Workpiece drilling path generation method and system based on artificial intelligence

Abstract

The invention relates to the technical field of data processing. The invention particularly relates to a workpiece drilling path generation method and system based on artificial intelligence. The method comprises the steps that a minimum action set is obtained according to information of a to-be-machined hole site and information of a drill bit on a drill package; sorting the drilling actions in the minimum action set by using an ant colony algorithm to obtain a first sorting result; sorting the drilling actions in the minimum action set by utilizing a genetic algorithm to obtain a second sorting result; and the total machining duration corresponding to the first sorting result is compared with the total machining duration corresponding to the second sorting result, the sorting result with the smaller total machining duration is selected as the optimal sorting result, and the moving path, corresponding to the optimal sorting result, of the machine head serves as the workpiece drilling path. By means of the workpiece drilling path generation method based on artificial intelligence, the workpiece drilling efficiency can be greatly improved, and the energy consumption of drilling equipment is greatly reduced.

Description

Technical Field

[0001] The present invention generally relates to the field of data processing technology. More specifically, the present invention relates to a method and system for generating a workpiece drilling path based on artificial intelligence. Background Art

[0002] As people's living standards continue to improve, the requirements for living environment and furniture are also increasing. Many developers' newly built residential buildings are delivered as finely decorated houses, and almost all of them use customized furniture of unified standards. Therefore, the customized panel furniture industry has flourished. A panel furniture is often assembled from multiple processed panels. Therefore, when manufacturing panel furniture, it is usually necessary to drill holes on each panel to facilitate the splicing of panels or the installation of other components.

[0003] At present, the method of drilling holes in plates is to use drilling equipment, such as Figure 1 As shown, the existing drilling equipment includes a support frame 3 for horizontally placing the plate, a clamp 4 for fixing the plate when drilling the plate, a plurality of gantries 1 arranged above the support frame, a plurality of machine heads 2 arranged on the gantries, a drill bag arranged on each machine head, a drill bag consisting of multiple groups of drill bits, and different groups of drill bits are used to drill different surfaces of the plate; when drilling, the drill bit is aligned with the corresponding hole position by moving the gantry and the drill bag, thereby completing the drilling action; with the surge in orders from panel furniture manufacturers and the continuous growth of production scale, higher and higher requirements are placed on the accuracy and efficiency of plate drilling. In order to improve the drilling efficiency, the commonly used approach is to optimize the drilling path. The existing optimization method is to first obtain all possible drilling actions based on the hole position information and the drill bit configuration information, and use a greedy algorithm to screen the drilling actions, that is, to give priority to the drilling actions with more corresponding hole positions, but the use of this method will result in the total number of drillings not being reduced to a minimum, and the processing efficiency is still low. Summary of the invention

[0004] In order to solve the technical problem of low drilling efficiency when drilling using the existing drilling path optimization method, the present invention provides solutions in the following aspects.

[0005] In a first aspect, the present invention provides a method for generating a workpiece drilling path based on artificial intelligence, comprising:

[0006] According to the hole position information to be processed and the drill bit information on the drill kit, a minimum action set is obtained; the minimum action set refers to a set of drilling actions with the least number of drilling actions and covering all the hole positions to be processed;

[0007] The drilling actions in the minimum action set are sorted by using an ant colony algorithm so that the total processing time of the drilling equipment head after each drilling action in the minimum action set is completed is the shortest, and a first sorting result is obtained; the drilling actions in the minimum action set are sorted by using a genetic algorithm so that the total processing time of the drilling equipment head after each drilling action in the minimum action set is completed is the shortest, and a second sorting result is obtained;

[0008] Compare the total processing time of the drilling equipment head corresponding to the first sorting result with the total processing time of the drilling equipment head corresponding to the second sorting result, select the sorting result with the smaller total processing time as the optimal sorting result, and use the movement path of the head corresponding to the optimal sorting result as the workpiece drilling path; the total processing time of the drilling equipment head is equal to the sum of the total time of the gantry movement, the total time of the head moving on the gantry, the total time of the clamping hand moving and the total drilling time of the drilling action.

[0009] The beneficial effects are as follows: when obtaining the workpiece drilling path, the method of the present invention obtains the minimum action set, thereby reducing the drilling action on the basis of ensuring that all the holes to be processed are covered, thereby improving the drilling efficiency; after obtaining the minimum action set, the ant colony algorithm and the genetic algorithm are used to sort the various drilling actions, and the best sorting result is selected as the final sorting result, thereby determining the best drilling path, so that after the drilling equipment head completes drilling all the holes, the total processing time of the drilling equipment head is the shortest, thereby further improving the drilling efficiency and reducing the energy consumption of the drilling equipment. Therefore, the use of the workpiece drilling path generation method based on artificial intelligence of the present invention can greatly improve the efficiency of workpiece drilling and greatly reduce the energy consumption of the drilling equipment.

[0010] In one embodiment, for a certain drilling action sorting result, the method for determining the total duration of the clamping hand movement includes:

[0011] Generate a gripper safety zone for each drilling action; the gripper is used to grip the plate to fix it during drilling;

[0012] Using the hand-clamping safety area corresponding to each drilling action, the hand-clamping safety areas of adjacent drilling actions in the drilling action sorting results are continuously intersected to obtain the optimal hand-clamping safety areas.

[0013] The optimal gripper safety area is used as the gripper layout position corresponding to the corresponding drilling action;

[0014] The total time of gripper movement is calculated based on the positions of each gripper and the gripper movement speed.

[0015] The beneficial effect is that by determining the optimal clamping safety area, the clamping hand is set in the optimal clamping safety area during drilling, thereby avoiding collision between the drill bit and the clamping hand, damaging the clamping hand, and ensuring the safety of the drilling action. The number of times the clamping hand position is changed can be reduced, thereby improving processing efficiency.

