Intelligent assembly robot assembly control method and control system

Abstract

The application discloses an intelligent assembly robot assembly control method and a control system, and the control method comprises the following steps: importing prior information; generating an assembly sequence and a process; generating an assembly action flow of each part according to the prior information and the assembly sequence and the process; performing assembly action flow disassembly, so that each action flow corresponds to movement control of one or more mechanical arms; and uniformly using a state machine model to perform movement control of each mechanical arm. The application simultaneously unifies a control model of a multi-robot bottom layer design into path planning, movement and operation state machine models, adopts a block chain type message sharing mechanism to synchronize respective state information, finally adopts an embodied intelligent framework to perform information collection and feedback processing, the collected information comprises three-dimensional point cloud data, sensor data, robot feedback state data and system historical record data, and finally intelligent assembly decision is completed. The application can realize autonomous decision operation, and high-precision standards suitable for various different assembly requirements.

Description

Technical Field

[0001] This invention belongs to the field of robot control, specifically relating to an assembly control method and control system for an intelligent assembly robot. Background Technology

[0002] The information integration process of intelligent assembly lines involves integrating physical objects (intelligent assembly equipment, products) with information systems (such as MES and ERP), and controlling the product assembly process through computer integration. Intelligent human-machine interaction fully leverages the advantages of both machines and humans, achieving intelligent and efficient product assembly. Intelligent final assembly lines integrate advanced technologies such as intelligent equipment, intelligent distribution, the Internet of Things, artificial intelligence, data mining, information system integration, and computer simulation. Utilizing modern information and network technologies, with "product processing and assembly" as the main thread, a system platform composed of computers, networks, databases, equipment, and software is constructed into a high-speed information network. This enables rapid plan issuance, work scheduling and control, process guidance, production statistics, equipment status monitoring, comprehensive quality control and traceability, and production information collaboration (material collaboration, just-in-time delivery, production preparation collaboration), ultimately achieving intelligent equipment and information-based production management.

[0003] While individual intelligent equipment possesses intelligent features, its functionality and efficiency are always the priority, which cannot meet the needs of large-scale development in modern manufacturing. Therefore, it is necessary to further develop intelligent equipment to establish intelligent assembly systems.

[0004] Currently, multiple intelligent equipment units at the bottom layer form a digital assembly production line, enabling connectivity between these intelligent equipment units. These multiple digital assembly lines further combine to form a digital workshop, connecting the digital production lines. Ultimately, the digital workshops together form an intelligent factory, connecting all the digital workshops. The top-level application layer consists of enabling technologies such as the Internet of Things, cloud computing, big data, machine learning, and remote operation and maintenance, providing technical support and services for the intelligent assembly system.

[0005] In the field of intelligent assembly, assembly line operations are very common. In this scenario, the robot's movement mode is fixed, and it cannot make autonomous decisions by analyzing the relationship between the environment, objects and its own state. The assembled parts are simple, so it cannot achieve high-precision standards that can be intelligently adapted to various different assembly needs. Summary of the Invention

[0006] The purpose of this invention is to address the problems in the prior art by providing an intelligent assembly robot assembly control method and control system. This system enables autonomous decision-making, intelligently adapts to various high-precision standards for different assembly needs, and uses a blockchain-based message sharing mechanism to synchronize the states of each robot, enabling parallel and collaborative control of each robot and improving work efficiency.

[0007] To achieve the above objectives, the present invention provides the following technical solution:

[0008] An assembly control method for an intelligent assembly robot includes:

[0009] Import prior information;

[0010] Generate assembly sequence and process;

[0011] Based on prior information, assembly sequence, and process, the assembly action flow for each part is generated.

[0012] The assembly process is broken down so that each process corresponds to the motion control of one or more robotic arms.

[0013] The motion control of each robotic arm is uniformly performed using a state machine model.

[0014] As a preferred embodiment, the prior information in the step of importing prior information includes product models, part models, assembly table models, and environment models.

[0015] As a preferred embodiment, in the step of generating assembly sequence and process, the output of assembly sequence analysis is expressed through a tree structure. The tree structure includes one or more trees, wherein in the structure of each tree, the leaf nodes are the initial parts, and the non-leaf nodes are the local parts formed by assembly. Each assembly action is to assemble multiple parts into a component, and the parts are either the initial parts or the components.

[0016] As a preferred approach, in the step of importing prior information, the structural features of the parts are first extracted, and the assembly relationship is determined through the relationship between the structural features;

[0017] Then, by decomposing the key feature elements of the structure, the position tolerance is obtained by micro-statistical analysis of the position and orientation size deviations, and the shape tolerance is obtained by micro-statistical analysis of the shape size deviations.

