Dynamic scheduling management method, system, equipment and medium

Abstract

The invention discloses a dynamic scheduling management method, system and device and a medium, and the method comprises the steps: carrying out the logical reasoning analysis according to a task scheduling instruction through a pre-trained neural network analysis model based on the historical information of a historical task scene similar to a task in the task scheduling instruction, task to-be-scheduled information of the current task scheduling instruction is obtained; wherein the task to-be-scheduled information comprises to-be-scheduled contents and to-be-scheduled positions; and searching a corresponding terminal according to the to-be-scheduled content and the to-be-scheduled position, and scheduling the terminal. The historical task scene similar to the current task state can be obtained according to the current task scheduling instruction according to the neural network analysis model with logical reasoning ability, the terminal to be scheduled is rapidly positioned through logical reasoning analysis, various terminals are dynamically scheduled, various terminal resources are scheduled in time, potential safety hazards are observed in time, and the scheduling efficiency is improved. The scheduling efficiency and flexibility are improved, and rapid linkage and communication can be realized.

Description

Technical Field

[0001] The present invention relates to the technical field of Internet of Things, and more specifically to a dynamic scheduling management method, system, device and medium. Background Art

[0002] Currently, converged communication systems leverage IP network technology to integrate traditionally independent and heterogeneous communication methods and applications (such as voice, video, data, and messaging) onto a unified platform, enabling comprehensive communications based on a unified interface, unified management, and unified scheduling. Converged communication systems often require the scheduling of various voice and video resources to timely detect safety hazards (such as traffic accidents, severe weather, and equipment failures) at critical moments. These resources vary over time and in different scenarios, and the amount of data required increases significantly. Traditional voice scheduling struggles to quickly manage these resources within this vast amount of data and changing scenarios, resulting in delayed communication and low scheduling efficiency. Summary of the Invention

[0003] The present invention provides a dynamic scheduling management method, system, device and medium to solve the technical problems of untimely communication and low scheduling efficiency in traditional voice scheduling.

[0004] In a first aspect, a dynamic scheduling management method is provided, comprising: Using a pre-trained neural network analysis model according to the task scheduling instruction, logical reasoning analysis is performed based on historical information of historical task scenarios similar to the task in the task scheduling instruction to obtain task scheduling information of the current task scheduling instruction; wherein the task scheduling information includes the task scheduling content and the task scheduling location; A corresponding terminal is searched according to the to-be-scheduled content and the to-be-scheduled location, and the terminal is scheduled.

[0005] In a second aspect, a dynamic scheduling management system is provided, comprising a unit for executing the above-mentioned dynamic scheduling management method.

[0006] The present invention also provides a dynamic scheduling management system, comprising a server and multiple terminals, wherein the server is configured with a call intelligence model, and the call intelligence model comprises multiple neural network analysis models for task scheduling management, wherein: The server is configured to use a pre-trained neural network analysis model according to the task scheduling instruction to perform logical reasoning analysis based on historical information of historical task scenarios similar to the task in the task scheduling instruction, thereby obtaining task scheduling information of the current task scheduling instruction; wherein the task scheduling information includes a content to be scheduled and a location to be scheduled; and searching for a corresponding terminal according to the content to be scheduled and the location to be scheduled to schedule the terminal; or The server is used to perform logical reasoning analysis based on the call information using a pre-trained neural network analysis model to obtain a call number identification result and a communication link identification result; if the call number identification result is an alarm or emergency call, the server obtains task scheduling information including the content to be scheduled and the location to be scheduled based on the call number identification result and the communication link identification result; searches for a corresponding terminal based on the content to be scheduled and the location to be scheduled, schedules the terminal, generates a SIP instruction, and transmits the SIP instruction to a receiving object associated with the call number.

[0007] In a third aspect, a computer device is provided, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the above-mentioned dynamic scheduling management method when executing the computer program.

[0008] In a fourth aspect, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the above-mentioned dynamic scheduling management method are implemented.

