating path is used as the basis for generating the initial group. However, if a single individual scheme (generated coating path) is used as the basis and the initial population is derived in a predetermined manner, the randomness of the initial population will be affected, and it is easy to fall into the local optimum during the subsequent optimization process. This scheme does not consider the drying problem of the coating material in combination with the generated coverage path. For cells that require multi-layer coating, there are multiple nodes in the vertical direction, and the distance is close, so it is easy to generate a continuous path and perform multiple coatings on a cell. However, if the coating material is not dry, the coating effect will be greatly affected. Therefore, this scheme identifies unreasonable path nodes by judging whether the coating material on the coating point that has been coated at the current time is dry, and then forms three initial populations by making the unreasonable path nodes form new paths, optimizing the position of the unreasonable path nodes in the path, and randomly changing the position of the unreasonable path nodes. This increases the randomness of the initial population and maintains the superiority of the gene. At the same time, it also solves the problem of the coating material drying affecting the coating effect, thereby ensuring that the optimized coating path is better.

[0073] During the iterative optimization process, multiple populations are optimized from different evaluation directions to form several new populations. Individual solutions are then selected from the new populations to form a new population that covers a variety of situations and guarantees randomness. Optimizing them separately helps the populations quickly find the optimal solution in each direction. After the combination, the advantages of each direction are absorbed and combined into a new population, which selects excellent genes while providing comprehensive coverage to prevent falling into local optimality. When selecting individual solutions from several new populations, the number of solutions selected in each population can be selected according to different needs, so that the optimization direction in the formed new population has a tendency, which is more conducive to quickly optimizing the optimal individual solution in the required direction. The new population is iteratively optimized through a comprehensive evaluation function to select the optimal individual solution. The evaluation function performs a comprehensive evaluation to ensure the comprehensiveness and accuracy of the optimization.

[0074] In summary, this solution can optimize the coating path, generate a coating path that best meets user needs, ensure coating quality, improve coating efficiency, and reduce coating costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0075] Figure 1 This is a flow chart of an embodiment of a path planning method for a building coating robot according to the present invention;

[0076] Figure 2 This is a logic block diagram of an embodiment of a path planning system for a building coating robot according to the present invention. DETAILED DESCRIPTION

[0077] The following is further described in detail through specific implementation methods:

[0078] Example 1

[0079] This embodiment is basically as shown in the attached Figure 1 Figure 2 shows a path planning method for a building coating robot, including the following:

[0080] Task acquisition step: obtain coating task information;

[0081] The coating task information includes: coating area information and coating parameters;

[0082] Coating area information, including: shape of the coating area, size of the coating area, and center coordinates of the minimum coating area;

[0083] Coating parameters include: coating movement speed and coating material drying speed.

[0084] Specifically, an image acquisition device is used to capture an image of the plate to be coated;

[0085] Use image recognition technology to extract the coated area on the plate in the image and identify the shape and size of the coated area;

[0086] The identified coating area is divided into a plurality of cells according to a minimum coating area of ​​a preset size. In other embodiments, for an irregular coating area, a grid consisting of a minimum coating area of ​​a preset size can be used to cover and divide the area into a plurality of cells, and the coating area in the cell may not be completely filled. During the coating process, the coating robot can first identify the coating area in the cell, adjust the coating range accordingly, and then perform coating.

[0087] Set the coating thickness of each cell. When setting it specifically, you can select an area in the identified coating area and then set the coating thickness of the selected area;

[0088] Establish a spatial coordinate system and mark the center coordinates of each cell as the node of the cell;

[0089] Set the coating parameters, where the coating movement speed is generally set in combination with the drying speed of the coating material. If the coating material is not dried before the second coating, it will cause problems in the first coating.

