Multi-modal ai micro server based on robot task group and control method thereof

Abstract

The application discloses a multi-modal AI micro server based on a robot working group and a control method thereof, relates to the technical field of robot working, and comprises the following: a data transmission interface unit connected with an external device, which acquires external data of the external device, controls the external device based on control data, and the external device comprises a working device based on the robot working group; a gateway unit connected with the data transmission interface unit and an artificial intelligence calculation and storage unit; the artificial intelligence calculation and storage unit receives the external data, obtains control data output by real-time calibration based on a preset control algorithm for the external data, and the external data comprises demand management data, resource management data and distribution management data; and a programmable logic control unit connected with the artificial intelligence calculation and storage unit and the data transmission interface unit, which receives the control data, sends the control data to the external device after programming. The application solves the problem of low transportation efficiency of dynamic path planning of an AGV logistics distribution system.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of robot operation, and in particular to a multi-modal AI micro server based on a robot operation group and a control method thereof. BACKGROUND

[0002] The logistics dynamic management system of the Internet of Things manufacturing factory is an intelligent management system for the operation process of the factory combined with the Internet of Things.

[0003] However, the existing logistics dynamic management system of the Internet of Things manufacturing factory only plans offline paths, and lacks a global coordination mechanism, which is prone to local congestion or resource waste, such as deadlock and collision when multiple AGVs (Automated Guided Vehicle) work together. SUMMARY

[0004] The present application provides a multi-modal AI micro server based on a robot operation group and a control method thereof to at least solve the problem of low transportation efficiency of the AGV logistics distribution system dynamic path planning in the related art.

[0005] The present application provides a multi-modal AI micro server based on a robot operation group, comprising: a programmable logic control unit, a data transmission interface unit, an artificial intelligence computing and storage unit, and a gateway unit, wherein,

[0006] The data transmission interface unit is connected with an external device, and is configured to obtain external data of the external device and control the external device based on control data, wherein the external device comprises an operation device based on a robot operation group.

[0007] The gateway unit is connected with the data transmission interface unit and the artificial intelligence computing and storage unit, respectively, and is configured to perform protocol conversion between data of the connected external device and the data transmission interface unit and the artificial intelligence computing and storage unit, so as to transmit the data.

[0008] The artificial intelligence computing and storage unit receives external data, performs real-time calibration on the external data based on a preset control algorithm to obtain control data, and outputs the control data, wherein the external data comprises demand management data, resource management data, and distribution management data.

[0009] The programmable logic control unit is connected with the artificial intelligence computing and storage unit and the data transmission interface unit, respectively, and is configured to receive the control data, program the control data, and send the programmed control data to the external device.

[0010] In some optional embodiments, the external device comprises:

[0011] An intelligent building system connected to the data transmission interface unit, for sending building data to the data transmission interface unit, and receiving building control data and operating according to the building control data;

[0012] A robot system connected to the data transmission interface unit, for sending robot data to the data transmission interface unit, and receiving robot control data and operating according to the robot control data;

[0013] A factory operation system connected to the data transmission interface unit, for sending factory operation data to the data transmission interface unit, and receiving factory operation control data and operating according to the factory operation control data.

[0014] The application also provides a control method of a multi-modal AI micro server based on a robot work group, which is applied to the multi-modal AI micro server based on a robot work group as above, and includes:

[0015] Pre-calibration of work requirements in each work step according to the work order of the workstations, optimal planning calculation of the pre-calibrated data combined with real-time dynamic data, and confirmation of the optimal planning result of the current work step until the calibration of all work steps is completed;

[0016] Confirmation of the optimal planning result of all work steps, and control according to the optimal planning result to deliver the required items of the workstations from the shelves to the workstations.

[0017] In some optional embodiments, the step of pre-calibration of work requirements in each work step according to the work order of the workstations includes:

[0018] Confirmation of the workstation required item configuration in the work;

[0019] Pre-calibration of the shelf work instruction based on the confirmed target workstation required item;

[0020] Pre-calibration of the robot configuration based on the confirmed target shelf;

[0021] Pre-calibration of the elevator and path based on the confirmed target robot configuration;

[0022] Pre-calibration of the work timing dispatching based on the confirmed target elevator and target path.

[0023] In some optional embodiments, the step of optimal planning calculation of the pre-calibrated data combined with real-time dynamic data includes:

[0024] Acquisition of the workstation required item configuration, and confirmation of the target workstation required item combined with real-time item dynamics;

[0025] acquire pre-designated shelf operation guide, combine real-time shelf dynamic to confirm target shelf;

[0026] acquire pre-designated robot configuration, combine real-time robot dynamic to confirm target robot configuration;

[0027] acquire pre-designated elevator and path, combine real-time elevator position and path to confirm target elevator and target path;

[0028] acquire pre-designated operation timing dispatch, combine operation timing dynamic to confirm operation timing.

[0029] In some optional embodiments, the combining real-time dynamic data optimally plans the pre-designated data until completing the step of pre-designating all operation steps, including:

[0030] when the pre-designated data and real-time dynamic data do not match, re-designating the operation demand of the current step until confirming the optimal planning result of the current operation step.