[0016] In one embodiment, the use of the ant colony algorithm to sort the drilling actions in the minimum action set includes: selecting a starting point, allowing N ants to traverse all nodes respectively, and counting the best results of the ant colony; the pheromone matrix of the ant colony algorithm is of N order, and the nodes that the ants pass through when searching for food are the position coordinates of the machine head of each drilling action; the value of N is the total number of actions in the minimum action set; the position coordinates of the machine head of the drilling action refer to the position coordinates of the machine head when the drilling action is performed; the best result of the ant colony refers to the result corresponding to the ant with the shortest total processing time of the drilling equipment machine head;

[0017] If the best score of the ant colony does not meet the standard, the path of traversing all nodes is changed, the best score of the ant colony is recalculated, and the pheromone matrix is ​​updated; otherwise, the first ranking result is obtained according to the ant crawling path corresponding to the best score.

[0018] The beneficial effect is that when the ant colony algorithm is used to sort the drilling actions in the minimum action set, by sending N ants to search for paths at the same time, the efficiency of obtaining the first sorting result can be greatly improved, thereby improving the efficiency of generating the workpiece drilling path.

[0019] In one embodiment, the sorting of the drilling actions in the minimum action set by using a genetic algorithm comprises:

[0020] According to the minimum action set, an initial population is generated according to a preset generation rule, wherein the initial population includes at least two individuals, one drilling action sorting scheme is one individual, and each drilling action sorting scheme includes all drilling actions; and the fitness value of each individual in the population is calculated;

[0021] Iteratively perform selection, crossover and mutation operations on individuals in the initial population until a termination condition is met; after the iteration is terminated, the individual with the highest fitness is output, and the drilling action sorting scheme corresponding to the individual with the highest fitness is used as the drilling action sorting result.

[0022] In one embodiment, the generation rule includes at least one of the following rules:

[0023] Generation rule 1: the first action among all drilling actions required to fix the workpiece is selected, and then subsequent actions are randomly selected in sequence until all actions are selected;

[0024] Generate rule 2: select the first action among all drilling actions required to fix the workpiece, and then use the greedy algorithm to select subsequent actions in sequence until all actions are selected;

[0025] Generation rule three: randomly order all drilling actions required for the workpiece;

[0026] Generation rule 4: randomly select the first action among all drilling actions required for the workpiece, and then use the greedy algorithm to select subsequent actions in sequence until all actions are selected.

[0027] In one embodiment, obtaining the minimum action set includes:

[0028] Generate a hole position and matching drill bit correspondence table based on the hole position information to be processed and the drill bit information on the drill pack, wherein the hole position and matching drill bit correspondence table is used to characterize all drill bits that are compatible with each hole position to be processed;

[0029] The hole positions and the hole positions with the same drill bag coordinates in the matching drill bit correspondence table are combined into the same drilling action, thereby obtaining an action set table;

[0030] The action set table is simplified using the minimum set cover problem to obtain the minimum action set.

[0031] Its beneficial effects are as follows: when obtaining the minimum action set, the correspondence between the hole position and the matching drill bit is first obtained, and the drilling actions are merged according to the correspondence between the hole position and the matching drill bit, thereby greatly reducing the number of drilling actions, and finally the minimum set cover problem is used to screen the drilling actions and remove redundant drilling actions, thereby further reducing the number of drilling actions and improving the drilling efficiency.

[0032] In one embodiment, simplifying the action set table using the minimum set cover problem includes:

[0033] S701, if only the Yth column of the Mth row of the table is 1, and the other columns are all 0, then the Yth column is a member of the optimal solution, and the column and the row corresponding to the 1 of the column are locked; if all the columns in a column are 1, then the column is a member of the optimal solution; the table is a simplified action-hole position correspondence table obtained according to the action set table;

[0034] S702, if a column contains the largest number of 1s, take the column as one of the optimal solutions, and lock the column and the row corresponding to the 1 in the column;

[0035] S703, repeating steps S701 and S702 until the set consisting of all optimal solutions covers all hole positions, and the set consisting of the optimal solutions is recorded as the minimum action set.

[0036] In one embodiment, the simplified action-hole position correspondence table acquisition method includes:

[0037] Check the action set table for duplicates, and if the action set table contains several actions with exactly the same hole positions, perform a deduplication operation;

[0038] The hole positions to be processed are taken as rows, and the punching actions in the action set table are taken as columns, so as to obtain an initial action-hole position correspondence table. For a certain column of the table, if the hole position corresponding to a certain row of the column is the hole position included in the punching action corresponding to the column, the value of the row is set to 1, otherwise the value of the row is set to 0;

[0039] If all 1s in column M appear in column N, delete column M; if all 1s in row X appear in row Y, delete row Y.

[0040] In one embodiment, generating a table of correspondence between hole positions and matching drill bits includes:

[0041] Read the attribute information of each hole to be processed from the external processing file and generate a hole table. The attribute information of the hole to be processed includes the working surface where the hole is located, the hole type, the hole diameter, the hole depth, and whether a specific tool is specified;

[0042] Read the attribute information of each drill bit from the equipment table to generate a drill bit table, wherein the attribute information of the drill bit includes the drill bit type, the relative position coordinates of the drill bit on the drill bag, the drill bit type, the drill bit working surface, the maximum drilling depth and the travel coordinate limit of the drill bit;

[0043] For each hole to be processed, a drill bit that matches the working surface, hole type, hole diameter and hole depth of the hole is obtained according to the hole table and the drill bit table, thereby generating a corresponding table of hole positions and matching drill bits.

