System and method for optimizing part taking and placing process

Abstract

The invention provides a system and a method for optimizing a part taking and placing process. According to the system, a plurality of parameters input by a user are processed through an optimization calculation module, and solutions obtained after processing are processed by the optimization calculation module to output a plurality of final parameters. The optimization calculation module adopts a gene algorithm and integrates a precisely designed punishment mechanism, so that a user only needs to provide specific constraint conditions and material taking and placing data needing to be optimized, and it is ensured that the optimal machine layout and the optimal taking and placing sequence are found under the given constraint conditions.

Description

Technical Field

[0001] The present invention relates to the technical field of object picking and placing, and in particular to an optimized parts picking and placing process system. Background Art

[0002] There are numerous automated application areas involving object picking and placing, such as the picking and placing of packages in the warehousing and logistics industry, the picking and placing of components in the electronics assembly industry, and the picking and placing and dispensing of medicines in the medical industry. Other fields such as manufacturing, food processing and packaging also have application scenarios for object picking and placing.

[0003] In the field of electronics assembly, printed circuit board (PCB) manufacturing currently requires the use of automated insertion machines, which automatically install electronic components onto the PCB according to a pre-programmed procedure. However, conventional insertion machines rely heavily on manual planning by experts regarding the feeding strategy and the order in which electronic component materials are placed. Regarding the feeding strategy, experts must consider the material placement configuration of the feeder and appropriately select a matching pick-and-place mechanism. This selection involves several mechanical constraints, such as the fact that certain pick-and-place mechanisms can only be used with specific feeders, or that certain insertion points can only be reached by specific pick-and-place mechanisms. Furthermore, because different types of materials may vary in size, a single pick-and-place mechanism can only handle materials of the same type throughout each complete insertion process. Therefore, experts must consider these constraints during the planning process.

[0004] When planning the loading and unloading sequence, experts must pay special attention to the order of loading and unloading and the order of the insertion points, as these factors directly affect the insertion cycle time and overall operational smoothness. After planning, experts must input this information into a simulation environment to calculate the insertion time and select the configuration combination with the best performance based on multiple simulation results. It is important to note that these planning results only apply to specific types of PCBs. Once a different type of PCB is introduced, the point distribution may also differ, requiring experts to re-plan for the new case, a process that is not only time-consuming but also requires considerable effort.

[0005] Therefore, how to find the optimal machine layout and pick-and-place sequence based on the machines and the environment to achieve a more efficient and economical production process has become one of the problems that need to be solved in this field. Summary of the Invention

[0006] In order to achieve a more efficient and economical production process, the present application provides a method and system that can optimize the order and configuration of parts picking and placing based on user constraints. This system includes an optimization algorithm module, which focuses on determining the order and configuration of parts picking and placing based on user-defined constraints and integrates a penalty mechanism. The user only needs to provide specific constraints and the picking and placing material data that needs to be optimized to ensure that the optimal machine layout and picking and placing sequence are found under the given constraints.

[0007] An object of the present invention is to provide a method for optimizing a parts pick-and-place process, which is performed by an optimization algorithm module and includes the following steps:

[0008] (a) receiving a plurality of parameters and at least one constraint, wherein the plurality of parameters and the constraint are related to at least one of a feeding mechanism, a picking mechanism, a material type, and a material picking and placing sequence;

[0009] (b) executing a genetic algorithm based on the parameters and constraints;

[0010] (c) outputting a plurality of final parameters representing an optimized part placement sequence and configuration based on the results obtained after executing the genetic algorithm;

[0011] The genetic algorithm includes: (i) randomly generating a plurality of chromosomes according to the parameter encoding, the plurality of chromosomes forming an initial chromosome pool; (ii) decoding the plurality of chromosomes in the initial chromosome pool according to the at least one constraint; (iii) scoring the decoded plurality of chromosomes to obtain a fitness score, wherein the scoring calculation includes feedback of a penalty value when the decoded chromosome exceeds the constraint; and (iv) selecting the chromosome with the minimum fitness score, and generating a chromosome representing an optimized component placement sequence and configuration through multiple iterations and selections.

[0012] In an embodiment of the present invention, after step (b) and before step (c), the method further includes (b1): simulating actual physical movement of a machine based on the decoded chromosomes to evaluate the production cycle time of the decoded chromosomes.

[0013] In an embodiment of the present invention, after step (c), the method further includes (c1): presenting the multiple final parameters in a graphical form.

