Intelligent robot labor dispatch management method and system

Abstract

The present invention belongs to the field of intelligent manufacturing, and provides an intelligent robot labor dispatch management method and system, which obtains the characteristic information of the customer's workpiece, and the characteristic information of the workpiece includes: workpiece material, workpiece shape, workpiece size, and workpiece surface treatment requirements; obtains the processing requirements of the customer's workpiece, and the processing requirements of the workpiece include: processing task type, precision requirements, processing speed, working environment, and processing quantity; selects the required manipulator and peripheral equipment types based on the characteristic information and processing requirements of the workpiece using a first model; and determines the number of robots to be dispatched based on the processing requirements of the workpiece and the types of the manipulator and peripheral equipment using a second model. The above scheme can efficiently perform intelligent robot labor dispatch.

Description

Technical Field

[0001] The present invention belongs to the field of intelligent manufacturing, and specifically relates to an intelligent robot labor dispatch management method and system. Background Art

[0002] Intelligent robot labor dispatch is a new industrial service model that uses intelligent technology to dispatch industrial robots with certain autonomous decision-making capabilities and related ancillary equipment to the customer's workplace to complete specific production tasks. This model is similar to human labor dispatch, but the target is robots instead of human workers.

[0003] Intelligent robot labor dispatch is very different from conventional human worker labor dispatch and has many problems, which are mainly reflected in the following aspects:

[0004] Complex and ever-changing workpiece requirements: The workpiece's material, shape, size, surface treatment and other characteristic information as well as the type of processing task, precision requirements, processing speed, working environment and other requirements vary, which makes it difficult for fixed rules to meet the selection of manipulators, peripheral equipment and robots in all scenarios.

[0005] Difficulty in optimizing resource allocation: On a multi-process production line, different types of robots, manipulators and peripheral equipment may be required at different stages. How to complete tasks within a limited time and minimize total costs becomes a challenge.

[0006] Dynamic adjustment of customer feedback: In actual production, customer feedback will constantly affect the production process. If the system cannot make dynamic adjustments based on the feedback, it may lead to decreased production efficiency, increased resource waste, and even failure to meet customer needs. Summary of the invention

[0007] In order to solve the problems in the prior art, the present invention provides a method for managing intelligent robot labor dispatch, which includes the following steps: obtaining characteristic information of a customer's workpiece, wherein the characteristic information of the workpiece includes: workpiece material, workpiece shape, workpiece size, and workpiece surface treatment requirements; obtaining processing requirements of the customer's workpiece, wherein the processing requirements of the workpiece include: processing task type, precision requirements, processing speed, working environment, and processing quantity; selecting the required manipulator and peripheral equipment types using a first model based on the characteristic information of the workpiece and the processing requirements; and determining the number of robots to be dispatched using a second model based on the processing requirements of the workpiece and the types of the manipulator and peripheral equipment.

[0008] Another aspect of the present invention further provides an intelligent robot labor dispatch management system, the system comprising the following modules:

[0009] The first acquisition module is used to acquire feature information of a customer's workpiece, wherein the feature information of the workpiece includes: workpiece material, workpiece shape, workpiece size, and workpiece surface treatment requirements;

[0010] The second acquisition module is used to obtain the processing requirements of the customer's workpiece, and the processing requirements of the workpiece include: processing task type, precision requirement, processing speed, working environment, and processing quantity;

[0011] The classification module is used to select the required manipulator and peripheral equipment types using the first model based on the feature information and processing requirements of the workpiece;

[0012] The planning module is used to determine the number of robots that need to be dispatched using the second model according to the processing requirements of the workpiece and the types of the manipulator and peripheral equipment.

[0013] The present invention uses a preliminary classifier to perform a preliminary analysis of workpiece feature information and processing requirements, calculates the optimal type of manipulator and peripheral equipment, realizes intelligent decision-making, and avoids the rigidity and inflexibility of traditional rule systems. Combined with the reinforcement learning algorithm, the system continuously optimizes the selection strategy of manipulators and peripheral equipment through interaction with the actual production environment on the basis of preliminary selection, ensuring that the system can adapt to changes in production requirements and gradually improve the accuracy of selection.