[0031] In some optional embodiments, the step of acquiring pre-designated elevator and path, combining real-time elevator position and path to confirm target elevator and target path includes:

[0032] acquiring real-time elevator position and operation parameters, detecting whether the operation capacity of the elevator matches the operation item demand; when detecting mismatch, confirming the target elevator after disassembling the operation item demand;

[0033] acquiring pre-designated path, detecting whether the conflict probability of the pre-designated path is lower than a preset conflict value;

[0034] when detecting higher than or equal to the preset conflict value, re-planning the path.

[0035] In some optional embodiments, the step of controlling according to the optimal planning result to distribute the demand item of the station from the shelf to the station includes:

[0036] controlling the factory operation system to perform light-on operation of the target shelf;

[0037] controlling the robot system to make the target robot move to the target shelf to acquire the target station demand item and then move to the target elevator;

[0038] controlling the intelligent building system to make the target elevator operate, and controlling the target robot to distribute the target station demand item to the target station according to the target path.

[0039] In some optional embodiments, the control method further includes:

[0040] Obtaining transportation cost, inventory cost, and energy consumption cost, adding the transportation cost, the inventory cost, and the energy consumption cost to obtain a total cost;

[0041] Obtaining a delivery punctuality rate and a delivery utilization rate, adding the delivery punctuality rate and the delivery utilization rate to obtain a total efficiency;

[0042] Subtracting the total cost and the total efficiency to obtain an operation model, and confirming a minimum value of the operation model as an optimal planning calculation.

[0043] The application also provides a computer device, comprising:

[0044] A memory and a processor, which are in communication connection with each other, the memory stores computer instructions, and the processor executes the computer instructions to perform the control method of the multi-modal AI micro server based on the robot operation group. BRIEF DESCRIPTION OF DRAWINGS

[0045] In order to more clearly illustrate the embodiments of the present application, the drawings needed in the embodiments will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0046] Figure 1 A structural diagram of a multi-modal AI micro server based on a robot operation group is provided for the embodiments of the present application;

[0047] Figure 2 An application schematic diagram of a multi-modal AI micro server based on a robot operation group is provided for the embodiments of the present application;

[0048] Figure 3 A flowchart of a control method of a multi-modal AI micro server based on a robot operation group is provided for the embodiments of the present application;

[0049] Figure 4 A flowchart of another control method of a multi-modal AI micro server based on a robot operation group is provided for the embodiments of the present application;

[0050] Figure 5 A flowchart of still another control method of a multi-modal AI micro server based on a robot operation group is provided for the embodiments of the present application;

[0051] Figure 6 A flowchart of yet another control method of a multi-modal AI micro server based on a robot operation group is provided for the embodiments of the present application;

[0052] Figure 7It is a structural schematic diagram of a computer device. DETAILED DESCRIPTION

[0053] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the protection scope of the present application.

[0054] It should be noted that, in the description of the present application, the terms "comprise", "contain" or any other variants thereof are intended to cover non-exclusive inclusion, so that the process, method, article or device comprising a series of elements not only includes those elements, but also includes other elements not explicitly listed or inherent to such process, method, article or device. The terms "first", "second" and the like in the present application are used to distinguish similar objects, and are not used to describe a specific order or sequence.

[0055] In order to enable those skilled in the art to better understand the present application, the present application will be further described in detail below with reference to the drawings and specific embodiments.

[0056] The logistics dynamic management system of the Internet of Things manufacturing factory is for the operation process of the factory, and intelligently manages in combination with the Internet of Things.

[0057] However, the existing logistics dynamic management system of the Internet of Things manufacturing factory only plans offline paths, and lacks a global coordination mechanism, which is prone to local congestion or resource waste, such as deadlock and collision of multiple AGVs (Automated Guided Vehicle) in collaborative operation.

[0058] In addition, the related art also has the following technical defects: first, the limitations of static distribution dynamic line planning and fixed AGV vehicles: relying on offline path planning, unable to respond quickly to dynamic orders, equipment failures and other real-time situations; second, multi-objective conflict: difficult to balance optimization objectives such as task completion time, AGV energy consumption, path conflict; third, low efficiency of multi-AGV cooperation: lack of global coordination mechanism, prone to local congestion or resource waste; fourth, poor model generalization: traditional algorithm parameters are fixed, difficult to adapt to different workshop layouts and process flows; fifth, production operation and industrial infrastructure (factory transportation elevator, fire elevator) have no synergy benefits and other problems.

[0059] Embodiments of the present application provide a multi-modal AI micro server based on a robot operation group, such as Figure 1 and Figure 2As shown, the multi-modal AI micro-server based on the robot working group comprises a programmable logic control unit, a data transmission interface unit, an artificial intelligence calculation storage unit, and a gateway unit, wherein,

[0060] The data transmission interface unit is connected with an external device, and is used to acquire external data of the external device and control the external device based on control data. The external device comprises a working device based on the robot working group.

[0061] The gateway unit is connected with the data transmission interface unit and the artificial intelligence calculation storage unit, respectively, and is used to protocol convert between data of the connected external device and the data transmission interface unit and the artificial intelligence calculation storage unit, so as to transmit data.