[0014] In an embodiment of the present invention, the at least one constraint condition includes a configuration order of the feeding mechanism, a correspondence between the picking mechanism and the pieces to be picked, and a quantity limit of the pieces on the feeding mechanism.

[0015] In an embodiment of the present invention, the score calculation is performed based on M, F, D, weight 1, and weight 2 to obtain the fitness score.

[0016] Where M represents a fixed value for calculating the benefit score, weight 1 represents the penalty value, F represents the number of constraints exceeded, weight 2 represents the distance weight of the decoded chromosome, and D represents the distance.

[0017] In an embodiment of the present invention, the multiple iterations and selection steps include selecting at least two chromosomes with minimum values ​​based on the multiple parameters for mating and mutation to produce new chromosome offspring; calculating scores for the new chromosome offspring, selecting at least two chromosome offspring with minimum values, and again mating and mutation to produce new chromosome grandchildren; and when the number of iterations exceeds a preset value, selecting the chromosome with the minimum value as the chromosome representing the optimized component placement sequence and configuration.

[0018] In an embodiment of the present invention, the preset value may be, for example, a value between 10 and 100. Preferably, the preset value is 20-50, and more preferably 30, which means that the process of mating, mutation, and producing offspring will be repeated 30 times.

[0019] In an embodiment of the present invention, the plurality of parameters include candidate positions of the feeding mechanism, end point information of the picking mechanism, positions of the discharge point, and physical limitations of the actuator.

[0020] In an embodiment of the present invention, the plurality of parameters further include a correspondence between a material taking mechanism and a material piece, a configuration sequence of a specific feeding mechanism, and a configuration quantity range of a specific material piece on the feeding mechanism.

[0021] In an embodiment of the present invention, the multiple final parameters include the total number of feeding mechanisms, material types and feeding mechanism configurations, material picking mechanisms and feeding mechanism configurations, material picking mechanisms and material configurations, material picking and unloading sequence, predicted cycle time, initial material mechanism relationship diagram, feeding position and unloading configurations and usage frequency distribution diagram, material picking and unloading path diagram, and path length distribution diagram.

[0022] Another object of the present invention is to provide a system for optimizing a parts pick-and-place process, comprising an optimization algorithm module configured to process multiple parameters and at least one constraint condition input by a user, and output multiple final parameters representing an optimized parts pick-and-place sequence and arrangement based on the processed results, wherein the multiple parameters and constraints are related to at least one of a feeder mechanism, a picker mechanism, a part type, and a part pick-and-place sequence. The optimization algorithm module comprises:

[0023] an optimization core module configured to execute a genetic algorithm;

[0024] The genetic algorithm includes: (i) randomly generating a plurality of chromosomes according to the parameter encoding, the plurality of chromosomes forming an initial chromosome pool; (ii) decoding the plurality of chromosomes in the initial chromosome pool according to the at least one constraint; (iii) scoring the decoded plurality of chromosomes to obtain a fitness score, wherein the scoring calculation includes feedback of a penalty value when the decoded chromosome exceeds the constraint; and (iv) selecting the chromosome with the minimum fitness score, and generating a chromosome representing an optimized component placement sequence and configuration through multiple iterations and selections.

[0025] In an embodiment of the present invention, the optimization calculation module further includes a data verification module connected to the optimization core module and configured to perform further score calculations on the chromosomes decoded by the optimization core module to accelerate iterative convergence.

[0026] In an embodiment of the present invention, the optimization algorithm module further includes a physics module connected to the optimization core module and configured to simulate actual physical motion of a machine according to the decoded chromosomes to evaluate the production cycle time of the decoded chromosomes.

[0027] In an embodiment of the present invention, the optimization calculation module further includes a visualization module connected to the optimization core module and configured to present the multiple final parameters in a graphical form.

[0028] In an embodiment of the present invention, the at least one constraint condition includes a configuration order of the feeding mechanism, a correspondence between the picking mechanism and the pieces to be picked, and a quantity limit of the pieces on the feeding mechanism.

[0029] In an embodiment of the present invention, the score calculation is performed based on M, F, D, weight 1, and weight 2 to obtain the fitness score.

[0030] Where M represents a fixed value for calculating the benefit score, weight 1 represents the penalty value, F represents the number of constraints exceeded, weight 2 represents the distance weight of the decoded chromosome, and D represents the distance.