[0014] The system introduces a loop feedback mechanism that can adjust the classifier's prior probability and conditional probability, as well as the reward function and Q value of reinforcement learning based on customer feedback during the production process, to further optimize the configuration of the robot, manipulator and peripheral equipment.

[0015] The present invention comprehensively considers the task requirements of each stage in the production process, determines the optimal number of robots through an integer linear programming model, and ensures that the total cost is minimized while meeting the production time requirements. Through the dynamic adjustment of the optimization model, the present invention can reasonably allocate production resources, avoid robot overload or idleness, and maximize the work efficiency and economic benefits of the production line. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0017] Figure 1 is a flow chart of the method of the present invention. DETAILED DESCRIPTION

[0018] Below, the invention is preferably described in conjunction with the accompanying drawings and specific implementation methods.

[0019] This embodiment solves the above problem through the following steps:

[0020] In one embodiment, reference Figure 1 The present invention provides an intelligent robot labor dispatch management method. The method system realizes efficient management and dispatch of multiple robots and their ancillary equipment through intelligent technical means, so that they can perform corresponding tasks at a specified time and place according to the specific needs of customers. The method not only takes into account the working ability of the robot itself, but also includes factors such as the complexity of the task, time urgency, and environmental adaptability to ensure the optimal configuration and use of robot resources.

[0021] The intelligent robot in the present invention refers to an industrial robot with certain adaptive functions. It is usually equipped with a variety of sensors and actuators, and can independently or collaboratively complete various manufacturing, processing, assembly and other tasks in a complex industrial environment. Intelligent robots can realize real-time data processing and task adjustment through embedded systems or connecting to external control platforms, so as to adapt to changing working conditions and requirements.

[0022] In the present invention, the intelligent robot has the ability to replace manipulators and peripheral equipment according to task requirements. Specifically, the intelligent robot is equipped with a modular manipulator and peripheral equipment interface design, so that different types of manipulators and peripheral equipment can be quickly replaced according to the specific requirements of the task. This design allows the robot to flexibly switch between different tasks, thereby adapting to a variety of workpiece processing requirements, such as welding, assembly, handling, grinding, etc. Peripheral equipment refers to supporting hardware equipment that assists the manipulator to complete the task. These devices may include:

[0023] Sensor: used to detect and feedback the workpiece's status, position, temperature, humidity and other environmental information.

[0024] Workbench and support structure: provide a stable working platform for the manipulator to ensure stability during work.

[0025] Vision system: used to identify, locate and detect workpieces to improve the accuracy of grasping and operation.

[0026] Safety system: including light barriers, guardrails, etc. to ensure safety during operation.

[0027] In the present invention, the manipulator and peripheral equipment can be used in a set, such as a type A manipulator combined with types 1, 2, and 3 peripheral equipment to form a set of solutions, a type A manipulator combined with types 2, 3, and 4 peripheral equipment to form a set of solutions, and a type B manipulator combined with types 2 and 3 peripheral equipment to form a set of solutions, etc.

[0028] In the present invention, labor dispatch specifically refers to that a robot service provider dispatches robots and their related equipment to a workplace designated by a customer to perform predetermined tasks based on customer needs.

[0029] Acquire the characteristic information of the customer's workpiece, wherein the characteristic information of the workpiece includes: workpiece material, workpiece shape, workpiece size, and workpiece surface treatment requirements.

