CN116300772B - Full-industry-chain manufacturing method and system based on robot cluster - Google Patents
Full-industry-chain manufacturing method and system based on robot cluster Download PDFInfo
- Publication number
- CN116300772B CN116300772B CN202310561301.0A CN202310561301A CN116300772B CN 116300772 B CN116300772 B CN 116300772B CN 202310561301 A CN202310561301 A CN 202310561301A CN 116300772 B CN116300772 B CN 116300772B
- Authority
- CN
- China
- Prior art keywords
- controller
- network
- production
- intelligent
- platform
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000004519 manufacturing process Methods 0.000 title claims abstract description 201
- 238000000034 method Methods 0.000 claims abstract description 28
- 239000003795 chemical substances by application Substances 0.000 claims description 78
- 230000009471 action Effects 0.000 claims description 44
- 239000013598 vector Substances 0.000 claims description 34
- 230000007246 mechanism Effects 0.000 claims description 15
- 230000006870 function Effects 0.000 claims description 14
- 238000010276 construction Methods 0.000 claims description 11
- 238000004590 computer program Methods 0.000 claims description 10
- 230000008569 process Effects 0.000 claims description 9
- 230000000007 visual effect Effects 0.000 claims description 6
- 238000003860 storage Methods 0.000 claims description 4
- 210000004556 brain Anatomy 0.000 description 13
- 238000013528 artificial neural network Methods 0.000 description 8
- 230000004048 modification Effects 0.000 description 5
- 238000012986 modification Methods 0.000 description 5
- 230000010354 integration Effects 0.000 description 4
- 230000006855 networking Effects 0.000 description 4
- 238000012937 correction Methods 0.000 description 3
- 238000001514 detection method Methods 0.000 description 3
- 230000006872 improvement Effects 0.000 description 3
- 238000005457 optimization Methods 0.000 description 3
- 238000006467 substitution reaction Methods 0.000 description 3
- 230000004075 alteration Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 230000008901 benefit Effects 0.000 description 2
- 230000008878 coupling Effects 0.000 description 2
- 238000010168 coupling process Methods 0.000 description 2
- 238000005859 coupling reaction Methods 0.000 description 2
- 238000007405 data analysis Methods 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 230000009467 reduction Effects 0.000 description 2
- 230000002123 temporal effect Effects 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- 241000238413 Octopus Species 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 239000008358 core component Substances 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000003754 machining Methods 0.000 description 1
- 230000008447 perception Effects 0.000 description 1
- 230000002787 reinforcement Effects 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 230000002195 synergetic effect Effects 0.000 description 1
- 238000011144 upstream manufacturing Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B19/00—Programme-control systems
- G05B19/02—Programme-control systems electric
- G05B19/418—Total factory control, i.e. centrally controlling a plurality of machines, e.g. direct or distributed numerical control [DNC], flexible manufacturing systems [FMS], integrated manufacturing systems [IMS] or computer integrated manufacturing [CIM]
- G05B19/41865—Total factory control, i.e. centrally controlling a plurality of machines, e.g. direct or distributed numerical control [DNC], flexible manufacturing systems [FMS], integrated manufacturing systems [IMS] or computer integrated manufacturing [CIM] characterised by job scheduling, process planning, material flow
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B2219/00—Program-control systems
- G05B2219/30—Nc systems
- G05B2219/32—Operator till task planning
- G05B2219/32252—Scheduling production, machining, job shop
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/02—Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS]
Landscapes
- Engineering & Computer Science (AREA)
- General Engineering & Computer Science (AREA)
- Manufacturing & Machinery (AREA)
- Quality & Reliability (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- General Factory Administration (AREA)
Abstract
The application provides a method and a system for manufacturing a full industrial chain based on a robot cluster, wherein the method comprises the following steps: s1: constructing a first controller, managing an enterprise supply chain through a supply chain intelligent link platform, managing and analyzing industrial data through the industrial intelligent link platform and generating production information; s2: constructing a second controller, connecting the micro-control cloud platform with the first controller and receiving production information, connecting the intelligent network controller with the micro-control cloud platform, and scheduling and controlling the production line according to the production information; s3: and constructing a third controller, connecting the third controller to the intelligent network controller, receiving the production task information of the intelligent network controller, and controlling the manufacturing robot according to the production task information. The application can form the cluster robot mode into the personalized manufacturing of the whole industrial chain through the three-layer controller, can endow the transformed factory with a new business mode, and can provide more flexible products and services.
Description
Technical Field
The application relates to the technical field of intelligent manufacturing, in particular to a full-industry-chain manufacturing method and system based on a robot cluster.
Background
In the face of the development of the current international intelligent manufacturing technology and the reconstruction of the global supply chain, china must develop own intelligent manufacturing new mode, thereby meeting the increasingly personalized consumption demands of people and improving the competitiveness of the whole industrial chain. From the united states automation pipeline to japan lean production to german industry 4.0, there are all future modes of production that have their advantages but are not suitable for large scale personalized manufacturing. Intelligent manufacturing is moving from traditional single robot automated production to clustered full-industry-chain personalized manufacturing production.