[0062] The artificial intelligence calculation storage unit receives external data, and outputs control data obtained by real-time calibration of the external data based on a preset control algorithm. The external data comprises demand management data, resource management data, and distribution management data.

[0063] The programmable logic control unit is connected with the artificial intelligence calculation storage unit and the data transmission interface unit, respectively, and is used to receive control data, program the control data, and send the programmed control data to the external device.

[0064] Specifically, AI (Artificial Intelligence) refers to artificial intelligence. The programmable logic control unit is the core of the control system, and is connected to the external device through the data transmission interface unit. The gateway unit communicates with the server and the external device. The artificial intelligence calculation storage unit communicates with the external device through the data transmission interface unit, and controls the external device based on control data obtained by calculating control data based on a preset control algorithm. Thus, the demand management data, the resource management data, and the distribution management data are calibrated in real time, the external device is controlled in cooperation, the paths of the required items and the distribution items in the working process are planned in real time, and the required items of the working position are distributed from the shelves to the working position. Thus, the robot is prevented from being deadlocked and collided during cooperative work, and the resource conflict problem of cooperative work is greatly solved. Moreover, through real-time planning, the robot is prevented from moving in front of the shelf where the required item is located, and the item is taken away by other working positions, thereby causing energy waste and reducing efficiency of the robot.

[0065] Optionally, the chassis is installed as a whole unit on the base, and the programmable logic control unit is the core of the control system, connected to various sensors and actuators (i.e. external devices) of the field devices through the data transmission interface unit, and at the same time, through the gateway unit, communicates with the server and AGV (Automated Guided Vehicle). The artificial intelligence computing storage unit is connected with the operation database, SCADA system (Supervisory Control And Data Acquisition, i.e. data acquisition and monitoring control system), PLC (programmable logic control unit), MES (manufacturing execution system), and building intelligent system (building environment, energy, and elevator management) through the data transmission interface unit (such as switches and concentrators) for data acquisition, analysis, and production management. The power supply system unit introduces external AC power, which is rectified and stabilized to supply power to other units through the bus in the server.

[0066] In addition, with reference to Figure 1 The Internet of Things line related to the external device is introduced into the server all-in-one machine through the bus slot. Through a unified specification connector, such as PCIe (peripheral component interconnect express, a high-speed serial computer expansion bus standard), USB-C (a USB interface appearance standard), backplane bus, etc., the line integration mode of electrical and communication connection is realized, and the technology follows the requirements of heat dissipation, EMC (electromagnetic compatibility), etc.

[0067] Through the data interface unit, it is connected to various sensors and actuators of field devices. For different interfaces, protocol conversion can be used, and a master-slave architecture is adopted, i.e. a programmable logic control unit as a master station, responsible for sending and receiving data frames, while field devices (such as sensors, actuators, and drivers) as slave stations, which operate according to the instructions of the master station.

[0068] The multi-modal AI micro-server based on the robot job group supports a hardware device structure that can be quickly installed, disassembled, and functionally expanded, and is characterized by: a factory device registration verification mechanism, an artificial intelligence automatic comparison of the fingerprint library of the connected device, and identification of the compliance of the device.

[0069] In some feasible real-time ways, as shown in Figure 1 and Figure 2 The external device includes:

[0070] The intelligent building system is connected with the data transmission interface unit, used for sending building data to the data transmission interface unit, and receiving building control data and operating according to the building control data;

[0071] a robot system connected with the data transmission interface unit for sending robot data to the data transmission interface unit and receiving robot control data and operating according to the robot control data;

[0072] a factory operation system connected with the data transmission interface unit for sending factory operation data to the data transmission interface unit and receiving factory operation control data and operating according to the factory operation control data.

[0073] Reference Figure 1 、 Figure 2 and Figure 6 , the intelligent building system is mainly used for elevator position management, path management and timing management, the robot system is mainly an AGV (Automated Guided Vehicle) system, and the factory operation system is mainly a station requirement and shelf management. At the same time, real-time dynamic data of the intelligent building system, the robot system and the factory operation system are acquired, and job scene data is generated in real time based on a preset control method through artificial intelligence to plan the required articles and the path for delivering the articles in real time, and the robot and the elevator are controlled to work in linkage to deliver the required articles of the station from the shelf to the station.

[0074] Embodiments of the present application provide a control method of a multi-modal AI micro server based on a robot job group, which is applied to the multi-modal AI micro server based on the robot job group as described above, and the control method comprises:

[0075] Step one: pre-designating the job requirement in each job step according to the working order of the station, performing optimal planning calculation on the pre-designated data in combination with real-time dynamic data, and confirming the optimal planning result of the current job step until the designation of all job steps is completed;

[0076] Specifically, the required articles are acquired according to the working order of the station, and the job step for delivering the articles is confirmed, then the job requirement in each job step is pre-designated, and then optimal planning calculation is performed on the pre-designated data in real time, and the optimal planning result of the current job step is confirmed.