[0090] Path generation step: generating a coating path according to coating task information;

[0091] Specifically, a coating path is randomly generated based on the nodes, or a coating path is generated by starting from a certain node and traversing other nodes in a clockwise or counterclockwise rotation. However, the coating path optimization generated in this way only generates an executable path and has no advantages. In addition, the optimization is more difficult. Therefore, a new coating path method is constructed in this embodiment to generate a better coating path. Whether it is directly adopted or used after optimization, it is better than the existing direct generation. The specific content is as follows:

[0092] S201, taking the initial position of the coating robot as the starting point of the coating path;

[0093] S202, dividing the coating area into n coating sub-areas according to the size of the coating area, and sorting the coating sub-areas according to the distance between the center point of the coating sub-area and the starting point of the coating path; if there are multiple coating sub-areas whose center points are at the same distance from the starting point of the coating path, randomly sorting the coating sub-areas with the same distance; wherein the coating area is divided mainly by judging the size of the coating area, and if the size of the coating area belongs to a preset size interval, deciding whether to divide it, and dividing it into several coating sub-areas, and adopting a corresponding setting of division method, such as dividing it into several coating sub-areas with the same area size;

[0094] S203: For each coating sub-region, any point is used as the starting point of the sub-region coating path, and other nodes except the starting point are sorted according to the distance from the node to the starting point of the sub-region coating path; if there are multiple nodes with the same distance from the starting point of the sub-region coating path, the nodes with the same distance are randomly encoded in sequence; in this embodiment, any corner point is used as the starting point of the sub-region coating path;

[0095] S204: Obtain the node closest to the current node as the next node. If there are multiple nodes with the same and closest distance to the current node, randomly select one node as the next node using a roulette wheel method and add it to the sub-region coating path.

[0096] S206, determining whether the sub-region coating path includes all nodes of the coating sub-region, if so, executing S207; if not, executing S208;

[0097] S207, output the sub-area coating path, and execute S209;

[0098] S208, update the next node to the current node, and execute S204;

[0099] S209 , according to the sorting result of the coating sub-areas, sequentially connect the starting point and end point of the sub-area coating path generated by each coating sub-area, and use the initial position of the coating robot as the starting point of the coating path to generate a coating path.

[0100] Path optimization step: using the constructed path optimization model to optimize the generated coating path;

[0101] The specific steps are as follows:

[0102] S301, based on the generated coating path, coating task information, and the initial position of the coating robot, the generated coating path is expanded in different ways to generate several individual coating path plans, which are combined into several initial groups;

[0103] Specifically, based on the generated coating path and the set coating parameters, it is determined whether the node to be coated is located directly below a coated coating point (i.e., coating material exists). If so, based on the difference between the coating time of the coated coating point and the current time, and the drying speed of the coating material, it is determined whether the coating material on the coated coating point at the current time has dried. If not, the node currently to be coated is determined to be an unreasonable path node.

[0104] Eliminate unreasonable path nodes, and randomly generate several unreasonable path node coating paths for unreasonable path nodes. Each unreasonable path node coating path is combined with the coating path after excluding the unreasonable path nodes to form a new coating path. The coating path is used as an individual solution to form the first initial group.

[0105] Extract unreasonable path nodes, and for the previous node of the unreasonable path node, reselect other nodes that are not unreasonable path nodes and are closest to it as the next node to form a new coating path. Then, cross the new coating paths to form several new coating paths. The coating paths are used as individual solutions to form the second initial group.

[0106] Extract unreasonable path nodes, and randomly insert unreasonable path nodes into the coating path from which unreasonable path nodes have been removed multiple times to form several new coating paths. The coating paths are used as individual solutions to form the third initial group.

[0107] In this scheme, the coating path is divided into several nodes, and the number of nodes is very large. Compared with directly randomly generating several individual schemes to form a group, optimizing and combining a large number of nodes to form the optimal coating path, there is no established optimization direction and the optimization efficiency is low. This scheme takes into account that the initial planned coating path takes into account the distance between nodes and plans a relatively better coating path as much as possible. Its gene itself has certain advantages, so the generated coating path is used as the basis for generating the initial group. However, if a single individual scheme (generated coating path) is used as the basis and the initial population is derived in a fixed manner, the randomness of the initial population will be affected, and it is easy to fall into local optimality in the subsequent optimization process. Therefore, this scheme does not consider the drying problem of the coating material in combination with the generated coverage path. For cells that require multi-layer coating, there are multiple nodes in the z-axis direction, and the distance is close, so it is easy to generate a continuous path and perform multiple coatings on a cell. However, if the coating material is not dry, the coating effect will be greatly affected. Therefore, this scheme identifies unreasonable path nodes by judging whether the coating material on the coating point that has been coated at the current time is dry, and then forms three initial populations by making the unreasonable path nodes form new paths, optimizing the position of unreasonable path nodes in the path, and randomly changing the position of unreasonable path nodes. This increases the randomness of the initial population and maintains the superiority of the gene. At the same time, it also solves the problem of the coating material drying affecting the coating effect.

