A robotic matrix production scheduling system and method

Abstract

The application discloses a kind of production deployment system and method of robot matrix, comprising: construction module, for obtaining the equipment information of mobile collaborative robot, and constructing robot matrix according to equipment information;Internet connection docking module, for supporting and docking with the existing WMS / MES data system of warehouse and factory, actively responding to the dynamic adjustment of production appeal, generating deployment task according to the dynamic adjustment;Signal transceiver module, for supporting a variety of network protocol adaptation, and integrated by industrial equipment and rule engine module, task engine module;Rule engine module, for supporting internal and external data correlation in system, low-code visual self-defined rule mechanism logic;Task engine module, for executing deployment task, realizing dynamic allocation, dynamic switching and dynamic adjustment of robot matrix under the support of global optimal strategy.It realizes dynamic adjustment based on production appeal, generates deployment task, optimally adjusts robot matrix, and is convenient for improving production efficiency.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of robot production scheduling, in particular to a robot matrix production scheduling system and method. BACKGROUND

[0002] Currently, mobile robots are used to assist in performing simple and single production tasks in industrial production processes, which cannot be used to perform complex production tasks by configuring various types of mobile robots, and there is no command system that can dynamically adjust and schedule various types of mobile robots according to production tasks, which is not conducive to improving production rate and automation. SUMMARY

[0003] The present application aims to at least one of the above technical problems in the technical field. To this end, the first object of the present application is to provide a robot matrix production scheduling system to solve the complex operational problems of task allocation, path planning, traffic control, group cooperation and resource conservation in the scene of unmanned factory, so that the whole unmanned factory system can accurately meet the changes of customer reservation tasks or dynamic production rhythm and respond to demand in real time, while the operation success rate, work efficiency and energy consumption of all related equipment are optimized, and the whole system can realize efficient, full-time and autonomous operation.

[0004] The second object of the present application is to provide a robot matrix production scheduling method.

[0005] To achieve the above object, the first aspect of the present application provides a robot matrix production scheduling system, comprising:

[0006] A construction module is configured to obtain device information of a mobile collaborative robot and construct a robot matrix according to the device information.

[0007] A material connection module is configured to support connection with an existing WMS / MES data system of a warehouse and a factory, actively respond to dynamic adjustment of production demands, and generate a scheduling task according to the dynamic adjustment; the dynamic adjustment includes position, specification, rhythm and intention.

[0008] A signal transceiver module is configured to support multiple network protocol adaptation and integration with an industrial device and a rule engine module, a task engine module.

[0009] A rule engine module is configured to support internal and external data association, low-code visual self-defined rule mechanism logic.

[0010] A task engine module is configured to execute a scheduling task and realize dynamic allocation, dynamic switching and dynamic adjustment of a robot matrix under the support of a global optimal strategy.

[0011] According to some embodiments of the present application, the mobile collaborative robot comprises a differential single-arm mobile collaborative robot, a differential double-arm mobile collaborative robot, an omnidirectional double-arm mobile collaborative robot and an omnidirectional single-arm mobile collaborative robot.

[0012] According to some embodiments of the present application, further comprising: a group collaboration module, configured to support multi-IoT docking, dynamically match the device systems existing in the system with each other, realize precise control of multi-scene, multi-device and multi-demand, and maximize utilization and dynamic matching.

[0013] According to some embodiments of the present application, further comprising: a traffic control module, configured to support flexible configuration and automatic guidance of the robot matrix in the global drivable road traffic rules.

[0014] According to some embodiments of the present application, further comprising: a monitoring and alarming module, configured to monitor the action of the robot matrix, and issue an alarm prompt when an abnormal situation is determined according to the monitoring result.