[0009] Compared with the prior art, the present invention utilizes a pre-trained neural network analysis model according to the task scheduling instruction, performs logical reasoning analysis based on historical information of historical task scenarios similar to the tasks in the task scheduling instruction, and obtains the task scheduling information of the current task scheduling instruction; then searches for the corresponding terminal according to the content to be scheduled and the location to be scheduled in the task scheduling information, and schedules the terminal to obtain various resource data obtained by the scheduling terminal; it can be seen that the present invention can obtain historical task scenarios similar to the current task status according to the current task scheduling instruction based on the neural network analysis model with logical reasoning ability, quickly locate the terminal to be scheduled through logical reasoning analysis, dynamically schedule various terminals, and timely schedule various terminal resources, and timely observe safety hazards, thereby greatly improving the efficiency and flexibility of scheduling, and can quickly link communication and improve decision-making efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] Figure 1 A schematic flow chart of a dynamic scheduling management method provided in the first embodiment of the present invention; Figure 2 for Figure 1 A flow chart of a specific implementation of step S120; Figure 3 for Figure 2 A schematic flow chart of a specific implementation of step S123; Figure 4 A schematic flow chart of a dynamic scheduling management method provided in a second embodiment of the present invention; Figure 5is a structural diagram of a dynamic scheduling management system provided by a first embodiment of the present invention; Figure 6 is a structural diagram of a dynamic scheduling management system provided by a second embodiment of the present invention; Figure 7 It is a structural diagram of a computer device in one embodiment of the present invention. DETAILED DESCRIPTION

[0011] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0012] It will be understood that when used in this specification and the appended claims, the terms “comprises” and “comprising” indicate the presence of described features, integers, steps, operations, elements and / or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or groups thereof.

[0013] It will also be understood that the term "and / or" used in the present description and appended claims refers to and includes any and all possible combinations of one or more of the associated listed items.

[0014] See also Figure 1 , Figure 1 This is a schematic flow chart of a dynamic scheduling management method provided by the first embodiment of the present invention. In the embodiment shown in the accompanying drawings, the dynamic scheduling management method includes the following steps S110-130: S110: Acquire information of each terminal and generate terminal feature data associated with features of each terminal.

[0015] In the present invention, the terminal can be a variety of different types of terminals such as IP / video phones, surveillance balls, drone robots, video law enforcement devices, positioning equipment, etc.

[0016] In this step, information of the connected terminal is obtained. Various terminals can be accessed through a network (such as Wi-Fi, cellular network, etc.). Each connected terminal has a unique identifier. Terminal feature data is generated based on the terminal features of the terminal to further identify the terminal, so as to facilitate rapid search for the terminal to be scheduled in subsequent tasks; wherein, the terminal features include the terminal location and the target features of the generated content. For example, when the terminal is a control ball, the terminal features of the control ball may include the location of the control ball and the shooting target features of the control ball when shooting. The target features may include the target category (such as a car, a person, or certain specific objects, etc.); for example, the terminal feature data of control ball 1 may include the shot target features (car models and license plates of passing vehicles) and location, the terminal feature data of control ball 2 may include the shot target features (faces of people who do not obey traffic rules / appear in dangerous areas) and location, and the terminal feature data of drone 1 may include the shot target features (environmental markers) and location.

[0017] S120 , using a pre-trained neural network analysis model according to the task scheduling instruction, performing logical reasoning analysis based on historical information of historical task scenarios similar to the task in the task scheduling instruction, to obtain task scheduling information of the current task scheduling instruction.

[0018] In the present invention, the neural network analysis model can be pre-trained based on the scheduling data sets of different types of tasks that have occurred, so that it has logical reasoning capabilities. The neural network analysis model can call the historical information of historical task scenarios similar to its task based on the current task scheduling instructions to perform logical reasoning analysis, and obtain the task scheduling information of the current task scheduling instructions to dynamically call the resource data of different terminals.

[0019] In this embodiment, the task to be scheduled information includes the to-be-scheduled content and the to-be-scheduled location, and the task scheduling instruction may be a voice scheduling instruction generated by the user according to the task progress.

[0020] like Figure 2 As shown, the step S120 may specifically include steps S121-S123: S121. Analyze the input task scheduling instruction using a pre-trained neural network analysis model to obtain the current task.

[0021] In this step, the current task may be a resource calling instruction or a monitoring instruction for an emergency (such as a traffic accident, bad weather, equipment failure, or emergency management accident such as a leak of hazardous materials).

[0022] S122 , acquiring historical information of historical task scenarios similar to the current task in real time, and configuring instruction variables related to the current task based on the prompt words.