The current production mode is: 1. in the equipment level, the existing robot workstation scheme is to simply integrate a robot (motion) controller and a PLC logic unit; 2. in a (digital) plant, the common practice of the digital plant is through a scheme of connecting a bus PLC with other low-level PLCs; 3. at the industrial internet platform (industry) level, the current scheme is oriented to data and application, namely, data analysis and data-based application are provided for enterprises in a cloud software mode. The production mode has the following defects: the robot motion control and PLC logic unit integration scheme at the equipment level enables the robot workstation to be equivalent to an executing mechanism, and has no self-optimizing or learning ability, namely lacks autonomous intelligence. The traditional digital factory scheme has a very slow response speed, and cannot be quickly adjusted, scheduled and planned according to production state fluctuation and order change. The industrial internet platform is used for analyzing data in the production process, and enterprises can not adjust and control the production process by applying the data in the SaaS service form.
Disclosure of Invention
According to the full-industry-chain manufacturing method and system based on the robot clusters, the three-layer controllers are constructed, so that the industry, factories and manufacturing robots form full elements, full processes and quality improvement, cost reduction and synergy of the full-industry-chain on each level, the robot industry chain is reconstructed, the basic capacity of each level is improved, and the technical problems in the process can be solved.
The technical scheme for solving the technical problems is as follows:
in a first aspect, the present application provides a method for manufacturing an all-industrial chain based on a robot cluster, comprising the steps of:
s1: constructing a first controller, wherein the first controller comprises a supply chain intelligent link platform and an industry intelligent link platform, manages an enterprise supply chain through the supply chain intelligent link platform, manages and analyzes industrial data through the industry intelligent link platform and generates production information;
s2: constructing a second controller, wherein the second controller comprises a crowd-sourced network controller and a micro-controlled cloud platform, connecting the micro-controlled cloud platform with the first controller and receiving production information, connecting the crowd-sourced network controller with the micro-controlled cloud platform, and scheduling and controlling a production line according to the production information;
s3: and constructing a third controller, connecting the third controller to the intelligent network controller, receiving the production task information from the intelligent network controller, and controlling the manufacturing robot according to the production task information.
In some embodiments, the S1 comprises:
s11: connecting the supply chain intelligent link platform to different enterprises, planning and scheduling orders of the supply end, the production end and the sales end of the enterprises according to the supply chain of the enterprises, and connecting the supply chain intelligent link platform to the industry intelligent link platform;
s12: the industrial intelligent platform shares industrial information or resources through the server cluster.
In some embodiments, the S2 comprises:
s21: receiving supply chain information and production information from a supply chain intelligent link platform and an industry intelligent link platform through a micro control cloud platform;
s22: and generating production line production task planning and scheduling data through the intelligent network controller according to the supply chain information and the production information.
In some embodiments, the step S3 includes:
s31: the third controller receives production line production task planning and scheduling data of the intelligent network controller and controls the robot according to the production line production task planning and scheduling data;
s32: the robot cluster positions an execution part through a visual sensor and executes production operation through an execution mechanism;
s33: the robot cluster senses the working environment through a feedback sensor to obtain production data, and the production data is fed back to the second controller through the third controller.
In a second aspect, the present application provides a full industrial chain manufacturing system based on a robot cluster, comprising:
the first controller construction module is used for constructing a first controller and comprises a supply chain intelligent connection platform and an industry intelligent connection platform, wherein the supply chain of an enterprise is managed through the supply chain intelligent connection platform, and industrial data is managed and analyzed through the industry intelligent connection platform to generate production information;
the second controller construction module is used for constructing a second controller, comprises a crowd-sourced network controller and a micro-controlled cloud platform, connecting the micro-controlled cloud platform with the first controller and receiving production information, connecting the crowd-sourced network controller with the micro-controlled cloud platform, and scheduling and controlling the production line according to the production information;
and the third controller construction module is used for constructing a third controller, connecting the third controller to the intelligent network controller, receiving the production task information from the intelligent network controller and controlling the manufacturing robot according to the production task information.
In some embodiments, the first controller building module comprises:
the supply chain intelligent link platform sub-module is used for connecting the supply chain intelligent link platform to different enterprises, planning and scheduling orders for the supply end, the production end and the sales end of the enterprises according to the supply chain of the enterprises, and connecting the supply chain intelligent link platform to the industry intelligent link platform;
and the industrial intelligent link platform sub-module is used for sharing industrial information or resources through the server cluster by the industrial intelligent link platform.
In some embodiments, the second controller building module comprises:
the micro-control cloud platform sub-module is used for receiving supply chain information and production information from the supply chain intelligent link platform and the industry intelligent link platform through the micro-control cloud platform;
and the crowd-sourced network controller submodule is used for generating production line production task planning and scheduling data through the crowd-sourced network controller according to the supply chain information and the production information.
In some embodiments, the third controller building module comprises:
the cluster control sub-module is used for enabling the third controller to receive production line production task planning and scheduling data of the cluster intelligent network controller and controlling the robot according to the production line production task planning and scheduling data;
the production execution sub-module is used for enabling the robot cluster to position an execution part through the visual sensor and executing production operation through the execution mechanism;
and the data feedback sub-module is used for enabling the robot cluster to sense the working environment through the feedback sensor to obtain production data, and feeding the production data back to the second controller through the third controller.