[0015] According to some embodiments of the present application, the IoT docking module comprises:

[0016] The acquisition module is configured to dock with the existing WMS / MES data system of the warehouse and factory, and acquire warehouse data and production data of the factory;

[0017] The first determination module is configured to perform data fusion on the warehouse data and the production data, determine a target feature vector, and perform grid processing on the target feature vector;

[0018] The second determination module is configured to:

[0019] divide the grid-processed target feature vector into a plurality of sub-target feature vectors based on a quadtree algorithm;

[0020] determine the data type and the corresponding visualization processing requirement corresponding to each sub-target feature vector, and distribute each sub-target feature vector and the visualization processing requirement to a corresponding visualization processing node;

[0021] extract a plurality of visualization labels from the sub-target feature vectors and the visualization processing requirements based on the visualization processing node, each visualization label corresponds to a scene feature, match the scene feature with a preset scene feature in a preset scene feature database, determine a preset scene feature corresponding to the maximum matching degree as a target preset scene feature;

[0022] obtain a plurality of target preset scene features based on the plurality of visualization labels, eliminate the target preset scene features with a matching degree less than a preset threshold, and determine a target set;

[0023] The first fusion module is configured to:

[0024] determine a first fusion relationship of preset scene features of each target in the target set, and construct a sub-scene;

[0025] determine a second fusion relationship of each sub-scene, and determine a fusion visual scene;

[0026] a dynamic adjustment module, configured to actively respond to dynamic adjustment of production demands based on the fusion visual scene, and generate a deployment task according to the dynamic adjustment.

[0027] According to some embodiments of the present application, the first determining module comprises:

[0028] a extracting module, configured to input warehouse data and production data into a feature word extraction model, and determine first feature words and second feature words;

[0029] a second fusion module, configured to:

[0030] input the warehouse data and the first feature words into a word frequency statistical model, and determine first attribute information of the warehouse data; the word frequency statistical model is configured to determine the first attribute information of the input warehouse data according to word frequency statistics of each item of content in the input warehouse data and word frequency statistics of the input first feature words;

[0031] input the production data and the second feature words into a data time sequence model, and determine second attribute information of the production data; the data time sequence model is configured to determine the second attribute information of the input production data according to vector representation of each item of content in the input production data, position information and time sequence information of each item of content, and vector representation of the input second feature words;

[0032] extract first to-be-fused data in the warehouse data and second to-be-fused data in the production data according to the first attribute information and the second attribute information, perform data fusion according to the first to-be-fused data and the second to-be-fused data, determine a target feature vector, and perform grid processing on the target feature vector.

[0033] According to some embodiments of the present application, the monitoring and alarming module comprises:

[0034] a establishing module, configured to determine a monitoring video image of a monitoring area where the robot matrix is located, and establish a background model of a monitoring area scene according to the monitoring video image;

[0035] a third determining module, configured to:

[0036] extract a moving object foreground of the monitoring image from the background model based on a SILTP texture feature, perform connected domain analysis and target segmentation on the moving object foreground of the monitoring image, determine a segmentation result, perform feature classification based on a feature classifier to extract an outline of the moving object;

[0037] comparing the contour of the moving object with a preset contour, and determining each type of mobile collaborative robot in the robot matrix according to a comparison result;

[0038] a fourth determining module configured to:

[0039] editing and combining the monitored video images to determine a video frame corresponding to each mobile collaborative robot;

[0040] a fifth determining module configured to determine action information of the corresponding mobile collaborative robot based on the video frame;

[0041] an alarm module configured to compare the action information of each mobile collaborative robot with preset action information, determine whether the action of the mobile collaborative robot is abnormal according to a comparison result, mark the abnormal mobile collaborative robot, and issue an alarm prompt.

[0042] According to some embodiments of the present application, the fifth determining module comprises:

[0043] determining coordinates of a motion node of the mobile collaborative robot in each image of the video frame;

[0044] connecting all motion nodes of the mobile collaborative robot in each frame;

[0045] each motion node is connected to the same motion node in a continuous coordinate system;

[0046] constructing a time-space sequence graph of the motion node of the mobile collaborative robot based on time sequence information of the video frame, and determining the action information of the corresponding mobile collaborative robot according to the time-space sequence graph.