[0023] In this step, historical information of historical task scenarios similar to the current task is obtained in real time to synchronize the data required for the current task (such as audio, video, text, positioning and other multimodal data) in real time; wherein, the historical information may include historical variable data and historical scheduling information, etc.

[0024] In the present invention, machine learning methods are used to infer and configure instruction variables based on prompt words. Prompt words are associated with task scenarios. This step can call prompt words pre-associated with each task scenario, or obtain prompt words selected / entered by the user. Instruction variables are also task-related factors that can describe the characteristics of the task. For example, if the task scheduling instruction is "water pipe cracking situation" and the user enters prompt words representing the location and previous freezing data, the neural network analysis model can infer and configure instruction variables related to the current task (such as water pipe material, weather temperature conditions, construction age, location), etc. based on the prompt words using the characteristics of the target object, task purpose, and / or task environment. In some task scenarios, instruction variables may also include prompt words. For example, if the task scheduling instruction is "traffic accident handling", the prompt word is location, and the configured instruction variables can include road conditions, pedestrian flow, whether it is located at an intersection or a critical road, etc.

[0025] S123: Call the corresponding historical information according to the instruction variable to perform matching analysis to obtain the task scheduling information of the current task scheduling instruction.

[0026] In this step, the instruction variables are matched and analyzed with the corresponding historical information. The task scheduling information may include the content to be scheduled and the location to be scheduled.

[0027] Specifically, if Figure 3 As shown, step S123 may include steps S1231-S1233: S1231. Calling historical variable data corresponding to the instruction variable in the historical task scenario; S1232. Utilize a weighted linear algorithm to calculate, based on the historical variable data corresponding to the instruction variable in the historical task scenario, a matching value between the historical task scenario and the current task.

[0028] In this step, the matching value is calculated based on the called historical variable data and the instruction variable data of the current task using the formula y=b+a1*x1+a2*x2+a3*x3+…+an*xn, where y represents the matching value between the historical task scenario and the current task, a represents the weight, and x represents the deviation between the historical variable data corresponding to the instruction variable and the instruction variable data in the current task. It can be set manually or obtained through training based on the training set and validation set. For example, when the historical temperature is 1 degree lower than the temperature in the current task, x can be set to 1. The larger the deviation, the smaller x is. b represents the deviation constant.

[0029] S1233: If the matching value is greater than a preset threshold, infer the task scheduling information of the current task scheduling instruction based on the historical scheduling information of the historical task scenario.

[0030] In this embodiment, the scheduling location and scheduling content are selected based on the similarity compared with the task scenario that has occurred. When the matching value calculated by the model is greater than the preset threshold, it means that the historical task scenario and the current task are highly similar. The task scheduling information of the current task scheduling instruction is inferred based on the historical scheduling information containing the scheduling location and scheduling content of the historical task scenario.

[0031] S130: Search for a corresponding terminal according to the to-be-scheduled content and the to-be-scheduled location, and schedule the terminal.

[0032] In this step, the terminal feature data of each terminal is searched and compared based on the target features in the content to be scheduled and the location to be scheduled, to find the corresponding terminal and schedule the data resources obtained by the terminal; these data resources may include video, audio, and / or text data. That is, the terminal to be scheduled is reversely determined based on the target features in the content to be scheduled and the location to be scheduled, and a control command is sent to the terminal to retrieve the resource data of the terminal, such as captured video and recorded audio, thereby enabling the upload of audio, video, and location information, facilitating the timely detection of safety hazards.