[0077] Specifically, the job steps are divided into the first step to the nth step, the first step is pre-designated, and optimal planning calculation is performed on the pre-designated data of the first step according to real-time dynamic data, and the optimal planning result of the first step is confirmed, then the pre-designation of the job requirement of the second step is performed according to the optimal planning result of the first step, and optimal planning calculation is performed on the pre-designated data of the second step according to real-time dynamic data, and the optimal planning result of the second step is confirmed, and so on until the designation of the nth step is completed.

[0078] It should be noted that if the optimal planning result of the current operation step is confirmed for the pre-calibration data, it is confirmed that the calibration of the current step is completed.

[0079] Specifically, by performing pre-calibration first, and then performing optimal planning calculation on the pre-calibration data combined with real-time dynamic data, i.e., parallel calculation for the calibration of each operation step.

[0080] Step two: confirm the optimal planning result of all operation steps, and control according to the optimal planning result to deliver the required items of the workstations from the shelves to the workstations.

[0081] Specifically, after confirming the optimal planning result of all operation steps, the optimal planning result of all operation steps is taken as the calibration result, and the robot and the elevator are controlled to deliver the required items of the workstations from the shelves to the workstations.

[0082] Reference Figure 3 , steps S3 to S6 are the specific process of controlling the optimal planning result, including:

[0083] Step S3: artificial intelligence simulation calculation confirms the task or operation calibration; step S4: network collaborative operation control; step S5: shelf light guiding; step S6: robot operation driving.

[0084] Specifically, after confirming the optimal planning result of all operation steps, each operation process is calculated in a serial calculation mode, and the robot and the elevator are controlled to deliver the required items of the workstations from the shelves to the workstations. Thus, through the combination of parallel calculation and serial calculation, compared with the traditional serial calculation mode, the efficiency is greatly improved.

[0085] The present application uses AI large model technology to propose a control model and calculation formula for electronic manufacturing and similar discrete manufacturing production scenes. Real-time perception and analysis of environmental infrastructure (elevator working condition, workstation demand) are realized to achieve dynamic path planning of AGV logistics distribution system to improve transportation efficiency. Micro servers cooperate in decision-making and task scheduling to realize real-time communication and data sharing between warehouses and logistics, improve human-machine cooperation efficiency, and reduce operation difficulty and labor intensity. AI micro servers can be flexibly configured and can be flexibly installed in the original control system cabinet of the enterprise to realize upgrading and modification, or the standard configuration control cabinet can be quickly changed according to the needs of the enterprise to provide a whole solution.

[0086] In some feasible embodiments, the step of "pre-calibrating the operation demand in each operation step according to the work order of the workstations" in step one includes:

[0087] Step S11: confirming the station required item configuration in the task;

[0088] Step S12: pre-calibrating the shelf task guidance based on the confirmed target station required item;

[0089] Step S13: pre-calibrating the robot configuration based on the confirmed target shelf;

[0090] Step S14: pre-calibrating the elevator and path based on the confirmed target robot configuration;

[0091] Step S15: pre-calibrating the task timing assignment based on the confirmed target elevator and target path.

[0092] Specifically, referring to steps S11 to S15 in Figure 3 , the task steps are specifically to first confirm the items required by the station, then place the shelves of the items to be obtained where the confirmed items are located, then confirm the robot to be executed in the multiple robots, then plan the elevator and path of the robot, and finally confirm whether there is a conflict task timing.

[0093] In some possible embodiments, the step of "optimally planning and calculating the pre-calibrated data in combination with real-time dynamic data" in step one includes:

[0094] Step S21: obtaining the station required item configuration, and confirming the target station required item in combination with real-time item dynamics;

[0095] Step S22: obtaining the pre-calibrated shelf task guidance, and confirming the target shelf in combination with real-time shelf dynamics;

[0096] Step S23: obtaining the pre-calibrated robot configuration, and confirming the target robot configuration in combination with real-time robot dynamics;

[0097] Step S24: obtaining the pre-calibrated elevator and path, and confirming the target elevator and target path in combination with real-time elevator position and path;

[0098] Step S25: obtaining the pre-calibrated task timing assignment, and confirming the task timing in combination with task timing dynamics.

[0099] Specifically, referring to steps S21 to S25 in Figure 3 , that is, steps S1 and S2 are simultaneously and in parallel, constituting a Y-shaped structure, step S1 pre-calibrates each task, and step S2 calculates relevant data of the task requirements by using an artificial intelligence algorithm. Thus, steps S1 and S2 are parallel tasks.

[0100] For example, if the current required item is a wrench as pre-designated in step S11, step S21 is to obtain the real-time wrench condition and confirm whether there is a wrench in the spare parts; if the shelf for obtaining the wrench is the 5th shelf as pre-designated in step S12, step S22 is to confirm whether there is a wrench in the shelf based on the real-time dynamic of the shelf, and if there is, continue to execute; if the shelf for obtaining the wrench is the 5th shelf as determined in step S13, pre-designate the required robot, and step S23 is to confirm the target robot based on the real-time dynamic of the robot, such as the position and power of the robot; if the target robot is determined in step S14, pre-designate the elevator and path, and step S24 is to confirm the target elevator and target path in combination with the real-time elevator position and path, specifically, the current idle elevator and the elevator close to the target robot, and the path with the shortest delivery distance; if the work sequence dispatching pre-designation is made in step S15, step S25 is to dynamically confirm the work sequence, for example, dynamic anti-collision calculation.