[0047] To achieve the above-mentioned purpose, the second aspect of the embodiments of the present application proposes a production deployment method of a robot matrix, comprising:

[0048] obtaining equipment information of the mobile collaborative robot, and constructing a robot matrix according to the equipment information;

[0049] supporting docking with an existing WMS / MES data system of a warehouse and a factory based on an Internet of Things docking module, actively responding to dynamic adjustment of production demands, generating a deployment task according to the dynamic adjustment; the dynamic adjustment comprises position, specification, beat, and intention;

[0050] supporting multiple network protocol adaptation based on a signal transceiver module, and integrating with an industrial device and a rule engine module, a task engine module;

[0051] supporting internal and external data association of the system, and low-code visual self-defined rule mechanism logic based on the rule engine module;

[0052] The task engine module is used to perform the deployment task, and the robot matrix is dynamically allocated, switched and adjusted under the support of the global optimal strategy.

[0053] The present application provides a production deployment system and method of robot matrix, which solves the complex operation problems such as task allocation, path planning, traffic control, group cooperation and resource saving in the unmanned factory scene, so that the whole unmanned factory system can accurately meet the changes of customer appointment tasks or dynamic production rhythm to respond to the demand in real time, while the operation success rate, work efficiency and energy consumption of all related equipment are adjusted to the optimum, and the whole system can realize efficient, full-time and autonomous operation.

[0054] Other features and advantages of the present application will be set forth in the following description, and in part will become apparent to those skilled in the art from the description, or can be learned by practice of the present application. The objects and other advantages of the present application can be achieved and obtained by the structure particularly pointed out in the written description and the accompanying drawings.

[0055] The technical solutions of the present application will be further described in detail below with the help of the accompanying drawings and examples. BRIEF DESCRIPTION OF DRAWINGS

[0056] The accompanying drawings are used to provide a further understanding of the present application, and constitute a part of the specification, together with the embodiments of the present application, to explain the present application, and do not constitute a limitation on the present application. In the drawings:

[0057] Figure 1 is a block diagram of a production deployment system of a robot matrix according to an embodiment of the present application;

[0058] Figure 2 is a schematic diagram of a robot matrix according to an embodiment of the present application;

[0059] Figure 3 is a flowchart of a production deployment method of a robot matrix according to an embodiment of the present application. DETAILED DESCRIPTION

[0060] The preferred embodiments of the present application will be described below in conjunction with the accompanying drawings, and it should be understood that the preferred embodiments described herein are only used to illustrate and explain the present application, and do not limit the present application.

[0061] As shown in Figure 1 , the first aspect embodiment of the present application provides a production deployment system of a robot matrix, which comprises:

[0062] A construction module is configured to obtain device information of a mobile collaborative robot, and construct a robot matrix according to the device information.

[0063] The Internet of Things docking module is used for supporting docking with the existing WMS / MES data system of the warehouse and factory, actively responding to dynamic adjustment of production demands, and generating a deployment task according to the dynamic adjustment; the dynamic adjustment includes position, specification, beat and intention.

[0064] The signal transceiver module is used for supporting adaptation of various network protocols and integrating with the rule engine module and the task engine module through the industrial equipment.

[0065] The rule engine module is used for supporting internal and external data association, low-code visual self-defined rule mechanism logic.

[0066] The task engine module is used for executing the deployment task, realizing dynamic allocation, dynamic switching and dynamic adjustment of the robot matrix under the support of the global optimal strategy.

[0067] In the embodiment, WMS is the abbreviation of Warehouse Management System, which is a management system that comprehensively uses functions such as warehouse entry business, warehouse exit business, warehouse allocation, inventory allocation and virtual warehouse management to manage batch management, material correspondence, inventory checking, quality inspection management, virtual warehouse management and real-time inventory management, effectively controls and tracks the whole process of logistics and cost management of warehouse business, and realizes or perfects the warehouse information management of an enterprise. MES (Manufacturing Execution System) is a production information management system for the execution layer of a manufacturing enterprise workshop.

[0068] The above technical scheme has the beneficial effects: the production deployment system of the robot matrix is provided, complex operation problems such as task allocation, path planning, traffic control, group cooperation and resource saving in the unmanned factory scene are solved, the whole unmanned factory system can accurately meet the change of customer appointment tasks or dynamic production beat to respond to the demand in real time while the operation success rate, work efficiency and energy loss of all related equipment are adjusted to the optimum, and the whole system can realize efficient, full-time and autonomous operation.