In a third aspect, the application provides an electronic device,
in some embodiments, the route planning module further comprises: comprising the following steps:
a memory for storing a computer program;
a processor for implementing the steps of the robot cluster-based whole industry chain manufacturing method according to any one of the above, when executing the computer program.
In a fourth aspect, the present application provides an improvement of a computer readable storage medium, on which a computer program is stored, the computer program implementing the steps of the robot cluster-based whole industry chain manufacturing method according to any one of the above, when being executed by a processor.
The beneficial effects of the application are as follows:
the whole industrial chain manufacturing method and system based on the robot cluster provided by the application have the following beneficial effects: 1. the cluster robot mode can form personalized manufacturing of a whole industrial chain, can endow a transformed factory with a new business mode, and can provide more flexible products and services; 2. all-element, whole-flow and whole-industry chain quality improvement, cost reduction and synergy are formed on different levels by the controllers of all layers, and the robot industry chain is reconstructed, so that the basic capability of all levels is improved; 3. after construction is completed, the accumulated industrial data provides intelligent services based on the data, and service value is generated.
Drawings
FIG. 1 is a flow chart of a method for manufacturing a full industrial chain based on a robot cluster according to the present application;
FIG. 2 is a sub-flowchart of step S1 of the present application;
FIG. 3 is a sub-flowchart of step S2 of the present application;
fig. 4 is a sub-flowchart of step S3 of the present application.
Detailed Description
The principles and features of the present application are described below with reference to the drawings, the examples are illustrated for the purpose of illustrating the application and are not to be construed as limiting the scope of the application.
In order that the above-recited objects, features and advantages of the present application can be more clearly understood, a more particular description of the application will be rendered by reference to specific embodiments thereof which are illustrated in the appended drawings. It is to be understood that the depicted embodiments are some, but not all, embodiments of the present application. The specific embodiments described herein are to be considered in an illustrative rather than a restrictive sense. All other embodiments, which are obtained by a person skilled in the art based on the described embodiments of the application, fall within the scope of protection of the application.
It should be noted that in this document, relational terms such as "first" and "second" and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions.
Fig. 1 is a flowchart of a method for manufacturing a full industrial chain based on a robot cluster according to the present application.
The manufacturing method of the whole industrial chain based on the robot cluster, combined with fig. 1, comprises the following steps:
s1: constructing a first controller, wherein the first controller comprises a supply chain intelligent link platform and an industry intelligent link platform, manages an enterprise supply chain through the supply chain intelligent link platform, manages and analyzes industrial data through the industry intelligent link platform and generates production information;
in some embodiments, in combination with the sub-flowchart of fig. 2, i.e. step S1 of the present scheme, the step S1 includes:
s11: connecting the supply chain intelligent link platform to different enterprises, planning and scheduling orders of the supply end, the production end and the sales end of the enterprises according to the supply chain of the enterprises, and connecting the supply chain intelligent link platform to the industry intelligent link platform;
s12: the industrial intelligent platform shares industrial information or resources through the server cluster.
Specifically, the first controller of this scheme has its core components of two-stage industrial brain-linked network, namely supply chain intelligent-linked platform and industrial intelligent-linked platform. The supply chain intelligent connection platform mainly aims at tap enterprises and covers upstream and downstream ring joints in the enterprise supply chain, namely a CMS system from a supply end, a production end to a sales end, so that the intelligent connection platform on informatization software of the tap enterprises is built. The supply chain intelligent link platform can be connected with the industry intelligent link platform upwards to form linkage between the clustered enterprises and the larger industry platform, and the concrete expression form is a server cluster. The traditional industrial internet platform is oriented to data and application, namely, data analysis application is provided for enterprises in a cloud software mode, and the first controller of the scheme is directly oriented to production control and penetrates into each layer of a factory, so that personalized manufacturing of a whole industrial chain is realized. The first controllers are downwards connected with all the second controllers, all the second controllers are connected with all the third controllers, and equipment, production lines, factories and the like of each stage are all provided with autonomous intelligence, so that the multi-stage distributed brain structure enables the whole industrial chain to truly have real-time personalized manufacturing capacity of the whole industrial chain.
The first controller further includes an eye-plurality of factory level quality detection centers; hand-multiple factory level production machining centers; a plurality of factory internal and external sensing units; foot-multiple factory level logistics center. By means of three-stream fusion of logistics, information flow and value flow, the parts such as brain eyes, hands and feet are subjected to nonlinear coupling, so that the tight coupling of the whole industrial chain is realized, the robot and the manufacturing mode are further reconstructed, and real-time personalized manufacturing of the industrial layer is realized.
S2: constructing a second controller, wherein the second controller comprises a crowd-sourced network controller and a micro-controlled cloud platform, connecting the micro-controlled cloud platform with the first controller and receiving production information, connecting the crowd-sourced network controller with the micro-controlled cloud platform, and scheduling and controlling a production line according to the production information;
in some embodiments, in conjunction with fig. 3, which is a sub-flowchart of step S2 of the present application, the step S2 includes:
s21: receiving supply chain information and production information from a supply chain intelligent link platform and an industry intelligent link platform through a micro control cloud platform;
s22: and generating production line production task planning and scheduling data through the intelligent network controller according to the supply chain information and the production information.