[0031] In an embodiment of the present invention, the multiple iterations and selection steps include selecting at least two chromosomes with minimum values ​​based on the multiple parameters for mating and mutation to produce new chromosome offspring; calculating scores for the new chromosome offspring, selecting at least two chromosome offspring with minimum values, and again mating and mutation to produce new chromosome grandchildren; and when the number of iterations exceeds a preset value, selecting the chromosome with the minimum value as the chromosome representing the optimized component placement sequence and configuration.

[0032] In an embodiment of the present invention, the preset value may be, for example, a value between 10 and 100. Preferably, the preset value is 20-50, and more preferably 30, which means that the process of mating, mutation, and producing offspring will be repeated 30 times.

[0033] In an embodiment of the present invention, the plurality of parameters include candidate positions of the feeding mechanism, end point information of the picking mechanism, positions of the discharge point, and physical limitations of the actuator.

[0034] In an embodiment of the present invention, the plurality of parameters further include a correspondence between a material taking mechanism and a material piece, a configuration sequence of a specific feeding mechanism, and a configuration quantity range of a specific material piece on the feeding mechanism.

[0035] In an embodiment of the present invention, the multiple final parameters include the total number of feeding mechanisms, material types and feeding mechanism configurations, material picking mechanisms and feeding mechanism configurations, material picking mechanisms and material configurations, material picking and unloading sequence, predicted cycle time, initial material mechanism relationship diagram, feeding position and unloading configurations and usage frequency distribution diagram, material picking and unloading path diagram, and path length distribution diagram.

[0036] Compared with the related art, the advantages of the present invention are:

[0037] (1) Provide highly intuitive diagrams to help users quickly understand the optimal machine layout and pick-and-place sequence;

[0038] (2) By establishing an actuator model, the system can accurately predict cycle time based solely on calculations, avoiding tedious and time-consuming actual simulations, significantly reducing the computational burden while maintaining prediction accuracy; and

[0039] (3) This system allows users to customize various constraints, which not only increases the flexibility of the optimization process but also ensures that the final results meet actual needs and rules. It also significantly reduces the time cost of experts in machine adjustment and trial and error.

[0040] The aforementioned advantages enable the system of the present invention to significantly reduce the time required for expert planning and personnel to adjust machine configurations, thereby improving overall production efficiency and economic benefits. The technology of the present invention can be applied in a variety of fields. For example, in the field of surface mount technology (SMT), it can be used to optimize the pick-up and placement of components, improving production speed and accuracy; in robotics, it can be used to guide robotic arms or other automatic equipment to perform more efficient and accurate pick-up and placement activities. Different industries and business scenarios can customize and apply the technology of the present invention according to their specific needs. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1A and Figure 1BThey are respectively a flow chart of the system for optimizing the parts pick-and-place process of the present invention and a composition diagram of the optimization calculation module 11 in one embodiment of the system of the present invention.

[0042] Figure 2 This is the iterative flow chart of the genetic algorithm of the present invention.

[0043] Figure 3A and Figure 3B A cross-sectional view of an insert processing device and a three-dimensional view of an insert processing device respectively.

[0044] Figure 4 Displays multiple types of insertable parts and multiple insertion positions of a PCB, with the goal of inserting the multiple types of parts into the multiple insertion positions to minimize the total cycle time for completing the PCB insertion work.

[0045] Figure 5 Schematic diagram of the feeder configuration for optimizing the parts pick-and-place process system planning of the present invention in scenario one.

[0046] Figure 6 Schematic diagram of the feeder configuration planned by the expert in Scenario 1.

[0047] Figure 7 Schematic diagram of the material placement sequence planned by the optimized parts picking and placing process system of the present invention in scenario one.

[0048] Figure 8 Schematic diagram of the material placement sequence planned by the experts in Scenario 1.

[0049] Figure 9 Schematic diagram of the feeder configuration for the system planning of the optimized parts pick-and-place process of the present invention in scenario 2.

[0050] Figure 10 Schematic diagram of the feeder configuration planned by the expert in Scenario 2.

[0051] Figure 11 Schematic diagram of the material placement sequence planned by the optimized parts picking and placing process system of the present invention in scenario 2.

[0052] Figure 12 Schematic diagram of the material placement sequence planned by the experts in Scenario 2.

[0053] Description of reference numerals:

[0054] 11: Optimization calculation module

[0055] 111: Optimizing core modules

[0056] 112: Data verification module

[0057] 113: Physics Module

[0058] 114: Visualization Module

[0059] P1 to P24: Part number

[0060] A to C: Material type DETAILED DESCRIPTION

[0061] The following describes some specific embodiments of the present invention. However, the present invention may be practiced in a variety of different forms without departing from the spirit of the present invention, and the scope of protection of the present invention should not be construed as limited to that set forth in this specification. Furthermore, unless otherwise specified, the use of "a," "an," "the," and similar terms in this specification (especially in the patent claims described below) should be understood to include both the singular and the plural.