[0069] In an embodiment, different types of mobile collaborative robots are included in the robot matrix, for example, the robot matrix A includes differential single-arm mobile collaborative robots and differential double-arm mobile collaborative robots; the robot matrix B includes differential single-arm mobile collaborative robots, differential double-arm mobile collaborative robots, and omnidirectional double-arm mobile collaborative robots; the robot matrix C includes differential single-arm mobile collaborative robots, differential double-arm mobile collaborative robots, omnidirectional double-arm mobile collaborative robots, and omnidirectional single-arm mobile collaborative robots, and the like. Each robot matrix is configured with different types of mobile collaborative robots to complete corresponding processes or corresponding production tasks, and matrix production can be performed based on the robot matrix. In the production process, the mobile collaborative robots in the robot matrix and other robot matrices are dynamically adjusted according to the adjustment of the production task, and the optimal adjustment of the robot matrix facilitates the improvement of production efficiency.

[0070] According to some embodiments of the present application, the device management module is further configured to support management and control of the product line and the devices of the mobile collaborative robots connected to the product line.

[0071] As shown in Figure 2 According to some embodiments of the present application, the mobile collaborative robots include differential single-arm mobile collaborative robots, differential double-arm mobile collaborative robots, omnidirectional double-arm mobile collaborative robots, and omnidirectional single-arm mobile collaborative robots.

[0072] According to some embodiments of the present application, the group collaboration module is further configured to support multi-thing connection, dynamically match the device systems existing in the system, and realize precise control, maximum utilization, and dynamic matching of multiple scenes, multiple devices, and multiple needs.

[0073] According to some embodiments of the present application, the traffic control module is further configured to support flexible configuration and automatic guidance of the robot matrix in the global drivable road traffic rules.

[0074] According to some embodiments of the present application, the monitoring and alarming module is further configured to monitor the actions of the robot matrix, and issue an alarm prompt when an abnormal situation is determined according to the monitoring result.

[0075] The above technical solution has the following beneficial effects: the abnormal self-triggering, response, and self-processing of the automatic robot and system are supported, and the entire process can be truly and independently executed in a complete manner without interruption for 7*24 hours.

[0076] According to some embodiments of the present application, the thing connection module includes:

[0077] The acquisition module is configured to connect with the existing WMS / MES data system of the warehouse and the factory, and acquire warehouse data and production data of the factory.

[0078] The first determining module is configured to perform data fusion on the warehouse data and the production data, determine a target feature vector, and perform grid processing on the target feature vector.

[0079] The second determining module is configured to:

[0080] divide the grid-processed target feature vector into a plurality of sub-target feature vectors based on a quadtree algorithm;

[0081] determine a data type and a corresponding visual processing requirement corresponding to each sub-target feature vector, and distribute each sub-target feature vector and the visual processing requirement to a corresponding visual processing node;

[0082] extract a plurality of visual labels from the sub-target feature vectors and the visual processing requirements based on the visual processing node, each visual label corresponding to a scene feature, match the scene feature with a preset scene feature in a preset scene feature database, determine a preset scene feature corresponding to a maximum matching degree as a target preset scene feature;

[0083] obtain a plurality of target preset scene features based on the plurality of visual labels, eliminate target preset scene features with a matching degree less than a preset threshold, and determine a target set;

[0084] The first fusion module is configured to:

[0085] determine a first fusion relationship of each target preset scene feature in the target set, and construct a sub-scene;

[0086] determine a second fusion relationship of each sub-scene, and determine a fused visual scene;

[0087] The dynamic adjustment module is configured to actively respond to dynamic adjustments of production demands based on the fused visual scene, and generate a deployment task according to the dynamic adjustments.

[0088] The working principle of the above technical solution is as follows: In this embodiment, the warehouse data includes the model, quantity, product eligibility judgment information, and storage location of each product.

[0089] In this embodiment, the production data includes information of each production process, and each production process corresponds to product information.

