Sky-ground integrated monitoring task planning method and system based on agent

Abstract

The present application relates to the technical field of sky-ground integrated monitoring, and proposes a sky-ground integrated monitoring task planning method and system based on an intelligent agent, which comprises: analyzing overall monitoring task targets corresponding to task planning elements between multiple source monitoring platforms, and decomposing multiple subtasks in the overall monitoring task targets and dependency relationships between the multiple subtasks; merging the multiple subtasks into at least one candidate task chain, and after performing feasibility checking and optimization sorting on the at least one candidate task chain, calculating a synergistic effective coverage rate of a target task chain on the overall monitoring task targets; and when the synergistic effective coverage rate reaches a preset threshold, performing task planning of the multiple source monitoring platforms based on the target task chain. The present application can effectively handle dependency relationships and resource conflicts between subtasks, and improve the reliability of sky-ground integrated monitoring task planning.

Description

Technical Field

[0001] This invention relates to a method and system for planning integrated air-ground monitoring tasks based on intelligent agents, belonging to the field of integrated air-ground monitoring technology. Background Technology

[0002] With the increase in heterogeneous nodes and the improvement in task response time requirements in integrated air-ground monitoring, the application scope of monitoring task planning system continues to expand. The integrated monitoring platform realizes the unified scheduling and conflict resolution of multi-source sensing resources by building in and dynamically updating the collaborative task planning model. This is not only the key to achieving regional coverage and target allocation coordination, but also directly affects the efficiency of task execution and the rationality of resource utilization.

[0003] Traditional integrated air-ground monitoring mission planning methods rely on static rules or preset priorities to match tasks with resources. While this approach can achieve basic scheduling functions, it lacks the ability to autonomously decompose tasks and form executable links based on mission planning elements and resource status when facing the needs of heterogeneous node collaboration and dynamic response. Therefore, it is difficult to handle the dependencies and resource conflicts between subtasks.

[0004] Therefore, existing technologies lack a solution that can effectively handle dependencies and resource conflicts between subtasks and improve the reliability of integrated air-ground monitoring mission planning. Summary of the Invention

[0005] This invention provides a method and system for planning integrated air-ground monitoring tasks based on intelligent agents. Its main purpose is to effectively handle the dependencies and resource conflicts between subtasks and improve the reliability of integrated air-ground monitoring task planning.

[0006] To achieve the above objectives, the present invention provides a space-air-ground integrated monitoring task planning method based on intelligent agents, comprising: By deploying intelligent agents at least one of the monitoring center, edge nodes, cloud servers, or monitoring resource terminals, spatiotemporal monitoring data corresponding to the multi-source monitoring platform is obtained, so as to determine the task planning elements between the multi-source monitoring platforms. The multi-source monitoring platform includes one or more of the following: space-based platform, air-based platform, ground-based platform, mobile platform, or artificial platform. The overall monitoring task objective corresponding to the task planning elements is parsed out. In response to the task triggering event of the multi-source monitoring platform, the agent decomposes multiple sub-tasks in the overall monitoring task objective and the dependencies between the multiple sub-tasks. Based on the task planning elements, a resource status vector corresponding to each monitoring platform is established. According to the dependency relationship, the resource status vector and the preset business rules, the multiple sub-tasks are merged into at least one candidate task chain. The agent is used to perform feasibility verification and optimization ranking of the at least one candidate task chain to obtain the target task chain, and the collaborative effective coverage rate of the target task chain to the overall monitoring task target is calculated. When the effective coverage of the collaboration reaches a preset threshold, the task planning of the multi-source monitoring platform is executed based on the target task chain.

[0007] Optionally, based on the dependencies, the resource state vector, and preset business rules, the multiple subtasks are merged into at least one candidate task chain, including: Based on the dependencies, determine the execution order constraints among the multiple subtasks; By combining the resource state vector and the execution order constraint, a corresponding target monitoring platform is assigned to each subtask, and an initial allocation scheme is generated. Based on the preset business rules, the initial allocation scheme is processed to resolve conflicts, thereby obtaining the target allocation scheme; According to the target allocation scheme, the multiple subtasks are combined into at least one candidate task chain according to the execution order constraint.

[0008] Optionally, by combining the resource state vector and the execution order constraint, a corresponding target monitoring platform is assigned to each subtask to generate an initial allocation scheme, including: Based on the execution order constraints, determine the resource requirement type and required time window for each subtask; The resource type, available time window, and load rate of each monitoring platform are obtained from the resource status vector. When the resource type matches the resource demand type, the available time window covers the required time window, and the load rate is lower than the preset load rate threshold, the target monitoring platform is selected, and a corresponding sub-task is assigned to the target monitoring platform to generate an initial allocation scheme.

[0009] Optionally, based on the preset business rules, conflict resolution processing is performed on the initial allocation scheme to obtain the target allocation scheme, including: Conflicting subtasks were detected in the initial allocation scheme; Based on the resource priority rules in the preset business rules, the low-priority subtasks in the conflicting subtasks are selected. The low-priority subtasks are processed by resource reallocation or time window delay to obtain the target allocation scheme. When no other available monitoring platform meets the resource requirement type and the time window, the low-priority subtask is marked as a waiting task, and the task degradation process of the low-priority subtask is triggered.

[0010] Optionally, the task planning elements among the multi-source monitoring platforms are determined, including: The spatiotemporal monitoring data of the multi-source monitoring platform is transformed into a unified spatiotemporal reference to generate unified spatiotemporal reference data. The unified spatiotemporal reference data includes the position coordinates of each monitoring platform in the same coordinate system and the time window in the same time reference. Spatiotemporal correlation processing is performed on the location coordinates and time windows in the unified spatiotemporal reference data to obtain the spatiotemporal overlap relationship between each monitoring platform; Based on the spatiotemporal overlap relationship, a three-dimensional spatiotemporal diagram of the multi-source monitoring platform in the task collaboration process is constructed, and the collaboration tightness between the multi-source monitoring platforms is calculated. Extract the spatiotemporal coupling nodes and resource contention edges in the three-dimensional spatiotemporal graph; Based on the spatiotemporal coupling nodes, the resource competition edges, and the degree of collaboration, the task planning elements among the multi-source monitoring platforms are determined.

[0011] Optionally, the unified spatiotemporal reference conversion of the spatiotemporal monitoring data is achieved based on BeiDou / GNSS timing and positioning information, and is used to convert the monitoring data of space-based platforms, air-based platforms, ground-based platforms, mobile platforms and artificial platforms to the same coordinate system and the same time reference.

[0012] Optionally, based on the target task chain, the task planning of the multi-source monitoring platform is executed, including: Based on the task execution order in the target task chain, generate the task execution sequence of each monitoring platform of the multi-source monitoring platform; Based on the resource binding relationship in the task execution sequence and the target task chain, sub-tasks and task execution time periods are allocated to each monitoring platform, and a task allocation table is generated; Based on the task allocation table, task execution instructions for each monitoring platform are generated and sent to the corresponding monitoring platform.

[0013] Optionally, based on the task allocation table, task execution instructions for each monitoring platform are generated, including: Identify the monitoring platform type corresponding to each task in the task allocation table; If the monitoring platform type is a space-based platform, then a task execution instruction containing the monitoring area, start time, data format, and backhaul link is generated; If the monitoring platform type is an airborne platform, then a task execution command containing the flight path, observation angle, duration and sampling frequency is generated; If the monitoring platform type is a ground-based platform or a mobile platform, then a task execution instruction containing the monitoring area, start time, duration and sampling frequency is generated; If the monitoring platform type is a manual platform, then a task execution instruction containing the monitoring area, review requirements, and alarm thresholds will be generated.

[0014] Optionally, the method further includes: When the effective coverage of the collaboration is lower than a preset threshold, the agent triggers the task replanning process of the multi-source monitoring platform. The task replanning process includes adjusting the decomposition granularity of the overall monitoring task objective, re-screening the candidate task chain, or modifying the resource priority of the conflicting sub-tasks.

[0015] Optionally, calculating the collaborative effective coverage rate of the target task chain to the overall monitoring task objective includes: Extract the key monitoring indicators and indicator achievement thresholds for each sub-task from the overall monitoring task objectives; Obtain the actual measurement values ​​of each task node in the target task chain and the corresponding key monitoring indicators; Based on the actual measured values ​​and the target thresholds, calculate the achievement coefficient for each sub-task; Calculate the coordination effectiveness coefficient between each task node in the target task chain; By combining the compliance coefficient and the collaborative effectiveness coefficient, the collaborative effective coverage rate of the target task chain on the overall monitoring task target is calculated.