[0033] In the present invention, the specific scheduling operation process of the current task can refer to the historical scheduling information of historical task scenes (such as some emergency management accidents), or in some specific object monitoring tasks, the historical scheduling information of almost identical historical task scenes can be used as the task scheduling information of the current task scheduling instruction. For example, in the above-mentioned "water pipe freezing and cracking situation" task scheduling instruction, when the instruction variable data of the water pipe at a certain position is weighted by the weighted linear algorithm formula variable to obtain a matching value greater than the preset threshold, the model infers that the water pipe at this location is the water pipe currently to be monitored for freezing and cracking, and the resource data of the monitoring equipment such as the control ball for photographing and observing the water pipe is scheduled; and when the task scheduling instruction is a task scheduling instruction related to emergency management accidents, that is, when the current task type is an emergency management accident, it usually involves demand scheduling in multiple dimensions such as on-site perception, personnel safety and / or vehicle tracking, then the task scheduling information of the current task scheduling instruction is obtained by referring to the historical scheduling information of historical task scenes with high similarity obtained by the above steps, and is usually task scheduling information including multiple dimensions of demand, such as In a joint emergency rescue mission involving a chemical plant explosion and leakage, the most similar historical mission scenario is obtained based on a neural network analysis model. The historical dispatch information for this historical mission scenario may include the dispatch content and dispatch location in three dimensions: quickly perceiving and locating the accident to achieve on-site perception; querying whether personnel were in the most dangerous position (leakage source) at the time of the explosion and whether personnel had been safely evacuated to achieve personnel safety awareness; and tracking and screening hazardous chemical delivery vehicles during a specific period to quickly identify the suspected vehicle causing the leak. If the on-site perceived location is the leak site, and the dispatch content includes photographed environmental markers at the leak site, then based on the target features and location of the dispatch content, it can be seen that the on-site perception of the current mission also requires obtaining the markers at the leak site. The terminal set at the leak site of the current mission scenario is searched and compared, and the target features of the terminal feature data are queried and compared with the target features of the dispatch content to reversely obtain the terminal to be dispatched (i.e., the terminal that photographed the markers at the current mission leak site, such as drone 1 that photographed the markers at that location). The terminal's photographed data is obtained to quickly perceive and locate the accident.

[0034] As can be seen in the above scheme, the dynamic scheduling management method of the present invention uses a pre-trained neural network analysis model to perform historical task scenario reasoning and matching based on the task scheduling instruction, obtains historical task scenarios with a similarity exceeding a preset threshold, performs logical reasoning based on the historical scheduling information of the historical task scenarios, obtains the task to be scheduled information of the current task scheduling instruction, and then searches for the corresponding terminal based on the to-be-scheduled content and to-be-scheduled location in the task to be scheduled information, and schedules the terminal. It can be seen that the present invention can obtain historical task scenarios with a high similarity to the current task state based on the current task scheduling instruction using a neural network analysis model with logical reasoning capabilities, quickly locate the terminal to be scheduled through logical reasoning analysis, dynamically schedule various terminals, and timely schedule various terminal resources, and promptly observe safety hazards, greatly improving the efficiency and flexibility of scheduling, enabling rapid linkage and communication, and being applicable to various scheduling scenarios (such as public transportation management, intelligent management, emergency management, etc. in various scenarios such as scenic spots, factory areas, commercial areas, and venues). Dynamic scheduling of various terminal resources can break down communication barriers in various scheduling scenarios, and the scheduled resource data can be transmitted back immediately, facilitating the immediate issuance of instructions on demand, helping the command center to make quick decisions, and effectively reducing risk losses.

[0035] It should be noted that the size of the serial numbers of the steps in the above embodiment does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiment of the present invention.

[0036] Reference Figure 4 , Figure 4 This is a flow chart of a dynamic scheduling management method provided by the second embodiment of the present invention. As shown in the figure, the method includes the following steps S210-S280: S210: Acquire information of each terminal and generate terminal feature data associated with features of each terminal.

[0037] Initially, the information of each terminal is obtained. When there is a scheduling requirement, step S220 and step S230 can be executed respectively according to the input task scheduling instruction or call information. Since step S210 is the same or similar to step S110, it will not be repeated here.

[0038] S220: Utilize a pre-trained neural network analysis model according to the task scheduling instruction and perform logical reasoning analysis based on historical information of historical task scenarios similar to the task in the task scheduling instruction to obtain task scheduling information of the current task scheduling instruction.

[0039] This step is the same as or similar to step S120 and will not be described again here.

[0040] S230: Receive call information, and perform logical reasoning analysis based on the call information using a pre-trained neural network analysis model to obtain a call number identification result and a communication link identification result.

[0041] In this step, the call information includes the number of the dialed mobile terminal, the calling number and the communication link. A pre-trained neural network analysis model is used to perform logical reasoning analysis based on the calling number and the communication link to obtain the calling number identification result and the communication link identification result. The calling number identification result is the type of the calling number, such as whether the call is an emergency call or an alarm call. If the calling number is 110, it is inferred to be an alarm call. If the calling number is 120 or 119, it is inferred to be an emergency call.

[0042] S240: If the calling number identification result is an alarm or emergency call, obtain task scheduling information including the task content and the task location according to the calling number identification result and the communication link identification result.