[0090] In this embodiment, the warehouse data and the production data are fused, which facilitates the fusion of corresponding data in the warehouse data and the production data, realizes integrated processing of data, improves the correlation between the warehouse data and the production data, and avoids the issuance of incorrect production tasks due to the inequality and non-sharing of the warehouse data and the production data.

[0091] In the embodiment, the target feature vector is an overall data model generated based on the fused data, so as to facilitate efficient processing of the data.

[0092] In the embodiment, the target feature vector is subjected to grid processing, so as to facilitate overall quantization of the target feature vector and determination of the local features.

[0093] In the embodiment, each sub-target feature vector corresponds to a data type.

[0094] In the embodiment, the different types of sub-target feature vectors are processed based on the visualization processing requirements by the corresponding visualization processing nodes.

[0095] The above technical solution has the following beneficial effects: the warehouse data and the production data are fused to improve the correlation between the two and obtain the target feature vector; the target feature vector is subjected to grid processing to facilitate determination of the local features; in the determination of the local features, the sub-target feature vectors are processed based on the visualization processing requirements by the corresponding visualization processing nodes, the first fusion relationship of the target preset scene features in the target set is determined based on the first fusion module, the sub-scenes are constructed, the second fusion relationship of each sub-scene is determined, the fusion visualization scene is determined, the target feature vector is subjected to visualization processing, the analysis accuracy of the data is improved, and when the dynamic adjustment of the production demand is obtained, the adjustment parameters can be accurately and quickly determined, and the allocation task is generated.

[0096] According to some embodiments of the application, the first determination module comprises:

[0097] The extraction module is configured to input the warehouse data and the production data into a feature word extraction model to determine first feature words and second feature words.

[0098] The second fusion module is configured to:

[0099] input the warehouse data and the first feature words into a word frequency statistical model to determine first attribute information of the warehouse data; the word frequency statistical model is configured to determine the first attribute information of the input warehouse data according to the word frequency statistics of each item of content in the input warehouse data and the word frequency statistics of the input first feature words;

[0100] input the production data and the second feature words into a data time sequence model to determine second attribute information of the generated data; the data time sequence model is configured to determine the second attribute information of the input production data according to the vector representation of each item of content in the input production data, the position information and time sequence information of each item of content, and the vector representation of the input second feature words;

[0101] The first to-be-fused data of the middle warehouse data and the second to-be-fused data in the production data are extracted according to the first attribute information and the second attribute information, data fusion is performed according to the first to-be-fused data and the second to-be-fused data, a target feature vector is determined, and the target feature vector is subjected to grid processing.

[0102] The technical scheme has the beneficial effects that: based on the characteristics of the warehouse data, more product quantity statistical data are subjected to word frequency statistics, overall classification of the warehouse data is realized. Based on the characteristics of the production process information of the production data, processing is performed based on a data time sequence model, efficient arrangement and statistics of the production data are realized, the first attribute information and the second attribute information are determined, the first to-be-fused data of the middle warehouse data and the second to-be-fused data in the production data are more accurately and quickly extracted according to the first attribute information and the second attribute information, and the accuracy of the target feature vector is improved.

[0103] According to some embodiments of the application, the monitoring and alarming module comprises:

[0104] The establishing module is configured to determine a monitoring video image of a monitoring area where the robot matrix is located, and establish a background model of a monitoring area scene according to the monitoring video image;

[0105] The third determining module is configured to:

[0106] The third determining module is configured to:

[0107] The third determining module is configured to:

[0108] The fourth determining module is configured to:

[0109] The fourth determining module is configured to:

[0110] The fifth determining module is configured to determine the action information of the corresponding mobile collaborative robot based on the video frame.

[0111] The alarming module is configured to compare the action information of each mobile collaborative robot with preset action information, judge whether the action of the mobile collaborative robot is abnormal according to a comparison result, mark the abnormal mobile collaborative robot, and issue an alarm prompt.

[0112] The working principle of the technical scheme is as follows: in the embodiment, the video monitoring device collects monitoring video images of a monitoring area, and a background model of a monitoring area scene is established according to the monitoring video images.