[0108] S302, constructing an evaluation function;

[0109] Time evaluation function, used to evaluate the time required to execute the coating path:

[0110] ;

[0111] in is the time at which the function is evaluated, To preset standard time, is the number of nodes, including the node with the initial position of the coating robot as the starting point, is the coating time from node i to node i+1, is the path length from node i to node i+1, is the coating speed, Time for coating equipment setup and other non-coating operations (such as starting, stopping, reversing, etc.);

[0112] Length evaluation function, used to evaluate the total length of the coating path:

[0113] ;

[0114] in Evaluate the function value for the length, It is the preset standard length;

[0115] Comprehensive evaluation function:

[0116] ;

[0117] in is the comprehensive evaluation value, is the weight constant, is the normalized time function evaluation value, Evaluate the function value for the normalized length to achieve uniform dimension;

[0118] S303, iteratively optimizing several initial groups according to the constructed evaluation function, and selecting several individual solutions from each optimized group to form a new group, and iteratively optimizing the new group according to the evaluation function to select the optimal coating path as the optimized coating path;

[0119] Specifically, S30301, initialize the number of iterations k=1;

[0120] S30302. Calculate the time evaluation function values ​​and length evaluation function values ​​of the individual solutions in the first initial group, the second initial group, and the third initial group according to the constructed evaluation function;

[0121] S30303: Sort the individual solutions in the first initial group, the second initial group, and the third initial group according to the time evaluation function value; and sort the individual solutions in the first initial group, the second initial group, and the third initial group according to the length evaluation function value;

[0122] S30304. Sort the individual solutions in each group based on the time evaluation function value and the length evaluation function value of each individual solution, and select x individual solutions from the initial group using a roulette wheel method, as follows:

[0123] For the first, second, and third initial populations, select x individual solutions from the initial populations using a roulette wheel method based on the time evaluation function value of each individual solution. The selection probability is proportional to the value of the time evaluation function.

[0124] For the first, second, and third initial populations, the length evaluation function value of each individual solution is evaluated, and x individual solutions are selected from the initial population using a roulette wheel method; the selection probability is proportional to the length evaluation function value;

[0125] S30305. Perform mutation and crossover on the selected individual solutions, and eliminate unfeasible individual solutions to form two new first groups, two new second groups, and two new third groups, including: a new first group selected based on the time evaluation function value, a new first group selected based on the length evaluation function value, a new second group selected based on the time evaluation function value, a new second group selected based on the length evaluation function value, a new third group selected based on the time evaluation function value, and a new third group selected based on the length evaluation function value;

[0126] S30306. Rank the individual solutions of each ethnic group separately;

[0127] For the new first group selected according to the time evaluation function value, calculate the time evaluation function value of the individual solutions in the first group according to the time evaluation function, and sort them from large to small according to the time evaluation function value;

[0128] For the new second group selected according to the time evaluation function value, calculate the time evaluation function value of the individual solutions in the second group according to the time evaluation function, and sort them from large to small according to the time evaluation function value;

[0129] For the new third group selected based on the time evaluation function value, calculate the time evaluation function value of the individual solutions in the third group based on the time evaluation function, and sort them from large to small based on the time evaluation function value;

[0130] For the new first group selected according to the length evaluation function value, calculate the length evaluation function values ​​of the individual solutions in the first group according to the length evaluation function, and sort them from large to small according to the length evaluation function values;

[0131] For the new second group selected according to the length evaluation function value, calculate the length evaluation function value of the individual solutions in the second group according to the length evaluation function, and sort them from large to small according to the length evaluation function value;

[0132] For the new third group selected according to the length evaluation function value, the length evaluation function values ​​of the individual solutions in the third group are calculated according to the length evaluation function, and the solutions are sorted from large to small according to the length evaluation function values;

[0133] S30307. Obtain the first r individual solutions from each group to form a new group;

[0134] S30308. Perform mutation and crossover on the new comprehensive population, eliminate unfeasible individual solutions, and form a new population. Calculate the comprehensive evaluation value of each individual solution based on the comprehensive evaluation function. r can be set to a different value for each population, depending on the specific needs.