[0043] In this step, when the call number identification result is an alarm or emergency call, the location to be dispatched can be obtained based on the communication link identification result, and the content to be dispatched can be obtained based on the call number identification result. If it is an alarm or emergency call, the content to be dispatched is the current status of the owner of the dialing mobile terminal, which can be obtained by shooting with a terminal such as a surveillance ball or a drone.

[0044] S250: Search for a corresponding terminal according to the to-be-scheduled content and the to-be-scheduled location, and schedule the terminal.

[0045] In this step, when the input is call information, a terminal for shooting or generating content targeting the target in the content to be scheduled is searched for at the location according to the content to be scheduled and the location to be scheduled. Since the specific implementation of this step is the same or similar to step S130, it will not be repeated here.

[0046] S260: Acquire a receiving object associated with the task scheduling instruction / call number.

[0047] In this step, when the input is call information, the receiving object associated with the call number (such as police, hospital, etc.) is obtained. If the input is a task scheduling instruction, and the completion of the current task corresponding to the task scheduling instruction requires cooperation from other departments, the receiving object associated with the current task cooperation can be obtained to facilitate instant communication and cooperation.

[0048] In this embodiment, step S270 or step S280 may be executed after obtaining the associated receiving object.

[0049] S270: Generate a SIP instruction according to the location, and schedule a communication link to deliver the SIP instruction to a receiving object.

[0050] In this step, a SIP instruction may be generated according to the location to be scheduled, and the SIP instruction may be delivered to the recipient via the IMS communication link.

[0051] S280: Generate a SIP instruction according to the location, schedule a communication link to transmit the SIP instruction to a receiving object, and push the scheduled data resources to the receiving object.

[0052] In this step, SIP instructions can be generated according to the location and delivered to the recipient through the IMS communication link. The data resources of the scheduled terminal can also be pushed to the recipient, so that the recipient can understand the situation in a timely manner, and more efficiently link with external units such as public security, fire protection, and medical care to achieve real-time information sharing, more efficient cross-departmental collaboration, significantly improve scheduling management efficiency, and provide strong function expansion and remote management capabilities.

[0053] It can be seen that in the above scheme, the present invention can obtain historical task scenarios with high similarity to the current task status according to the current task scheduling instructions based on the neural network analysis model with logical reasoning ability, quickly locate the terminal to be scheduled through logical reasoning analysis, and dynamically schedule various terminals, or analyze the call information according to the neural network analysis model to obtain the call number identification result and the communication link identification result. When the call number identification result is an alarm or emergency call, the task scheduling information including the content to be scheduled and the location to be scheduled is obtained according to the call number identification result and the communication link identification result, thereby searching for the terminal to be scheduled, scheduling the resource data of the corresponding terminal, and can be based on The location generates a SIP instruction, which is transmitted to the receiving object associated with the current task scheduling instruction / call number through the IMS communication link. The scheduled data resources can further be pushed to the receiving object. It can be seen that the dynamic scheduling management method of this embodiment can timely schedule various terminal resources, timely observe safety hazards, and realize real-time information sharing, more efficiently realize cross-departmental collaboration, and efficiently link external units such as public security, fire protection, and medical care to realize real-time information sharing, more efficient cross-departmental collaboration, and significantly improve scheduling management efficiency. In emergencies, high-definition video return and instant voice communication can be realized, helping the command center to make quick decisions and dispatch resources, effectively reduce risk losses, and enhance emergency response capabilities.

[0054] Reference Figure 5 , Figure 5 This is a schematic diagram of the structure of the dynamic scheduling management system provided by the first embodiment of the present invention. In the embodiment shown in the figure, the dynamic scheduling management system includes a terminal feature collection unit 101, a model inference unit 102, and a scheduling unit 103. The functional units are described in detail as follows: The terminal feature collection unit 101 is used to obtain information about each terminal and generate terminal feature data associated with each terminal feature; wherein the terminal feature includes the terminal location and target features of the generated content; The model reasoning unit 102 is configured to use a pre-trained neural network analysis model to perform logical reasoning analysis based on the task scheduling instruction and historical information of historical task scenarios similar to the task in the task scheduling instruction, thereby obtaining the task to be scheduled information of the current task scheduling instruction; wherein the task to be scheduled information includes the task to be scheduled content and the task to be scheduled location; The scheduling unit 103 is configured to search for a corresponding terminal according to the to-be-scheduled content and the to-be-scheduled location, and schedule the terminal.