[0030] In the present invention, obtaining the characteristic information of the customer's workpiece is a key step in the intelligent robot labor dispatch management method. The characteristic information of the workpiece includes but is not limited to the following aspects, which can be directly obtained through the information provided by the customer, or by processing the workpiece image provided by the customer using image processing:

[0031] Workpiece material: refers to the type of material used for the workpiece, such as metal, plastic, ceramic, composite material, etc. The workpiece material has an important impact on the selection of manipulators and peripheral equipment. For example, metal materials may require the use of manipulators with higher rigidity and wear resistance, while plastic materials may be more suitable for the use of manipulators with flexible grasping capabilities. When using image processing to obtain the workpiece material, different peripheral equipment such as visual systems and measurement sensors are used, and color space conversion (such as from RGB to HSV) is used to separate color information, and color features of the workpiece are extracted through methods such as color histograms and principal component analysis (PCA). These features can help distinguish workpieces of different materials such as metal, plastic, and ceramic.

[0032] Workpiece shape: refers to the geometric shape and structural features of the workpiece, such as square, round, irregular, polyhedron, etc. The shape of the workpiece directly affects the gripping method of the robot and the planning of the work path. For example, workpieces with regular geometric shapes may be easier to handle, while irregular workpieces may require specially designed robots or customized gripping methods. When using image processing to obtain the shape of the workpiece, clear edge contours are extracted from the workpiece image through algorithms such as the Sobel operator and Canny edge detection.

[0033] Workpiece size: refers to the physical size of the workpiece, including length, width, height and weight. The size of the workpiece determines the robot's operating range, gripping force and motion path planning. For large or heavy workpieces, you may need to choose a robot with greater gripping force and operating space to ensure safe and efficient operation. At the same time, you also need to cooperate with different peripheral equipment to detect shape, size, etc.

[0034] Workpiece surface treatment requirements: refers to the surface treatment that the workpiece needs to undergo during the processing, such as polishing, grinding, coating, spraying, etc. These requirements are directly related to the function selection of the manipulator and peripheral equipment and the execution order of the tasks. For example, polishing or grinding may require a dedicated surface treatment manipulator, while coating or spraying may require professional equipment with high-precision control capabilities.

[0035] By acquiring and analyzing the above workpiece feature information in detail, the present invention can accurately identify the specific needs of customer tasks, thereby providing a solid foundation for subsequent manipulator selection and robot scheduling. Based on this information, the system can automatically match the most suitable manipulator and robot configuration to ensure the efficiency, accuracy and consistency of the workpiece processing process, and ultimately achieve accurate satisfaction of customer needs.

[0036] Obtain the processing requirements of the customer's workpiece, which include: processing task type, precision requirements, processing speed, working environment, and processing quantity.

[0037] In the present invention, obtaining the processing requirements of the customer's workpiece is an important step to ensure that the intelligent robot can accurately perform the task. The processing requirements of the workpiece include the following key aspects:

[0038] Processing task type: refers to the specific processing operation or task category that the customer wants the intelligent robot to perform. These tasks may include but are not limited to welding, assembly, handling, grinding, polishing, cutting, painting, etc. Different types of processing tasks require the intelligent robot to have different functional modules and manipulators and peripheral equipment. For example, welding tasks may require the installation of dedicated welding manipulators and peripheral equipment, while handling tasks may require the installation of manipulators and peripheral equipment with strong gripping functions. By accurately identifying the type of processing task, the system can select the most suitable robot configuration to ensure the smooth execution of the task.

[0039] Precision requirements: refers to the customer's precision standards for workpiece processing, that is, the specific requirements for the size, tolerance, surface finish, etc. of the workpiece after processing. The precision requirements directly affect the selection of manipulators and peripheral equipment and the motion control parameters of the robot. For example, high-precision processing tasks may require the use of precision manipulators and peripheral equipment, and fine-tuning of the robot's motion path and speed to ensure that the workpiece meets the expected precision standards. The present invention can effectively match the appropriate robot configuration and ensure the quality of the processing results by acquiring and analyzing the precision requirements.

[0040] Processing speed: refers to the workpiece processing speed or production rhythm required by the customer, that is, the number of workpieces that need to be completed per unit time. Processing speed is an important factor affecting robot scheduling and resource allocation. For the demand for high processing speed, the system may need to schedule multiple robots to work simultaneously, or optimize the robot's working path and action time to improve overall production efficiency. By analyzing the processing speed, the present invention can maximize the processing efficiency and meet the customer's time requirements while ensuring quality.