[0135] Among them, individual solutions that cannot be realized are eliminated, that is, individual solutions with unreasonable path nodes are eliminated;

[0136] S30308. Determine whether the current number of iterations is equal to the preset number. If so, obtain the individual solution with the largest comprehensive evaluation value as the optimal coating path. If not, merge two new first groups, two new second groups, and two new third groups respectively, update the first initial group, the second initial group, and the third initial group, update the number of iterations k=k+1, and then execute S30302.

[0137] During the optimization process, the three populations were optimized from the perspectives of time optimization and path optimization, thus forming six new populations. Individual solutions were then selected from the six new populations and combined into a new population that covers a variety of situations and ensures randomness. Optimizing each population individually helps the populations quickly find the optimal solution in each direction. After combining, the advantages of each direction are absorbed and combined into a new population. This not only selects excellent genes but also provides comprehensive coverage and prevents falling into local optimality.

[0138] Furthermore, when selecting individual solutions in the six new groups, the number of solutions selected in each group can be varied based on different needs, thus biasing the optimization direction in the formed new groups and making it easier to quickly optimize the optimal individual solution in the required direction.

[0139] The new population is iteratively optimized through a comprehensive evaluation function to select the optimal individual solution. The comprehensive evaluation function performs a comprehensive evaluation to ensure the comprehensiveness and accuracy of the optimization.

[0140] The first node in the above iterative optimization process of the individual solution is the initial position of the coating robot, without mutation and crossover;

[0141] Coating execution steps: coating is performed according to the optimized coating path.

[0142] This embodiment also provides a path planning system for a building coating robot, which is used to execute the above-mentioned path planning method for the building coating robot and is connected to the coating robot, including:

[0143] Task acquisition module: obtain coating task information;

[0144] Path generation module: generates coating paths according to coating task information;

[0145] Path optimization module: Use the constructed path optimization model to optimize the generated coating path;

[0146] Coating execution module: triggers the coating robot to perform coating according to the optimized coating path.

[0147] Example 2

[0148] This embodiment is basically the same as the above embodiment, except that:

[0149] The S301 further includes:

[0150] According to the nodes, several coating paths are randomly generated. The coating paths are used as individual solutions to form the fourth initial group, thereby increasing the randomness of the initial group, better preventing it from falling into the local optimum, and reducing the impact on the improvement of optimization speed;

[0151] Correspondingly, S30302, according to the constructed evaluation function, calculate the time evaluation function value and the length evaluation function value of the individual solutions in the first initial group, the second initial group, the third initial group, and the fourth initial group;

[0152] S30303. Sort the individual solutions in the first initial group, the second initial group, the third initial group, and the fourth initial group according to the time evaluation function value; and sort the individual solutions in the first initial group, the second initial group, the third initial group, and the fourth initial group according to the length evaluation function value;

[0153] S30304. Sort the individual solutions in each group based on the time evaluation function value and the length evaluation function value of each individual solution, and select x individual solutions from the initial group using a roulette wheel method, as follows:

[0154] For the first, second, third, and fourth initial populations, select x individual solutions from the initial populations using a roulette wheel method based on the time evaluation function value of each individual solution. The selection probability is proportional to the value of the time evaluation function.

[0155] For the first, second, third, and fourth initial populations, the length evaluation function value of each individual solution is evaluated, and x individual solutions are selected from the initial population using a roulette wheel method; the selection probability is proportional to the length evaluation function value;

[0156] S30305. Perform mutation and crossover on the selected individual solutions, and eliminate unfeasible individual solutions to form two new first groups, two new second groups, two new third groups, and two new fourth groups;

[0157] S30306. Sort the individual solutions of each group separately, including

[0158] For the new fourth group selected based on the time evaluation function value, calculate the time evaluation function value of the individual solutions in the first group based on the time evaluation function, and sort them from large to small based on the time evaluation function value;

[0159] For the new fourth group selected based on the length evaluation function value, calculate the length evaluation function values ​​of the individual solutions in the first group based on the length evaluation function, and sort them from large to small based on the length evaluation function values;

[0160] S30307. Obtain the first r individual solutions from each group to form a new group;

[0161] S30308. Perform mutation and crossover on the new comprehensive population, eliminate unfeasible individual solutions, and form a new population. Calculate the comprehensive evaluation value of each individual solution based on the comprehensive evaluation function. r can be set to a different value for each population, depending on the specific needs.