[0018] As a preferred approach, in the steps of generating the assembly process flow for each part based on prior information, assembly sequence, and process, and in the steps of decomposing the assembly process flow so that each process flow corresponds to the motion control of one or more robotic arms, the process is decomposed step by step according to the meta-operation hierarchy. The first-level meta-operation nodes are obtained by decomposing the business process, and the second-level meta-operation nodes are obtained by analyzing the basic operations that the robot can perform. Each first-level meta-operation corresponds to several second-level meta-operations, and the meta-operations are a reusable set of nodes.

[0019] As a preferred approach, the core algorithm for generating first-order meta-operations is matrix / model decision-making. It collects field data to form input features, iterates sequentially, and forms a sequence of first-order meta-operations. The collected field data includes: 3D point cloud data, sensor data, robot feedback status data, and system historical data. Through continuous perception, continuous movement, and interaction with the environment, updated information is obtained from these data.

[0020] As a preferred approach, if the two-level meta-operation library covers all actions that the robot can perform in a single instruction, then regardless of the task defined by the first-level meta-operation, as long as it is something the robot can complete, it can be found in the two-level meta-operation library, and a corresponding two-level meta-operation solution can be formed.

[0021] As a preferred approach, the step of using a state machine model to uniformly control the motion of each robotic arm is described. Each robot corresponds to an independent state machine, and each state machine includes 6 states: start, end, exception, path planning, motion, and operation.

[0022] As a preferred solution, the path planning process includes: moving the robot from its starting position to the sequence of operation points, visually confirming the deviation from the operation pose, and determining whether to proceed with the operation or re-plan the path and adjust the movement.

[0023] Two robots moving simultaneously, regardless of whether they have a common task, constitute continuous cooperation, relying on path planning to control their movements according to constraints given by time or position; discrete cooperation uses cooperation conditions to trigger or terminate the cooperation.

[0024] When multiple robots move simultaneously, it is considered cooperative control, and path planning is considered based on time or position. When multiple robots operate on the same target, it is considered collaborative operation in coordination, and multiple interactions occur during state machine control. When robots share data states, they use a blockchain-style message sharing mechanism to synchronize their respective state information, which is a distributed control system.

[0025] An intelligent assembly robot assembly control system includes:

[0026] The prior information import module is used to import prior information;

[0027] The assembly sequence and process generation module is used to generate assembly sequences and processes.

[0028] The assembly action flow generation module is used to generate the assembly action flow for each part based on prior information, assembly sequence, and process.

[0029] The assembly motion process decomposition module is used to decompose the assembly motion process so that each motion process corresponds to the motion control of one or more robotic arms.

[0030] The state machine representation module is used to unify the motion control of each robotic arm using a state machine model.

[0031] Compared with the prior art, the present invention has at least the following beneficial effects:

[0032] By analyzing the environment, objects, and ontology models, the assembly sequence and process can be intelligently generated. Simultaneously, a unified control model for the underlying design of multiple robots is established as a path planning, motion, and operational state machine model. A blockchain-based message sharing mechanism is used to synchronize their respective state information. Finally, an embodied intelligence framework is employed for information collection and feedback processing. The collected information includes 3D point cloud data, sensor data, robot feedback state data, and system historical data, ultimately completing intelligent assembly decisions. This invention analyzes the overall assembly model and part models to obtain a set of assembly relationships and assembly methods, forming a table of part assembly relationships and installation positions. Based on this, the part assembly sequence and process are derived. The assembly process is automatically generated for each part by traversing the assembly sequence, and then the state machine is invoked in conjunction with the assembly process to complete the specific robotic arm control actions. This invention enables autonomous robot assembly decision-making, intelligently adapts to various assembly requirements, and uses a blockchain-based message sharing mechanism to synchronize the states of each robot, enabling parallel and collaborative control of all robots and improving operational efficiency. Attached Figure Description

[0033] To more clearly illustrate the technical solutions in this application, the drawings used in the application description will be briefly introduced below. The drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0034] Figure 1 This is a system hierarchy diagram of the intelligent assembly robot assembly control method according to an embodiment of the present invention;

[0035] Figure 2 This is a flowchart of the assembly control system for the intelligent assembly robot according to an embodiment of the present invention;

[0036] Figure 3 This is a schematic diagram showing the assembly relationship and geometric tolerances in an embodiment of the present invention;

[0037] Figure 4 This is a schematic diagram of the assembly tree according to an embodiment of the present invention;

[0038] Figure 5 This is a flowchart illustrating the assembly process generation of the intelligent assembly robot assembly control method according to an embodiment of the present invention.