[0044] In a second aspect, the present invention provides an artificial intelligence-based workpiece drilling path generation system, comprising a memory and a processor, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the artificial intelligence-based workpiece drilling path generation method of the present invention is implemented. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] By reading the following detailed description with reference to the accompanying drawings, the above and other objects, features and advantages of the exemplary embodiments of the present invention will become readily understood. In the accompanying drawings, several embodiments of the present invention are shown in an exemplary and non-limiting manner, and the same or corresponding reference numerals represent the same or corresponding parts, wherein:

[0046] Figure 1 is a schematic diagram schematically showing a drilling device in the prior art;

[0047] Figure 2The present invention schematically illustrates the process of the method for generating a workpiece drilling path based on artificial intelligence. Figure 1 ;

[0048] Figure 3 The present invention schematically illustrates the process of the method for generating a workpiece drilling path based on artificial intelligence. Figure 2 ;

[0049] Figure 4 is a schematic diagram schematically showing the sequencing of drilling actions in an embodiment of the present invention;

[0050] Figure 5 It is a schematic diagram schematically showing the structure of a workpiece drilling path generation system based on artificial intelligence according to an embodiment of the present invention. DETAILED DESCRIPTION

[0051] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0052] The specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0053] Embodiment of a method for generating a workpiece drilling path based on artificial intelligence:

[0054] like Figure 2 As shown, the workpiece drilling path generation method based on artificial intelligence of the present invention includes:

[0055] S101, obtaining a minimum action set, specifically: obtaining a minimum action set based on information of the hole positions to be processed and information of the drill bit on the drill kit; the minimum action set refers to a set of drilling actions with the least number of drilling actions and covering all the hole positions to be processed.

[0056] The information of the hole positions to be processed includes the working surface where the hole positions to be processed are located, the hole position type, the hole diameter, the hole depth, and whether a specific tool is specified. The drill information on the drill package includes the drill type, the relative position coordinates of the drill on the drill package, the drill type, the drill working surface, the maximum drilling depth, and the travel coordinate limit of the drill. The drill bits corresponding to each hole position to be processed can be obtained based on the hole position information to be processed and the drill information on the drill package. The drill bits corresponding to the hole positions to be processed refer to the drill bits that can be used to process the hole positions. For example, the drill bits that can be used to process hole position A are drill a, drill b, and drill c, and the drill bits corresponding to hole position A are drill a, drill b, and drill c.

[0057] Usually, a workpiece has 6 sides. There are multiple sets of drill bits on the same drill package. A set of drill bits includes multiple drill bits. Different sets of drill bits are used to process holes on different sides of the workpiece. After determining the drill bits corresponding to each hole to be processed, the hole position combination that can be processed simultaneously under the same drill package can be screened out. One hole position combination corresponds to one drilling action. For example: hole position 1, hole position 2, and hole position 3 of the workpiece can be processed by drill bit a, drill bit b, and drill bit c respectively. Then hole position 1, hole position 2, and hole position 3 can be processed at the same time and merged into one drilling action. During processing, drill bit a, drill bit b, and drill bit c can be drilled at the same time.

[0058] After filtering out the hole combinations that can be processed simultaneously under the same drill package, the combination with the least number of holes and covering all the holes to be processed can be filtered out from all the hole combinations, thereby obtaining the minimum action set. For example, assuming that the hole combinations include: {hole 1, hole 2, hole 3}, {hole 2, hole 3}, {hole 4}, then the combination with the least number of holes and covering all the holes to be processed is: {hole 1, hole 2, hole 3}, {hole 4}.

[0059] S102, obtaining a first sorting result and a second sorting result, specifically: using an ant colony algorithm to sort the drilling actions in the minimum action set so that the total processing time of the drilling equipment head after each drilling action in the minimum action set is completed is the shortest, and obtaining a first sorting result; using a genetic algorithm to sort the drilling actions in the minimum action set so that the total processing time of the drilling equipment head after each drilling action in the minimum action set is completed is the shortest, and obtaining a second sorting result.

[0060] In nature, when ants are looking for food, they release pheromones, which are volatile chemicals, on the paths they pass. Other ants behind them will choose paths based on the concentration of pheromones left by the leading ants on the road. The increase in pheromone concentration will attract more ants to choose this path, forming a positive feedback loop, and eventually all ants will choose the same shortest path. The ant colony algorithm is a bionic algorithm that simulates the foraging behavior of the above ants. It guides other ants to choose paths by simulating the pheromones left by ants in the process of looking for food. The more advanced ants choose the path, the more random it is. The farther the path, the longer it takes. The pheromone evaporates over time. The later ants tend to choose paths with higher pheromone concentrations. This behavior causes the pheromone to gradually accumulate on the shortest path over time, and the algorithm finally obtains the optimal path. The characteristics of the ant colony algorithm include parallelism, self-organization, robustness, positive feedback, etc. These characteristics make the ant colony algorithm perform well in solving combinatorial optimization problems.

[0061] When applying the principle of ant colony algorithm to solve the problem of optimal drilling path of machines, the coordinate position of the machine head of each drilling action needs to be regarded as the waypoint (or city) when the ants are looking for food. According to the different design characteristics of different machines, it is stipulated whether the starting point and end point of ant foraging are fixed or must comply with certain rules, such as whether it is necessary to return to the starting point to form a closed trajectory after processing. Each ant must traverse all cities before the foraging process is considered to be over. During this process, each ant will release pheromones in the cities it passes through. While other ants tend to choose paths based on the concentration of pheromones, they also have a certain degree of random selectivity. Pheromones evaporate according to the set rules. The increase in the accumulated pheromone concentration in the city will attract more ants to choose the city, forming a positive feedback loop, and finally the algorithm obtains an optimal path composed of the city sequence with the highest pheromone concentration. This path is the processing sequence of the optimal drilling action.

[0062] After the dispatched ants complete the path-finding work, if the algorithm termination conditions are not met, the next batch of ants must be dispatched to repeat a new round of path-finding work. This process is called iteration. The iteration termination conditions can be: the number of iterations reaches the maximum number of iterations, the ant colony algorithm reaches the longest algorithm usage time, the best result of the ant colony meets the preset conditions, or the difference between the best results of multiple consecutive iterations is within the set range.