[0162] The comprehensive evaluation function is:

[0163] ;

[0164] in is the comprehensive evaluation value, is the weight constant, is the normalized time function evaluation value, Evaluate the function for the normalized length;

[0165] S30308. Determine whether the current number of iterations is equal to the preset number. If so, obtain the individual solution with the largest comprehensive evaluation value as the optimal coating path. If not, merge two new first groups, two new second groups, two new third groups, and two new fourth groups respectively, update the first initial group, the second initial group, the third initial group, and the fourth initial group, update the number of iterations k=k+1, and then execute S30302.

[0166] This embodiment also provides a path planning system for a building coating robot, which is used to execute the above-mentioned path planning method for the building coating robot.

[0167] Example 3

[0168] This embodiment is basically the same as the above embodiment, except that:

[0169] S302: constructing an evaluation function;

[0170] The time evaluation function is used to evaluate the time required to execute the coating path:

[0171] ;

[0172] in is the time at which the function is evaluated, To preset standard time, is the number of nodes, including the node with the initial position of the coating robot as the starting point, is the coating time from node i to node i+1, is the path length from node i to node i+1, is the coating speed, Time for coating equipment setup and other non-coating operations (such as starting, stopping, reversing, etc.), The time required for the end point of the coating path to return to the starting point;

[0173] Length evaluation function, used to evaluate the total length of the coating path:

[0174] ;

[0175] in Evaluate the function value for the length, For the preset standard length, It is the length of the path from the end point of the coating path back to the starting point. After the coating path is planned, it is generally used for coating the plates on the assembly line. The same path will be repeated to coat the plates. Therefore, the time consumed by the coating robot to reset and the travel path must be considered to ensure the accuracy and comprehensiveness of the evaluation.

[0176] This embodiment also provides a path planning system for a building coating robot, which is used to execute the above-mentioned path planning method for the building coating robot.

[0177] Example 4

[0178] This embodiment is basically the same as the above embodiment, except that:

[0179] The coating execution step further includes: collecting information of the applied coating, such as a coating thickness sensor to monitor coating thickness data;

[0180] It also includes: coating feedback step: dynamically adjust the coating path and coating parameters according to the coating information, such as judging whether the node coating meets the requirements based on the coating thickness data. If not, the coating speed can be adjusted, and nodes can be added to adjust the coating path for re-coating to ensure coating quality.

[0181] This embodiment also provides a path planning system for a building coating robot, which is used to execute the above-mentioned path planning method for the building coating robot.

[0182] Example 5

[0183] This embodiment is basically the same as the above embodiment, except that: a penalty term is set in the comprehensive evaluation function;

[0184] ;

[0185] in is the weight constant of the penalty term, is the penalty function value after normalization;

[0186] Penalty function, used to evaluate the number of times the path passes through the coated set nodes in the coating path;

[0187] ;

[0188]

[0189] From the coating node To the next coating node If the coating path passes through a coated node, and the coating time of the coated node is Less than the drying time threshold , the count is increased by one, thereby increasing the penalty term to avoid affecting the coated points in the optimized coating path.

[0190] Example 6

[0191] This embodiment provides a path planning system for a building coating robot, which is used to execute the path planning method for the building coating robot, including: a control platform and a coating robot, such as Figure 2 As shown;

[0192] A control platform is used to control the coating robot to perform coating according to the coating path;

[0193] Control platform, including: task acquisition module, path generation module, path optimization module, and control module;

[0194] Task acquisition module, used to obtain coating task information;

[0195] A path generation module is used to generate a coating path according to coating task information;

[0196] A path optimization module is used to optimize the generated coating path using the constructed path optimization model;

[0197] The control module is used to control the coating robot to perform coating according to the optimized coating path.