[0062] First, please refer to Figure 1A and Figure 1B , which are respectively a flow chart of a system for optimizing a part pick-and-place process according to the present invention and a diagram illustrating the composition of an optimization calculation module 11 in one embodiment of the system according to the present invention. The present invention primarily discloses a system for optimizing a part pick-and-place process, comprising an optimization calculation module 11 configured to process multiple parameters and at least one constraint condition input by a user, and output multiple final parameters representing an optimized part pick-and-place sequence and configuration based on the processed results. The multiple parameters and constraints are related to at least one of a feeding mechanism, a retrieving mechanism, a material type, and a material pick-and-place sequence.

[0063] like Figure 1A and Figure 1B As shown, the optimization calculation module 11 may include an optimization core module 111 , a data verification module 112 , a physics module 113 , and a visualization module 114 .

[0064] The optimization core module 111 is used to execute a genetic algorithm to perform optimization based on multiple parameters and constraints input by the user, wherein the multiple parameters and constraints are related to the material placement information, such as at least one of the feeding mechanism, the material picking mechanism, the material type, and the material placement sequence. The genetic algorithm can find the best solution to the problem by simulating the natural selection process. In this embodiment, the genetic algorithm includes: (i) randomly generating multiple chromosomes based on the parameter encoding, and the multiple chromosomes form an initial chromosome pool; (ii) decoding the multiple chromosomes in the initial chromosome pool based on the at least one constraint; (iii) scoring the decoded multiple chromosomes to obtain a fitness score, and when the decoded chromosome exceeds the constraint, the scoring calculation includes a penalty value feedback; and (iv) selecting the chromosome with the minimum fitness score, and generating the chromosome representing the optimized part placement sequence and configuration through multiple iterations and selections.

[0065] In one embodiment, the step of randomly generating multiple chromosomes based on the parameter encoding is to generate multiple chromosomes by encoding various variables such as the feeding mechanism, the material picking mechanism, the material type, and the material picking and placing order as chromosome genes, thereby generating multiple chromosomes and ultimately forming an initial chromosome pool (or initial population).

[0066] In one embodiment, the at least one constraint condition may include a configuration order of the feeding mechanism, a correspondence between the picking mechanism and the pieces to be picked, and a quantity limit of the pieces on the feeding mechanism.

[0067] The data verification module 112 is connected to the optimization core module 111 and is used to further perform score calculations on the decoded chromosomes to accelerate the speed of iterative convergence. The data verification module 112 reviews the feasibility of the decoded chromosomes in actual applications based on multiple factors. For example, the multiple factors include: a) Feeding mechanism requirements are met: checking whether the number of feeding mechanisms meets the user's needs; b) Quantity limit: verifying whether the materials provided by the feeding mechanism are within the quantity limit of each single material; c) Gripper selection: ensuring that the gripper is only used to grip the materials that can be gripped at the moment; d) Physical constraint compliance: verifying whether the gripper number used for picking up and placing materials complies with the physical limitations of the machine, such as stroke limits.

[0068] The physics module 113 is connected to the optimization core module 111 and is used to simulate the actual physical motion of a machine based on the decoded chromosomes to estimate the production cycle time of the decoded chromosomes. In one embodiment, the physics module 113 includes an actuator model that simulates the dynamic motion of the machine. The process includes: a) coordinate system maintenance: continuously tracking the coordinate information of all feeders, grippers, standby points, etc.; b) dynamic simulation: using the physical model to simulate the motion of each axis and estimate the movement time based on the movement distance; and c) time cost evaluation: comprehensively considering the pick and place process time corresponding to the entire chromosome and feeding this information back to the optimization core module 111.

[0069] In one embodiment, the step of scoring the decoded chromosomes to obtain fitness scores is performed based on M, F, D, weight 1, and weight 2.

[0070] Where M represents a fixed value calculated by the fitness score, weight 1 represents the penalty value, F represents the number of constraints exceeded, weight 2 represents the distance weight of the decoded chromosome, and D represents the distance; and the smaller the fitness score, the fewer constraints violated and the shorter the distance, the better the performance.