[0063] Genetic algorithms are based on the biological evolution process of natural selection and genetic mechanisms in Darwin's theory of biological evolution, thereby establishing a computational model and simulating the process of natural evolution to continuously produce better offspring, thereby achieving the purpose of searching for the optimal solution to the problem. The algorithm uses mathematical methods and computer simulation operations to convert the problem-solving process into a process similar to the gene crossover and mutation of chromosomes in biological evolution. When solving more complex combinatorial optimization problems, it can usually obtain better optimization results faster than some conventional optimization algorithms. Genetic algorithms have been widely used in combinatorial optimization, machine learning, signal processing, adaptive control, artificial life and other fields.

[0064] When using genetic algorithms to optimize drilling processing paths, each drilling action in the minimum action set needs to be numbered when generating the initial population. Any ordering of all processing actions is a chromosome in the solution set, and a chromosome corresponds to an individual in the population. For example, there are a total of 6 processing actions, numbered 1 to 6, then 325416 is considered a chromosome, and 623145 is also a chromosome.

[0065] The main conditions that the fitness function should meet in its design include: 1. single value, continuous, non-negative, and maximized; 2. reasonable and consistent; 3. small amount of calculation; 4. strong versatility.

[0066] When using genetic algorithms to optimize drilling processing paths, the pros and cons of a path can be judged with different emphases, such as the shortest total distance, the lowest total energy consumption, or the shortest total processing time; the shortest total distance means the shortest processing time and the highest efficiency; in addition, because the energy consumption of gantry movement and drill package movement during machine processing are different, the movement path can be split into gantry travel and drill package travel according to the processing action sequence, and the total processing energy consumption can be calculated separately. The path with the lowest energy consumption is ranked best.

[0067] S103, generating a workpiece drilling path, specifically: comparing the total processing time of the drilling equipment head corresponding to the first sorting result with the total processing time of the drilling equipment head corresponding to the second sorting result, selecting the sorting result with the smaller total processing time as the optimal sorting result, and using the movement path of the head corresponding to the optimal sorting result as the workpiece drilling path; the total processing time of the drilling equipment head is equal to the sum of the total time of the gantry movement, the total time of the head moving on the gantry, the total time of the clamping hand moving and the total drilling time of the drilling action.

[0068] Assuming that the optimal sorting result is drilling action Q, drilling action W, and drilling action E, the movement path of the machine head corresponding to the optimal sorting result is drilling action Q-drilling action W-drilling action E; the time for the machine head to move on the gantry from the initial position to the machine head position corresponding to drilling action Q is a, the time for the gantry to move is b, and the drilling time of drilling action Q is c; the time for the machine head to move from the machine head position corresponding to drilling action Q to the machine head position corresponding to drilling action W is c. The duration of the movement on the gantry is d, the duration of the gantry movement is e, and the drilling time of drilling action W is f; the time when the machine head moves from the machine head position corresponding to drilling action W to the machine head position corresponding to drilling action E, the time when the machine head moves on the gantry is g, the time when the gantry moves is h, and the drilling time of drilling action E is i; after all drilling actions are completed, the total duration of the clamping movement is j, and the total processing time of the corresponding drilling equipment machine head is equal to a+b+c+d+e+f+g+h+i+j.

[0069] Because the workpiece needs to be fixed with a clamp during drilling, if the workpiece may be touched during a drilling action, the clamp position needs to be changed before the drilling action is performed. Changing the clamp position takes time, so the clamp movement time needs to be considered when calculating the total processing time of the drilling equipment head.

[0070] After obtaining the drilling path of the workpiece, the workpiece is drilled according to the drilling path, thereby improving the drilling efficiency and reducing the energy consumption of the drilling equipment.

[0071] The method of the present invention obtains the minimum action set when obtaining the workpiece drilling path, thereby reducing the drilling action on the basis of ensuring that all the holes to be processed are covered, thereby improving the drilling efficiency; after obtaining the minimum action set, the ant colony algorithm and the genetic algorithm are used to sort the various drilling actions, and the best sorting result is selected as the final sorting result, thereby determining the best drilling path, so that after the drilling equipment head completes drilling all the holes, the total processing time of the drilling equipment head is the shortest, thereby further improving the drilling efficiency and reducing the energy consumption of the drilling equipment. Therefore, the use of the workpiece drilling path generation method based on artificial intelligence of the present invention can greatly improve the efficiency of workpiece drilling and greatly reduce the energy consumption of the drilling equipment.

[0072] like Figure 3 As shown, in one embodiment, for a certain drilling action sorting result, the method for determining the total duration of the clamping hand movement includes:

[0073] S201, generating a gripper safety area for each drilling action; the gripper is used to grip the plate to fix it during drilling;

[0074] S202, using the hand-clamping safety area corresponding to each drilling action, continuously calculating the intersection of the hand-clamping safety areas of adjacent drilling actions in the optimal sorting result to obtain each optimal hand-clamping safety area;

[0075] S203, using the optimal gripper safety area as the gripper layout position corresponding to the corresponding drilling action, and calculating the total gripper movement time according to each gripper layout position and gripper movement speed.