[0113] In the embodiment, a scene-based adaptive threshold (t) updating strategy is added on the basis of the SILTP texture feature, the threshold t is adaptively adjusted within a certain range according to the statistical change of scene pixels, and the background learning and updating strategy of the codebook symbol weighting of time is combined to realize video detection of the monitoring video images.

[0114] In the embodiment, the moving object includes various types of mobile collaborative robots.

[0115] In the embodiment, the preset contour includes standard contour parameters of various types of mobile collaborative robots.

[0116] In the embodiment, only the corresponding mobile collaborative robot is included in the video frame, and other types of mobile collaborative robots are not included.

[0117] The beneficial effects of the technical scheme are as follows: a background model of a monitoring area scene is established according to monitoring video images, a moving object foreground of a monitoring image is extracted from the background model based on a SILTP texture feature, connected domain analysis and target segmentation are performed on the moving object foreground of the monitoring image, a segmentation result is determined, a contour of the moving object is extracted based on feature classification by a feature classifier; and accurate segmentation of each moving object is facilitated. The contour of the moving object is compared with a preset contour, and each type of mobile collaborative robot in the robot matrix is determined according to a comparison result; each type of mobile collaborative robot after segmentation is identified and marked. For example, it is determined that contour A corresponds to a differential single-arm mobile collaborative robot, and contour B corresponds to a differential double-arm mobile collaborative robot. The monitoring video images are edited and combined to determine the video frame corresponding to each mobile collaborative robot; the video frame corresponding to each mobile collaborative robot is accurately obtained, and the action information of the corresponding mobile collaborative robot is more comprehensively and accurately analyzed. The preset action information of each mobile collaborative robot is different, so the alarm module is used to compare the action information of each mobile collaborative robot with the preset action information, to determine whether the action of the mobile collaborative robot is abnormal according to a comparison result, to mark the abnormal mobile collaborative robot, and to issue an alarm prompt. The abnormal mobile collaborative robot is accurately determined, and the abnormal situation is timely handled.

[0118] According to some embodiments of the application, the fifth determination module comprises:

[0119] The coordinates of the motion nodes of the mobile collaborative robot in each image of the video frame are determined.

[0120] connecting all motion nodes of the mobile collaborative robot within each frame;

[0121] each motion node is connected to the same motion node in a continuous coordinate system;

[0122] based on the time sequence information of the video frames, constructing a time-space sequence graph of the motion nodes of the mobile collaborative robot, and determining the action information of the corresponding mobile collaborative robot according to the time-space sequence graph.

[0123] The beneficial effects of the above technical solutions are: based on the coordinates of the motion nodes of the mobile collaborative robot in each image of the video frames; connecting all motion nodes of the mobile collaborative robot within each frame; each motion node is connected to the same motion node in a continuous coordinate system; based on the time sequence information of the video frames, constructing a time-space sequence graph of the motion nodes of the mobile collaborative robot, and determining the action information of the corresponding mobile collaborative robot according to the time-space sequence graph. Based on the combination of time-space information and the connection of the motion nodes of the mobile collaborative robot, the action information of the mobile collaborative robot can be accurately determined.

[0124] As shown in Figure 3 the second aspect embodiment of the present application proposes a production scheduling method of a robot matrix, comprising steps S1-S5:

[0125] S1, obtaining device information of a mobile collaborative robot, and constructing a robot matrix according to the device information;

[0126] S2, based on the Internet connection module, supporting the connection with the existing WMS / MES data system of the warehouse and factory, actively responding to the dynamic adjustment of the production demand, generating a scheduling task according to the dynamic adjustment; the dynamic adjustment includes position, specification, beat and intention;

[0127] S3, based on the signal transceiver module, supporting multiple network protocol adaptation, and integrating with the industrial device and the rule engine module, the task engine module;

[0128] S4, based on the rule engine module, supporting internal and external data association, low-code visual self-defined rule mechanism logic;

[0129] S5, based on the task engine module, executing the scheduling task, realizing the dynamic allocation, dynamic switching and dynamic adjustment of the robot matrix under the support of the global optimal strategy.