[0039] Figure 6This is a schematic diagram illustrating the mapping relationship between first-level and second-level meta-operations in an embodiment of the present invention;

[0040] Figure 7 This is a schematic diagram of a single state machine design according to an embodiment of the present invention. Detailed Implementation

[0041] To make the objectives, technical solutions, and advantages of this application clearer, the application is described and illustrated below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application. All other embodiments obtained by those skilled in the art based on the embodiments provided in this application without inventive effort are within the scope of protection of this application.

[0042] Please see Figure 1 The intelligent assembly robot assembly control method of this invention includes the following steps:

[0043] S1. Import prior information;

[0044] S2. Generation assembly sequence and process;

[0045] S3. Based on prior information, assembly sequence and process, generate the assembly action flow for each part.

[0046] S4. Decompose the assembly process so that each process corresponds to the motion control of one or more robotic arms.

[0047] S5. Use a state machine model to uniformly control the motion of each robotic arm.

[0048] The intelligent assembly robot assembly control method corresponding to the above embodiments of the present invention comprises five levels from top to bottom: Level 1, prior information import, which includes product model, part model, assembly table model, and environment model; Level 2, assembly sequence and process generation, which generates assembly tree, connection semantics, process parameters, and process methods through analysis of various models; Level 3, assembly workflow generation, which generates the assembly action flow for each part based on the information from the first two levels and common sense of operation; Level 4, process action decomposition, where each action flow needs to be mapped to the motion control of one or more robotic arms; and Level 5, module action decomposition, which uses a state machine model to uniformly control the motion of each robotic arm.

[0049] like Figure 2As shown, in the workflow of the intelligent assembly robot assembly control system of this invention embodiment, the assembly model and part model in the system model are first analyzed to obtain the assembly relationship and assembly method set, forming a part assembly relationship and installation position table. Based on this, the part assembly sequence and part assembly process are further derived. The part assembly sequence is traversed, and the assembly process is automatically generated for each part. Then, the state machine is called in combination with the assembly process to complete the specific robotic arm control action.

[0050] It should be noted that, in level 1 of the control system, the analysis process for the assembly relationships and installation positions of parts is as follows: Figure 3 As shown, firstly, the structural features of the parts are extracted, and the assembly relationships are determined through the relationships between the structural features. Then, the key feature elements are decomposed from the structural features, and the positional tolerance is obtained through micro-statistics of the positional and orientation dimensional deviations, and the form tolerance is obtained through micro-statistics of the shape dimensional deviations.

[0051] It should be further noted that the assembly sequence is often not sequential, such as... Figure 4 As shown, the output of the assembly sequence analysis is expressed through a tree structure (the sequence is one or more trees, where the structure of each tree is as follows). Figure 4 As shown, leaf nodes are the initial parts, and non-leaf nodes are the assembled partial parts (hereinafter referred to as components). Each assembly action involves assembling multiple parts into a component, where a part is either an initial part or a component. The main reasons are as follows:

[0052] 1. Considering the entire assembly sequence, there may be one or more independent local assembly tasks, which can be represented as multiple trees and multiple state machines executing the assembly. It is necessary to consider the resource contention problem of multiple state machines and realize the blocking, ready, and execution of tasks.

[0053] 2. When assembling several parts into a component, there is no strict order requirement. Therefore, the child nodes of a certain node have no order requirement. Thus, the assembly of a certain component can be represented by a multi-way tree.

[0054] 3. The specific assembly method for certain parts is determined by analyzing the shape using a 3D algorithm and based on prior knowledge of the assembly process. It is necessary to consider how to operate the assembly method through a state machine.

[0055] 4. Different components or parts on the same layer can be assembled in parallel. If it is a non-leaf node, it is necessary to recursively go to the next layer to complete the assembly of the component first. The assembly process is equivalent to the reverse growth process from leaf to root, and the assembly is completed step by step.

[0056] During the downward propagation process from level 3 and level 4, the operation is decomposed level by level according to the meta-operation hierarchy, such as... Figure 5As shown. First-level meta-operation nodes are obtained by breaking down the business process. Second-level meta-operation nodes are obtained by analyzing the basic operations that the robot can perform.