[0076] like Figure 4As shown, it is assumed that the workpiece drilling actions include 5: drilling action A1, drilling action A2, drilling action A3, drilling action A4, drilling action A5, and the drilling action sequence is drilling action A1-drilling action A2-drilling action A3-drilling action A4-drilling action A5. In the figure, the circle represents the workpiece drilling action position, and the shadow represents the clamping safety area (the clamping safety area is used to characterize the area where the clamping hand can be set. Setting the clamping hand in the clamping safety area can avoid the collision between the drill bit and the clamping hand). According to the intersection of the hand clamp safety areas of drilling action A1 and drilling action A2, the intersection area B1 can be obtained (the intersection area B1 means that the hand clamp is in the intersection area B1, and the drill bit will never drill into the hand clamp when the machine head performs drilling action A1 and drilling action A2). According to the intersection of the intersection area B1 and the hand clamp safety area of ​​drilling action A3, the intersection area B2 can be obtained (the intersection area B2 means that the hand clamp is in the intersection area B2, and the drill bit will never drill into the hand clamp when the machine head performs drilling action A1, drilling action A2 and drilling action A3, that is, the hand clamp can be arranged in the intersection area B2, so that the hand clamp position does not need to be changed when the machine head performs drilling action A1, drilling action A2 and drilling action A3). According to the intersection of the intersection area B2 and the hand clamp safety area of ​​drilling action A4, the intersection area B3 can be obtained. Figure 4 It can be seen that the intersection area B3 is a blank area (the blank area means that a hand clamping area cannot be found, which allows the machine head to complete drilling actions A1, A2 and A3, and still continue to complete drilling action A4 without changing the hand clamping position. In other words, the number of times the intersection area is a blank area can be used to determine the number of times the hand clamping position is changed, and then the time to change the hand clamp is determined). Therefore, the hand clamping position needs to be changed at this time. At this time, the hand clamping safety area of ​​drilling action A4 can be used as the intersection area B4, and the intersection with the hand clamping safety area of ​​drilling action A5 is obtained to obtain the intersection area B5.

[0077] Depend on Figure 4 It can be seen that by placing the gripper in the intersection area B2, drilling actions A1, A2 and A3 can be performed without changing the gripper position, and by placing the gripper in the intersection area B5, drilling actions A4 and A5 can be performed without changing the gripper position. During the entire workpiece drilling process, the gripper only needs to be changed once, that is, the gripper is moved to the intersection area B5 after the drilling action A3 is completed. The total gripper movement time can be determined based on the gripper movement speed and the path length of the gripper moving from the intersection area B2 to the intersection area B5.

[0078] When performing each drilling action, the clamp is set in the optimal clamp safety area to fix the workpiece, thereby completing the drilling action.

[0079] By determining the optimal clamping safety area, the clamp is set in the optimal clamping safety area during drilling, thereby avoiding collision between the drill bit and the clamp, damaging the clamp, and reducing the number of times the clamp position is changed while ensuring the safety of the drilling action, thereby improving processing efficiency.

[0080] From the above embodiments, it can be known that the iteration termination condition may be: the best performance of the ant colony meets the preset condition. In one embodiment, the use of the ant colony algorithm to sort the punching actions in the minimum action set includes:

[0081] S301, select a starting point, make N ants traverse all nodes respectively, and count the best score of the ant colony; the pheromone matrix of the ant colony algorithm is of N order, and the nodes that the ants pass through when searching for food are the position coordinates of the machine head of each drilling action; the value of N is the total number of actions in the minimum action set; the position coordinates of the machine head of the drilling action refer to the position coordinates of the machine head when the drilling action is performed; the best score of the ant colony refers to the score corresponding to the ant with the shortest total processing time of the drilling equipment head;

[0082] When performing the drilling action, the position of the machine head remains unchanged, and only the drill bit needs to be extended to drill. The machine head position coordinates of the drilling action refer to the position where the machine head of the drilling device should be when performing the corresponding drilling action.

[0083] S302: If the best score of the ant colony does not meet the standard, change the path of traversing all nodes, recalculate the best score of the ant colony, and update the pheromone matrix; otherwise, obtain the first ranking result according to the ant crawling path corresponding to the best score.

[0084] N ants are sent out at the same time to search for paths. The path-finding calculation process of each ant is completely independent. This algorithm is suitable for efficient parallel operation in a multi-core or multi-threaded environment.

[0085] By sending out N ants to search for paths at the same time, the efficiency of obtaining the first sorting result can be greatly improved, thereby improving the efficiency of generating the workpiece drilling path.

[0086] For ant A, its path-finding process includes:

[0087] (1) Select a fixed or random starting point;

[0088] (2) Taking a random number Q1, in response to the random number Q1 being less than a preset value Q0, taking a random number Q2, and selecting the next node using a roulette algorithm based on the random number Q2; in response to the random number Q1 being greater than or equal to the preset value Q0, selecting the next node based on the distance from the current node to each of the remaining nodes and the pheromone amount of each of the remaining nodes;

[0089] Combining the distance from the current node to each of the remaining nodes and the pheromone amount of each of the remaining nodes to select the next node includes calculating the probability of each node being selected. For a certain node, the probability P1 of being selected is calculated as follows:

[0090]

[0091] In the formula, ρ represents the pheromone concentration of the node, and l represents the distance between the node and the current node.

[0092] (3) Matrix pheromone volatilization and local pheromone enhancement;

[0093] Pheromone volatilization: Pheromones volatilize at a speed ratio set by the system at regular intervals.

[0094] Local pheromone enhancement: During the path-finding process, ants leave pheromones at the destination node at every step. The amount of pheromones accumulates according to the set value. The formula is as follows:

[0095] a i,j =a i ′ ,j +(1-Rho)×Tau0(2)

[0096] Where Rho is the local pheromone volatility factor, (1-Rho) is the residual amount, Tau0 is the pheromone increment, and a i,j is the pheromone concentration from node i to node j, a i ′ ,j is the pheromone concentration from node i to node j before the current accumulation.

[0097] (4) In response to ant A completing its traversal of all nodes, it calculates the current path length and ends the path-finding process.

[0098] In one embodiment, the sorting of the punching actions in the minimum action set by using a genetic algorithm comprises:

[0099] S401. Generate an initial population according to the minimum action set and preset generation rules, wherein the initial population includes at least two individuals, one punching action sorting scheme is one individual, and each punching action sorting scheme includes all punching actions; calculate the fitness value of each individual in the population.

[0100] The core of Darwin's theory of evolution is the survival of the fittest. Fitness is the ability of an individual to adapt to the environment. The better the individual, the more offspring he or she will have. The poorer the individual, the more likely he or she will be eliminated in the process of evolution. The fitness function of the genetic algorithm is also called the evaluation function. It is an indicator used to evaluate the quality of each individual in the group. It is evaluated based on the objective function of the problem being solved. The fitness value is calculated based on the fitness function. The design of the fitness function mainly meets the following conditions:

[0101] 1. Single-valued, continuous, non-negative, and maximized.