[0016] Optionally, in response to a task triggering event from the multi-source monitoring platform, the agent decomposes the overall monitoring task objective into multiple sub-tasks, including: Extract the event type, event level, event location, and scope of influence of the task-triggered events to determine the decomposition granularity of the overall monitoring task objective; Analyze the spatial distribution characteristics and temporal urgency of the overall monitoring task objectives to identify the spatial sub-regions and temporal sub-stages of the overall monitoring task objectives; Based on the decomposition granularity, the spatial sub-regions, and the temporal sub-stages, the overall monitoring task objective is broken down into multiple sub-tasks.

[0017] Optionally, the dependencies between the multiple subtasks include data dependencies, resource dependencies, and timing dependencies. Data dependencies refer to the sequential constraints arising from input data dependencies between subtasks, determined by analyzing the input and output data flows between each subtask; one subtask depends on the data results generated by another subtask before execution. Resource dependencies refer to the competition or mutual exclusion constraints arising from sharing the same monitoring resources between subtasks, determined by analyzing the contention of each subtask for monitoring resources corresponding to space-based, air-based, ground-based, mobile, or artificial platforms; multiple subtasks have conflicting or shared constraints regarding the same resource. Timing dependencies refer to the execution order constraints arising from time window constraints between subtasks, determined by analyzing the sequential connection between the available time windows and task durations of each subtask.

[0018] Optionally, based on the task planning elements, a resource status vector corresponding to each monitoring platform is established, including: Based on the available resource list in the task planning elements, extract the resource identifier, current location, and available time window of each monitoring platform to generate static attribute components; Based on the business rules and real-time monitoring data in the task planning elements, the load rate, energy consumption status and task occupancy status of each monitoring platform are extracted synchronously to generate dynamic status components. The static attribute components are combined with the dynamic state components to obtain the resource state vector.

[0019] Optionally, the agent is used to perform feasibility verification and optimization ranking of the at least one candidate task chain, including: The system calls upon the agent's historical case library and extracts the event types, task planning schemes, resource allocation records, and execution feedback results of historical tasks from the historical case library. Based on the event type and the task planning scheme, the feasibility of the candidate task chain is verified, wherein the feasibility verification includes one or more of resource conflict verification, timing conflict verification, or communication reachability verification. Based on the resource allocation record and the execution feedback result, the candidate task chains that have passed the verification are optimized and sorted. The optimized sorting is based on at least two of the following: task response time, resource load balancing degree, or coverage completeness of each candidate task chain.

[0020] Optionally, the intelligent agent includes a task understanding module, a resource state perception module, a task chain planning module, a conflict resolution module, an execution feedback module, and a replanning module. The task chain planning module generates candidate task chains based on a large language model, knowledge graph, or rule engine, according to task triggering events, historical task cases, resource state vectors, and preset business rules. The replanning module updates subsequent planning strategies based on task execution feedback through a memory retrieval module or a reinforcement learning model.

[0021] This invention provides an agent-based integrated air-ground monitoring task planning system for executing the aforementioned agent-based integrated air-ground monitoring task planning method, the system comprising: The planning element determination module is used to obtain spatiotemporal monitoring data corresponding to the multi-source monitoring platform through an intelligent agent deployed at least one of the monitoring center, edge node, cloud server or monitoring resource terminal, so as to determine the task planning elements between the multi-source monitoring platforms. The multi-source monitoring platform includes one or more of the following: space-based platform, air-based platform, ground-based platform, mobile platform or artificial platform. The task decomposition module is used to parse out the overall monitoring task objective corresponding to the task planning elements, and in response to the task triggering event of the multi-source monitoring platform, the agent decomposes the overall monitoring task objective into multiple sub-tasks and the dependencies between the multiple sub-tasks. The candidate chain generation module is used to establish resource status vectors corresponding to each monitoring platform based on the task planning elements, and to merge the multiple subtasks into at least one candidate task chain according to the dependency relationship, the resource status vectors and preset business rules. The task chain optimization module is used to use the intelligent agent to perform feasibility verification and optimization ranking of the at least one candidate task chain to obtain the target task chain and calculate the collaborative effective coverage rate of the target task chain to the overall monitoring task target. The planning and execution module is used to execute the task planning of the multi-source monitoring platform based on the target task chain when the effective coverage of the collaboration reaches a preset threshold.

[0022] The present invention provides an electronic device for executing the above-described intelligent agent-based integrated air-ground monitoring task planning method, including a processor, a memory, and a computer program stored in the memory and capable of running on the processor.

[0023] The present invention provides a computer-readable storage medium for executing the above-described agent-based integrated air-ground monitoring task planning method, wherein a computer program is stored thereon.

[0024] Compared to the problems described in the background technology, this invention, through an intelligent agent deployed at least one of the monitoring center, edge nodes, cloud servers, or monitoring resource terminals, acquires spatiotemporal monitoring data corresponding to multi-source monitoring platforms, determines the task planning elements between the multi-source monitoring platforms, and can perceive the dynamic changes of multi-source monitoring platforms in real time, improving the accuracy of task planning in response to the spatiotemporal coordination needs of heterogeneous nodes. Furthermore, in response to the task triggering events of the multi-source monitoring platforms, this invention, through the intelligent agent, decomposes multiple sub-tasks in the overall monitoring task objective and the dependencies between these sub-tasks, dynamically responding to the real-time monitoring needs of integrated air-ground systems, automatically decomposing macro-level task objectives into structured sub-tasks and clarifying their dependencies, improving the real-time performance and executability of task planning. This invention, by using the dependencies, resource state vectors, and preset business rules, further... By merging multiple subtasks into at least one candidate task chain, the logical dependencies between subtasks, the real-time status of resources, and business constraints can be integrated to automatically generate an executable task chain, avoiding planning failures caused by traditional static preset sequences or ignoring changes in resource status. Furthermore, by utilizing the intelligent agent to perform feasibility verification and optimization sorting of the at least one candidate task chain to obtain the target task chain, the quality and executability of task planning results can be improved, avoiding execution failures or resource waste caused by selecting infeasible or suboptimal task chains. Finally, by executing the task planning of the multi-source monitoring platform based on the target task chain when the effective collaborative coverage reaches a preset threshold, the present invention avoids blind execution leading to resource waste or task failure when collaborative coverage is insufficient, improving the effectiveness and reliability of the integrated sky-ground monitoring task planning execution. Therefore, the present invention can effectively handle dependencies and resource conflicts between subtasks, improving the reliability of integrated sky-ground monitoring task planning. Attached Figure Description

[0025] Figure 1 This is a diagram illustrating the overall architecture of an integrated air-ground system for a space-ground integrated monitoring task planning method based on an intelligent agent, as provided in an embodiment of the present invention. Figure 2 This is a flowchart illustrating an embodiment of the intelligent agent-based integrated air-ground monitoring task planning method provided by the present invention. Figure 3 A diagram illustrating the composition of a multi-source monitoring platform for an agent-based integrated air-ground monitoring task planning method according to an embodiment of the present invention. Figure 4 This is a diagram illustrating the agent module structure of an agent-based integrated air-ground monitoring task planning method according to an embodiment of the present invention. Figure 5A unified spatiotemporal reference transformation diagram for an agent-based integrated air-ground monitoring task planning method provided in an embodiment of the present invention; Figure 6 This is a subtask decomposition and dependency graph of an agent-based integrated air-ground monitoring task planning method provided in an embodiment of the present invention. Figure 7 This is a resource state vector construction diagram of an agent-based integrated air-ground monitoring task planning method provided in an embodiment of the present invention; Figure 8 This is a closed-loop diagram of candidate task chain generation, conflict resolution, and replanning of a space-ground integrated monitoring task planning method based on an agent provided in an embodiment of the present invention. Figure 9 This is a diagram illustrating the generation and distribution of task execution instructions for a space-ground integrated monitoring task planning method based on an intelligent agent, provided in an embodiment of the present invention. Figure 10 This is a schematic diagram of a module for implementing an agent-based integrated air-ground monitoring task planning system according to an embodiment of the present invention. Figure 11 A schematic diagram of a computer device for implementing an agent-based integrated air-ground monitoring task planning method according to an embodiment of the present invention; The objectives, features, and advantages of this invention will be further explained in conjunction with the embodiments and the accompanying drawings. Detailed Implementation

[0026] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention. This application provides a method for planning integrated air-ground monitoring tasks based on intelligent agents. The executing entity of this method includes, but is not limited to, at least one of the following electronic devices that can be configured to execute the method provided in this application: a server, a terminal, etc. In other words, this method can be executed by software or hardware installed on a terminal device or a server device. The server includes, but is not limited to, a single server, a server cluster, a cloud server, or a cloud server cluster.