[0055] In some embodiments, the model inference unit 102 is specifically configured to: Use the pre-trained neural network analysis model to analyze the input task scheduling instructions and obtain the current task; Acquire historical information of historical task scenarios similar to the current task in real time, and infer and configure instruction variables related to the current task based on prompt words; The corresponding historical information is called according to the instruction variable to perform matching analysis to obtain the task scheduling information of the current task scheduling instruction.

[0056] In some embodiments, the model inference unit 102 is further configured to: Calling historical variable data corresponding to the instruction variable in the historical task scenario; A weighted linear algorithm is used to calculate the matching value between the historical task scenario and the current task based on the historical variable data corresponding to the instruction variable in the historical task scenario; If the matching value is greater than a preset threshold, the task scheduling information of the current task scheduling instruction is inferred based on the historical scheduling information of the historical task scenario; wherein the historical information includes historical variable data and historical scheduling information.

[0057] In some embodiments, the model inference unit 102 is specifically configured to: The matching value between the historical task scenario and the current task is calculated based on the historical variable data corresponding to the instruction variable in the historical task scenario and the instruction variable data of the current task using the formula y=b+a1*x1+a2*x2+a3*x3+……+an*xn; wherein a represents the weight, x represents the deviation between the historical variable data corresponding to the instruction variable and the instruction variable data in the current task, and b represents the deviation constant.

[0058] In some embodiments, the model inference unit 102 is further configured to: Receive call information, and perform logical reasoning analysis based on the call information using a pre-trained neural network analysis model to obtain a call number identification result and a communication link identification result; If the call number identification result is an alarm or emergency call, obtaining task dispatch information including the content to be dispatched and the location to be dispatched according to the call number identification result and the communication link identification result; The corresponding terminal is searched according to the content to be scheduled and the location to be scheduled, and the terminal is scheduled, and a SIP instruction is generated and transmitted to a receiving object associated with the call number.

[0059] In some embodiments, the scheduling unit 103 is specifically configured to: The terminal feature data of each terminal is searched and compared according to the target features in the content to be scheduled and the location to be scheduled to find the corresponding terminal, and the data resources obtained by the terminal are scheduled; wherein the data resources include video, audio and / or text data.

[0060] In some embodiments, the scheduling unit 103 is further configured to: Acquire a receiving object associated with the task scheduling instruction according to the task scheduling instruction; Generate a SIP instruction according to the location, and schedule a communication link to deliver the SIP instruction to a recipient; or A SIP instruction is generated according to the position, a communication link is scheduled to transmit the SIP instruction to a receiving object, and the scheduled data resources are pushed to the receiving object.

[0061] It can be seen that the present invention provides a dynamic scheduling and management system, which can be responsible for the operation of similar tasks based on a neural network analysis model with logical reasoning capabilities, and can dynamically allocate and adjust resources according to the requirements input into the system (task scheduling instructions or call information), timely observe safety hazards, and realize real-time information sharing, more efficiently realize cross-departmental collaboration, and efficiently link external units such as public security, fire protection, and medical care to realize real-time information sharing, more efficient cross-departmental collaboration, and significantly improve scheduling and management efficiency. In emergencies, high-definition video feedback and instant voice communication can be realized to help the command center make quick decisions and dispatch resources, effectively reduce risk losses, and enhance emergency response capabilities.

[0062] Reference Figure 6 , Figure 6 This is a schematic block diagram of a dynamic scheduling management system provided by the second embodiment of the present invention. Figure 6As shown, the dynamic scheduling management system of this embodiment includes a server 301 and multiple terminals 302. The server 301 is configured with a call intelligence model 3011. The call intelligence model 3011 includes multiple neural network analysis models for task scheduling management to schedule multiple task requirements.