[0102] 2. Reasonable and consistent.

[0103] 3. Small amount of calculation.

[0104] 4. Strong versatility.

[0105] When using genetic algorithms to optimize drilling processing paths, the pros and cons of a path can be judged with different emphases, such as the shortest total distance (i.e., an efficiency-first evaluation method) or the lowest total energy consumption (i.e., an energy consumption-first evaluation method); the shortest total distance means the shortest processing time and the highest efficiency; in addition, because the energy consumption of gantry movement and drill package movement are different during machine processing, the movement path can be split into gantry travel and drill package travel according to the processing action sequence, and the total processing energy consumption can be calculated separately. The path with the lowest energy consumption is ranked best.

[0106] If the efficiency-first evaluation method is selected, the fitness function expression is:

[0107]

[0108] If the energy consumption priority evaluation method is selected, the fitness function expression is:

[0109]

[0110] In formula (3) to formula (4), f x represents the fitness of an individual, l1 represents the total length of the moving path of the machine head corresponding to the individual, l a represents the total length of the gantry moving path corresponding to the individual, E1 represents the unit energy consumption of the gantry moving, l b It represents the total length of the drilling package moving path, and E2 represents the unit energy consumption of the drilling package moving.

[0111] S402, iteratively performing selection, crossover and mutation operations on individuals in the initial population until a termination condition is met; after the iteration is terminated, the individual with the highest fitness is output, and the punching action sorting scheme corresponding to the individual with the highest fitness is used as the punching action sorting result.

[0112] Selection: The process of selecting excellent individuals in a population for combination, pairing and mating to produce better offspring is called selection. The purpose of selection is to directly pass on the optimized individuals (or solutions) to the next generation or to generate new individuals through pairing and crossover and then pass on to the next generation. The selection operation is based on the fitness evaluation of individuals in the population. Commonly used selection operators are as follows: fitness ratio method, random traversal sampling method, and local selection method. The selection process can be relatively flexible. The best chromosomes can be paired with each other in a way that is well-matched. At the same time, inbreeding should be avoided, which will lead to the convergence of offspring and the inability to be better. Sometimes, in order to prevent the current optimal chromosome from being destroyed, it can be directly arranged to enter the next generation of evolution.

[0113] The core role in the biological evolution process in nature is the recombination (i.e., mutation) of biological genetic genes. Similarly, the core role in genetic algorithms is the crossover operator of genetic operations. The so-called crossover refers to the operation of replacing and recombining the partial structures of two parent individuals to generate new individuals. Through crossover, the search ability of genetic algorithms has been greatly improved. In the long time since the emergence of genetic algorithms, domestic and foreign scientific researchers have been continuously studying crossover operators, with the aim of how to perform chromosome gene crossover for the same or similar problems, so that the algorithm can converge to the local optimal result faster and better. At present, the better crossover methods are all based on the statistical summary of the experimental data of classical problems. In this embodiment, the ordered crossover method is selected when solving the drilling path optimization problem.

[0114] The basic content of the mutation operator is to change the gene values ​​at certain loci of the individual strings in the population. Depending on the different individual coding representation methods, the mutation operator includes: real-valued mutation and binary mutation.

[0115] In one embodiment, the generation rule may be any one of the following:

[0116] Generation rule 1: the first action among all the punching actions required to fix the workpiece is selected, and then the subsequent actions are randomly selected in sequence until all the actions are selected;

[0117] Generate rule 2: select the first action among all the punching actions required to fix the workpiece, and then use the greedy algorithm to select the subsequent actions in sequence until all the actions are selected;

[0118] Generation rule three: randomly order all punching actions required for the workpiece;

[0119] Generation rule 4: randomly select the first action among all the punching actions required for the workpiece, and then use the greedy algorithm to select subsequent actions in sequence until all actions are selected.

[0120] The initial population can be quickly generated according to the generation rules in this embodiment.

[0121] In one embodiment, obtaining the minimum action set includes:

[0122] S501, generating a correspondence table between hole positions and adapted drill bits, wherein the correspondence table between hole positions and adapted drill bits is used to characterize all drill bits adapted to each hole position to be processed;

[0123] S502, merging the hole positions with the same drill bag coordinates in the matching drill bit correspondence table into the same drilling action, thereby obtaining an action set table;

[0124] S503: Simplify the action set table using the minimum set cover problem, so as to obtain a minimum action set.

[0125] When obtaining the minimum action set, the correspondence between the hole position and the suitable drill bit is first obtained, and the drilling actions are merged according to the correspondence between the hole position and the suitable drill bit, thereby greatly reducing the number of drilling actions. Finally, the drilling actions are screened according to the minimum set coverage problem, and redundant drilling actions are removed, thereby further reducing the number of drilling actions and improving the drilling efficiency.

[0126] In one embodiment, simplifying the action set table using the minimum set cover problem includes:

[0127] S701, check a single row and a column with all 1s, including: if only the Yth column of the Mth row of the table is 1, and the other columns are all 0, then the Yth column is a member of the optimal solution, and the column and the row corresponding to the 1 of the column are locked; if all the columns in a column are 1, then the column is a member of the optimal solution; the table is a simplified action-hole position correspondence table obtained according to the action set table;

[0128] The actions in the simplified action-hole position correspondence table cover all the holes to be processed and have the least number of actions.

[0129] S702, using a greedy algorithm to find the optimal solution: if a column contains the largest number of 1s, take the column as a member of the optimal solution, and lock the column and the row corresponding to the 1 in the column;

[0130] S703, repeating steps S701 and S702 until the set consisting of all optimal solutions covers all hole positions, and the set consisting of the optimal solutions is recorded as the minimum action set.