[0041] Working environment: refers to the specific environmental conditions required for workpiece processing, including temperature, humidity, cleanliness, light intensity, etc. These environmental conditions have an important impact on the selection and operation of the robot. For example, a high-temperature environment may require the use of high-temperature resistant manipulators and peripheral equipment, and a clean room environment may require the robot to have dust and pollution prevention functions. By obtaining working environment information, the present invention can optimize the robot configuration to ensure its stable and efficient operation in various complex environments.

[0042] Processing quantity: refers to the total number of workpieces that the customer requires to be processed. This requirement is directly related to the robot's workload and scheduling arrangements. When the processing quantity is large, the system may need to schedule multiple robots to work in parallel, and formulate reasonable task allocation and shift plans to avoid excessive use of a single robot or idle equipment. At the same time, the processing quantity is also the basic data for calculating task completion time and resource allocation. By accurately obtaining the processing quantity, the present invention can provide customers with more flexible and efficient production solutions.

[0043] By obtaining and analyzing the above processing requirements in detail, the present invention can fully understand the specific requirements of customers and then formulate the optimal robot configuration and scheduling plan. This process ensures that the robot can complete the processing efficiently and accurately when performing tasks, and ultimately meet the expectations and standards of customers.

[0044] The first model is used to select the required robot and peripheral equipment types based on the feature information and processing requirements of the workpiece.

[0045] In view of the wide variety of feature information and processing requirements of workpieces, the requirements of workpieces for manipulators and peripheral equipment are very complex and changeable. It is indeed difficult to directly determine the types of manipulators and peripheral equipment through simple preset rules. In addition, during the production process, it is necessary to gradually improve the selection strategy based on customer feedback. In order to solve this problem, the present invention uses the following method to implement the first model.

[0046] First, the feature information and processing requirements of the workpiece are organized into feature vectors ;

[0047] ;

[0048] in, It is the first type of feature. It is the first type of feature. It is the first type of feature. It is the number of features. Feature is one of the characteristic information and processing requirements of the workpiece. Specifically, it includes workpiece material, workpiece shape, workpiece size, workpiece surface treatment requirements, processing task type, accuracy requirements, processing speed, working environment, and processing quantity.

[0049] Furthermore, in order to reduce the amount of data processing, each feature is classified.

[0050] For example, Indicates the size of the workpiece. Workpieces smaller than 10 cm are classified as 1, workpieces between 10 cm and 20 cm are classified as 2, and workpieces larger than 20 cm are classified as 3. The value of is 1, 2, 3.

[0051] Similarly, similar operations can be performed on other features, and each type of feature can be divided into different types according to actual conditions.

[0052] For each type of robot and peripheral equipment , the preliminary classifier calculates the feature vector for this type of robot and peripheral equipment Conditional probability of occurrence :

[0053] ;

[0054] in, Is the type of robot and peripheral equipment Under the conditions, characteristics The probability of occurrence and specific data can be obtained based on historical data statistics, and can also be updated based on real-time data during the model operation process.

[0055] Calculate the number of robots and peripherals for each type of robot The posterior probability :

[0056] ;

[0057] in, Is the type of robot and peripheral equipment The prior probability of is obtained based on historical data statistics. is the total probability of the feature vector 𝑋, which can be obtained by the joint probability of all possible types of manipulators and peripheral devices:

[0058] ;

[0059] in, It is the general type of robot and peripheral equipment.

[0060] For a given feature vector X, select the type of manipulator and peripheral equipment with the maximum posterior probability As a preliminary selection of manipulators and peripheral equipment:

[0061] ;

[0062] After making a preliminary selection, a detailed selection is made through reinforcement learning, and the robot and peripheral equipment selection strategy is continuously optimized through interaction with the environment;

[0063] The state of reinforcement learning Defined as the current artifact feature vector X and the initial selection of the classifier constitute:

[0064] ;

[0065] The action of reinforcement learning is defined as the system in state Select the type of robot and peripheral equipment ,The action space includes all possible types of manipulators and peripherals;

[0066] The reward function for reinforcement learning is defined as , the reward function is used to measure the selection of a certain type of manipulator and peripheral equipment The effects afterward;

[0067] ;

[0068] in, To adjust the parameters, make adjustments based on actual needs (such as adjustments based on economic effects).