[0071] For example, if M is 1000, weight 1 is 10, and weight 2 is 1.5,

[0072] When these chromosomes do not violate any constraints:

[0073] The performance of the first group of chromosomes is: distance 4.5, then the fitness score should be: -1000+10×0+1.5×4.5=-993.25,

[0074] The second set of chromosomes shows a distance of 6.75, so the fitness score should be: -1000+10×0+1.5×6.75=-989.875,

[0075] Since -993.25 is less than -989.875, the first group of chromosomes performs better. Judging from the distance, the first group of chromosomes has a shorter distance.

[0076] When all of these chromosomes violate 1 constraint:

[0077] The third set of chromosomes shows a distance of 2.5, so the fitness score should be: -1000 + 10 × 1 + 1.5 × 2.5 = -986.25

[0078] The fourth set of chromosomes has a distance of 4.5, so the fitness score should be: -1000 + 10 × 1 + 1.5 × 4.5 = -983.25

[0079] The fifth set of chromosomes has a distance of 7.75, so the fitness score should be: -1000 + 10 × 1 + 1.5 × 7.75 = -978.375

[0080] Since -986.25 is the smallest, the third group of chromosomes performs better, and judging from the distance, the third group of chromosomes is also shorter.

[0081] At Figure 2 In a specific embodiment shown, the above-mentioned step of selecting the chromosome with the minimum fitness score can be to select the chromosome with the minimum fitness score from the multiple chromosomes through multiple iterations and selections, wherein the chromosome with the minimum fitness score can refer to the chromosome with the lowest score or two or more chromosomes with lower scores. Specifically, this step can include selecting at least two chromosomes with the minimum fitness scores according to the multiple parameters, mating and mutating each other to produce new chromosome offspring; scoring the new chromosome offspring to obtain fitness scores, selecting at least two chromosome offspring with the minimum fitness scores (i.e., at least two chromosome offspring with relatively the lowest scores), mating and mutating again to produce new chromosome grandchildren; and when the number of iterations exceeds a preset value, taking the chromosome with the minimum score as the chromosome representing the optimized parts placement order and configuration.

[0082] In one embodiment of the present invention, the preset value may be a value between 10 and 100. For example, a value of 100 indicates that the process of mating, mutation, and offspring generation will be repeated 100 times. After 100 iterations, the chromosome with the lowest fitness score is selected as the chromosome representing the optimized component placement sequence and configuration. In other embodiments, the preset value may preferably be between 20 and 50, and more preferably 30, indicating that the process of mating, mutation, and offspring generation will be repeated 30 times. The setting of the preset value depends on the context.

[0083] The visualization module 114 is connected to the optimization core module 111 and is used to present the multiple final parameters in the form of a chart. In one embodiment, the optimization core module 111 can output the chromosome decoding representing the optimized part pick-up and placement sequence and configuration into a JSON format file, and then the visualization module 114 generates a corresponding chart for the user based on the JSON format file. The visualization module 114 can provide a multi-level visual presentation function, and its process includes: a) problem description visualization: clearly showing the relative relationship between the material picking mechanism, the feeding mechanism and the placement point; b) optimization result display: visualizing the material picking and placement sequence and the material picking mechanism used for each trip; and c) feeding mechanism load analysis: providing a visual tool to evaluate the material picking balance of each feeding mechanism for field experts to review and reference.

[0084] In one embodiment, the multiple parameters may include, but are not limited to, candidate positions of the feeding mechanism, end point information of the picking mechanism, the position of the discharge point, physical limitations of the actuator, the correspondence between the picking mechanism and the material, the configuration order of a specific feeding mechanism, and the configuration quantity range of a specific material on the feeding mechanism.

[0085] In one embodiment, the multiple final parameters may include, but are not limited to, the total number of feeding mechanisms, the type of materials and the feeding mechanism configuration, the material picking mechanism and the feeding mechanism configuration, the material picking mechanism and the material configuration, the material picking and unloading sequence, the predicted cycle time, the initial material mechanism relationship diagram, the feeding position and unloading configuration and the usage frequency distribution diagram, the material picking and unloading path diagram, and the path length distribution diagram.