[0130] The application provides a production scheduling method of a robot matrix, solves complex operation problems such as task allocation, path planning, traffic control, group cooperation and resource saving in an unmanned factory scene, makes the whole unmanned factory system accurately satisfy the change of a customer reservation task or a dynamic production rhythm and respond to the demand in real time, meanwhile, adjusts the operation success rate, work efficiency and energy loss of all related equipment to be optimal, and guarantees that the whole system can realize efficient, full-time and autonomous operation.

[0131] Obviously, various modifications and changes can be made to the present application by those skilled in the art without departing from the spirit and scope of the present application. Thus, if these modifications and changes are within the scope of the present application, the present application is intended to include these modifications and changes.

Claims

1. A production and allocation system for a robot matrix, characterized in that, include: The module is used to acquire device information of mobile collaborative robots and build a robot matrix based on the device information. The IoT docking module is used to support docking with the existing WMS / MES data systems in warehouses and factories, proactively respond to dynamic adjustments in production demands, and generate allocation tasks based on the dynamic adjustments; the dynamic adjustments include location, specifications, cycle time, and intent; The signal transceiver module is used to support the adaptation of multiple network protocols and is integrated with the rule engine module and task engine module through industrial equipment; The rules engine module is used to support the association of internal and external data and low-code visual custom rule mechanism logic; The task engine module is used to execute dispatch tasks, enabling the robot matrix to be dynamically allocated, switched, and adjusted under the support of a globally optimal strategy. The IoT docking module includes: The acquisition module is used to interface with the existing WMS / MES data systems in the warehouse and factory to acquire warehouse data and factory production data; The first determining module is used to fuse the warehouse data with the production data, determine the target feature vector, and perform gridding processing on the target feature vector; The second determining module is used for: The gridded target feature vector is divided into several sub-target feature vectors based on the quadtree algorithm; Determine the data type and visualization requirements corresponding to each sub-target feature vector, and assign each sub-target feature vector and visualization requirements to the corresponding visualization processing node; Based on the visualization processing node, several visualization labels are extracted according to the sub-target feature vector and visualization processing requirements. Each visualization label corresponds to a scene feature. The scene feature is matched with the preset scene features in the preset scene feature database to determine the preset scene feature corresponding to the maximum matching degree, which is used as the target preset scene feature. Based on several visual labels, several target preset scene features are obtained. Target preset scene features with a matching degree less than a preset threshold are removed to determine the target set. The first fusion module is used for: Determine the first fusion relationship of the preset scene features of each target in the target set, and construct sub-scenes; Determine the second fusion relationship of each sub-scene, and determine the fusion visualization scene; The dynamic adjustment module is used to proactively respond to production demands based on the integrated visualization scenario and generate allocation tasks according to the dynamic adjustment.

2. The robot matrix production and allocation system as described in claim 1, characterized in that, The mobile collaborative robots include differential single-arm mobile collaborative robots, differential dual-arm mobile collaborative robots, omnidirectional dual-arm mobile collaborative robots, and omnidirectional single-arm mobile collaborative robots.

3. The robot matrix production and allocation system as described in claim 1, characterized in that, Also includes: The group collaboration module is used to support multi-IoT connection and dynamically match the devices in the system to achieve precise control, maximize utilization and dynamic matching of multiple scenarios, multiple devices and multiple needs.

4. The robot matrix production and allocation system as described in claim 1, characterized in that, Also includes: The traffic control module supports the flexible configuration and automatic guidance of robot matrices on globally drivable roads.

5. The robot matrix production and allocation system as described in claim 1, characterized in that, Also includes: The monitoring and alarm module is used to monitor the actions of the robot matrix and issue alarm prompts when abnormal situations are identified based on the monitoring results.

6. The robot matrix production and allocation system as described in claim 1, characterized in that, The first determining module includes: The extraction module is used to input warehouse data and production data into the feature word extraction model to determine the first feature word and the second feature word; The second fusion module is used for: The warehouse data and the first feature word are input into the word frequency statistics model to determine the first attribute information of the warehouse data; the word frequency statistics model is used to determine the first attribute information of the input warehouse data based on the word frequency statistics of each item in the input warehouse data and the word frequency statistics of the input first feature word. The production data and the second feature word are input into the data time series model to determine the second attribute information of the generated data; the data time series model is used to determine the second attribute information of the input production data based on the vector representation of each item in the input production data, the position information and time series information of each item, and the vector representation of the input second feature word. Based on the first attribute information and the second attribute information, extract the first data to be merged from the warehouse data and the second data to be merged from the production data. Perform data fusion based on the first data to be merged and the second data to be merged to determine the target feature vector, and perform gridding processing on the target feature vector.