[0069] Reinforcement learning uses the Q learning algorithm to update the Q value, which represents the state Next select action Expected cumulative rewards:

[0070] ;

[0071] in, is the learning rate, which determines the step size of the Q value update;

[0072] is the discount factor used to measure the impact of future rewards;

[0073] Indicates the next state Select Robot and Peripheral Equipment (Action) The Q value of the optimal action;

[0074] Each time the type of manipulator and peripherals is selected, the reinforcement learning model selects the best action (that is, the best manipulator and peripherals) based on the current Q value:

[0075] ;

[0076] in, Indicates the best choice of robot and peripheral equipment.

[0077] After each production process, the system collects customer feedback information and uses it as part of the reward function, which directly affects the Q value update of the reinforcement learning model. In addition, this feedback is also used to adjust the prior probability of the preliminary classifier. and conditional probability , so that the classifier can gradually adapt to new environmental changes.

[0078] By combining preliminary classification with reinforcement learning and introducing a loop feedback mechanism, the system can dynamically adapt to changing production needs and continuously optimize the selection strategy of the robot and peripheral equipment. This approach not only improves the accuracy of decision-making, but also enables the system to have the ability to self-improve and adapt to new environments.

[0079] The second model is used to determine the number of robots that need to be dispatched according to the processing requirements of the workpiece and the types of the manipulator and peripheral equipment.

[0080] In modern manufacturing, the processing requirements of workpieces are usually complex and changeable, including not only a variety of process flows, but also different manipulators and peripheral equipment and robot types. Each process stage may require a different combination of manipulators and peripheral equipment, which makes simple preset rules ineffective. To this end, the present invention uses the second model to determine the number of robots that need to be dispatched, and the implementation method of the second model is as follows:

[0081] The second model objective is to determine how to select and dispatch robots to minimize the total cost while meeting the workpiece processing needs and production time requirements.

[0082] The following factors need to be considered:

[0083] Process flow: Different process stages may require different types of manipulators and peripheral equipment and robots.

[0084] Robot cost: Different types of robots and manipulators and peripheral equipment combinations have different costs.

[0085] Production time requirement: All workpieces must be processed within a given time.

[0086] Model input parameters

[0087] Artifact characteristics and requirements:

[0088] : Number of workpieces.

[0089] : Task completion time requirement (production time limit).

[0090] : Processing complexity, expressed as a numerical value, is used to quantify the difficulty of processing the workpiece.

[0091] Process stage:

[0092] : Number of process stages (e.g. cutting, welding, grinding, etc.).

[0093] Each process stage Processing complexity , used to indicate the difficulty of the task at this stage.

[0094] Robots, manipulators and peripheral equipment:

[0095] : No. Types of robots and peripheral equipment.

[0096] : No. The number of robots required for each type of manipulator and peripheral equipment is variable.

[0097] :Manipulator and peripheral equipment types The processing time to complete a single workpiece at a specific process stage 𝑖.

[0098] Cost and performance:

[0099] :Equipped with manipulator and peripheral equipment The cost of a single robot.

[0100] :Manipulator and peripheral equipment types Efficiency refers to the amount of work per unit time.

[0101] The goal is to minimize the cost of the total number of robots selected:

[0102] ;

[0103] Where n is the total number of robot and peripheral equipment types.

[0104] To ensure production time requirements, constraints need to include:

[0105] For each process stage , the total processing time cannot exceed the given task completion time ;

[0106] Assume that the time of each process stage is , the total time constraint is:

[0107] ;

[0108] Each process stage Processing time The total workload of this stage and processing time for each type of robot and peripheral equipment Determine. For each process stage 𝑖:

[0109] ;

[0110] in, ;

[0111] The workload of each process stage should be evenly distributed among all participating robots to avoid overloading some robots.