Specifically, the second controller of the scheme utilizes the two-stage distributed brain structure of octopus, and corresponds to a two-stage control-namely overall control and production line/workshop sub-control structure in a factory, and a group intelligent network controller and micro-control cloud platform two-stage architecture is used for modifying the control mode of the existing digital factory integrated overall control PLC and production line PLC, so that intelligent engineering is manufactured in real time in a personalized manner. The intelligent network controller is connected with each device which is transformed into to form an intelligent production line; after all the intelligent production lines are transformed, the intelligent production lines are uploaded to a micro-control cloud platform through a brain networking gateway, so that networking of the whole factory at a second controller level is formed. The intelligent planning, scheduling and control capability of the whole factory are realized by utilizing the autonomous intelligence of each level of brain, the intelligent planning, scheduling and control capability is decomposed to a production line and an equipment level step by step, a real-time corresponding personalized manufacturing system is realized, and the defects of long corresponding period and poor real-time performance of the traditional digital factory are overcome. The micro-control cloud platform is used as a whole factory control center to control all production lines and equipment, so that a large factory brain network is integrally formed, real-time personalized manufacturing capacity is realized, different orders of different products are faced, the production process is switched seamlessly, and efficiency is not lost.
The intelligent network controller adopts an intelligent agent group deep learning reinforcement mechanism, aiming at the production line production task, intelligent agents which are distributed with responsibility for processing the task in the intelligent agent group obtain a deterministic strategy through the intelligent agent group learning mechanism, namely the strategy directly outputs an optimal production line task action value, and the production task information of the production line production task planning and scheduling is carried out based on the optimal action value.
The intelligent network controller builds a strategy network, an action value network, a target strategy network and a target action value network for each intelligent agent in the group based on the intelligent agent group, and the total number of the neural networks is 4. The strategy network and the target strategy network are neural networks with identical structures and different parameters; the action value network and the target action value network are also neural networks of identical structures but different parameters, and the above neural network structures are used for improving training stability. Based on an agent group learning mechanism, an optimal strategy obtained through learning can be used for giving out optimal actions only by utilizing local information when the agent group learning mechanism is applied, and the decentralization training and executing mechanism is realized.
Further, the crowd-sourced network controller defines, for each agent, a reward function for an ith agentHere, a->Defining an optimal deterministic policy function for the ith agent +.>Wherein->Is->Action value of individual agent, < >>For the status observations of the ith agent, < +.>And optimizing the optimal parameter vector for the strategy network.
The number of agents (agents) contained in an Agent group preset in the intelligent network controller is N, and specifically defines the group description parameters of the Agent group as follows:
state spaceThe environments where N agents are located are described;
action spaceWherein->Representing the action space of the i-th agent,；
observation setWherein->Representing a partial observation of the status by the ith agent.
The swarm intelligence network controller performs an initialization process for the swarm of agents, specifically, the initialization is used as iteration 0, and the state of all agents in round 0 is observed asThe method comprises the steps of carrying out a first treatment on the surface of the Initializing->Deterministic policy network of individual agents>Here, a->Is a deep neural network initialization parameter vector that forms a policy network,in order to implement the decentralization execution, the policy of each agent only needs to observe +.>Take action->The method comprises the steps of carrying out a first treatment on the surface of the Initializing->Target policy network of individual agent->Here, a->Is a parameter vector for deep neural network initialization of the target policy network, +.>The method comprises the steps of carrying out a first treatment on the surface of the Initializing->Action value network of individual agent +.>，/>Representing observations of all agents, +.> Action set representing all agents, +.>Is an initialization parameter vector of a deep neural network as a target policy network,the method comprises the steps of carrying out a first treatment on the surface of the Initializing->Target action value network of individual agent +.>Here, a->Is an initialization parameter vector of a deep neural network as a target action value network, +.>。
Further, the crowd-sourced network controller performs m rounds of iterations for the crowd of agents, thereby performing parameter optimization on the strategic network of the crowd of agents. Wherein in each round of iteration M (here, m=1, 2, … M), an initialized state observation is received:the method comprises the steps of carrying out a first treatment on the surface of the At each time step t of the mth iteration (here, t=1, 2, …, < >>，/>Is->Maximum number of time steps in each round), for each agent +.>Generating and executing actions using the current policy function and action exploration (execution), i.e. +.>Rewards +.>And new state observations->The method comprises the steps of carrying out a first treatment on the surface of the Will->Store to experience playback buffer pool->In, here, < >>，The method comprises the steps of carrying out a first treatment on the surface of the Buffer pool for experience playbackThe medium random decimation capacity S is a small batch (mini-batch) sample: />The method comprises the steps of carrying out a first treatment on the surface of the Next, use->Individual samples Calculating TD targets (Temporal Difference Target), i.eHere, a->The method comprises the steps of carrying out a first treatment on the surface of the Calculate timing difference (Temporal Difference Error): />Here, the number of the first and second electrodes, here,the method comprises the steps of carrying out a first treatment on the surface of the Update->Action value network parameter vector of individual agent:here, a->Is the super-parameter of the learning rate,is action value network->About parameter vector->Is a gradient of (2); update->Policy network parameter vector for individual agents: /> Here, a->Is a learning rate super parameter; />The method comprises the steps of carrying out a first treatment on the surface of the Updating two target networks, i.e. updating the target action value network parameter vector +.>And updating the target policy network parameter vectorHere, a->Is a weight coefficient, +.>。
Finally, after the iteration, the crowd-sourced network controller outputs N pieces ofOptimal parameter vector after policy network optimization of intelligent agentObtaining the optimal deterministic strategy function of each agent>. Thus, for the production task from the production line, the intelligent agent which is input to the trained distribution process the task generates each continuous task action value according to the deterministic strategy function by the intelligent agent, the continuous task action value belongs to the optimal production line task action value which is output after passing through the intelligent agent group learning mechanism, and the production task information of the production line production task planning and scheduling is carried out based on the optimal action value.