[0057] The core algorithm for generating first-level meta-operations is matrix / model decision-making. It collects on-site data to form input features, iterates sequentially, and generates a sequence of first-level meta-operations. The collected on-site data includes 3D point cloud data, sensor data, robot feedback state data, and system historical data. Through continuous perception, movement, and interaction with the environment, updated information is obtained. Its embodied nature is manifested in the continuous decision-making and movement process. Each first-level meta-operation corresponds to several second-level meta-operations, which are reusable sets of nodes. Assuming the second-level meta-operation library can cover all actions the robot can perform in a single command issuance (one state machine transition), then regardless of the task defined for a first-level meta-operation, as long as it is achievable by the robot, it can be found in the second-level meta-operation library, and a corresponding second-level meta-operation solution can be generated.

[0058] The logic for expanding the knowledge base then becomes:

[0059] (1) When faced with new business problems (assembling new products, assembling new parts, updating processes, etc.), the first-level meta-operation knowledge base will increase accordingly.

[0060] (2) Analyze the tasks that the robot can perform based on business needs or action types, and expand the secondary meta-operation knowledge base.

[0061] Such a knowledge base will evolve into something similar to a knowledge graph, such as Figure 6 As shown. The knowledge base is subsequently expanded through operations such as adding and classifying knowledge node attributes. After the data accumulates to a certain level, knowledge reasoning can be performed, i.e., intelligently generating second-level meta-operation sequences. The generated sequences are then manually corrected, thereby improving work efficiency.

[0062] Furthermore, in the action decomposition of the lowest-level module at level 5, each robot corresponds to an independent state machine, such as... Figure 7 As shown, each state machine includes six states: Start, End, Exception, Path Planning, Motion, and Operation. The first three states are relatively easy to understand. Path planning involves the robot moving from its starting position to the sequence of operation points, visually confirming the deviation from the operation pose, confirming whether to proceed with the operation or re-plan the path and adjust the motion. The operation process is mainly an I / O control process. If robot control is required, a linear interpolation method is used, and path planning is not necessary.

[0063] When two robots move simultaneously, regardless of whether they have a common task, it is a continuous collaboration. The robot's movement can be controlled separately by path planning based on constraints given by time or position, without the need to judge node conditions. Discrete collaboration requires the use of collaboration conditions to trigger or terminate the collaboration.

[0064] In a multi-robot system, parallel execution of multiple threads is supported. The threads within the multiple state machines may only have sequential requirements during the task state. When multiple robots move simultaneously, it constitutes cooperative control, requiring path planning based on time or location. When multiple robots operate on the same target, it involves collaborative work within a cooperative framework, resulting in multiple interactions during state machine control. Data state sharing between robots utilizes a blockchain-style message sharing mechanism to synchronize their respective state information, constituting a distributed control system.

[0065] This invention also proposes an intelligent assembly robot assembly control system, comprising:

[0066] The prior information import module is used to import prior information;

[0067] The assembly sequence and process generation module is used to generate assembly sequences and processes.

[0068] The assembly action flow generation module is used to generate the assembly action flow for each part based on prior information, assembly sequence, and process.

[0069] The assembly motion process decomposition module is used to decompose the assembly motion process so that each motion process corresponds to the motion control of one or more robotic arms.

[0070] The state machine representation module is used to unify the motion control of each robotic arm using a state machine model.

[0071] This invention also proposes an electronic device, comprising: a memory storing at least one instruction; and a processor executing the instructions stored in the memory to implement the intelligent assembly robot assembly control method of this invention.

[0072] This invention also proposes a computer-readable storage medium storing at least one instruction, which is executed by a processor in an electronic device to implement the intelligent assembly robot assembly control method of this invention.

[0073] For example, the instructions stored in the memory can be divided into one or more modules / units. These modules / units are stored in a computer-readable storage medium and executed by the processor to complete the intelligent assembly robot assembly control method of the present invention. The one or more modules / units can be a series of computer-readable instruction segments capable of performing specific functions, which describe the execution process of the computer program on the server.

[0074] The electronic device may be a smartphone, laptop, PDA, or cloud server, among other computing devices. It may include, but is not limited to, a processor and memory. Those skilled in the art will understand that the electronic device may also include more or fewer components, or combinations of certain components, or different components; for example, it may also include input / output devices, network access devices, buses, etc.

[0075] The processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor or any conventional processor.

[0076] The memory can be an internal storage unit of the server, such as a hard drive or RAM. Alternatively, it can be an external storage device, such as a plug-in hard drive, Smart Media Card (SMC), Secure Digital (SD) card, or Flash Card. Furthermore, the memory can include both internal and external storage units. The memory is used to store computer-readable instructions and other programs and data required by the server. It can also be used to temporarily store data that has been output or will be output.