[0086] For example, the specific parameters input according to a specific scenario of the present invention include but are not limited to the following: whether to specify the placement of the feed; the material name (key value) and the upper limit (ub) and lower limit (lb) of the number of material placements, for example, "A": {"lb": 1, "ub": 6}, then material A must be placed at least 1 and at most 6; when the parameter "whether to specify the placement of the feed" is true, the given placement is the main one, and the upper and lower limits are not considered; whether to limit all end tools to only take one type of material in all strokes, if it is false, the end tool can take material A in the first trip and material B in the second trip. ; The restrictions on the gripper picking up materials only work when the parameter "Whether to specify the feeding placement" is true. More than one feasible option can be listed. After the algorithm, the gripper will only select one material. The plug-in position relative to the product coordinates is the position, orientation and plug-in material number, where "index" is the material number, "part" is the material to be inserted in this plug-in position, "x, y, z" is the plug-in point coordinates (based on the PCB origin, unit: mm), "theta" is the plug-in rotation angle (degree), and "theta_pick" is the material picking rotation angle (degree).

[0087] The optimized parts pick-and-place process system of the present invention is used to optimize PCB plug-in production. Please refer to Figure 3A and Figure 3B In this embodiment, a clamp is used as a material picking mechanism and a feeder is used as a material feeding mechanism. Figure 3A This is a cross-sectional view of a plug-in processing device. Figure 3B This is a three-dimensional diagram of a plug-in processing equipment, used to illustrate the order and configuration of the material feeding mechanism and the material taking mechanism. The plug-in processing equipment is equipped with 6 feeding positions (such as Figure 3A 1 to 6 are shown as feeder slots), 4 grippers (such as Figure 3B As shown) and a visual camera. The gripper must move to the visual camera position to detect pin anomalies after taking a round of materials. In this embodiment, different types of materials can be placed on the feeding position, but once selected, the feeding position can only be used for that type of material. The goal is to insert multiple types of materials into the waiting insertion position on a PCB and minimize the total cycle time to complete the PCB insertion work (as shown). Figure 4 shown).

[0088] The following will further illustrate this embodiment using two scenarios: Scenario 1 is a scenario in which each gripper is configured to only take one type of workpiece, and Scenario 2 is a scenario in which one gripper is configured to take multiple types of workpieces.

[0089] Scenario 1: Retrieving a single part

[0090] This is a scenario where the gripper can only pick up one type of part. The same picker mechanism can only pick up the same type of part between runs. For example, if the first run picks up part A, the second run must also pick up part A. In this scenario, there are 24 insert parts, including 6 parts of part A, 6 parts of part B, and 12 parts of part C. A detailed description of the insert parts is shown in Table 1.

[0091] Table 1: Scenario 1 plug-in material description

[0092]

[0093] Figure 5 This is a schematic diagram of the feeder configuration planned for the optimized parts pick-and-place process system of the present invention. Three feed positions are used: the second feed position is for material A, the third feed position is for material C, and the fourth feed position is for material B. Figure 6 This is a schematic diagram of the feeder configuration planned by an expert. Material A is placed at the second feeder position, material B at the third feeder position, material C at the fourth feeder position, and material C at the fifth feeder position. The horizontal axis represents the feeder position, and the vertical axis represents the number of times it has been used. Feeder positions that have been used for material discharge are indicated by the material type in parentheses. First, the user's input data is checked for accuracy. Then, the optimization core module 111 in the system of the present invention executes a genetic algorithm to determine the optimal feeder configuration and material loading and unloading sequence. The data verification module 112 evaluates the feasibility of this solution in practical applications to minimize the total cycle time for the feeder configuration and material loading and unloading sequence.

[0094] The movement of the plug-in processing equipment is simulated by the physical module, as shown in Table 2, and the results are presented in the form of graphs through the visualization module 114 (such as Figure 7 and Figure 8 The placement position is indicated by the gripper to the right (i.e., P1, P2, etc.), and the arrows of different colors represent the first to sixth passes, respectively. The above results show that the planning time and expected cycle time of the optimized part pick-and-place process system of the present invention are both shorter than those of the expert. Compared to the expert's results, the proposed system reduces one feeding position and 7% of the movement time.

[0095] Table 2: Comparison of results for scenario 1

[0096]

[0097] Scenario 2: Retrieving multiple materials

[0098] This is a scenario where the gripper can only pick up one type of part. The same picker mechanism can only pick up the same type of part between runs. For example, if the first run picks up part A, the second run must also pick up part A. In this scenario, there are 24 insert parts, including 6 parts of part A, 6 parts of part B, and 12 parts of part C. A detailed description of the insert parts is shown in Table 3.