[0131] In one embodiment, the simplified method for obtaining the action-hole position correspondence table includes:

[0132] S801, checking the action set table for duplicates, and if the action set table contains several actions with exactly the same hole positions, performing a deduplication operation;

[0133] S802, taking the hole positions to be processed as rows and the punching actions in the action set table as columns, thereby obtaining an initial action-hole position correspondence table, and for a certain column of the table, if the hole positions corresponding to a certain row of the column are the hole positions included in the punching action corresponding to the column, then the value of the row is set to 1, otherwise the value of the row is set to 0;

[0134] For example, assuming there are three punching actions, among which the hole positions to be processed corresponding to punching action 1 are hole 1 and hole 2, the hole positions to be processed corresponding to punching action 2 are hole 1, hole 2 and hole 3, and the hole position to be processed corresponding to punching action 3 is hole 4, then the initial action-hole position correspondence table is as shown in the following table:

[0135] Table 1

[0136] Punching action 1 Punch Action 2 Punch Action 3 Hole 1 1 1 0 Hole 2 1 1 0 Hole 3 0 1 0 Hole 4 0 0 1

[0137] S803, perform column simplification and column simplification on the matrix, including: if all 1s in the Mth column appear in the Nth column, delete the Mth column; if all 1s in the Xth row appear in the Yth row, delete the Yth row.

[0138] In one embodiment, generating a table of correspondence between hole positions and matching drill bits includes:

[0139] S601, reading the attribute information of each hole to be processed from the external processing file and generating a hole table, wherein the attribute information of the hole to be processed includes the working surface where the hole is located, the hole type, the hole diameter, the hole depth, and whether a specific tool is specified;

[0140] S602, reading the attribute information of each drill bit from the equipment table to generate a drill bit table, wherein the attribute information of the drill bit includes the drill bit type, the relative position coordinates of the drill bit on the drill bag, the drill bit type, the drill bit working surface, the maximum drilling depth, and the travel coordinate limit of the drill bit;

[0141] S603, for each hole to be processed, obtain a drill bit that matches the working surface, hole type, hole diameter and hole depth of the hole according to the hole table and the drill bit table, thereby generating a corresponding table of hole positions and matching drill bits.

[0142] In one embodiment, if the drilling equipment has multiple drill packages, the attribute information of the drill bit also includes the drill package number of the drill bit; before simplifying the action set table, it also includes: merging actions that meet the drill package safety spacing into one action.

[0143] In one embodiment, if the drilling equipment has multiple drill kits and multiple gantries, after combining the actions that meet the drill kit safety spacing into one action, it also includes: combining the actions that meet the gantries safety spacing into one action.

[0144] Embodiment of workpiece drilling path generation system based on artificial intelligence:

[0145] The present invention also provides a workpiece drilling path generation system based on artificial intelligence. Figure 5 As shown, the artificial intelligence-based workpiece drilling path generation system includes a processor and a memory, and the memory stores computer program instructions. When the computer program instructions are executed by the processor, the artificial intelligence-based workpiece drilling path generation method in the above artificial intelligence-based workpiece drilling path generation method embodiment is implemented.

[0146] The artificial intelligence-based workpiece drilling path generation system also includes other components familiar to those skilled in the art, such as a communication bus and a communication interface, whose settings and functions are known in the art and will not be described in detail here.

[0147] In the present invention, the aforementioned memory may be any tangible medium containing or storing a program that may be used by or in combination with an instruction execution system, apparatus or device. For example, a computer-readable storage medium may be any suitable magnetic storage medium or magneto-optical storage medium, such as a resistive random access memory RRAM (Resistive Random Access Memory), a dynamic random access memory DRAM (Dynamic Random Access Memory), a static random access memory SRAM (Static Random-Access Memory), an enhanced dynamic random access memory EDRAM (Enhanced Dynamic Random Access Memory), a high-bandwidth memory HBM (High-Bandwidth Memory), a hybrid memory cube HMC (Hybrid Memory Cube), etc., or any other medium that may be used to store the required information and may be accessed by an application, a module, or both. Any such computer storage medium may be part of a device or accessible or connectable to a device. Any application or module described in the present invention may be implemented using computer-readable / executable instructions that may be stored or otherwise maintained by such a computer-readable medium.

[0148] In the description of this specification, "plurality" or "several" means at least two, such as two, three or more, etc., unless otherwise clearly defined.

[0149] Although this specification has shown and described a number of embodiments of the present invention, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Those skilled in the art will conceive of many modifications, changes and alternatives without departing from the ideas and spirit of the present invention. It should be understood that in the practice of the present invention, various alternatives to the embodiments of the present invention described herein may be employed.

Claims

1. A method for generating a workpiece drilling path based on artificial intelligence, characterized in that: include: According to the hole position information to be processed and the drill bit information on the drill kit, a minimum action set is obtained; the minimum action set refers to a set of drilling actions with the least number of drilling actions and covering all the hole positions to be processed; The drilling actions in the minimum action set are sorted by using an ant colony algorithm so that the total processing time of the drilling equipment head after each drilling action in the minimum action set is completed is the shortest, and a first sorting result is obtained; the drilling actions in the minimum action set are sorted by using a genetic algorithm so that the total processing time of the drilling equipment head after each drilling action in the minimum action set is completed is the shortest, and a second sorting result is obtained; Compare the total processing time of the drilling equipment head corresponding to the first sorting result with the total processing time of the drilling equipment head corresponding to the second sorting result, select the sorting result with the smaller total processing time as the optimal sorting result, and use the movement path of the head corresponding to the optimal sorting result as the workpiece drilling path; the total processing time of the drilling equipment head is equal to the sum of the total time of the gantry movement, the total time of the head moving on the gantry, the total time of the clamping hand moving and the total drilling time of the drilling action.