S3: constructing a third controller, connecting the third controller to the intelligent network controller, receiving production task information from the intelligent network controller, and controlling the manufacturing robot according to the production task information;
in some embodiments, in conjunction with fig. 4, which is a sub-flowchart of step S3 of the present application, the step S3 includes:
s31: the third controller receives production line production task planning and scheduling data of the intelligent network controller and controls the robot according to the production line production task planning and scheduling data;
s32: the robot cluster positions an execution part through a visual sensor and executes production operation through an execution mechanism;
s33: the robot cluster senses the working environment through a feedback sensor to obtain production data, and feeds the production data back to the second controller through the third controller;
specifically, the third controller of the scheme realizes planning, scheduling and control of the equipment layer, is at the bottommost end of industrial brain networking, is also a foundation and a root for establishing connection, and is a ubiquitous controller. The ubiquitous controller integrally realizes robot motion control, logic control based on PLC and intelligent control with feedback correction as a target, and realizes equipment-level self-detection, self-correction, self-adaption and self-organization capacity by simplifying hardware connection, optimizing software and hardware means such as a man-machine interaction interface and the like.
The ubiquitous controller comprises two implementation modes, namely, integrated motion control, logic control and intelligent control, namely, integration of self-detection, self-correction and self-adaptation functions, and independent optimization of a single process or processing engineering; secondly, by utilizing a simplified ubiquitous controller with intelligent control capability, the traditional process equipment can be intelligently modified to form a ubiquitous robot system with intelligent control capability. The ubiquitous controller can be upwards connected with the crowd-sourced network controller to form a production line-level brain networking system with a synergistic effect.
Besides, besides the third controller, the whole personalized manufacturing workstation also needs a brain, eyes, hands, bodies and feet integrated manufacturing robot, namely, the brain and the body jointly solve the complete work task on the production line; the hand-an executing mechanism for completing a specific process, which is composed of a plurality of motors, drives and mechanical constructions, wherein the common executing mechanism comprises a robot and a customized tool clamp thereof; body-various sensors including temperature, pressure and other sensing devices which transmit internal and external information to the brain in real time; the brain, eye, hand, body, foot integration core is to form decision, execute, perception autonomous closed loop, avoid simple integration and assembly, and can reconstruct the personalized manufacturing of equipment level.
The second aspect of the present application also provides a full industrial chain manufacturing system based on a robot cluster, comprising:
the first controller construction module is used for constructing a first controller and comprises a supply chain intelligent connection platform and an industry intelligent connection platform, wherein the supply chain of an enterprise is managed through the supply chain intelligent connection platform, and industrial data is managed and analyzed through the industry intelligent connection platform to generate production information;
the second controller construction module is used for constructing a second controller, comprises a crowd-sourced network controller and a micro-controlled cloud platform, connecting the micro-controlled cloud platform with the first controller and receiving production information, connecting the crowd-sourced network controller with the micro-controlled cloud platform, and scheduling and controlling the production line according to the production information;
and the third controller construction module is used for constructing a third controller, connecting the third controller to the intelligent network controller, receiving the production task information from the intelligent network controller and controlling the manufacturing robot according to the production task information.
In some embodiments, the first controller building module comprises:
the supply chain intelligent link platform sub-module is used for connecting the supply chain intelligent link platform to different enterprises, planning and scheduling orders for the supply end, the production end and the sales end of the enterprises according to the supply chain of the enterprises, and connecting the supply chain intelligent link platform to the industry intelligent link platform;
and the industrial intelligent link platform sub-module is used for sharing industrial information or resources through the server cluster by the industrial intelligent link platform.
In some embodiments, the second controller building module comprises:
the micro-control cloud platform sub-module is used for receiving supply chain information and production information from the supply chain intelligent link platform and the industry intelligent link platform through the micro-control cloud platform;
and the crowd-sourced network controller submodule is used for generating production line production task planning and scheduling data through the crowd-sourced network controller according to the supply chain information and the production information.
In some embodiments, the third controller building module comprises:
the cluster control sub-module is used for enabling the third controller to receive production line production task planning and scheduling data of the cluster intelligent network controller and controlling the robot according to the production line production task planning and scheduling data;
the production execution sub-module is used for enabling the robot cluster to position an execution part through the visual sensor and executing production operation through the execution mechanism;
and the data feedback sub-module is used for enabling the robot cluster to sense the working environment through the feedback sensor to obtain production data, and feeding the production data back to the second controller through the third controller.
The third aspect of the present application also provides an electronic device, including:
a memory for storing a computer program;
a processor for implementing the steps of the robot cluster-based whole industry chain manufacturing method according to any one of the above, when executing the computer program.