[0198] The above is only an embodiment of the present invention. Common knowledge such as the known specific structures and characteristics in the scheme is not described in detail here. Ordinary technicians in the field are aware of all common technical knowledge in the technical field of the invention before the application date or priority date, can obtain all existing technologies in the field, and have the ability to apply conventional experimental means before that date. Ordinary technicians in the field can improve and implement this scheme in combination with their own abilities under the inspiration given by this application. Some typical known structures or known methods should not become obstacles for ordinary technicians in the field to implement this application. It should be pointed out that for those skilled in the art, without departing from the structure of the present invention, several variations and improvements can be made, which should also be regarded as the scope of protection of the present invention. These will not affect the effect of the implementation of the present invention and the practicality of the patent. The scope of protection required by this application shall be based on the content of its claims, and the specific implementation methods and other records in the specification can be used to interpret the content of the claims.

Claims

1. A path planning method for a building coating robot, characterized in that: Includes the following: Task acquisition step: obtain coating task information; Path generation step: generating a coating path according to coating task information; Path optimization step: using the constructed path optimization model to optimize the generated coating path; Coating execution steps: coating according to the optimized coating path; The path optimization step includes: S301, based on the generated coating path, coating task information, and the initial position of the coating robot, expanding the generated coating path in different ways to generate several individual coating path plans, and combining them into several initial groups; S302, constructing several evaluation functions; S303, iteratively optimizing several initial groups according to the constructed evaluation function, and selecting several individual solutions from each optimized group to form a new group, and iteratively optimizing the new group according to the evaluation function to select the optimal coating path as the optimized coating path; The S301 includes: According to the generated coating path and the set coating parameters, it is determined whether there is a coated coating point below the node to be coated. If so, it is determined whether the coating material on the coated coating point at the current time has dried according to the difference between the coating time of the coated coating point and the current time, as well as the drying speed of the coating material. If not, it is determined that the node currently to be coated is an unreasonable path node; wherein the node is divided into n coating sub-areas according to the size of the coating area; Eliminate unreasonable path nodes, and randomly generate several unreasonable path node coating paths for unreasonable path nodes. Each unreasonable path node coating path is combined with the coating path after excluding the unreasonable path nodes to form a new coating path. The coating path is used as an individual solution to form the first initial group. Extract unreasonable path nodes, and for the previous node of the unreasonable path node, reselect other nodes that are not unreasonable path nodes and are closest to it as the next node to form a new coating path. Then, cross the new coating paths to form several new coating paths. The coating paths are used as individual solutions to form the second initial group. Extract unreasonable path nodes, and randomly insert unreasonable path nodes into the coating path from which unreasonable path nodes have been removed multiple times to form several new coating paths. The coating paths are used as individual solutions to form the third initial group. The S302 includes: Time evaluation function, used to evaluate the time required to execute the coating path: ； in is the time at which the function is evaluated, To preset standard time, is the number of nodes, including the node with the initial position of the coating robot as the starting point, is the coating time from node i to node i+1, , is the path length from node i to node i+1, is the coating speed, Time for coating equipment setup and other non-coating operations; Length evaluation function, used to evaluate the total length of the coating path: ； in Evaluate the function value for the length, It is the preset standard length; Comprehensive evaluation function: ； in is the comprehensive evaluation value, is the weight constant, is the normalized time function evaluation value, Evaluate the function for the normalized length; The S303 includes: S30301, initialize the number of iterations k=1; S30302. Calculate the time evaluation function values ​​and length evaluation function values ​​of the individual solutions in the first initial group, the second initial group, and the third initial group according to the constructed evaluation function; S30303: Sort the individual solutions in the first initial group, the second initial group, and the third initial group according to the time evaluation function value; and sort the individual solutions in the first initial group, the second initial group, and the third initial group according to the length evaluation function value; S30304. Sort the individual solutions in each population according to the time evaluation function value and the length evaluation function value of each individual solution, and select x individual solutions from the initial population using a roulette wheel method; the selection probability is proportional to the magnitude of the time evaluation function value and the length evaluation function value; S30305. Perform mutation and crossover on the selected individual solutions, and eliminate unfeasible individual solutions to form two new first groups, two new second groups, and two new third groups, including: a new first group selected based on the time evaluation function value, a new first group selected based on the length evaluation function value, a new second group selected based on the time evaluation function value, a new second group selected based on the length evaluation function value, a new third group selected based on the time evaluation function value, and a new third group selected based on the length evaluation function value; S30306. Sort the individual solutions of each group respectively, and sort them from large to small according to the evaluation function value; S30307. Obtain the first r individual solutions from each group to form a new group; S30308. Perform mutation and crossover on the new comprehensive population, eliminate unfeasible individual solutions, form a new population, and calculate the comprehensive evaluation value of each individual solution based on the comprehensive evaluation function; S30308. Determine whether the current number of iterations is equal to the preset number. If so, obtain the individual solution with the largest comprehensive evaluation value as the optimal coating path. If not, merge two new first groups, two new second groups, and two new third groups respectively, update the first initial group, the second initial group, and the third initial group, update the number of iterations k=k+1, and then execute S30302.