2. The method for generating a workpiece drilling path based on artificial intelligence as claimed in claim 1, characterized in that: For a drilling action sorting result, the method for determining the total time duration of the clamping hand movement includes: Generate a gripper safety zone for each drilling action; the gripper is used to grip the plate to fix it during drilling; Using the hand-clamping safety area corresponding to each drilling action, the hand-clamping safety areas of adjacent drilling actions in the drilling action sorting results are continuously intersected to obtain the optimal hand-clamping safety areas. The optimal gripper safety area is used as the gripper layout position corresponding to the corresponding drilling action, and the total gripper movement time is calculated based on each gripper layout position and gripper movement speed.

3. The method for generating a workpiece drilling path based on artificial intelligence as claimed in claim 1, characterized in that: The method of using the ant colony algorithm to sort the drilling actions in the minimum action set includes: selecting a starting point, respectively making N ants traverse all nodes, and counting the best results of the ant colony; the pheromone matrix of the ant colony algorithm is of N order, and the nodes that the ants pass through when searching for food are the position coordinates of the machine heads of each drilling action; the value of N is the total number of actions in the minimum action set; the position coordinates of the machine head of the drilling action refer to the position coordinates of the machine head when the drilling action is performed; the best result of the ant colony refers to the result corresponding to the ant with the shortest total processing time of the drilling equipment machine head; If the best score of the ant colony does not meet the standard, the path of traversing all nodes is changed, the best score of the ant colony is recalculated, and the pheromone matrix is ​​updated; otherwise, the first ranking result is obtained according to the ant crawling path corresponding to the best score.

4. The method for generating a workpiece drilling path based on artificial intelligence as claimed in claim 1, characterized in that: The using of a genetic algorithm to sort the drilling actions in the minimum action set comprises: According to the minimum action set, an initial population is generated according to a preset generation rule, wherein the initial population includes at least two individuals, one drilling action sorting scheme is one individual, and each drilling action sorting scheme includes all drilling actions; and the fitness value of each individual in the population is calculated; Iteratively perform selection, crossover and mutation operations on individuals in the initial population until a termination condition is met; after the iteration is terminated, the individual with the highest fitness is output, and the drilling action sorting scheme corresponding to the individual with the highest fitness is used as the drilling action sorting result.

5. The method for generating a workpiece drilling path based on artificial intelligence as claimed in claim 4, characterized in that: The generation rules include at least one of the following rules: Generation rule 1: the first action among all drilling actions required to fix the workpiece is selected, and then subsequent actions are randomly selected in sequence until all actions are selected; Generate rule 2: select the first action among all drilling actions required to fix the workpiece, and then use the greedy algorithm to select subsequent actions in sequence until all actions are selected; Generation rule three: randomly order all drilling actions required for the workpiece; Generation rule 4: randomly select the first action among all drilling actions required for the workpiece, and then use the greedy algorithm to select subsequent actions in sequence until all actions are selected.

6. The method for generating a workpiece drilling path based on artificial intelligence according to any one of claims 1 to 5, characterized in that: The obtaining of the minimum action set includes: Generate a hole position and matching drill bit correspondence table based on the hole position information to be processed and the drill bit information on the drill pack, wherein the hole position and matching drill bit correspondence table is used to characterize all drill bits that are compatible with each hole position to be processed; The hole positions and the hole positions with the same drill bag coordinates in the matching drill bit correspondence table are combined into the same drilling action, thereby obtaining an action set table; The action set table is simplified using the minimum set cover problem to obtain the minimum action set.

7. The method for generating a workpiece drilling path based on artificial intelligence as claimed in claim 6, characterized in that: The method of simplifying the action set table by using the minimum set cover problem includes: S701, if only the Yth column of the Mth row of the table is 1, and the other columns are all 0, then the Yth column is a member of the optimal solution, and the column and the row corresponding to the 1 of the column are locked; if all the columns in a column are 1, then the column is a member of the optimal solution; the table is a simplified action-hole position correspondence table obtained according to the action set table; S702, if a column contains the largest number of 1s, take the column as one of the optimal solutions, and lock the column and the row corresponding to the 1 in the column; S703, repeating steps S701 and S702 until the set consisting of all optimal solutions covers all hole positions, and the set consisting of the optimal solutions is recorded as the minimum action set.

8. The method for generating a workpiece drilling path based on artificial intelligence as claimed in claim 7, characterized in that: The simplified action-hole position correspondence table acquisition method comprises: Check the action set table for duplicates, and if the action set table contains several actions with exactly the same hole positions, perform a deduplication operation; The hole positions to be processed are taken as rows, and the punching actions in the action set table are taken as columns, so as to obtain an initial action-hole position correspondence table. For a certain column of the table, if the hole position corresponding to a certain row of the column is the hole position included in the punching action corresponding to the column, the value of the row is set to 1, otherwise the value of the row is set to 0; If all 1s in column M appear in column N, delete column M; if all 1s in row X appear in row Y, delete row Y.

9. The method for generating a workpiece drilling path based on artificial intelligence as claimed in claim 6, characterized in that: The generating of the table of correspondence between hole positions and matching drill bits comprises: Read the attribute information of each hole to be processed from the external processing file and generate a hole table. The attribute information of the hole to be processed includes the working surface where the hole is located, the hole type, the hole diameter, the hole depth, and whether a specific tool is specified; Read the attribute information of each drill bit from the equipment table to generate a drill bit table, wherein the attribute information of the drill bit includes the drill bit type, the relative position coordinates of the drill bit on the drill bag, the drill bit type, the drill bit working surface, the maximum drilling depth and the travel coordinate limit of the drill bit; For each hole to be processed, a drill bit that matches the working surface, hole type, hole diameter and hole depth of the hole is obtained according to the hole table and the drill bit table, thereby generating a corresponding table of hole positions and matching drill bits.

10. A workpiece drilling path generation system based on artificial intelligence, comprising a memory and a processor, wherein the memory stores computer program instructions, characterized in that: When the computer program instructions are executed by the processor, the method for generating a workpiece drilling path based on artificial intelligence as described in any one of claims 1 to 9 is implemented.

Images