[0099] Table 3: Scenario 2 plug-in material description

[0100]

[0101]

[0102] Figure 9 This is a schematic diagram of the feeder configuration planned for the optimized parts pick-and-place process system of the present invention. Five feeder positions are used: material A is placed at the first feeder position, material B is placed at the second feeder position, material B is placed at the third feeder position, material C is placed at the fourth feeder position, and material A is placed at the fifth feeder position. Figure 10 This is a schematic diagram of the feeder configuration planned by the expert. Material B is placed at the second feeder position, material B at the third feeder position, material C at the fourth feeder position, and material A at the fifth feeder position. The horizontal axis represents the feeder position, and the vertical axis represents the number of times it has been used. Feeder positions that have been used for material discharge are indicated by the material type in parentheses. As in scenario one, the user's input data is first checked for accuracy. The optimization core module 111 in the system of the present invention then executes a genetic algorithm to determine the optimal feeder configuration and material loading and unloading sequence. After decoding, the data verification module 111 evaluates the feasibility of the solution in practical applications to minimize the total cycle time for the feeder configuration and material loading and unloading sequence.

[0103] The movement of the plug-in processing equipment is simulated by the physical module 113, as shown in Table 4, and the results are presented in the form of charts by the visualization module 114 (as shown in Table 4). Figure 11 and Figure 12 As shown), the right side of the discharge position will indicate which clamp is used to discharge the material (i.e., P1, P2, etc.), and the arrows of different colors represent the first to third trips respectively. Figure 11 P5 and Figure 12 P5 in Figure 5 has no arrow-marked path because it only places one part. The above results show that the planning time and expected cycle time of the optimized part pick-and-place process system of the present invention are both shorter than those of the expert, and compared to the expert's results, the movement time is reduced by 29%.

[0104] Table 4: Comparison of results of scenario 2

[0105]

[0106] In summary, the optimized parts pick-and-place process system of the present invention can effectively reduce cycle time in most scenarios and significantly reduce the time required for expert planning and personnel to adjust machine configurations, thereby improving overall production efficiency and economic benefits.

[0107] Although the present invention has been disclosed in terms of preferred embodiments, this is not intended to limit the present invention. Anyone skilled in the art will be able to make various changes and modifications without departing from the concept and scope of the present invention. Therefore, the scope of protection of the present invention shall be determined by the scope of the claims.

Claims

1. A method for optimizing a parts pick-and-place process, executed by an optimization algorithm module, comprising the following steps: (a) receiving a plurality of parameters and at least one constraint, wherein the plurality of parameters and the constraint are related to at least one of a feeding mechanism, a picking mechanism, a material type, and a material picking and placing sequence; (b) executing a genetic algorithm based on the plurality of parameters and constraints; (c) outputting a plurality of final parameters representing an optimized part placement sequence and configuration based on the results obtained after executing the genetic algorithm; The genetic algorithm includes: (i) randomly generating a plurality of chromosomes according to the plurality of parameter codes, wherein the plurality of chromosomes form an initial chromosome pool; (ii) decoding the plurality of chromosomes in the initial chromosome pool according to the at least one constraint; (iii) scoring the decoded plurality of chromosomes to obtain fitness scores, wherein the scoring calculation includes feedback of a penalty value when the decoded chromosome exceeds the constraint; and (iv) selecting the chromosome with the minimum fitness score, and generating a chromosome representing the optimized part placement sequence and configuration through multiple iterations and selections.

2. The method of claim 1, wherein after step (b) and before step (c), the method further comprises (b1): simulating actual physical movement of a machine based on the decoded chromosomes to evaluate the production cycle time of the decoded chromosomes. 3 . The method according to claim 1 , wherein after step (c), the method further comprises ( c1 ): presenting the plurality of final parameters in a graphical form.

4. The method according to claim 1, wherein the at least one constraint condition includes a configuration order of the feeding mechanism, a correspondence between the picking mechanism and the parts to be picked, and a quantity limit of the parts on the feeding mechanism.

5. The method of claim 1, wherein the score calculation is performed based on M, F, D, weight 1, and weight 2 to obtain the fitness score, Where M represents a fixed value for calculating the benefit score, weight 1 represents the penalty value, F represents the number of constraints exceeded, weight 2 represents the distance weight of the decoded chromosome, and D represents the distance.

6. The method of claim 1 , wherein the multiple iterations and selection steps include selecting at least two chromosomes with minimum values ​​based on the multiple parameters for mating and mutation to produce new chromosome offspring; calculating scores for the new chromosome offspring, selecting at least two chromosome offspring with minimum values, and again mating and mutation to produce new chromosome grandchildren; and when the number of iterations exceeds a preset value, selecting the chromosome with the minimum value as the chromosome representing the optimized parts placement sequence and configuration.

7. The method of claim 1, wherein the plurality of parameters include candidate positions of a feeding mechanism, end point information of a picking mechanism, a position of a discharge point, and physical limitations of an actuator.

8. The method of claim 7, wherein the plurality of parameters further comprises a correspondence between a material taking mechanism and a material piece, a configuration sequence of a specific material feeding mechanism, and a configuration quantity range of a specific material piece on the material feeding mechanism.

9. The method as claimed in claim 1, wherein the multiple final parameters include the total number of feeding mechanisms, the type of materials and the feeding mechanism configuration, the material picking mechanism and the feeding mechanism configuration, the material picking mechanism and the material configuration, the material picking and unloading sequence, the predicted cycle time, the initial material mechanism relationship diagram, the feeding position and unloading configuration and the usage frequency distribution diagram, the material picking and unloading path diagram, and the path length distribution diagram.

10. A system for optimizing a parts pick-and-place process, comprising an optimization algorithm module configured to process multiple parameters and at least one constraint condition input by a user, and output multiple final parameters representing an optimized parts pick-and-place sequence and arrangement based on the processed results, wherein the multiple parameters and constraints are related to at least one of a feeding mechanism, a retrieving mechanism, a material type, and a material pick-and-place sequence. The optimization algorithm module comprises: an optimization core module configured to execute a genetic algorithm; The genetic algorithm includes: (i) randomly generating a plurality of chromosomes according to the plurality of parameter codes, wherein the plurality of chromosomes form an initial chromosome pool; (ii) decoding the plurality of chromosomes in the initial chromosome pool according to the at least one constraint; (iii) scoring the decoded plurality of chromosomes to obtain fitness scores, wherein the scoring calculation includes feedback of a penalty value when the decoded chromosome exceeds the constraint; and (iv) selecting the chromosome with the minimum fitness score, and generating a chromosome representing the optimized part placement sequence and configuration through multiple iterations and selections.

11. The system of claim 10, wherein the optimization calculation module further comprises a data verification module connected to the optimization core module and configured to perform further score calculation on the chromosomes decoded by the optimization core module to accelerate the speed of iterative convergence.

12. The system of claim 11, wherein the optimization algorithm module further comprises a physics module connected to the optimization core module and configured to simulate actual physical motion of a machine according to the decoded chromosomes to estimate the production cycle time of the decoded chromosomes. 13 . The system of claim 11 , wherein the optimization calculation module further comprises a visualization module connected to the optimization core module and configured to present the plurality of final parameters in a graphical form.

14. The system of claim 11, wherein the at least one constraint condition includes a configuration order of the feeding mechanism, a correspondence between the picking mechanism and the parts to be picked, and a quantity limit of the parts on the feeding mechanism.

15. The system of claim 11, wherein the score calculation is performed based on M, F, D, weight 1, and weight 2 to obtain a fitness score. Where M represents a fixed value for calculating the benefit score, weight 1 represents the penalty value, F represents the number of constraints exceeded, weight 2 represents the distance weight of the decoded chromosome, and D represents the distance.

16. The system of claim 11, wherein the multiple iterations and selection steps include selecting at least two chromosomes with minimum values ​​based on the multiple parameters for mating and mutation to produce new chromosome offspring; calculating scores for the new chromosome offspring, selecting at least two chromosome offspring with minimum values, and mating and mutation again to produce new chromosome grandchildren; and when the number of iterations exceeds a preset value, selecting the chromosome with the minimum value as the chromosome representing the optimized parts placement sequence and configuration.

17. The system of claim 11, wherein the plurality of parameters include candidate positions of a feeding mechanism, end point information of a picking mechanism, a position of a discharge point, and physical limitations of an actuator.

18. The system of claim 17, wherein the plurality of parameters further include a correspondence between a material taking mechanism and a material piece, a configuration sequence of a specific material feeding mechanism, and a configuration quantity range of a specific material piece on a material feeding mechanism.

19. A system as described in claim 11, wherein the multiple final parameters include the total number of feeding mechanisms, material types and feeding mechanism configurations, material picking mechanisms and feeding mechanism configurations, material picking mechanisms and material configurations, material picking and unloading sequence, predicted cycle time, initial material mechanism relationship diagram, feeding position and unloading configurations and usage frequency distribution diagram, material picking and unloading path diagram, and path length distribution diagram.

Images