2. The path planning method for a building coating robot according to claim 1, characterized in that: The coating task information includes: coating area information and coating parameters; Coating area information, including: shape of the coating area, size of the coating area, and center coordinates of the minimum coating area; Coating parameters include: coating movement speed and coating material drying speed.

3. The path planning method for a building coating robot according to claim 2, characterized in that: The task acquisition step includes: An image acquisition device is used to capture an image of the plate to be coated; Use image recognition technology to extract the coated area on the plate in the image and identify the shape and size of the coated area; Splitting the identified coating area into a plurality of cells according to a minimum coating area of ​​a preset size; Set the coating thickness for each cell; Establish a spatial coordinate system and mark the center coordinates of each cell as the node of the cell; Set coating parameters.

4. The path planning method for a building coating robot according to claim 3, characterized in that: The path generation step includes: S201, taking the initial position of the coating robot as the starting point of the coating path; S202, dividing the coating area into n coating sub-areas according to the size of the coating area, and sorting the coating sub-areas according to the distance between the center point of the coating sub-area and the starting point of the coating path; if there are multiple coating sub-areas whose center points are at the same distance from the starting point of the coating path, randomly sorting the coating sub-areas with the same distance; S203: For each coating sub-region, any point is used as the starting point of the sub-region coating path, and the nodes other than the starting point are sorted according to their distance from the starting point of the sub-region coating path; if there are multiple nodes with the same distance from the starting point of the sub-region coating path, the nodes with the same distance are randomly encoded in sequence; S204: Obtain the node closest to the current node as the next node. If there are multiple nodes with the same and closest distance to the current node, randomly select one node as the next node using a roulette wheel method and add it to the sub-region coating path. S206, determining whether the sub-region coating path includes all nodes of the coating sub-region, if so, executing S207; if not, executing S208; S207, output the sub-area coating path, and execute S209; S208, update the next node to the current node, and execute S204; S209 , according to the sorting result of the coating sub-areas, sequentially connect the starting point and end point of the sub-area coating path generated by each coating sub-area, and use the initial position of the coating robot as the starting point of the coating path to generate a coating path.

5. The path planning method for a building coating robot according to claim 1, characterized in that: The r is set accordingly according to the ethnic group.

6. The path planning method for a building coating robot according to claim 1, characterized in that: The S302 includes: Time evaluation function, used to evaluate the time required to execute the coating path: ； in is the time at which the function is evaluated, To preset standard time, is the coating time from node i to node i+1, , is the path length from node i to node i+1, is the coating speed, Time for coating equipment setup and other non-coating operations, The time required for the end point of the coating path to return to the starting point; Length evaluation function, used to evaluate the total length of the coating path: ； in Evaluate the function value for the length, For the preset standard length, The length of the path from the end point of the coating path back to the starting point; Comprehensive evaluation function: ； in is the comprehensive evaluation value, is the weight constant, is the normalized time function evaluation value, Evaluate the function for the normalized length.

7. The path planning system of the building coating robot is characterized by: A method for executing a path planning method for a building coating robot according to any one of claims 1 to 6, connected to the coating robot, comprising: Task acquisition module: obtain coating task information; Path generation module: generates coating paths according to coating task information; Path optimization module: Use the constructed path optimization model to optimize the generated coating path; Coating execution module: triggers the coating robot to perform coating according to the optimized coating path.

Images