The fourth aspect of the present application also provides a computer-readable storage medium, on which a computer program is stored, which when being executed by a processor, implements the steps of the robot cluster-based all-industry-chain manufacturing method as described in any one of the above.
Those skilled in the art will appreciate that while some embodiments described herein include some features but not others included in other embodiments, combinations of features of different embodiments are meant to be within the scope of the application and form different embodiments.
Those skilled in the art will appreciate that the descriptions of the various embodiments are each focused on, and that portions of one embodiment that are not described in detail may be referred to as related descriptions of other embodiments.
Although the embodiments of the present application have been described with reference to the accompanying drawings, those skilled in the art may make various modifications and alterations without departing from the spirit and scope of the present application, and such modifications and alterations fall within the scope of the appended claims, which are to be construed as merely illustrative of the present application, but the scope of the application is not limited thereto, and various equivalent modifications and substitutions will be readily apparent to those skilled in the art within the scope of the present application, and are intended to be included within the scope of the present application. Therefore, the protection scope of the application is subject to the protection scope of the claims.
The present application is not limited to the above embodiments, and various equivalent modifications and substitutions can be easily made by those skilled in the art within the technical scope of the present application, and these modifications and substitutions are intended to be included in the scope of the present application. Therefore, the protection scope of the application is subject to the protection scope of the claims.
Claims (8)
1. The manufacturing method of the whole industrial chain based on the robot cluster is characterized by comprising the following steps of:
s1: constructing a first controller, wherein the first controller comprises a supply chain intelligent link platform and an industry intelligent link platform, manages an enterprise supply chain through the supply chain intelligent link platform, manages and analyzes industrial data through the industry intelligent link platform and generates production information;
s2: constructing a second controller, wherein the second controller comprises a crowd-sourced network controller and a micro-controlled cloud platform, connecting the micro-controlled cloud platform with the first controller and receiving production information, connecting the crowd-sourced network controller with the micro-controlled cloud platform, and scheduling and controlling a production line according to the production information;
s3: constructing a third controller, connecting the third controller to the intelligent network controller, receiving production task information from the intelligent network controller, and controlling the manufacturing robot according to the production task information;
the step S2 comprises the following steps:
s21: receiving supply chain information and production information from a supply chain intelligent link platform and an industry intelligent link platform through a micro control cloud platform;
s22: generating production line production task planning and scheduling data through a crowd-sourced network controller according to the supply chain information and the production information;
the crowd-sourced network controller includes a plurality of agents defining a reward function for an ith agentHere, the number of the first and second electrodes, here,,/>the method comprises the steps of carrying out a first treatment on the surface of the Defining an optimal deterministic policy function for the ith agent +.> Wherein->Is->Action value of individual agent, < >>For the status observations of the ith agent, < +.>An optimal parameter vector after the strategy network is optimized;
the specific process for obtaining the optimal parameter vector is as follows:
initializing a plurality of agents, taking the initialization as iteration of the 0 th round, and observing the states of all the agents of the 0 th round asWherein->Representing the state observation set of all agents for iteration round 0,/->Representing a state observation set of the agent N in the 0 th iteration;
initialize the firstDeterministic policy network of individual agents>Wherein->Initializing a parameter vector for a deterministic policy network;
initialize the firstTarget policy network of individual agent->Wherein->Initializing a parameter vector for a target policy network;
initialize the firstAction value network of individual agent +.>Wherein->Initializing a parameter vector for the action value network;
initialize the firstTarget action value network of individual agent +.>Wherein->Initializing a parameter vector for a target action value network;
performing m rounds of iteration on a plurality of agents, receiving initialization state observations of the agents, and generating actions for each agent using a current policy function and action explorationWherein->Action of agent i representing the current time step t,/->A target policy network parameter vector representing a current time step t;
will beStore to experience playback buffer pool->In (1)/(2)>,/>,,/>；
Buffer pool for experience playbackSmall lot sample with medium random decimation capacity S +.>And use->Sample->(/>) Calculating TD target, i.e.)>Wherein，/>；
Calculating a time sequence difference:wherein->；
Update the firstAction value network parameter vector of individual agent +.>Wherein->Is learning rate superparameter->Is action value network->About parameter vector->Is a gradient of (2);
update the firstPolicy network parameter vector for individual agents:
wherein->Is a learning rate super parameter;；
updating target action value network parameter vectorAnd updating the target policy network parameter vector +.>Wherein->Is a weight coefficient;
outputting the optimized optimal parameter vectors of the strategy network of the N intelligent agentsObtaining the optimal deterministic strategy function of each agent> 。
2. The all-industry-chain manufacturing method based on the robot cluster according to claim 1, wherein S1 comprises:
s11: connecting the supply chain intelligent link platform to different enterprises, planning and scheduling orders of the supply end, the production end and the sales end of the enterprises according to the supply chain of the enterprises, and connecting the supply chain intelligent link platform to the industry intelligent link platform;
s12: the industrial intelligent platform shares industrial information or resources through the server cluster.
3. The method for manufacturing a full industrial chain based on a robot cluster according to claim 2, wherein S3 includes:
s31: the third controller receives production line production task planning and scheduling data of the intelligent network controller and controls the robot according to the production line production task planning and scheduling data;
s32: the robot cluster positions an execution part through a visual sensor and executes production operation through an execution mechanism;
s33: the robot cluster senses the working environment through a feedback sensor to obtain production data, and the production data is fed back to the second controller through the third controller.