[0112] ;

[0113] This constraint ensures the minimum number of robots required to complete the task within a given production time.

[0114] All variables All are non-negative integers:

[0115] , and is an integer;

[0116] The above model is an integer linear programming, which can be solved by branch and bound method, mixed integer linear programming (MILP) solver: such as CPLEX, Gurobi and other professional solving tools.

[0117] In actual production, input parameters can be adjusted regularly based on production feedback , and , ensuring the accuracy and practicality of the optimization model. Through the feedback mechanism, the system can dynamically adjust the number of dispatched robots to meet the needs of different batches of workpieces and further optimize the total cost.

[0118] On the other hand, the present invention also provides an intelligent robot labor dispatch management system, comprising:

[0119] The first acquisition module is used to acquire feature information of a customer's workpiece, wherein the feature information of the workpiece includes: workpiece material, workpiece shape, workpiece size, and workpiece surface treatment requirements;

[0120] The second acquisition module is used to obtain the processing requirements of the customer's workpiece, and the processing requirements of the workpiece include: processing task type, precision requirement, processing speed, working environment, and processing quantity;

[0121] The classification module is used to select the required manipulator and peripheral equipment types using the first model based on the feature information and processing requirements of the workpiece;

[0122] The planning module is used to determine the number of robots that need to be dispatched using the second model according to the processing requirements of the workpiece and the types of the manipulator and peripheral equipment.

[0123] The prior art mentioned in the above background technology section and specific embodiment section of the present invention can be used as part of the present invention to understand the meaning of some technical features or parameters.

Claims

1. A method for managing intelligent robot labor dispatch, characterized in that: The method comprises the following steps: Obtain feature information of customer workpieces; Obtain the processing requirements of customer workpieces; Select the required robot and peripheral equipment types using the first model based on the feature information and processing requirements of the workpiece; The second model is used to determine the number of robots that need to be dispatched according to the processing requirements of the workpiece and the types of manipulators and peripheral equipment; The implementation steps of the first model include: Organize the feature information and processing requirements of the workpiece into a feature vector; According to historical data, the conditional probability of each feature vector appearing under each type of robot and peripheral equipment is calculated, where the conditional probability is calculated using the prior probability of each type of robot and peripheral equipment appearing; The posterior probability of each type of manipulator and peripheral equipment is calculated based on the conditional probability of each characteristic vector appearing under each type of manipulator and peripheral equipment; For a given feature vector, the type of manipulator and peripheral equipment with the maximum posterior probability is selected as the preliminary manipulator and peripheral equipment selection; The state of reinforcement learning is defined as the current artifact feature vector and the initial selection of the preliminary classifier; The reinforcement learning action is defined as the type of manipulator and peripherals selected by the system in the state; Determine the reward function for reinforcement learning using weight distribution based on customer feedback, processing time, and processing quality; The Q-value is updated through the Q-learning algorithm. The Q-value represents the expected cumulative reward for selecting the type of manipulator and peripheral equipment in a certain state. Each time the type of manipulator and peripheral equipment is selected, the reinforcement learning model selects the best manipulator and peripheral equipment based on the current Q value; After each production process, the system collects customer feedback information and updates the reward function, prior probability, and conditional probability based on the feedback information; The second model includes: The input parameters of the model include workpiece characteristics and requirements, process stages, robots and manipulators and peripheral equipment, costs and performance; The goal is to minimize the cost of the total number of robots selected: ， Where n is the total number of robot and peripheral equipment types, Equipped with manipulators and peripheral equipment The cost of a single robot, Variable for the number of robots required for the 𝑗th type of manipulator and peripheral equipment; Constraints include: For each process stage 𝑖, the total processing time cannot exceed the given task completion time Assume that the time of each process stage is , the total time constraint is: ， The processing time of each process stage 𝑖 The total workload of this stage and processing time for each type of robot and peripheral equipment Determine, 𝑘 is the number of process stages; For each process stage 𝑖: ， in, , For robot and peripheral equipment type The efficiency is the workload per unit time. The workload of each process stage should be evenly distributed to all participating robots. is the processing complexity of each process stage i; ， This ensures the minimum number of robots required to complete a task within a given production time.