[0101] Reference Figure 4 The work step further comprises: controlling the robot to deliver the work station completed semi-finished product or the required temporary storage item to the shelf, in the process, confirming the target item in the work station, the target item comprising the semi-finished product or the temporary storage item; based on the confirmed target item, pre-designating the shelf work instruction; based on the confirmed target shelf, pre-designating the robot configuration; based on the confirmed target robot configuration, pre-designating the elevator and path; based on the confirmed target elevator and target path, pre-designating the work sequence dispatching.

[0102] In some possible embodiments, the step in step one, "the combination of real-time dynamic data for pre-designated data optimal planning calculation, until the completion of all work step designation", comprises:

[0103] Step (1): when the pre-designated data and real-time dynamic data do not match, re-designate the work requirement of the current step until the optimal planning result of the current work step is confirmed.

[0104] Specifically, in the real-time dynamic calculation process, if the pre-designated data has been unable to be confirmed or is about to be used in other work station operations, re-designate until the optimal planning result of the current work step is confirmed.

[0105] For example, if the current pre-designated robot power is insufficient to support the delivery energy consumption or the robot is about to be used in other work station operations, re-designate another robot. If the current pre-designated elevator is designated in real time to be used in other work station operations, re-designate another elevator until the optimal planning result of the elevator is confirmed.

[0106] In some possible embodiments, step S24 comprises:

[0107] Step (1): Obtain real-time elevator position and running parameters, and detect whether the capacity of the elevator and the demand for running articles match;

[0108] Step (2): When a mismatch is detected, disassemble the running article demand to confirm the target elevator;

[0109] Step (3): Obtain the pre-designated path, and detect whether the conflict probability of the pre-designated path is lower than the preset conflict value;

[0110] Step (4): When it is detected that the conflict probability is higher than or equal to the preset conflict value, re-plan the path.

[0111] Specifically, referring to Figure 4 , the server obtains the robot state, shelf inventory scanning, elevator running data, and elevator running parameters through physical layer input data, and then, after data preprocessing, generates a navigation path when the conflict probability is detected to be lower than the preset conflict threshold (15%), and dynamically re-plans the path when the conflict probability is detected to be higher than the preset conflict threshold (15%). By matching the transportation demand with the capacity of the elevator (goods elevator), when the elevator is overloaded, the task is disassembled by the task disassembly strategy, and when it is not overloaded, the transportation instruction is generated.

[0112] In some possible embodiments, the step of "controlling according to the optimal planning result to distribute the demand article of the station from the shelf to the station" in step two includes:

[0113] Step (1): Control the factory operation system to perform the light-on operation of the target shelf;

[0114] Step (2): Control the robot system to move the target robot to the target shelf to obtain the target station demand article and then move to the target elevator;

[0115] Step (3): Control the intelligent building system to make the target elevator run, and control the target robot to distribute the target station demand article to the target station according to the target path.

[0116] Specifically, by means of serial operation, the work of each step is performed. First, the factory operation system performs the light-on operation of the target shelf, then the robot system moves the target robot to the target shelf to obtain the target station demand article and then moves to the target elevator, and the intelligent building system is controlled to make the target elevator run, and the target robot is controlled to distribute the target station demand article to the target station according to the target path, so as to complete the task of all steps.

[0117] It should be noted that in the control process, if there is an emergency, the emergency is uploaded to the emergency center for separate processing. If it can be controlled in normal timing, the normal timing control condition is uploaded to the background.

[0118] In some possible embodiments, the control method of the multi-modal AI micro server based on the robot working group further comprises:

[0119] Step one: obtain transportation cost, inventory cost, and energy consumption cost, and add the transportation cost, inventory cost, and energy consumption cost to obtain the total cost;

[0120] Specifically, the formula of the total cost is as follows: w1·(∑i transportation cost i+∑j inventory cost j+∑k energy consumption cost k).

[0121] Step two: obtain the delivery punctuality rate and the delivery utilization rate, and add the delivery punctuality rate and the delivery utilization rate to obtain the total efficiency;

[0122] Specifically, the formula of the total efficiency is as follows: w2·(∑m punctuality rate m+∑n utilization rate n).

[0123] Step three: subtract the total cost and the total efficiency to obtain the working model, and confirm that the minimum value of the working model is the optimal planning calculation.

[0124] Specifically, the optimal planning calculation model is minZ=w1·(∑i transportation cost i+∑j inventory cost j+∑k energy consumption cost k)-w2·(∑m punctuality rate m+∑n utilization rate n). Thus, the planning calculation with the highest efficiency and utilization rate at the minimum cost is obtained. Wherein, transportation includes the distance of the path, energy consumption includes the work of elevator up and down, and inventory includes item management. The punctuality rate is the punctuality rate of the delivery demand item, and the rate utilization rate is the utilization rate of the robot and the elevator. It should be noted that the linear calculation part is the fixed cost in the transportation cost, including the depreciation of the robot, and the non-linear calculation part includes the cubic relationship between the energy consumption cost and the speed, because there is resistance in the running process. At the same time, the integer constraint is that the shelf adjustment times need to be an integer, specifically, the maximum adjustment preset times (5 times) per month.

[0125] Exemplarily, the calculation method of transportation timeliness is the time standard deviation of order completion, the calculation method of equipment utilization rate is AGV effective driving time / total time, the calculation method of energy efficiency is (total transportation quantity×distance) / total power consumption, the calculation method of elevator cooperation degree is elevator full load rate×time window matching degree, and the calculation method of abnormal recovery time is the average time consumption from conflict detection to path re-planning. The punctuality rate is the transportation timeliness. The utilization rate includes the equipment utilization rate.

[0126] Specifically, the solving process is as follows:

[0127] Data preprocessing and hierarchical modeling, specifically input data, workstation demand matrix, AGV initial position, elevator status, ring path topology; then hierarchical modeling, upper layer (task allocation): use integer programming to allocate tasks to AGV, determine the shelf adjustment scheme, middle layer (path planning), generate cross-path based on improved artificial intelligence algorithm, integrate elevator reservation mechanism, lower layer (speed optimization), dynamically adjust speed to balance energy consumption and on-time rate.

[0128] Improved genetic algorithm solution, specifically coding design, chromosome contains task allocation sequence (binary), shelf adjustment frequency (integer), speed segmentation value (real number); fitness function, reciprocal of objective function Z, add penalty term to handle constraint violation (such as elevator overload penalty); crossover and mutation, task allocation part uses two-point crossover, shelf adjustment part uses uniform mutation, speed gene uses simulated annealing strategy for disturbance, avoiding local optimum.

[0129] Dynamic scheduling and re-planning, specifically real-time monitoring, AGV position and elevator queue length are updated every 5 seconds, workstation demand changes. Trigger mechanism, when path conflict or elevator waiting time is greater than 30 seconds, start local re-planning (based on rolling horizon optimization).

[0130] Table 1 below is the specific data of the optimal planning result of calibration:

[0131]

[0132] Table 1

[0133] Specifically, according to the target function combined with artificial intelligence data collection function, such as data collection and classification, the work scene operation process is realized.

[0134] Table 2 below is the economic benefit comparison table with the traditional scheme:

[0135]

[0136] Table 2

[0137] Through path planning, global optimization and dynamic adjustment, further performance standard quantification system. Performance standard quantification efficiency indicators, single machine performance: such as task completion time (standard value ≤ 5 minutes / single), empty running rate (target ≤ 15%); Cluster performance: unit time throughput (such as a certain storage piece / hour), task balance degree (variance ≤ 0.2); Data synchronization consistency (5G network packet loss rate ≤ 0.01%). AI based on performance standard quantification efficiency index AGV operation route guidance, namely S6 link. S6 link, AI algorithm server realizes the standardized control of AGV operation through the following process: initial path: assign energy-efficient benchmark route to each AGV, AI server real-time correction: through edge computing node monitoring AGV position to calculate multi-machine cooperative operation system.

[0138] The control method of the multi-modal AI micro server based on the robot operation group further comprises:

[0139] Step one: detect whether the power of the robot is lower than the preset power value;

[0140] Step two: if yes, control the robot to move to the charging position for charging.

[0141] Reference Figure 5 by real-time detection of the power of the robot, and when the power is detected to be lower than 20%, the robot is controlled to move to the charging position for charging.

[0142] Specifically, reference Figure 5 , Figure 5 is a control logic flowchart of the robot. First, it is in standby state, then if it receives a task instruction, it detects the validity of the task, and when it detects that the path is conflict, it re-plans the path and moves to the target point, if it has not reached the target point, it real-time positioning corrects to the target point, then it performs loading and unloading operation, if it is in interrupt state, it detects whether the power of the robot is lower than the preset value (20%), if yes, it performs charging navigation, controls the robot to move to the charging position, and automatically charges to charge, after detecting that the charging is completed, it controls the robot to move to the standby position for standby.

[0143] In some possible embodiments, the control method of the multi-modal AI micro server based on the robot operation group comprises:

[0144] Step one: scan the inventory of goods on the shelf, and detect whether the inventory of goods is lower than the inventory safety threshold;

[0145] Step two: when it is detected that the inventory is lower than the inventory safety threshold, calculate the replenishment priority, and control the replenishment according to the replenishment priority.

[0146] Specifically, through the inventory security value detection, when the inventory is detected to be insufficient, the replenishment priority calculation is performed to trigger the replenishment instruction, and when the inventory is sufficient, the current inventory is maintained. Then, after the control instruction fusion, it is sent to the robot, the elevator controller and the goods adjusting device for control, and the current state is fed back in real time.

[0147] It is worth noting that the replenishment instruction can be sent to the robot that replenishes goods, or to the user to make the user replenish.

[0148] In addition, the shelf layout optimization can also be used to adjust the high-frequency access shelves to the elevator nearby every month, such as adjusting the top 5 high-frequency access shelves to the elevator nearby, so that the cross-layer transportation distance is reduced by more than 20%. Through the task allocation strategy, the AGV with more than a preset power is preferentially allocated to the emergency order, such as the order with less than 10 minutes of remaining time is allocated to the AGV with more than 80% of power. Through the elevator reservation mechanism, the AGV applies for the elevator 30 seconds in advance, and the system allocates a time window, for example, elevator A is idle at 10:00:00-10:00:15. Through dynamic shape and dynamic line optimization, the clockwise and counterclockwise dual-channel design is used, and the direction is dynamically switched in congestion. Specifically, the flow is detected by laser radar, virtual passes are set at key nodes (such as intersections) to avoid deadlock, and the conflict rate is reduced by 90%. Through speed control and energy consumption management, the speed is 1.8 m / s (maximizing the on-time rate) in the straight line segment, and the speed is 0.8 m / s (reducing energy consumption and collision risk) in the turning / congestion area. Compared with the uniform speed 1.5 m / s scheme, the energy consumption is reduced by 27%, and the on-time rate is only reduced by 3%. Through factory implementation verification, the AGV empty running rate is reduced from 35% to less than 10%. Cost optimization: through the transportation demand matching module, the invalid start and stop of the elevator is reduced, and the energy consumption is reduced by less than 20%. Reliability enhancement: the response speed of inventory anomaly problem processing is shortened from 25 minutes of artificial inspection to 2 seconds of system automatic triggering.

[0149] The embodiment of the present application also provides a computer device with the above-mentioned logistics dynamic management system of the Internet of Things manufacturing factory.

[0150] Please refer to Figure 7 , Figure 7 is a structural schematic diagram of a computer device provided by an optional embodiment of the present application, as Figure 7As shown, the computer device includes one or more processors 10, memory 20, and interfaces 30 for external devices such as a keyboard and a mouse and a disk drive. One or more of the interfaces 30 enable a user to interact with the computer device. In some embodiments, the interface 30 also includes an input device, such as a microphone, or output device, such as a speaker. Figure 7 The processor 10 is used in the description as an example.

[0151] The processor 10 can be a central processing unit, a network processor, or a combination thereof. The processor 10 can further include a hardware chip. The hardware chip can be an application specific integrated circuit, a programmable logic device, or a combination thereof. The programmable logic device can be a complex programmable logic device, a field programmable logic device, a general array logic, or any combination thereof.

[0152] The memory 20 stores instructions that can be executed by the at least one processor 10 to cause the at least one processor 10 to perform the methods described in the above embodiments.

[0153] The memory 20 can include a program storage area and a data storage area. The program storage area can store an operating system, application programs, and the like for use by the at least one processor 10. The data storage area can store data created by the computer device, etc. Additionally, the memory 20 can include a volatile memory, such as a random access memory, and a non-volatile memory, such as at least one disk storage device, flash memory, or other non-volatile solid state storage device. In some embodiments, the memory 20 can optionally include a memory that is remote from the processor 10, such as a network storage device connected to the computer device via a network. Examples of the network include, but are not limited to, the Internet, an enterprise intranet, a local area network, a mobile communication network, and a combination thereof.

[0154] The memory 20 can include a volatile memory, such as a random access memory, and a non-volatile memory, such as at least one disk storage device, flash memory, or other non-volatile solid state storage device.

[0155] The computer device also includes input device 30 and output device 40. The processor 10, memory 20, input device 30 and output device 40 can be connected through a bus or other means, Figure 7 The bus connection is taken as an example in the above description.

[0156] The input device 30 can receive inputted digital or character information, and generate key signal input relating to user settings and function controls of the computer device, such as touch screen, keypad, mouse, trackpad, touchpad, pointing stick, one or more mouse buttons, trackball, joystick, etc. The output device 40 can include display device, auxiliary lighting device (e.g. LED), and tactile feedback device (e.g. vibration motor), etc. The display device includes but is not limited to liquid crystal display, light emitting diode, display and plasma display. In some optional embodiments, the display device can be a touch screen.

[0157] Those skilled in the art will further appreciate that the individual steps of the example described in connection with the embodiments disclosed herein can be realized by electronic hardware, computer software, or a combination of both. The disclosure has been described in relation to particular examples, which are intended to be illustrative only and changes can be made to the specific examples described without departing from the scope of the disclosure. The disclosure is not limited to the described examples, and changes can be made to the described examples without departing from the scope of the disclosure. The scope of the disclosure is limited only by the claims.

[0158] The above provides a kind of multi-modal AI micro server based on robot working group and its control method of the principle and implementation of the present application are described in detail in the present application. The principle and implementation of the present application are described in the above examples, which are only used to help understand the method and core idea of the present application. It should be pointed out that, for the ordinary skilled in the art, without departing from the principle of the present application, the present application can be improved and modified in several ways, and these improvements and modifications also fall within the scope of the claims of the present application.

Claims

1. A multimodal AI microserver based on robot swarms, characterized in that, include: Programmable logic control unit, data transmission interface unit, artificial intelligence computing and storage unit, gateway unit, among which, A data transmission interface unit is connected to an external device for acquiring external data from the external device and controlling the external device based on control data. The external device includes a work device based on a robot work group. The gateway unit is connected to the data transmission interface unit and the artificial intelligence computing and storage unit respectively, and is used to convert the protocol between the connected external device and the data transmission interface unit and the artificial intelligence computing and storage unit respectively, so as to transmit data; The artificial intelligence computing and storage unit receives external data, performs real-time calibration on the external data based on a preset control algorithm to obtain control data, and then outputs it to plan the items needed in the operation process and the path for delivering the items, and controls the robot and elevator to work together to deliver the required items from the shelves to the workstation. The external data includes demand management data, resource management data and delivery management data. The programmable logic control unit is connected to the artificial intelligence computing and storage unit and the data transmission interface unit, respectively, and is used to receive control data, program the control data, and send it to the external device. The external device includes: The intelligent building system is connected to a data transmission interface unit and is used to send building data to the data transmission interface unit, as well as to receive building control data and operate according to the building control data. The intelligent building system is used for elevator location management, path management and timing management. The robot system is connected to a data transmission interface unit and is used to send robot data to the data transmission interface unit, as well as to receive robot control data and operate according to the robot control data. The robot system is an AGV system. The factory operation system is connected to the data transmission interface unit and is used to send factory operation data to the data transmission interface unit, and to receive factory operation control data and operate according to the factory operation control data. The factory operation system is a system for workstation demand and shelf management.

2. A control method for a multimodal AI microserver based on a robot swarm, characterized in that, The method is applied to the multimodal AI microserver based on robot task groups as described in claim 1, and the method includes: According to the work sequence of the workstation, the work requirements of each work step are pre-calibrated. The optimal planning calculation is performed on the pre-calibrated data in combination with real-time dynamic data, and the optimal planning result of the current work step is confirmed until all work steps are calibrated. Confirm the optimal planning result for all work steps, and control the process according to the optimal planning result to deliver the required items from the shelf to the workstation.

3. The control method of a multimodal AI microserver based on a robot work group according to claim 2, characterized in that, The step of pre-calibrating the work requirements for each work step according to the work sequence of the workstation includes: Confirm the required item configuration for each workstation in the operation; Based on the confirmed items required for the target workstations, pre-calibrate the shelf operation guidelines; Based on the confirmed target shelf, perform robot configuration pre-calibration; Based on the confirmed target robot configuration, elevator and path pre-calibration is performed; Based on the confirmed target elevator and target path, the operation sequence dispatch is pre-calibrated.

4. The control method of a multimodal AI microserver based on a robot work group according to claim 3, characterized in that, The step of performing optimal planning calculations based on pre-calibrated data using real-time dynamic data includes: Obtain the required item configuration for each workstation, and combine it with real-time item dynamics to confirm the required items for the target workstation. Obtain pre-calibrated shelving operation instructions and confirm the target shelving in conjunction with real-time shelving dynamics; Obtain the pre-calibrated robot configuration and combine it with the real-time robot dynamics to confirm the target robot configuration; Obtain pre-calibrated elevators and paths, and combine them with real-time elevator locations and paths to confirm the target elevator and target path; Obtain the pre-calibrated job sequence assignment and dynamically confirm the job sequence based on the job sequence.

5. The control method for a multimodal AI microserver based on a robot work group according to claim 4, characterized in that, The step of combining real-time dynamic data to perform optimal planning calculations on pre-calibrated data until all work steps are calibrated includes: When the pre-calibrated data and the real-time dynamic data do not match, the data is re-calibrated according to the work requirements of the current step until the optimal planning result of the current work step is confirmed.

6. The control method for a multimodal AI microserver based on a robot work group according to claim 5, characterized in that, The steps of obtaining the pre-calibrated elevators and paths, and confirming the target elevator and target path by combining the real-time elevator location and path, include: Obtain real-time elevator location and operating parameters, and check whether the elevator's carrying capacity matches the demand for the transported goods; if a mismatch is detected, disassemble the transported goods demand and then confirm the target elevator; Obtain the pre-calibrated path and check whether the conflict probability of the pre-calibrated path is lower than the preset conflict value; When a conflict value higher than or equal to a preset value is detected, the path is replanned.

7. The control method of a multimodal AI microserver based on a robot work group according to claim 2, characterized in that, The step of controlling the delivery of required items from the shelf to the workstation according to the optimal planning result includes: Control the factory operating system to execute the lighting operation of the target shelf; The robot control system enables the target robot to move to the target shelf, retrieve the required items from the target workstation, and then move to the target elevator. Control the intelligent building system to make the target elevator run, and control the target robot to deliver the required items to the target workstation according to the target path.

8. The control method of a multimodal AI microserver based on a robot work group according to claim 7, characterized in that, The control method further includes: Obtain transportation costs, inventory costs, and energy costs, and add them together to get the total cost; Obtain the on-time delivery rate and the delivery utilization rate, and add the on-time delivery rate and the delivery utilization rate to obtain the total efficiency; The operation model is obtained by subtracting the total cost from the total efficiency, and the minimum value of the operation model is confirmed to be the optimal planning calculation.

9. A computer device, characterized in that, include: The system includes a memory and a processor, which are interconnected. The memory stores computer instructions, and the processor executes the computer instructions to perform the control method of the multimodal AI microserver based on robot workgroups as described in any one of claims 2 to 8.

Images