[0063] In this embodiment, the server 301 is configured to use a pre-trained neural network analysis model according to the task scheduling instruction to perform logical reasoning analysis based on historical information of historical task scenarios similar to the task in the task scheduling instruction, thereby obtaining task scheduling information of the current task scheduling instruction; wherein the task scheduling information includes the scheduled content and the scheduled location, and preferably, the task scheduling information can also be output to the user device screen for display; and search for the corresponding terminal 302 according to the scheduled content and the scheduled location to schedule the terminal 302; Alternatively, the server 301 is used to perform logical reasoning analysis based on the call information using a pre-trained neural network analysis model to obtain a call number identification result and a communication link identification result; if the call number identification result is an alarm or emergency call, the task scheduling information including the content to be scheduled and the location to be scheduled is obtained based on the call number identification result and the communication link identification result; the corresponding terminal 302 is searched for based on the content to be scheduled and the location to be scheduled, and the terminal 302 is scheduled, and a SIP instruction is generated, and the SIP instruction is passed to the receiving object associated with the call number.

[0064] In the present invention, the neural network analysis model has the ability of logical reasoning, which can judge and match tasks and perform operations responsible for similar tasks. That is, when an emergency occurs (traffic accident, bad weather, equipment failure), the neural network analysis model in the dynamic scheduling management system of this embodiment can obtain historical scheduling data of historical task scenarios with high similarity, and then infer the corresponding emergency plan for the current task (that is, the task scheduling information) based on the historical scheduling data. It can also be output and displayed on the user device, and the user can start the emergency plan with one click, making the scheduling more intelligent, efficient and flexible.

[0065] It is understandable that a neural network analysis model in the call intelligent model 3011 can schedule and manage a task, or schedule and manage a dimensional requirement in the task. For example, in scenarios such as preset patrol routes and check-in points, accident monitoring and scheduling at different patrol points can use different neural network analysis models for monitoring and scheduling, so as to achieve key inspections at the patrol points where accidents have occurred.

[0066] It should be noted that those skilled in the art can clearly understand that the specific implementation process of the above-mentioned dynamic scheduling management system and each unit or component can refer to the corresponding description in the aforementioned method embodiment. For the convenience and brevity of the description, it will not be repeated here.

[0067] In one embodiment, a computer device is provided, whose internal structure diagram can be as follows: Figure 7 As shown. The computer device includes a processor, memory, network interface, and database connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes non-volatile and / or volatile storage media and internal memory. The non-volatile storage medium stores an operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The network interface of the computer device is used to communicate with external devices via a network connection. Furthermore, the computer device may also include a display screen and input devices (such as a mouse, keyboard, etc.) for interaction.

[0068] Specifically, the processor in the computer device implements the steps of the dynamic scheduling management method provided in the first embodiment and the second embodiment when executing the computer program.

[0069] In one embodiment, the present invention may further provide a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the steps of the dynamic scheduling management method provided in the first and second embodiments are implemented.

[0070] Those skilled in the art will understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), Synchronous Link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0071] Those skilled in the art will clearly understand that for the sake of convenience and brevity in description, only the division of the above-mentioned functional units and modules is used as an example. In actual applications, the above-mentioned functions can be distributed and completed by different functional units and modules as needed, that is, the internal structure of the system can be divided into different functional units or modules to complete all or part of the functions described above.

[0072] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the scope of the technical solutions of the embodiments of the present invention, and should all be included in the scope of protection of the present invention.

Claims

1. A dynamic scheduling management method, characterized in that: The dynamic scheduling management method includes: Using a pre-trained neural network analysis model according to the task scheduling instruction, logical reasoning analysis is performed based on historical information of historical task scenarios similar to the task in the task scheduling instruction to obtain task scheduling information of the current task scheduling instruction; wherein the task scheduling information includes the task scheduling content and the task scheduling location; A corresponding terminal is searched according to the to-be-scheduled content and the to-be-scheduled location, and the terminal is scheduled.

2. The dynamic scheduling management method according to claim 1, wherein: The task scheduling instruction is obtained by using a pre-trained neural network analysis model and performing logical reasoning analysis based on historical information of historical task scenarios similar to the task in the task scheduling instruction, specifically including: Use the pre-trained neural network analysis model to analyze the input task scheduling instructions and obtain the current task; Acquire historical information of historical task scenarios similar to the current task in real time, and infer and configure instruction variables related to the current task based on prompt words; The corresponding historical information is called according to the instruction variable to perform matching analysis to obtain the task scheduling information of the current task scheduling instruction.

3. The dynamic scheduling management method according to claim 2, characterized in that: The step of calling the corresponding historical information according to the instruction variable to perform matching analysis to obtain the task scheduling information of the current task scheduling instruction specifically includes: Calling historical variable data corresponding to the instruction variable in the historical task scenario; A weighted linear algorithm is used to calculate the matching value between the historical task scenario and the current task based on the historical variable data corresponding to the instruction variable in the historical task scenario; If the matching value is greater than a preset threshold, the task scheduling information of the current task scheduling instruction is inferred based on the historical scheduling information of the historical task scenario; wherein the historical information includes historical variable data and historical scheduling information.

4. The dynamic scheduling management method according to claim 3, wherein: The method of using a weighted linear algorithm to calculate the matching value between the historical task scenario and the current task based on the historical variable data corresponding to the instruction variable in the historical task scenario specifically includes: The matching value between the historical task scenario and the current task is calculated based on the historical variable data corresponding to the instruction variable in the historical task scenario and the instruction variable data of the current task using the formula y=b+a1*x1+a2*x2+a3*x3+……+an*xn; wherein a represents the weight, x represents the deviation between the historical variable data corresponding to the instruction variable and the instruction variable data in the current task, and b represents the deviation constant.

5. The dynamic scheduling management method according to claim 1, wherein: The dynamic scheduling management method further includes: Receive call information, and perform logical reasoning analysis based on the call information using a pre-trained neural network analysis model to obtain a call number identification result and a communication link identification result; If the call number identification result is an alarm or emergency call, obtaining task dispatch information including the content to be dispatched and the location to be dispatched according to the call number identification result and the communication link identification result; The corresponding terminal is searched according to the content to be scheduled and the location to be scheduled, and the terminal is scheduled, and a SIP instruction is generated and transmitted to a receiving object associated with the call number.

6. The dynamic scheduling management method according to claim 1, wherein: The dynamic scheduling management method further includes: Acquire information of each terminal and generate terminal feature data associated with features of each terminal; wherein the terminal features include terminal location and target features of generated content.

7. The dynamic scheduling management method according to claim 6, characterized in that: The searching for a corresponding terminal according to the to-be-scheduled content and the to-be-scheduled location, and scheduling the terminal, specifically includes: The terminal feature data of each terminal is searched and compared according to the target features in the content to be scheduled and the location to be scheduled to find the corresponding terminal, and the data resources obtained by the terminal are scheduled; wherein the data resources include video, audio and / or text data.

8. The dynamic scheduling management method according to claim 7, wherein: After searching for a corresponding terminal according to the to-be-scheduled content and the to-be-scheduled location and scheduling the terminal, the method further includes: Acquire a receiving object associated with the task scheduling instruction according to the task scheduling instruction; Generate a SIP instruction according to the location, and schedule a communication link to deliver the SIP instruction to a recipient; or A SIP instruction is generated according to the position, a communication link is scheduled to transmit the SIP instruction to a receiving object, and the scheduled data resources are pushed to the receiving object.

9. A dynamic scheduling management system, characterized in that: The method comprises a unit for executing the dynamic scheduling management method according to any one of claims 1 to 8.

10. A dynamic scheduling management system, characterized in that: The system comprises a server and multiple terminals, wherein the server is configured with a call intelligence model, and the call intelligence model comprises multiple neural network analysis models for task scheduling management, wherein: The server is configured to use a pre-trained neural network analysis model according to the task scheduling instruction to perform logical reasoning analysis based on historical information of historical task scenarios similar to the task in the task scheduling instruction, thereby obtaining task scheduling information of the current task scheduling instruction; wherein the task scheduling information includes a content to be scheduled and a location to be scheduled; and searching for a corresponding terminal according to the content to be scheduled and the location to be scheduled to schedule the terminal; or The server is used to perform logical reasoning analysis based on the call information using a pre-trained neural network analysis model to obtain a call number identification result and a communication link identification result; if the call number identification result is an alarm or emergency call, the server obtains task scheduling information including the content to be scheduled and the location to be scheduled based on the call number identification result and the communication link identification result; searches for a corresponding terminal based on the content to be scheduled and the location to be scheduled, schedules the terminal, generates a SIP instruction, and transmits the SIP instruction to a receiving object associated with the call number.

11. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the steps of the dynamic scheduling management method according to any one of claims 1 to 8 are implemented.

12. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the dynamic scheduling management method according to any one of claims 1 to 8 are implemented.

Images