[0077] It should be noted that the information interaction and execution process between the above-mentioned module units are based on the same concept as the method embodiment. For details on their specific functions and technical effects, please refer to the method embodiment section. They will not be repeated here.

[0078] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is merely an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. Furthermore, the specific names of the functional units and modules are only for easy differentiation and are not intended to limit the scope of protection of this application. The specific working process of the units and modules in the above system can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0079] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments of this application can be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include at least: any entity or device capable of carrying the computer program code to a photographing device / terminal device, a recording medium, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium. Examples include USB flash drives, portable hard drives, magnetic disks, or optical disks.

[0080] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0081] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. An assembly control method for an intelligent assembly robot, characterized in that, include: Importing prior information; In the step of importing prior information, firstly, the structural features of the parts are extracted, and the assembly relationship is determined through the relationship between the structural features; then, the key feature elements are decomposed through the structural features, and the position tolerance is obtained by micro-statistical analysis of the position and orientation size deviations, and the shape tolerance is obtained by micro-statistical analysis of the shape size deviations. Generate assembly sequence and process; Based on prior information, assembly sequence, and process, the assembly process flow for each part is generated. Specifically, the assembly model and part model in the system model are analyzed to obtain the set of assembly relationships and assembly methods, forming a table of part assembly relationships and installation positions. Based on this, the part assembly sequence and part assembly process are further derived. The assembly process flow for each part is automatically generated by traversing the part assembly sequence. The assembly process is broken down so that each process corresponds to the motion control of one or more robotic arms. Specifically, it is broken down step by step according to the meta-operation level. The first-level meta-operation nodes are obtained by breaking down the business process, and the second-level meta-operation nodes are obtained by analyzing the basic operations that the robot can perform. Each first-level operation corresponds to several second-level operations, and the second-level operations are reusable node sets. When generating first-level operations, the core algorithm is matrix or model decision-making, collecting field data to form input features, and iterating sequentially to form a sequence of first-level operations. The motion control of each robotic arm is uniformly performed using a state machine model.

2. The intelligent assembly robot assembly control method according to claim 1, characterized in that, In the step of importing prior information, the prior information includes product model, part model, assembly table model and environment model.

3. The intelligent assembly robot assembly control method according to claim 1, characterized in that, In the step of generating assembly sequence and process, the output of assembly sequence analysis is expressed through a tree structure. The tree structure includes one or more trees. In each tree structure, the leaf nodes are the initial parts, and the non-leaf nodes are the local parts formed by assembly. Each assembly action is to assemble multiple parts into a component. The parts are either the initial parts or the components.

4. The intelligent assembly robot assembly control method according to claim 1, characterized in that, The collected field data includes: 3D point cloud data, sensor data, robot feedback status data, and system historical data. Through continuous perception, movement, and interaction with the environment, updated information is obtained from these data.

5. The intelligent assembly robot assembly control method according to claim 1, characterized in that, The step of using a state machine model to uniformly control the motion of each robotic arm means that each robot corresponds to an independent state machine, and each state machine includes 6 states: start, end, exception, path planning, motion, and operation.

6. The intelligent assembly robot assembly control method according to claim 5, characterized in that, The path planning process involves the robot moving from its starting position to the sequence of operation points, visually confirming the deviation from the operation pose, and determining whether to proceed with the operation or re-plan the path and adjust the movement. Two robots move simultaneously, regardless of whether they have a common task, and cooperate continuously. They rely on path planning to control the movement of the robots separately according to constraints given by time or position. Discrete collaboration is triggered or terminated using collaboration conditions; When multiple robots move simultaneously, it is considered cooperative control, and path planning is considered based on time or position. When multiple robots operate on the same target, it is considered collaborative operation in coordination, and multiple interactions occur during state machine control. When robots share data states, they use a blockchain-style message sharing mechanism to synchronize their respective state information, which is a distributed control system.

7. An intelligent assembly robot assembly control system, characterized in that, The method for implementing the assembly control of an intelligent assembly robot as described in any one of claims 1 to 6 includes: The prior information import module is used to import prior information; The assembly sequence and process generation module is used to generate assembly sequences and processes. The assembly action flow generation module is used to generate the assembly action flow for each part based on prior information, assembly sequence, and process. The assembly motion process decomposition module is used to decompose the assembly motion process so that each motion process corresponds to the motion control of one or more robotic arms. The state machine representation module is used to unify the motion control of each robotic arm using a state machine model.

Images