4. Full industry chain manufacturing system based on robot cluster, characterized by comprising:
the first controller construction module is used for constructing a first controller and comprises a supply chain intelligent connection platform and an industry intelligent connection platform, wherein the supply chain of an enterprise is managed through the supply chain intelligent connection platform, and industrial data is managed and analyzed through the industry intelligent connection platform to generate production information;
the second controller construction module is used for constructing a second controller, comprises a crowd-sourced network controller and a micro-controlled cloud platform, connecting the micro-controlled cloud platform with the first controller and receiving production information, connecting the crowd-sourced network controller with the micro-controlled cloud platform, and scheduling and controlling the production line according to the production information;
the third controller construction module is used for constructing a third controller, connecting the third controller to the intelligent network controller, receiving production task information from the intelligent network controller and controlling the manufacturing robot according to the production task information;
the second controller building module includes:
the micro-control cloud platform sub-module is used for receiving supply chain information and production information from the supply chain intelligent link platform and the industry intelligent link platform through the micro-control cloud platform;
the intelligent network controller sub-module is used for generating production line production task planning and scheduling data through the intelligent network controller according to the supply chain information and the production information;
the crowd-sourced network controller includes a plurality of agents defining a reward function for an ith agentHere, the number of the first and second electrodes, here,,/>the method comprises the steps of carrying out a first treatment on the surface of the Defining an optimal deterministic policy function for the ith agent +.> Wherein->Is->Action value of individual agent, < >>For the status observations of the ith agent, < +.>An optimal parameter vector after the strategy network is optimized;
the specific process for obtaining the optimal parameter vector is as follows:
initializing a plurality of agents, taking the initialization as iteration of the 0 th round, and observing the states of all the agents of the 0 th round asWherein->Representing the state observation set of all agents for iteration round 0,/->Representing a state observation set of the agent N in the 0 th iteration;
initialize the firstDeterministic policy network of individual agents>Wherein->Initializing a parameter vector for a deterministic policy network;
initialize the firstTarget policy network of individual agent->Wherein->Initializing a parameter vector for a target policy network;
initialize the firstAction value network of individual agent +.>Wherein->Initializing a parameter vector for the action value network;
initialize the firstTarget action value network of individual agent +.>Wherein->Initializing a parameter vector for a target action value network;
performing m rounds of iteration on a plurality of agents, receiving initialization state observations of the agents, and generating actions for each agent using a current policy function and action explorationWherein->Action of agent i representing the current time step t,/->A target policy network parameter vector representing a current time step t;
will beStore to experience playback buffer pool->In (1)/(2)>,/>,,/>；
Buffer pool for experience playbackSmall lot sample with medium random decimation capacity S +.>And use->Sample->(/>) Calculating TD target, i.e.)>Wherein->，/>；
Calculating a time sequence difference:wherein->；
Update the firstAction value network parameter vector of individual agent +.>WhereinIs learning rate superparameter->Is action value network->About parameter vector->Is a gradient of (2);
update the firstPolicy network parameter vector for individual agents:
wherein->Is a learning rate super parameter; />；
Updating target action value network parameter vectorAnd updating the target policy network parameter vector +.>Wherein->Is a weight coefficient;
outputting the optimized optimal parameter vectors of the strategy network of the N intelligent agentsObtaining the optimal deterministic strategy function of each agent> 。
5. The robot cluster-based full industry chain manufacturing system of claim 4, wherein the first controller building block comprises:
the supply chain intelligent link platform sub-module is used for connecting the supply chain intelligent link platform to different enterprises, planning and scheduling orders for the supply end, the production end and the sales end of the enterprises according to the supply chain of the enterprises, and connecting the supply chain intelligent link platform to the industry intelligent link platform;
and the industrial intelligent link platform sub-module is used for sharing industrial information or resources through the server cluster by the industrial intelligent link platform.
6. The robot-cluster-based full industry chain manufacturing system of claim 5, wherein the third controller building module comprises:
the cluster control sub-module is used for enabling the third controller to receive production line production task planning and scheduling data of the cluster intelligent network controller and controlling the robot according to the production line production task planning and scheduling data;
the production execution sub-module is used for enabling the robot cluster to position an execution part through the visual sensor and executing production operation through the execution mechanism;
and the data feedback sub-module is used for enabling the robot cluster to sense the working environment through the feedback sensor to obtain production data, and feeding the production data back to the second controller through the third controller.
7. An electronic device, comprising:
a memory for storing a computer program;
a processor for implementing the steps of the robot cluster-based full industry chain manufacturing method according to any one of claims 1 to 3 when executing the computer program.