2. The intelligent robot labor dispatch management method according to claim 1 is characterized in that: The characteristic information of the customer's workpiece is obtained through image processing.

3. The intelligent robot labor dispatch management method according to claim 1, characterized in that: The second model is solved using a branch and bound method or a mixed integer linear programming solver.

4. An intelligent robot labor dispatch management system, the system implementing the method according to claim 1, characterized in that: The system includes the following modules: The first acquisition module is used to acquire feature information of a customer's workpiece, wherein the feature information of the workpiece includes: workpiece material, workpiece shape, workpiece size, and workpiece surface treatment requirements; The second acquisition module is used to obtain the processing requirements of the customer's workpiece, and the processing requirements of the workpiece include: processing task type, precision requirement, processing speed, working environment, and processing quantity; The classification module is used to select the required manipulator and peripheral equipment types using the first model based on the feature information and processing requirements of the workpiece; The planning module is used to determine the number of robots that need to be dispatched using the second model according to the processing requirements of the workpiece and the types of the manipulator and peripheral equipment.

5. The intelligent robot labor dispatch management system according to claim 4 is characterized in that: The characteristic information of the customer's workpiece is obtained through image processing.

6. The intelligent robot labor dispatch management system according to claim 5, characterized in that: The implementation steps of the first model include: Organize the feature information and processing requirements of the workpiece into a feature vector; According to historical data, the conditional probability of each feature vector appearing under each type of robot and peripheral equipment is calculated, where the conditional probability is calculated using the prior probability of each type of robot and peripheral equipment appearing; The posterior probability of each type of manipulator and peripheral equipment is calculated based on the conditional probability of each characteristic vector appearing under each type of manipulator and peripheral equipment; For a given feature vector, the type of manipulator and peripheral equipment with the maximum posterior probability is selected as the preliminary manipulator and peripheral equipment selection; The state of reinforcement learning is defined as the current artifact feature vector and the initial selection of the preliminary classifier; The reinforcement learning action is defined as the type of manipulator and peripherals selected by the system in the state; Determine the reward function for reinforcement learning using weight distribution based on customer feedback, processing time, and processing quality; The Q-value is updated through the Q-learning algorithm. The Q-value represents the expected cumulative reward for selecting the type of manipulator and peripheral equipment in a certain state. Each time the type of manipulator and peripheral equipment is selected, the reinforcement learning model selects the best manipulator and peripheral equipment based on the current Q value; After each production process, the system collects customer feedback information and updates the reward function, prior probability and conditional probability based on the feedback information.

7. The intelligent robot labor dispatch management system according to claim 6, characterized in that: The second model includes: The input parameters of the model include workpiece characteristics and requirements, process stages, robots and manipulators and peripheral equipment, costs and performance; The workpiece characteristics and requirements include the number of workpieces, task completion time requirements, and processing complexity; the process stages include the number of process stages and the processing complexity of each process stage; robots, manipulators and peripheral equipment include: manipulator types, the number of each type of manipulators, the processing time for each type of manipulator to complete a specified process stage, peripheral equipment types, and the number of peripheral equipment; cost and efficiency include the cost of a single robot and the efficiency of the manipulator; The goal is to minimize the cost of the total number of robots selected; Constraints include: For each process stage, the total processing time cannot exceed the given task completion time; the workload of each process stage should be evenly distributed to all participating robots; the number of robots of each type is a non-negative integer.

8. The intelligent robot labor dispatch management system according to claim 7, characterized in that: The second model is solved using a branch and bound method or a mixed integer linear programming solver.

Images