7. The robot matrix production and allocation system as described in claim 5, characterized in that, The monitoring and alarm module includes: A module is established to determine the monitoring video images of the monitoring area where the robot matrix is ​​located, and to build a background model of the monitoring area scene based on the monitoring video images; The third determining module is used for: Based on SILTP texture features, the moving object foreground of the monitoring image is extracted from the background model, and connected component analysis and target segmentation are performed on the moving object foreground of the monitoring image to determine the segmentation result. Feature classification is performed based on the feature classifier to extract the contour of the moving object. The contour of the moving object is compared with the preset contour, and the various types of mobile collaborative robots in the robot matrix are determined based on the comparison results. The fourth determining module is used for: Edit and combine the surveillance video images to determine the video frames corresponding to each mobile collaborative robot; The fifth determining module is used to determine the motion information of the corresponding mobile collaborative robot based on video frames; The alarm module is used to compare the motion information of each mobile collaborative robot with the preset motion information, determine whether the motion of the mobile collaborative robot is abnormal based on the comparison result, mark the abnormal mobile collaborative robot, and issue an alarm prompt.

8. The robot matrix production and allocation system as described in claim 7, wherein the fifth determining module comprises: Determine the coordinates of the motion nodes of the collaborative robot in each frame of the video frame; Connect all motion nodes of the mobile collaborative robot within each frame; Each motion node is connected to the same motion node in a continuous coordinate system; Based on the time-series information of video frames, a spatiotemporal sequence diagram of motion nodes of a mobile collaborative robot is constructed, and the corresponding motion information of the mobile collaborative robot is determined according to the spatiotemporal sequence diagram.

9. A method for the production and allocation of a robot matrix, characterized in that, include: Obtain device information for mobile collaborative robots and construct a robot matrix based on the device information; Based on the IoT docking module, it supports docking with the existing WMS / MES data systems in warehouses and factories, proactively responds to dynamic adjustments in production demands, and generates allocation tasks based on the dynamic adjustments; the dynamic adjustments include location, specifications, cycle time, and intent; The signal transceiver module supports adaptation to multiple network protocols and is integrated with industrial equipment, rule engine module, and task engine module. The rule engine module supports the association of internal and external data and low-code, visually customizable rule mechanism logic. Based on the task engine module, the robot matrix is ​​dynamically allocated, switched and adjusted under the support of the global optimal strategy. The IoT docking module includes: The acquisition module is used to interface with the existing WMS / MES data systems in the warehouse and factory to acquire warehouse data and factory production data; The first determining module is used to fuse the warehouse data with the production data, determine the target feature vector, and perform gridding processing on the target feature vector; The second determining module is used for: The gridded target feature vector is divided into several sub-target feature vectors based on the quadtree algorithm; Determine the data type and visualization requirements corresponding to each sub-target feature vector, and assign each sub-target feature vector and visualization requirements to the corresponding visualization processing node; Based on the visualization processing node, several visualization labels are extracted according to the sub-target feature vector and visualization processing requirements. Each visualization label corresponds to a scene feature. The scene feature is matched with the preset scene features in the preset scene feature database to determine the preset scene feature corresponding to the maximum matching degree, which is used as the target preset scene feature. Based on several visual labels, several target preset scene features are obtained. Target preset scene features with a matching degree less than a preset threshold are removed to determine the target set. The first fusion module is used for: Determine the first fusion relationship of the preset scene features of each target in the target set, and construct sub-scenes; Determine the second fusion relationship of each sub-scene, and determine the fusion visualization scene; The dynamic adjustment module is used to proactively respond to production demands based on the integrated visualization scenario and generate allocation tasks according to the dynamic adjustment.

Images