[0027] See Figure 1The diagram shown illustrates the overall architecture of an integrated air-ground system according to an embodiment of the present invention. From bottom to top, the system comprises: a multi-source monitoring platform layer, a communication and access layer, a unified spatiotemporal reference and data fusion layer, an intelligent agent collaborative decision-making layer, a task planning and execution layer, and an application and business scenario layer. Specifically, the multi-source monitoring platform layer provides five types of monitoring resources: space-based, air-based, ground-based, mobile, and manual; the communication and access layer achieves data access based on BeiDou / GNSS timing and positioning and multiple types of communication links; the unified spatiotemporal reference and data fusion layer completes the spatiotemporal alignment of multi-source data and the construction of a three-dimensional spatiotemporal map; the intelligent agent collaborative decision-making layer includes modules for task understanding, resource perception, task chain planning, conflict resolution, execution feedback, and replanning; the task planning and execution layer is responsible for task decomposition, candidate task chain generation, optimization sorting, and command issuance; and the application and business scenario layer supports services such as water environment monitoring, forest and grassland fire monitoring, geological disaster monitoring, urban safety monitoring, and emergency response command.

[0028] See Figure 2 The diagram shown is a flowchart illustrating a space-air-ground integrated monitoring task planning method based on an intelligent agent, according to an embodiment of the present invention. In this embodiment, the space-air-ground integrated monitoring task planning method based on an intelligent agent includes: S1. By deploying an intelligent agent at least one of the monitoring center, edge node, cloud server, or monitoring resource terminal, the spatiotemporal monitoring data corresponding to the multi-source monitoring platform is obtained to determine the task planning elements between the multi-source monitoring platforms. The multi-source monitoring platform includes one or more of the following: space-based platform, air-based platform, ground-based platform, mobile platform, or artificial platform.

[0029] This invention, through acquiring spatiotemporal monitoring data corresponding to multi-source monitoring platforms, enables task planning for monitoring data from space-based, airborne, ground-based, mobile, and artificial platforms under the same coordinate system and time reference. The spatiotemporal monitoring data includes satellite remote sensing data from space-based platforms, UAV survey data from airborne platforms, ground monitoring station data or video recognition data from ground-based platforms, mobile monitoring terminal data from mobile platforms, and reported information or historical event information from artificial platforms.

[0030] In detail, the monitoring center is a central cloud platform, the edge nodes are edge computing nodes, on-site intelligent terminals, UAV ground stations or monitoring gateways, and the cloud server is a public cloud, private cloud or hybrid cloud server; the monitoring resource terminal refers to the hardware device terminal deployed at the monitoring site for collecting monitoring data, and the monitoring platform corresponding to the hardware device terminal includes both software platform and hardware resources, device terminals, mobile carriers and manual terminals that carry out monitoring tasks.

[0031] See Figure 3The figure shows a multi-source monitoring platform composition diagram of the intelligent agent-based integrated space-ground monitoring task planning method provided in an embodiment of the present invention. As shown in the figure, the space-based platform includes remote sensing satellites, navigation satellites, meteorological satellites, and synthetic aperture radar satellites; the air-based platform includes unmanned aerial vehicles (UAVs), airships, high-altitude platforms, and unmanned helicopters; the ground-based platform includes ground monitoring stations, cameras, sensors, radars, and fixed terminals; the mobile platform includes unmanned vehicles, unmanned ships, vehicle-mounted terminals, and mobile sampling equipment; and the manual platform includes manual inspection terminals, verification and reporting terminals, expert judgment terminals, and handheld terminals. Each platform has common elements such as location coordinates, time window, coverage area, target attributes, equipment status, communication link, and task occupancy status, providing basic resource objects for the construction of resource status vectors.

[0032] Specifically, the intelligent agent includes a task understanding module, a resource state perception module, a task chain planning module, a conflict resolution module, an execution feedback module, and a replanning module. The task chain planning module generates candidate task chains based on a large language model, knowledge graph, or rule engine, according to task triggering events, historical task cases, resource state vectors, and preset business rules. The replanning module updates subsequent planning strategies based on task execution feedback through a memory retrieval module or a reinforcement learning model.

[0033] See Figure 4 The figure shows the intelligent agent module structure of an integrated air-ground monitoring task planning method based on an intelligent agent according to an embodiment of the present invention. As shown in the figure, the task triggering event and the status of the multi-source monitoring platform are input into the intelligent agent; the task understanding module parses the overall task objective and outputs the sub-task decomposition and dependency relationship; the resource status perception module outputs the resource status vector by combining unified spatiotemporal reference data; the task chain planning module generates candidate task chains based on the historical case library and preset business rules; the conflict resolution module handles resource, timing and priority conflicts; the execution feedback module receives the task execution status and results; and the replanning module adjusts the task decomposition granularity and resource allocation strategy when the effective coverage of collaboration is insufficient, and outputs the replanning result.

[0034] Furthermore, by determining the task planning elements among the multi-source monitoring platforms, the embodiments of the present invention can perceive the dynamic changes of the multi-source monitoring platforms in real time, thereby improving the accuracy of task planning in responding to the spatiotemporal coordination needs of heterogeneous nodes.

[0035] As an embodiment of the present invention, determining the task planning elements among the multi-source monitoring platforms includes: The spatiotemporal monitoring data of the multi-source monitoring platform is transformed into a unified spatiotemporal reference to generate unified spatiotemporal reference data. The unified spatiotemporal reference data includes the position coordinates of each monitoring platform in the same coordinate system and the time window in the same time reference. Spatiotemporal correlation processing is performed on the location coordinates and time windows in the unified spatiotemporal reference data to obtain the spatiotemporal overlap relationship between each monitoring platform; Based on the spatiotemporal overlap relationship, a three-dimensional spatiotemporal diagram of the multi-source monitoring platform in the task collaboration process is constructed, and the collaboration tightness between the multi-source monitoring platforms is calculated. Extract the spatiotemporal coupling nodes and resource contention edges in the three-dimensional spatiotemporal graph; Based on the spatiotemporal coupling nodes, the resource competition edges, and the degree of collaboration, the task planning elements among the multi-source monitoring platforms are determined.

[0036] Specifically, the unified spatiotemporal reference conversion of the spatiotemporal monitoring data is achieved based on BeiDou / GNSS timing and positioning information. It is used to convert the monitoring data of space-based platforms, air-based platforms, ground-based platforms, mobile platforms, and artificial platforms to the same coordinate system and the same time reference. The BeiDou / GNSS timing and positioning information is the original signal data obtained based on the BeiDou satellite navigation system or global navigation satellite system, used to achieve spatiotemporal synchronization of each monitoring platform, including timing pulse signals and positioning calculation data.

[0037] See Figure 5 The diagram shows a unified spatiotemporal reference transformation diagram of a space-air-ground integrated monitoring mission planning method based on an intelligent agent, provided in an embodiment of the present invention. The diagram shows the original spatiotemporal data collected by five types of platforms: space-based, air-based, ground-based, mobile, and artificial. These data are then connected to BeiDou / GNSS timing and positioning information. After being transformed by the position coordinate system and aligned with the time reference, standardized data containing unified position coordinates and a unified time window is generated. Through the association and fusion of multi-source spatiotemporal data, spatiotemporal overlap relationships and a three-dimensional spatiotemporal diagram are output, providing a standardized data foundation for subsequent determination of mission planning elements, construction of resource state vectors, and generation of candidate mission chains.

[0038] The location coordinates refer to the geographic or spatial three-dimensional coordinates of each monitoring platform in the unified spatiotemporal reference data; the time window refers to the available start and end times and duration for each monitoring platform to perform observation or communication tasks in the unified spatiotemporal reference data; the coverage area refers to the spatial region that each monitoring platform can cover in a single observation or detection task, including the observation swath, detection radius, or imaging area; the target attributes refer to the characteristic parameters of the monitored target, including target type, target size, movement speed, or spectral characteristics; and the spatiotemporal overlap relationship refers to the correlation formed by the common intersection of different monitoring platforms in terms of spatial coverage area and available time window. The three-dimensional spatiotemporal graph is a multi-dimensional graph structure constructed with monitoring platforms as nodes and spatiotemporal overlap relationships as edges. The spatiotemporal coupling node refers to a node in the three-dimensional spatiotemporal graph that has a spatiotemporal overlap relationship with at least two other monitoring platforms and whose overlap degree exceeds a preset threshold. The resource competition edge refers to an edge in the three-dimensional spatiotemporal graph that causes a conflict due to two or more monitoring platforms competing for the same spatial area or communication link within the same time window. The task planning elements refer to a set of characteristic parameters used to describe the task scenarios, available resources, and handling constraints faced by the multi-source monitoring platform, including monitoring event type, available resources, time window, spatial constraints, and handling objectives.

[0039] Optionally, a spatiotemporal overlay analysis algorithm can be used to perform spatiotemporal correlation processing on the position coordinates and time windows in the unified spatiotemporal reference data to obtain the spatiotemporal overlap relationship between each monitoring platform; a three-dimensional spatiotemporal map of the multi-source monitoring platform in the task collaboration process can be constructed by using the monitoring platform as a node, the spatiotemporal overlap relationship as an edge, and the spatial coordinates and time axis as three-dimensional coordinate axes; the collaboration tightness is calculated based on the weighted sum of the spatiotemporal intersection volume and the time overlap length.

[0040] S2. Parse the overall monitoring task objective corresponding to the task planning elements, and in response to the task triggering event of the multi-source monitoring platform, the agent decomposes the multiple sub-tasks in the overall monitoring task objective and the dependencies between the multiple sub-tasks.

[0041] This invention, by parsing the overall monitoring task objectives corresponding to the task planning elements, can extract and clarify the core intent and boundary conditions of the monitoring task from multi-source heterogeneous spatiotemporal monitoring data, providing a unified execution benchmark for subsequent task decomposition and resource scheduling. The overall monitoring task objectives include target location and tracking, regional change detection, abnormal event confirmation, and disaster scope assessment.

[0042] Furthermore, in response to task triggering events of the multi-source monitoring platform, the intelligent agent decomposes multiple sub-tasks in the overall monitoring task objective and the dependencies between these sub-tasks. This allows for dynamic response to the real-time monitoring needs of integrated air-ground systems, automatically decomposing macro-level task objectives into structured sub-tasks and clarifying their dependencies, thereby improving the real-time performance and executability of task planning. The task triggering events include abnormal alarms proactively reported by the monitoring platform, monitoring commands initiated by users, patrol tasks triggered by the system at regular intervals, and target change or threat escalation events identified by multi-source data fusion.

[0043] As an embodiment of the present invention, in response to the task triggering event of the multi-source monitoring platform, the intelligent agent decomposes multiple sub-tasks from the overall monitoring task objective, including: Extract the event type, event level, event location, and scope of influence of the task-triggered events to determine the decomposition granularity of the overall monitoring task objective; Analyze the spatial distribution characteristics and temporal urgency of the overall monitoring task objectives to identify the spatial sub-regions and temporal sub-stages of the overall monitoring task objectives; Based on the decomposition granularity, the spatial sub-regions, and the temporal sub-stages, the overall monitoring task objective is broken down into multiple sub-tasks.

[0044] The event types include abnormal alarms, user commands, scheduled tasks, or target changes; the event level refers to the urgency or importance of the task-triggered event, including multiple preset levels; the event location refers to the geographical coordinates or spatial area corresponding to the task-triggered event; the spatial sub-region refers to the sub-blocks divided into monitoring areas in the overall monitoring task objectives according to spatial distribution characteristics, including grid units, administrative division units, natural geographical units, or monitoring resource coverage units; the time sub-stages refer to the sub-intervals divided into monitoring periods in the overall monitoring task objectives according to the task execution sequence, including reconnaissance stage, confirmation stage, tracking stage, or evaluation stage.

[0045] Optionally, a text classification algorithm can be used to perform semantic parsing on the text description of the task-triggered event to extract the event type, event level, event location, and scope of influence.

[0046] Furthermore, the dependencies between the multiple subtasks include data dependencies, resource dependencies, and timing dependencies. Data dependencies refer to the sequential constraints arising from input data dependencies between subtasks, determined by analyzing the input and output data flows between each subtask; one subtask depends on the data results generated by another subtask before execution. Resource dependencies refer to the competition or mutual exclusion constraints arising from sharing the same monitoring resources between subtasks, determined by analyzing the contention of each subtask for monitoring resources corresponding to space-based, air-based, ground-based, mobile, or artificial platforms; multiple subtasks have conflicting or shared constraints regarding the same resource. Timing dependencies refer to the execution order constraints arising from time window constraints between subtasks, determined by analyzing the sequential connection between the available time windows and task durations of each subtask.

[0047] See Figure 6 The diagram shown illustrates the subtask decomposition and dependency relationship of an agent-based integrated space-ground monitoring task planning method according to an embodiment of the present invention. The agent decomposes the overall monitoring task objective into subtasks such as space-based initial screening, air-based close-range survey, ground-based continuous monitoring, mobile platform supplementary sampling, and manual verification and reporting. The subtasks are constrained by temporal dependencies represented by solid arrows, data dependencies represented by dashed arrows, and resource dependencies represented by bidirectional lines, providing a basis for the generation of candidate task chains.

[0048] S3. Based on the task planning elements, establish resource status vectors corresponding to each monitoring platform, and merge the multiple subtasks into at least one candidate task chain according to the dependency relationship, the resource status vectors and preset business rules.

[0049] The resource status vectors for each monitoring platform, established based on the task planning elements described in this embodiment of the invention, can structurally describe the real-time status of each monitoring platform in vector form, providing a calculable quantitative basis for subsequent task decomposition and resource matching. The resource status vectors include the resource identifier, current location, available time window, load rate, energy consumption status, and task occupancy status of the monitoring platform.

[0050] As an embodiment of the present invention, based on the task planning elements, a resource status vector corresponding to each monitoring platform is established, including: Based on the available resource list in the task planning elements, extract the resource identifier, current location, and available time window of each monitoring platform to generate static attribute components; Based on the business rules and real-time monitoring data in the task planning elements, the load rate, energy consumption status and task occupancy status of each monitoring platform are extracted synchronously to generate dynamic status components. The static attribute components are combined with the dynamic state components to obtain the resource state vector.

[0051] The available resource list includes the resource identifier, resource type, spatial location, time window, and current status information of each monitoring resource; the resource identifier refers to the number or name used to uniquely distinguish each monitoring platform; the business rules include preset constraint logic for resource priority, task preemption, and conflict resolution for each monitoring platform; the real-time monitoring data includes the target location, device status, and link quality of each monitoring platform; and the task occupancy status includes the current occupancy status and expected release time of each monitoring platform.

[0052] Optionally, the list of available resources can be obtained by sending a query command to the multi-source monitoring platform or by receiving status information actively reported by each monitoring platform; the business rules can be obtained by calling the rule strategy in the historical case library, wherein the historical case library refers to a database used to store the event types, task planning schemes, resource allocation records and execution result feedback of historical monitoring tasks.

[0053] See Figure 7 As shown, this is a resource state vector construction diagram of the intelligent agent-based integrated air-ground monitoring task planning method provided in an embodiment of the present invention. This diagram uniformly depicts the static attributes and dynamic states of each monitoring platform in vector form, and transforms the scattered platform information into structured data, providing quantitative input for subsequent platform selection, conflict judgment and task chain generation.

[0054] Furthermore, the present invention embodiment describes merging the multiple subtasks into at least one candidate task chain based on the dependency relationship, the resource state vector, and the preset business rules. This can integrate the logical dependencies between subtasks, the real-time status of resources, and business constraints to automatically generate an executable task chain, avoiding the planning failure problem caused by traditional static preset order or ignoring changes in resource status.

[0055] The preset business rules include resource priority rules, task preemption rules, time window allocation rules, and conflict resolution rules. The candidate task chain is a task sequence that can be delivered to the monitoring platform for execution, composed of multiple sub-tasks according to temporal dependencies and resource binding relationships.

[0056] As an embodiment of the present invention, based on the dependency relationship, the resource state vector, and preset business rules, the plurality of subtasks are merged into at least one candidate task chain, including: Based on the dependencies, determine the execution order constraints among the multiple subtasks; By combining the resource state vector and the execution order constraint, a corresponding target monitoring platform is assigned to each subtask, and an initial allocation scheme is generated. Based on the preset business rules, the initial allocation scheme is processed to resolve conflicts, thereby obtaining the target allocation scheme; According to the target allocation scheme, the multiple subtasks are combined into at least one candidate task chain according to the execution order constraint.

[0057] The execution order constraint refers to the order in which the subtasks are executed according to the dependency relationship, including time dependency constraints and resource dependency constraints; the target monitoring platform refers to the space-based platform, air-based platform, ground-based platform, mobile platform or artificial platform that is finally assigned to each subtask for execution after conflict resolution.

[0058] As an optional embodiment of the present invention, by combining the resource state vector and the execution order constraint, a corresponding target monitoring platform is assigned to each subtask to generate an initial allocation scheme, including: Based on the execution order constraints, determine the resource requirement type and required time window for each subtask; The resource type, available time window, and load rate of each monitoring platform are obtained from the resource status vector. When the resource type matches the resource demand type, the available time window covers the required time window, and the load rate is lower than the preset load rate threshold, the target monitoring platform is selected, and a corresponding sub-task is assigned to the target monitoring platform to generate an initial allocation scheme.

[0059] The resource requirements include space-based observation requirements, air-based reconnaissance requirements, ground-based detection requirements, mobile platform data collection requirements, or manual platform reporting requirements; the required time window refers to the expected execution time interval of each sub-task, including the start time and the end time.

[0060] As another optional embodiment of the present invention, based on the preset business rules, conflict resolution processing is performed on the initial allocation scheme to obtain the target allocation scheme, including: Conflicting subtasks were detected in the initial allocation scheme; Based on the resource priority rules in the preset business rules, the low-priority subtasks in the conflicting subtasks are selected. The low-priority subtasks are processed by resource reallocation or time window delay to obtain the target allocation scheme. When no other available monitoring platform meets the resource requirement type and the time window, the low-priority subtask is marked as a waiting task, and the task degradation process of the low-priority subtask is triggered.

[0061] The conflicting subtasks refer to multiple subtasks assigned to the same monitoring platform and whose required time windows overlap in the initial allocation scheme; the resource priority rule refers to the rule that determines the resource allocation order based on the priority level of each subtask when multiple conflicting subtasks compete for the same monitoring platform; the resource reallocation refers to releasing low-priority subtasks from their current assigned monitoring platform and reassigning them to other available monitoring platforms that meet their resource requirements and time window requirements; the time window delay refers to moving the execution time window of low-priority subtasks backward to avoid the time period occupied by high-priority subtasks on the same monitoring platform; and the task degradation refers to reducing the execution accuracy of the subtask, narrowing its coverage, or relaxing its completion time limit.

[0062] In one specific embodiment, when there are no other available monitoring platforms that meet the resource requirement type and time window requirements, the process further includes triggering the target replanning process of the overall monitoring task objective, wherein the target replanning refers to adjusting the decomposition method of the overall monitoring task objective or resetting the task priority.

[0063] Optionally, the priority level is determined based on one or more factors, including the urgency of the subtask, the user's permission level, the urgency of the time window, or the importance of historical tasks.

[0064] S4. Use the intelligent agent to perform feasibility verification and optimization sorting on the at least one candidate task chain to obtain the target task chain, and calculate the collaborative effective coverage rate of the target task chain on the overall monitoring task target.

[0065] The present invention utilizes the intelligent agent to perform feasibility verification and optimization sorting on at least one candidate task chain to obtain the target task chain, which can reduce the risk of execution failure or resource waste caused by selecting infeasible or suboptimal task chains.

[0066] As an embodiment of the present invention, the intelligent agent is used to perform feasibility verification and optimization ranking of the at least one candidate task chain, including: The system calls upon the agent's historical case library and extracts the event types, task planning schemes, resource allocation records, and execution feedback results of historical tasks from the historical case library. Based on the event type and the task planning scheme, the feasibility of the candidate task chain is verified, wherein the feasibility verification includes one or more of resource conflict verification, timing conflict verification, or communication reachability verification. Based on the resource allocation record and the execution feedback result, the candidate task chains that have passed the verification are optimized and sorted. The optimized sorting is based on at least two of the following: task response time, resource load balancing degree, or coverage completeness of each candidate task chain.

[0067] The historical case database refers to a database used to store event types, task planning schemes, resource allocation records, and execution result feedback for historical monitoring tasks. The event types include one or more of the following: water anomalies, forest and grassland fires, flash floods and landslides, illegal construction, pollution discharge, or urban safety incidents. The task planning scheme refers to the candidate task chain, target task chain, resource allocation scheme, and task execution order generated for historical events. The execution feedback results include task execution status, actual resource consumption, execution success or failure, reasons for failure, and deviation between the actual time window and the planned time window.

[0068] In one specific embodiment, the feasibility verification further includes one or more of the following: cost verification, risk level verification, task priority verification, or execution success probability verification.

[0069] In another specific embodiment, the optimized sorting is also based on one or more of the following: link reliability, energy consumption cost, task priority, or historical execution success rate of each candidate task chain.

[0070] See Figure 8 The diagram shown is a closed-loop diagram of candidate task chain generation, conflict resolution, and replanning of a space-air-ground integrated monitoring task planning method based on an embodiment of the present invention. The diagram describes the generation of candidate task chains from the sub-task set and dependencies, combined with resource state vectors, and the verification of resource conflicts, timing conflicts, and communication reachability. When a conflict occurs, resource reallocation, time window delay, or task degradation processing is performed. When the effective coverage of the collaboration is insufficient, the replanning process is triggered in a complete closed loop.

[0071] Furthermore, by calculating the effective collaborative coverage rate of the target task chain to the overall monitoring task objective, this embodiment of the invention can provide a measurable basis for determining whether the task planning results meet the overall monitoring task objective, thus avoiding the problem of incomplete monitoring task execution due to missing or incomplete coverage of factor tasks. The effective collaborative coverage rate is a comprehensive quantitative indicator used to measure the target task chain in two dimensions: the degree of compliance of sub-task monitoring indicators and the effectiveness of collaborative collaboration between task nodes. Its value is a normalized coefficient between 0 and 1.

[0072] As an embodiment of the present invention, calculating the cooperative effective coverage rate of the target task chain to the overall monitoring task target includes: Extract the key monitoring indicators and indicator achievement thresholds for each sub-task from the overall monitoring task objectives; Obtain the actual measurement values ​​of each task node in the target task chain and the corresponding key monitoring indicators; Based on the actual measured values ​​and the target thresholds, calculate the achievement coefficient for each sub-task; Calculate the coordination effectiveness coefficient between each task node in the target task chain; By combining the compliance coefficient and the collaborative effectiveness coefficient, the collaborative effective coverage rate of the target task chain on the overall monitoring task target is calculated.

[0073] The key monitoring indicators include target positioning accuracy, target identification accuracy, response time, coverage, and data update rate. The indicator compliance threshold is a pre-set critical value used to determine whether the monitoring results meet the task requirements. The compliance degree coefficient is a quantitative value used to measure the degree to which the actual monitoring results of a sub-task meet its key monitoring indicator requirements. The task node refers to the specific execution unit corresponding to each sub-task in the target task chain, including space-based satellites, airborne UAVs, ground-based monitoring stations, mobile terminals, or artificial platforms. The coordination effectiveness coefficient is a quantitative value used to measure the integrity of data transmission and the correctness of timing between adjacent task nodes in the target task chain, and its value ranges from 0 to 1, where 1 indicates that the coordination is completely effective and 0 indicates that the coordination is completely ineffective.

[0074] Optionally, the coordination effectiveness coefficient between each task node in the target task chain can be determined by multiplying the data transmission integrity coefficient between nodes and the timing connection correctness coefficient. The data transmission integrity coefficient can be determined by the ratio of the amount of data received to the amount of data sent by the node, with a value of [0,1]. The timing connection correctness coefficient can be determined by the ratio of the actual connection time to the maximum allowable interval, with a value of [0,1].

[0075] For example, the achievement coefficient of each sub-task is calculated using the following formula. It should be noted that this calculation method is only one possible method and does not affect the implementation of the basic scheme above: ; in, This represents the achievement level coefficient of the i-th subtask. This represents the actual measured value of the k-th metric in the i-th subtask. This represents the threshold for achieving the k-th metric in the i-th subtask. represents the index weight of the k-th indicator in the i-th subtask, m represents the number of key monitoring indicators included in each subtask, i represents the index of the subtask, and k represents the index of the key monitoring indicator.

[0076] Specifically, the actual measured value refers to the measured data output from the corresponding task node in the target task chain that is directly related to the key monitoring indicators; the index weight refers to the importance of each key monitoring indicator in the calculation of the compliance degree, which is determined by preset business rules or user configuration, and the value range is [0,1], and the sum of the weights of each indicator under the same sub-task is 1; the compliance degree coefficient The value range is [0,1]. When the actual measured values ​​of all indicators are better than or equal to the threshold, =1; when the actual measured value of any indicator is 0. =0.

[0077] It should be noted that the above formula is obtained by min The ratio of the actual measured value of each indicator to the compliance threshold is truncated to no more than 1, and a multiplication method is used for calculation, so that failure to meet any indicator will lower the compliance coefficient. This demonstrates that all key indicators in a monitoring task must meet requirements to determine the completion of a sub-task. For example, in a fire monitoring task, if the accuracy of satellite fire location meets the standard but the response time exceeds the time limit, the sub-task's guiding role for subsequent UAV close-in reconnaissance will be ineffective, as the insufficient response time cannot be compensated for by excessively meeting the accuracy standard.

[0078] Furthermore, the effective coverage rate of the collaboration is calculated using the following formula. It should be noted that this calculation method is only one possible method and does not affect the implementation of the basic scheme above: ; in, This indicates the effective collaborative coverage rate of the target task chain to the overall monitoring task objective. Indicates the degree of compliance coefficient. This represents the weight coefficient of the i-th subtask. This represents the collaborative penalty intensity coefficient, where n represents the total number of subtasks. This represents the collaboration effectiveness coefficient between the j-th task nodes, where j represents the collaboration relationship index between adjacent task nodes.

[0079] In detail, the weighting coefficients The value range of is [0,1], and =1, used to distinguish the importance of different subtasks in the overall task, determined by preset business rules or user configuration; the range of the collaborative penalty intensity coefficient λ is [0,1], used to control the penalty intensity of collaborative failure on the overall coverage. During system operation, the agent dynamically updates the λ value using the exponential weighted moving average method based on historical task execution feedback data, so that it can adapt to changes in different task scenarios and resource states. Optionally, the initial λ value can be set to 0.3~0.5 based on system operation and maintenance experience.

[0080] It should be noted that the above formula, by decomposing the effective collaborative coverage rate into the product of a weighted average achievement rate and a collaborative discount factor, can simultaneously evaluate the combined impact of the independent completion quality of subtasks and the collaborative quality between nodes on the overall coverage rate. Specifically: when the target task chain contains only one task node, the collaborative discount factor is 1; when the collaboration between all adjacent nodes is fully effective (i.e., ... When =1), the collaborative discount factor is 1, and the overall coverage rate equals the weighted average compliance rate; when collaboration completely fails (i.e., When the coefficient of performance is 0, the collaborative discount factor is 1-λ, and the overall coverage is reduced proportionally, thus avoiding a one-sided evaluation caused by ignoring collaborative issues while only considering the high degree of achievement of individual sub-tasks. For example, if the high-resolution images obtained by UAV close-range reconnaissance cannot be transmitted to the ground station in a timely and complete manner, the subsequent ground sensor verification tasks will lack the necessary input data. Even if the achievement coefficients of each sub-task are high, the overall task effect will be severely affected.

[0081] S5. When the effective coverage rate of the collaboration reaches a preset threshold, the task planning of the multi-source monitoring platform is executed based on the target task chain.

[0082] In this embodiment of the invention, when the effective collaborative coverage reaches a preset threshold, the task planning of the multi-source monitoring platform is executed based on the target task chain. This avoids resource waste or task failure caused by blind execution when the collaborative coverage is insufficient, and improves the effectiveness and reliability of the integrated air-ground monitoring task planning execution. The preset threshold is a pre-set critical value used to determine whether the effective collaborative coverage meets the task execution requirements. When the effective collaborative coverage is greater than or equal to the threshold, task planning execution based on the target task chain is triggered; when the effective collaborative coverage is lower than the threshold, task replanning or target adjustment process is triggered.

[0083] As an embodiment of the present invention, based on the target task chain, the task planning of the multi-source monitoring platform is executed, including: Based on the task execution order in the target task chain, generate the task execution sequence of each monitoring platform of the multi-source monitoring platform; Based on the resource binding relationship in the task execution sequence and the target task chain, sub-tasks and task execution time periods are allocated to each monitoring platform, and a task allocation table is generated; Based on the task allocation table, task execution instructions for each monitoring platform are generated and sent to the corresponding monitoring platform.

[0084] As another embodiment of the present invention, the method further includes: when the effective coverage rate of the collaboration is lower than a preset threshold, the agent triggers the task replanning process of the multi-source monitoring platform, the task replanning process including adjusting the decomposition granularity of the overall monitoring task objective, re-screening the candidate task chain, or modifying the resource priority of the conflicting sub-tasks.

[0085] The task execution sequence refers to the set of execution orders of the subtasks corresponding to each monitoring platform, including the execution order identifier of each subtask and their dependencies; the resource binding relationship refers to the correspondence between the monitoring platform and the subtask allocated to each subtask in the target task chain; the task allocation table is used to record the correspondence between each monitoring platform and its corresponding subtask identifier and task execution time period; the task execution instruction refers to the executable command used to control the monitoring platform to execute the subtask, including the subtask identifier, start time, duration, and execution parameters.

[0086] Optionally, the task allocation table generates a record by traversing each subtask in the task execution sequence, finding the corresponding monitoring platform in the resource binding relationship, and associating the subtask identifier, monitoring platform identifier, and task execution time period.

[0087] As an optional embodiment of the present invention, based on the task allocation table, task execution instructions for each monitoring platform are generated, including: Identify the monitoring platform type corresponding to each task in the task allocation table; If the monitoring platform type is a space-based platform, then a task execution instruction containing the monitoring area, start time, data format, and backhaul link is generated; If the monitoring platform type is an airborne platform, then a task execution command containing the flight path, observation angle, duration and sampling frequency is generated; If the monitoring platform type is a ground-based platform or a mobile platform, then a task execution instruction containing the monitoring area, start time, duration and sampling frequency is generated; If the monitoring platform type is a manual platform, then a task execution instruction containing the monitoring area, review requirements, and alarm thresholds will be generated.

[0088] The flight path refers to a three-dimensional spatial path consisting of a starting point, waypoints, an ending point, and flight altitude, used to guide the UAV to complete the survey mission; the observation angle refers to the spatial orientation of the optical axis of the monitoring equipment relative to the monitored target or ground plane, including pitch angle and yaw angle; the data format refers to the encoding method and encapsulation format of the monitoring data, including GeoTIFF, JSON, or Protobuf; the verification requirement refers to the instruction requiring manual verification of suspected abnormal points on-site, including the coordinates of the verification point, the verification content, and the result reporting method; the alarm threshold refers to the critical value of the condition that triggers an alarm, including pollutant concentration threshold or risk level threshold.

[0089] See Figure 9 The diagram shown illustrates the task execution instruction generation and distribution process of a space-air-ground integrated monitoring task planning method based on an intelligent agent, according to an embodiment of the present invention. The diagram demonstrates the complete process from starting with the target task chain and task allocation table, identifying the platform type (space-based / air-based / ground-based / mobile / human-based), generating platform-specific execution instructions, selecting the communication link (cellular communication / private network communication / satellite link / wireless link) for distribution, and receiving execution feedback through ACK confirmation.

[0090] See Figure 10 The diagram shown is a schematic diagram of a module of an intelligent agent-based integrated air-ground monitoring task planning system provided in an embodiment of the present invention.

[0091] The agent-based integrated air-ground monitoring task planning system 200 described in this invention can be installed in an electronic device. Depending on the functions implemented, the agent-based integrated air-ground monitoring task planning system includes a planning element determination module 201, a task decomposition module 202, a candidate chain generation module 203, a task chain optimization module 204, and a planning execution module 205. The module described in this invention can also be called a unit, which refers to a series of computer program segments that can be executed by the processor of an electronic device and can perform a fixed function, stored in the memory of the electronic device.

[0092] In this embodiment of the invention, the functions of each module / unit are as follows: The planning element determination module 201 is used to obtain spatiotemporal monitoring data corresponding to the multi-source monitoring platform through an intelligent agent deployed at least one of the monitoring center, edge node, cloud server or monitoring resource terminal, so as to determine the task planning elements between the multi-source monitoring platforms. The multi-source monitoring platform includes one or more of the following: space-based platform, air-based platform, ground-based platform, mobile platform or artificial platform. The task decomposition module 202 is used to parse out the overall monitoring task objective corresponding to the task planning elements, and in response to the task triggering event of the multi-source monitoring platform, the agent decomposes multiple sub-tasks in the overall monitoring task objective and the dependencies between the multiple sub-tasks. The candidate chain generation module 203 is used to establish resource status vectors corresponding to each monitoring platform based on the task planning elements, and to merge the multiple subtasks into at least one candidate task chain according to the dependency relationship, the resource status vectors and preset business rules. The task chain optimization module 204 is used to use the intelligent agent to perform feasibility verification and optimization sorting on the at least one candidate task chain to obtain the target task chain and calculate the collaborative effective coverage rate of the target task chain on the overall monitoring task target. The planning and execution module 205 is used to execute the task planning of the multi-source monitoring platform based on the target task chain when the effective coverage rate of the collaboration reaches a preset threshold.

[0093] In detail, the modules in the agent-based integrated air-ground monitoring task planning system 200 described in this embodiment of the invention employ the same methods as described above. Figure 1 The method used here is the same as the one described above for planning a space-air-ground integrated monitoring mission based on intelligent agents, and can produce the same technical effect. It will not be elaborated here.

[0094] In one embodiment, a computer device is provided, which may be a server or a client, and its internal structure diagram may be as follows: Figure 11 As shown, the computer device includes a processor, memory, network interface, and database connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile and / or volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and database. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The network interface is used for communication with external clients via a network connection. When the computer program is executed by the processor, it implements the functions or steps of a smart agent-based integrated air-ground monitoring mission planning method on the server or client side.

[0095] In one embodiment, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to perform the following steps: S1. By deploying an intelligent agent at least one of the monitoring center, edge node, cloud server, or monitoring resource terminal, the spatiotemporal monitoring data corresponding to the multi-source monitoring platform is obtained to determine the task planning elements between the multi-source monitoring platforms. The multi-source monitoring platform includes one or more of the following: space-based platform, air-based platform, ground-based platform, mobile platform, or artificial platform. S2. Parse the overall monitoring task objective corresponding to the task planning elements, and in response to the task triggering event of the multi-source monitoring platform, the agent decomposes the multiple sub-tasks in the overall monitoring task objective and the dependencies between the multiple sub-tasks. S3. Based on the task planning elements, establish resource status vectors corresponding to each monitoring platform, and merge the multiple subtasks into at least one candidate task chain according to the dependency relationship, the resource status vectors and preset business rules. S4. Use the intelligent agent to perform feasibility verification and optimization sorting on the at least one candidate task chain to obtain the target task chain, and calculate the cooperative effective coverage rate of the target task chain on the overall monitoring task target. S5. When the effective coverage rate of the collaboration reaches a preset threshold, the task planning of the multi-source monitoring platform is executed based on the target task chain.

[0096] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, the computer program performing the following steps when executed by a processor: S1. By deploying an intelligent agent at least one of the monitoring center, edge node, cloud server, or monitoring resource terminal, the spatiotemporal monitoring data corresponding to the multi-source monitoring platform is obtained to determine the task planning elements between the multi-source monitoring platforms. The multi-source monitoring platform includes one or more of the following: space-based platform, air-based platform, ground-based platform, mobile platform, or artificial platform. S2. Parse the overall monitoring task objective corresponding to the task planning elements, and in response to the task triggering event of the multi-source monitoring platform, the agent decomposes the multiple sub-tasks in the overall monitoring task objective and the dependencies between the multiple sub-tasks. S3. Based on the task planning elements, establish resource status vectors corresponding to each monitoring platform, and merge the multiple subtasks into at least one candidate task chain according to the dependency relationship, the resource status vectors and preset business rules. S4. Use the intelligent agent to perform feasibility verification and optimization sorting on the at least one candidate task chain to obtain the target task chain, and calculate the cooperative effective coverage rate of the target task chain on the overall monitoring task target. S5. When the effective coverage rate of the collaboration reaches a preset threshold, the task planning of the multi-source monitoring platform is executed based on the target task chain.

[0097] It should be noted that the functions or steps that can be implemented by the computer-readable storage medium or computer device described above can be referred to the relevant descriptions on the server side and client side in the foregoing method embodiments. To avoid repetition, they will not be described one by one here.

[0098] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Furthermore, any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory.

[0099] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is used as an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the system can be divided into different functional units or modules to complete all or part of the functions described above.

[0100] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0101] Finally, it should be noted that in the above embodiments, each embodiment can be combined with each other or independent. Deleting any one of them will not affect the technical implementation of other embodiments. The above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for planning integrated air-ground monitoring tasks based on intelligent agents, characterized in that, The method includes: By deploying intelligent agents at least one of the monitoring center, edge nodes, cloud servers, or monitoring resource terminals, spatiotemporal monitoring data corresponding to the multi-source monitoring platform is obtained, so as to determine the task planning elements between the multi-source monitoring platforms. The multi-source monitoring platform includes one or more of the following: space-based platform, air-based platform, ground-based platform, mobile platform, or artificial platform. The overall monitoring task objective corresponding to the task planning elements is parsed out. In response to the task triggering event of the multi-source monitoring platform, the agent decomposes multiple sub-tasks in the overall monitoring task objective and the dependencies between the multiple sub-tasks. Based on the task planning elements, a resource status vector corresponding to each monitoring platform is established. According to the dependency relationship, the resource status vector and the preset business rules, the multiple sub-tasks are merged into at least one candidate task chain. The agent is used to perform feasibility verification and optimization ranking of the at least one candidate task chain to obtain the target task chain, and the collaborative effective coverage rate of the target task chain to the overall monitoring task target is calculated. When the effective coverage of the collaboration reaches a preset threshold, the task planning of the multi-source monitoring platform is executed based on the target task chain.

2. The agent-based integrated air-ground monitoring task planning method as described in claim 1, characterized in that, Based on the dependencies, the resource state vector, and preset business rules, the multiple subtasks are merged into at least one candidate task chain, including: Based on the dependencies, determine the execution order constraints among the multiple subtasks; By combining the resource state vector and the execution order constraint, a corresponding target monitoring platform is assigned to each subtask, and an initial allocation scheme is generated. Based on the preset business rules, the initial allocation scheme is processed to resolve conflicts, thereby obtaining the target allocation scheme; According to the target allocation scheme, the multiple subtasks are combined into at least one candidate task chain according to the execution order constraint.

3. The agent-based integrated air-ground monitoring task planning method as described in claim 2, characterized in that, Combining the resource state vector and the execution order constraints, a corresponding target monitoring platform is assigned to each subtask, generating an initial allocation scheme, including: Based on the execution order constraints, determine the resource requirement type and required time window for each subtask; The resource type, available time window, and load rate of each monitoring platform are obtained from the resource status vector. When the resource type matches the resource demand type, the available time window covers the required time window, and the load rate is lower than the preset load rate threshold, the target monitoring platform is selected, and a corresponding sub-task is assigned to the target monitoring platform to generate an initial allocation scheme.

4. The agent-based integrated air-ground monitoring task planning method as described in claim 2, characterized in that, Based on the preset business rules, conflict resolution processing is performed on the initial allocation scheme to obtain the target allocation scheme, including: Conflicting subtasks were detected in the initial allocation scheme; Based on the resource priority rules in the preset business rules, the low-priority subtasks in the conflicting subtasks are selected. The low-priority subtasks are processed by resource reallocation or time window delay to obtain the target allocation scheme. When no other available monitoring platform meets the resource requirement type and the time window, the low-priority subtask is marked as a waiting task, and the task degradation process of the low-priority subtask is triggered.

5. The agent-based integrated air-ground monitoring task planning method as described in claim 1, characterized in that, Determine the task planning elements among the multi-source monitoring platforms, including: The spatiotemporal monitoring data of the multi-source monitoring platform is transformed into a unified spatiotemporal reference to generate unified spatiotemporal reference data. The unified spatiotemporal reference data includes the position coordinates of each monitoring platform in the same coordinate system and the time window in the same time reference. Spatiotemporal correlation processing is performed on the location coordinates and time windows in the unified spatiotemporal reference data to obtain the spatiotemporal overlap relationship between each monitoring platform; Based on the spatiotemporal overlap relationship, a three-dimensional spatiotemporal diagram of the multi-source monitoring platform in the task collaboration process is constructed, and the collaboration tightness between the multi-source monitoring platforms is calculated. Extract the spatiotemporal coupling nodes and resource contention edges in the three-dimensional spatiotemporal graph; Based on the spatiotemporal coupling nodes, the resource competition edges, and the degree of collaboration, the task planning elements among the multi-source monitoring platforms are determined.

6. The agent-based integrated air-ground monitoring task planning method as described in claim 5, characterized in that, The unified spatiotemporal reference conversion of the spatiotemporal monitoring data is achieved based on BeiDou / GNSS timing and positioning information, and is used to convert the monitoring data of space-based platforms, air-based platforms, ground-based platforms, mobile platforms and artificial platforms to the same coordinate system and the same time reference.

7. The agent-based integrated air-ground monitoring task planning method as described in claim 1, characterized in that, Based on the target task chain, the task planning of the multi-source monitoring platform is executed, including: Based on the task execution order in the target task chain, generate the task execution sequence of each monitoring platform of the multi-source monitoring platform; Based on the resource binding relationship in the task execution sequence and the target task chain, sub-tasks and task execution time periods are allocated to each monitoring platform, and a task allocation table is generated; Based on the task allocation table, task execution instructions for each monitoring platform are generated and sent to the corresponding monitoring platform.

8. The agent-based integrated air-ground monitoring task planning method as described in claim 7, characterized in that, Based on the task allocation table, task execution instructions for each monitoring platform are generated, including: Identify the monitoring platform type corresponding to each task in the task allocation table; If the monitoring platform type is a space-based platform, then a task execution instruction containing the monitoring area, start time, data format, and backhaul link is generated; If the monitoring platform type is an airborne platform, then a task execution command containing the flight path, observation angle, duration and sampling frequency is generated; If the monitoring platform type is a ground-based platform or a mobile platform, then a task execution instruction containing the monitoring area, start time, duration and sampling frequency is generated; If the monitoring platform type is a manual platform, then a task execution instruction containing the monitoring area, review requirements, and alarm thresholds will be generated.

9. The agent-based integrated air-ground monitoring task planning method as described in claim 1, characterized in that, The method further includes: When the effective coverage of the collaboration is lower than a preset threshold, the agent triggers the task replanning process of the multi-source monitoring platform. The task replanning process includes adjusting the decomposition granularity of the overall monitoring task objective, re-screening the candidate task chain, or modifying the resource priority of the conflicting sub-tasks.

10. The agent-based integrated air-ground monitoring task planning method as described in claim 1, characterized in that, Calculating the collaborative effective coverage rate of the target task chain to the overall monitoring task objective includes: Extract the key monitoring indicators and indicator achievement thresholds for each sub-task from the overall monitoring task objectives; Obtain the actual measurement values ​​of each task node in the target task chain and the corresponding key monitoring indicators; Based on the actual measured values ​​and the target thresholds, calculate the achievement coefficient for each sub-task; Calculate the coordination effectiveness coefficient between each task node in the target task chain; By combining the compliance coefficient and the collaborative effectiveness coefficient, the collaborative effective coverage rate of the target task chain on the overall monitoring task target is calculated.

11. The agent-based integrated air-ground monitoring task planning method as described in claim 1, characterized in that, In response to the task triggering event of the multi-source monitoring platform, the agent decomposes the overall monitoring task objective into multiple sub-tasks, including: Extract the event type, event level, event location, and scope of influence of the task-triggered events to determine the decomposition granularity of the overall monitoring task objective; Analyze the spatial distribution characteristics and temporal urgency of the overall monitoring task objectives to identify the spatial sub-regions and temporal sub-stages of the overall monitoring task objectives; Based on the decomposition granularity, the spatial sub-regions, and the temporal sub-stages, the overall monitoring task objective is broken down into multiple sub-tasks.

12. The agent-based integrated air-ground monitoring task planning method as described in claim 1, characterized in that, The dependencies between the multiple subtasks include data dependencies, resource dependencies, and timing dependencies. Data dependencies refer to the sequential constraints arising from input data dependencies between subtasks, determined by analyzing the input and output data flow between each subtask; one subtask depends on the data results generated by another subtask before execution. Resource dependencies refer to the competition or mutual exclusion constraints arising from sharing the same monitoring resources between subtasks, determined by analyzing the contention of each subtask for monitoring resources corresponding to space-based, air-based, ground-based, mobile, or artificial platforms; multiple subtasks may have conflicting or shared constraints regarding the same resource. Timing dependencies refer to the execution order constraints arising from time window constraints between subtasks, determined by analyzing the sequential relationship between the available time windows and task durations of each subtask.

13. The agent-based integrated air-ground monitoring task planning method as described in claim 1, characterized in that, Based on the aforementioned task planning elements, a resource status vector corresponding to each monitoring platform is established, including: Based on the available resource list in the task planning elements, extract the resource identifier, current location, and available time window of each monitoring platform to generate static attribute components; Based on the business rules and real-time monitoring data in the task planning elements, the load rate, energy consumption status and task occupancy status of each monitoring platform are extracted synchronously to generate dynamic status components. The static attribute components are combined with the dynamic state components to obtain the resource state vector.

14. The agent-based integrated air-ground monitoring task planning method as described in claim 1, characterized in that, The agent is used to perform feasibility verification and optimization ranking of the at least one candidate task chain, including: The system calls upon the agent's historical case library and extracts the event types, task planning schemes, resource allocation records, and execution feedback results of historical tasks from the historical case library. Based on the event type and the task planning scheme, the feasibility of the candidate task chain is verified, wherein the feasibility verification includes one or more of resource conflict verification, timing conflict verification, or communication reachability verification. Based on the resource allocation record and the execution feedback result, the candidate task chains that have passed the verification are optimized and sorted. The optimized sorting is based on at least two of the following: task response time, resource load balancing degree, or coverage completeness of each candidate task chain.

15. The agent-based integrated air-ground monitoring task planning method as described in claim 1, characterized in that, The intelligent agent includes a task understanding module, a resource state perception module, a task chain planning module, a conflict resolution module, an execution feedback module, and a replanning module. The task chain planning module generates candidate task chains based on a large language model, knowledge graph, or rule engine, according to task triggering events, historical task cases, resource state vectors, and preset business rules. The replanning module updates subsequent planning strategies based on task execution feedback through a memory retrieval module or a reinforcement learning model.

16. A space-air-ground integrated monitoring mission planning system based on intelligent agents, characterized in that, The system is used to execute the agent-based integrated air-ground monitoring mission planning method as described in any one of claims 1-15, the system comprising: The planning element determination module is used to obtain spatiotemporal monitoring data corresponding to the multi-source monitoring platform through an intelligent agent deployed at least one of the monitoring center, edge node, cloud server or monitoring resource terminal, so as to determine the task planning elements between the multi-source monitoring platforms. The multi-source monitoring platform includes one or more of the following: space-based platform, air-based platform, ground-based platform, mobile platform or artificial platform. The task decomposition module is used to parse out the overall monitoring task objective corresponding to the task planning elements, and in response to the task triggering event of the multi-source monitoring platform, the agent decomposes the overall monitoring task objective into multiple sub-tasks and the dependencies between the multiple sub-tasks. The candidate chain generation module is used to establish resource status vectors corresponding to each monitoring platform based on the task planning elements, and to merge the multiple subtasks into at least one candidate task chain according to the dependency relationship, the resource status vectors and preset business rules. The task chain optimization module is used to use the intelligent agent to perform feasibility verification and optimization ranking of the at least one candidate task chain to obtain the target task chain and calculate the collaborative effective coverage rate of the target task chain to the overall monitoring task target. The planning and execution module is used to execute the task planning of the multi-source monitoring platform based on the target task chain when the effective coverage of the collaboration reaches a preset threshold.

17. An electronic device comprising a processor, a memory, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the computer program, implements the method of any one of claims 1-15.

18. A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method of any one of claims 1-15.

Images