8. A computer-readable storage medium, characterized in that the computer-readable storage medium has stored thereon a computer program which, when executed by a processor, implements the steps of the robot cluster-based all industrial chain manufacturing method according to any of claims 1 to 3.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310561301.0A CN116300772B (en) | 2023-05-18 | 2023-05-18 | Full-industry-chain manufacturing method and system based on robot cluster |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310561301.0A CN116300772B (en) | 2023-05-18 | 2023-05-18 | Full-industry-chain manufacturing method and system based on robot cluster |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116300772A CN116300772A (en) | 2023-06-23 |
| CN116300772B true CN116300772B (en) | 2023-09-15 |
Family
ID=86818915
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310561301.0A Active CN116300772B (en) | 2023-05-18 | 2023-05-18 | Full-industry-chain manufacturing method and system based on robot cluster |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116300772B (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110837918A (en) * | 2019-10-08 | 2020-02-25 | 北京航空航天大学 | Multi-layer integrated scheduling mechanism |
| CN112180724A (en) * | 2020-09-25 | 2021-01-05 | 中国人民解放军军事科学院国防科技创新研究院 | Training method and system for multi-agent cooperative cooperation under interference condition |
| CN112508644A (en) * | 2020-12-02 | 2021-03-16 | 贵州航天云网科技有限公司 | Linkage supply intelligent cooperative scheduling system based on cloud platform |
| CN113361830A (en) * | 2021-08-09 | 2021-09-07 | 阿里巴巴（中国）有限公司 | Digital production control system, production scheduling method, device, equipment and storage medium |
| CN113433909A (en) * | 2021-06-28 | 2021-09-24 | 成都飞机工业（集团）有限责任公司 | Manufacturing process control system architecture for aviation industry cluster |
| CN114169766A (en) * | 2021-12-09 | 2022-03-11 | 丁业琴 | Production management method and system for industrial capacity allocation |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20220413455A1 (en) * | 2020-11-13 | 2022-12-29 | Zhejiang University | Adaptive-learning intelligent scheduling unified computing frame and system for industrial personalized customized production |
-
2023
- 2023-05-18 CN CN202310561301.0A patent/CN116300772B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110837918A (en) * | 2019-10-08 | 2020-02-25 | 北京航空航天大学 | Multi-layer integrated scheduling mechanism |
| CN112180724A (en) * | 2020-09-25 | 2021-01-05 | 中国人民解放军军事科学院国防科技创新研究院 | Training method and system for multi-agent cooperative cooperation under interference condition |
| CN112508644A (en) * | 2020-12-02 | 2021-03-16 | 贵州航天云网科技有限公司 | Linkage supply intelligent cooperative scheduling system based on cloud platform |
| CN113433909A (en) * | 2021-06-28 | 2021-09-24 | 成都飞机工业（集团）有限责任公司 | Manufacturing process control system architecture for aviation industry cluster |
| CN113361830A (en) * | 2021-08-09 | 2021-09-07 | 阿里巴巴（中国）有限公司 | Digital production control system, production scheduling method, device, equipment and storage medium |
| CN114169766A (en) * | 2021-12-09 | 2022-03-11 | 丁业琴 | Production management method and system for industrial capacity allocation |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116300772A (en) | 2023-06-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Zhou et al. | Multi-agent reinforcement learning for online scheduling in smart factories | |
| Xia et al. | A digital twin to train deep reinforcement learning agent for smart manufacturing plants: Environment, interfaces and intelligence | |
| Lee et al. | Digital twin-driven deep reinforcement learning for adaptive task allocation in robotic construction | |
| Kozák et al. | Research and education for industry 4.0: Present development | |
| Qiang et al. | Reinforcement learning model, algorithms and its application | |
| CN110023965A (en) | For selecting the neural network of the movement executed by intelligent robot body | |
| Lin et al. | Evolutionary digital twin: A new approach for intelligent industrial product development | |
| Vatankhah Barenji et al. | A digital twin-driven approach towards smart manufacturing: reduced energy consumption for a robotic cell | |
| Ouahabi et al. | A distributed digital twin architecture for shop floor monitoring based on edge-cloud collaboration | |
| Yu et al. | An adaptive neural network switching control approach of robotic manipulators for trajectory tracking | |
| Oyama et al. | Inverse kinematics learning for robotic arms with fewer degrees of freedom by modular neural network systems | |
| Latif et al. | A case study of digital twin for a manufacturing process involving human interactions | |
| CN116604532A (en) | Intelligent control method for upper limb rehabilitation robot | |
| Singh et al. | Energy-efficient gait optimization of snake-like modular robots by using multiobjective reinforcement learning and a fuzzy inference system | |
| Matsuoka et al. | Recovery planning of industrial robots based on semantic information of failures and time-dependent utility | |
| De Winter et al. | Autonomous assembly planning of demonstrated skills with reinforcement learning in simulation | |
| CN116300772B (en) | Full-industry-chain manufacturing method and system based on robot cluster | |
| Kemmerling et al. | Towards production-ready reinforcement learning scheduling agents: A hybrid two-step training approach based on discrete-event simulations | |
| Oniz et al. | Variable-structure-systems based approach for online learning of spiking neural networks and its experimental evaluation | |
| Hoover et al. | Industry 4.0 trends in intelligent manufacturing automation exploring machine learning | |
| Yang et al. | Least mean p-power extreme learning machine for obstacle avoidance of a mobile robot | |
| Eberle et al. | Synchronization-based control for a collaborative robot | |
| Atighehchian et al. | An environment-driven, function-based approach to dynamic single-machine scheduling | |
| Kaur et al. | Learning robotic skills through reinforcement learning | |
| Gheorghe | The development of intelligent industry in Romania |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |