Low-altitude inspection processing method and system, and storage medium

Abstract

The application provides a low-altitude inspection processing method and system and a storage medium, wherein the method comprises the following steps: when a task processing device detects a to-be-executed task, the task processing device acquires task search data of the to-be-executed task; the task processing device determines an execution plan corresponding to the to-be-executed task, and determines task execution data of the to-be-executed task based on the execution plan corresponding to the to-be-executed task and the task search data; the task processing device determines a target flight device for executing the to-be-executed task from at least one flight device, and issues a target flight instruction to the target flight device based on the task execution data; the target flight device collects data according to the target flight instruction to obtain target collection data; and the target flight device generates target collection evidence data based on the target collection data, so as to return the target collection evidence data to the task processing device; and the task processing device generates a disposal conclusion report based on the target collection evidence data. The application can conveniently perform inspection processing.

Description

Technical Field

[0001] This invention relates to the field of inspection technology, and in particular to a low-altitude inspection processing method, system and storage medium. Background Technology

[0002] Currently, with the maturity and widespread adoption of flight equipment (such as drones), their application in complex low-altitude scenarios such as urban governance, emergency management, and traffic inspection is becoming increasingly sophisticated. However, these technologies typically require manual configuration of tasks and plans, leading to cumbersome operations. Therefore, a satisfactory solution for convenient inspection and handling remains elusive. Summary of the Invention

[0003] In view of this, embodiments of the present invention provide a low-altitude inspection processing method, system, and storage medium to solve the problems of cumbersome operation in inspection processing (i.e., low-altitude inspection processing) caused by related technologies. In other words, embodiments of the present invention can treat daily inspection tasks and / or alarm event handling tasks as tasks to be executed, and can automatically determine task execution data through the execution plan corresponding to the tasks to be executed. Thus, the target flight equipment is driven to collect data through task execution data, thereby conveniently generating a handling conclusion report of the tasks to be executed. That is, embodiments of the present invention can integrate daily inspection and alarm event handling, and can realize the determination of task execution data of tasks to be executed without relying on manual planning and intervention, thus enabling convenient inspection processing.

[0004] According to one aspect of the present invention, a low-altitude inspection processing method is provided, the low-altitude inspection processing method being applied to a low-altitude inspection processing system, the low-altitude inspection processing system including mission processing equipment and at least one flight device, the method comprising: When the task processing device detects a task to be executed, it acquires the task retrieval data of the task to be executed; wherein, the task to be executed is a daily inspection task or an alarm event handling task. The task processing device determines the execution plan corresponding to the task to be executed, and determines the task execution data of the task to be executed based on the execution plan corresponding to the task to be executed and the task retrieval data; The task processing device determines the target flight device from the at least one flight device to perform the task to be performed, and issues a target flight command to the target flight device based on the task execution data; The target flight device collects data according to the target flight command to obtain target collection data; and generates target collection evidence data based on the target collection data, thereby returning the target collection evidence data to the task processing device; The task processing device generates a disposal conclusion report based on the evidence data collected from the target.

[0005] According to another aspect of the present invention, a low-altitude inspection and processing system is provided, the low-altitude inspection and processing system comprising mission processing equipment and at least one flight device; wherein, The task processing device is used to acquire task retrieval data of the task to be executed when a task to be executed is detected; wherein the task to be executed is a daily inspection task or an alarm event handling task. The task processing device is further configured to determine the execution plan corresponding to the task to be executed, and to determine the task execution data of the task to be executed based on the execution plan corresponding to the task to be executed and the task retrieval data; The task processing device is further configured to determine, from the at least one flight device, a target flight device for performing the task to be performed, and to issue a target flight command to the target flight device based on the task execution data; The target flight device is used to collect data according to the target flight command to obtain target collection data; and to generate target collection evidence data based on the target collection data, thereby returning the target collection evidence data to the task processing device; The task processing device is also used to generate a disposal conclusion report based on the evidence data collected from the target.

[0006] According to another aspect of the present invention, a non-transitory computer-readable storage medium is provided storing computer instructions for causing a computer to perform the methods mentioned above.

[0007] In this embodiment of the invention, the task processing device can acquire task retrieval data for a task to be executed upon detection of such a task. The task to be executed may be a routine inspection task or an alarm event handling task. Then, the task processing device can determine the execution plan corresponding to the task to be executed, and based on the execution plan and the task retrieval data, determine the task execution data. Further, the task processing device can identify a target flight device from at least one flight device to execute the task to be executed, and issue a target flight command to the target flight device based on the task execution data. Correspondingly, the target flight device can collect data according to the target flight command, obtaining target collected data; and generate target collected evidence data based on the target collected data, thereby returning the target collected evidence data to the task processing device. Based on this, the task processing device can generate a handling conclusion report based on the target collected evidence data. As can be seen, the embodiments of the present invention can treat daily inspection tasks and / or alarm event handling tasks as tasks to be executed, and can automatically determine task execution data through the execution plan corresponding to the tasks to be executed. In this way, the target flight equipment is driven to collect data through the task execution data, and a handling conclusion report of the tasks to be executed can be easily generated. In other words, the embodiments of the present invention can integrate daily inspection and alarm event handling, and can determine the task execution data of the tasks to be executed without relying on manual planning and intervention, thus enabling convenient inspection processing. Attached Figure Description

[0008] Further details, features, and advantages of the invention are disclosed in the following description of exemplary embodiments in conjunction with the accompanying drawings, in which: Figure 1 A flowchart illustrating a low-altitude inspection processing method according to an exemplary embodiment of the present invention is shown. Figure 2 A flowchart illustrating another low-altitude inspection processing method according to an exemplary embodiment of the present invention is shown; Figure 3 A flowchart illustrating another low-altitude inspection processing method according to an exemplary embodiment of the present invention is shown; Figure 4 A schematic block diagram of a low-altitude inspection and processing system according to an exemplary embodiment of the present invention is shown. Detailed Implementation

[0009] Embodiments of the present invention will now be described in more detail with reference to the accompanying drawings. While some embodiments of the invention are shown in the drawings, it should be understood that the invention can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided to provide a more thorough and complete understanding of the invention. It should be understood that the accompanying drawings and embodiments are for illustrative purposes only and are not intended to limit the scope of protection of the invention.

[0010] It should be understood that the various steps described in the method embodiments of the present invention may be performed in different orders and / or in parallel. Furthermore, the method embodiments may include additional steps and / or omit the steps shown. The scope of the present invention is not limited in this respect.

[0011] The term "comprising" and its variations as used herein are open-ended, meaning "including but not limited to". The term "based on" means "at least partially based on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". Definitions of other terms will be given in the following description. It should be noted that the concepts of "first", "second", etc., mentioned in this invention are used only to distinguish different devices, modules, or units, and are not intended to limit the order of functions performed by these devices, modules, or units or their interdependencies.

[0012] It should be noted that the terms "a" and "a plurality of" used in this invention are illustrative rather than restrictive. Those skilled in the art should understand that, unless otherwise expressly indicated in the context, they should be understood as "one or more".

[0013] The names of the messages or information exchanged between the multiple devices in the embodiments of the present invention are for illustrative purposes only and are not intended to limit the scope of these messages or information.

[0014] It should be noted that the embodiments of the present invention can provide a low-altitude inspection and processing system, which can be used to execute low-altitude inspection and processing methods. That is, the executing entity of the low-altitude inspection and processing method provided in the embodiments of the present invention can be a low-altitude inspection and processing system. The low-altitude inspection and processing system may include a task processing device and at least one flight device. Optionally, the task processing device may include one or more electronic devices, which may be a terminal (i.e., a client) or a server. Optionally, the terminal mentioned herein may include, but is not limited to, smartphones, tablets, laptops, desktop computers, etc. The server mentioned herein may be an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN (Content Delivery Network), and big data and artificial intelligence platforms, etc. Optionally, a flight device may be a drone, an unmanned airship, etc.; the embodiments of the present invention do not limit this.

[0015] Optionally, the low-altitude inspection processing method provided in this embodiment of the invention can be applied to any inspection processing scenario (i.e., low-altitude inspection processing scenario), and this embodiment of the invention does not limit it. For example, the inspection processing scenario can be an urban governance scenario, an emergency management scenario, a traffic management scenario, etc.

[0016] Based on the above description, this embodiment of the invention proposes a low-altitude inspection processing method. This low-altitude inspection processing method can be executed by the aforementioned low-altitude inspection processing system, that is, the low-altitude inspection processing method can be applied to the low-altitude inspection processing system, which may include mission processing equipment and at least one flight device. For example... Figure 1 As shown, the low-altitude inspection and processing method may include the following steps S101-S105: S101, when the task processing device detects a task to be executed, it obtains the task retrieval data of the task to be executed; wherein, the task to be executed is a daily inspection task or an alarm event handling task.

[0017] Optionally, the number of tasks to be executed can be one or more, and this embodiment of the invention does not limit this; when there are multiple tasks to be executed, the task retrieval data of each task to be executed can be obtained separately to realize the inspection processing for each task to be executed, and so on. For ease of explanation, the following description will use a single task to be executed as an example.

[0018] Optionally, the task to be executed can be any daily inspection task in the daily inspection task list, or it can be an alarm event handling task for any alarm event. This embodiment of the invention does not limit this. Based on this, this embodiment of the invention can connect daily inspections and alarm event handling, thereby breaking down task silos and achieving bidirectional empowerment and collaborative optimization between daily inspections and alarm event handling.

[0019] Optionally, the task processing device can also acquire daily inspection task generation guidance data. This data may include, but is not limited to, at least one of the following: risk indication information, coverage indication information, and inspected object attribute indication information for each preset inspection unit in at least one preset inspection unit. This embodiment of the invention does not limit the scope of this data. Based on the daily inspection task generation guidance data, a daily inspection task list can be generated. Any daily inspection task in the daily inspection task list can be used as a task to be executed when it meets the daily task triggering condition (i.e., the daily task triggering condition of any daily inspection task). Optionally, a preset inspection unit can be any grid (e.g., a 20×20 meter grid area). This embodiment of the invention does not limit the scope of this data. Optionally, the risk indication information for a preset inspection unit may include, but is not limited to, at least one of the following: the probability of risk occurrence under the corresponding preset inspection unit (e.g., the probability of the inspected object occurring), the number of risk occurrences (e.g., the number of times the inspected object appears within a historical time range, which can be any time range), the risk value, and the degree of harm (e.g., severity level). This embodiment of the invention does not limit the scope of this data. Optionally, the coverage indication information of a preset inspection unit may include, but is not limited to, at least one of the following: the coverage of inspection points under the corresponding preset inspection unit (such as the number of inspection points under the corresponding preset inspection unit / the total number of inspection points in all preset inspection units, etc.), the number of inspection blind spots, etc., which are not limited in this embodiment of the present invention. Optionally, the attribute indication information of the inspected object of a preset inspection unit may include, but is not limited to, at least one of the following: the historical change rate of the operating status of the inspected object under the corresponding preset inspection unit (which can be used to indicate the speed of change over time, such as the change between construction and non-construction when the inspected object is under construction), the historical change rate of parameters (such as the change of vehicle speed or vehicle density when the inspected object is under construction), the historical change rate of the environment (such as the change rate of weather conditions, etc.), etc., which are not limited in this embodiment of the present invention. Optionally, the inspected object in an inspection processing scenario can be set according to experience or actual needs (i.e., the inspected object corresponding to a task type can be set according to experience or actual needs, and an inspection processing scenario can correspond to a task type), and this embodiment of the invention does not limit this; for example, in a traffic management scenario, the inspected object can be road construction, etc. Optionally, a preset inspection unit may include, but is not limited to, at least one of the following: at least one inspection point, inspection time period, etc., and this embodiment of the invention does not limit this. Optionally, at least one preset inspection unit (also referred to as an inspection area) can be set according to experience or actual needs, and this embodiment of the invention does not limit this.

[0020] Optionally, the aforementioned daily inspection task generation guidance data can be daily inspection task generation guidance data under the target task type (i.e., the task type corresponding to the target inspection processing scenario), thereby generating a daily inspection task list under the target task type. Optionally, the target task type can be any task type, and this embodiment of the invention does not limit this; Optionally, when the number of task types involved in the low-altitude inspection processing system is multiple, daily inspection task generation guidance data under each of the multiple task types can be obtained separately, thereby generating a daily inspection task list under the corresponding task type based on the daily inspection task generation guidance data under each task type, and so on. For ease of explanation, the following description will use the target task type as an example, that is, the following description will use one task type as an example. Optionally, at least one preset inspection unit corresponding to different task types can be the same or different, and this embodiment of the invention does not limit this.

[0021] Optionally, when generating guidance data based on daily inspection tasks and generating a daily inspection task list, at least one preset inspection unit can be prioritized according to a preset priority sorting rule, resulting in a sorting result (which may include unit identifiers of inspection units from high to low priority, with one unit identifier indicating one inspection unit). A daily inspection task list can then be generated based on the sorting result. Alternatively, based on the guidance data generated from daily inspection tasks, priority scores for each evaluation indicator (including but not limited to risk indication information, coverage indication information, and inspected object attribute indication information, or each data point in an indication information (such as risk indication information) can be used as an evaluation indicator) can be determined for each preset inspection unit. A weighted sum of the priority scores for each evaluation indicator in any preset inspection unit can be obtained to achieve a total priority score for any preset inspection unit. A daily inspection task list can then be generated based on the total priority score of each preset unit, and so on. This embodiment of the invention does not limit this approach.

[0022] Optionally, the preset priority sorting rules can be set according to experience or actual needs, and this embodiment of the invention does not limit this. For example, risk indication information can be used as the core indicator for judging inspection priority (i.e., priority). At least one preset inspection unit can be sorted first according to the risk indication information (e.g., the higher the number of risk occurrences, the higher the priority), and then preset inspection units with the same risk indication information can be sorted according to the coverage indication information (e.g., the higher the coverage of the inspection point, the higher the priority), and so on. Optionally, when generating a daily inspection task list according to the inspection sorting results, the inspection cycle of each preset inspection unit's daily inspection task can be determined according to the order of the inspection unit sorting results from high to low (i.e., from front to back), thus obtaining a daily inspection task list. At this time, the inspection cycle of a preset inspection unit's daily inspection task can be determined based on the position of the corresponding preset inspection unit in the inspection unit sorting results. A preset inspection unit's daily inspection task can then be the daily inspection task of the corresponding preset inspection unit under the target task type. For example, each preset inspection unit can be determined according to the preset inspection cycle determination rules. The inspection cycle for the daily inspection tasks of a unit can be set according to experience or actual needs. For example, the daily inspection cycle of the pre-set inspection units that are in the top 50% of the inspection unit ranking results can be the first inspection cycle, and the daily inspection cycle of the pre-set inspection units that are in the bottom 50% can be the second inspection cycle. Both the first and second inspection cycles can be set according to experience or actual needs. The first inspection cycle can be shorter than the second inspection cycle (e.g., the first inspection cycle can be 1 day (i.e., inspected once a day), and the second inspection cycle can be 1 week (i.e., inspected once a week), etc.). Alternatively, the attribute indication information of the inspected object can also serve as a key basis for determining the inspection frequency. In this case, based on the attribute indication information of the inspected object of each preset inspection unit, the inspection cycle of the daily inspection task of each preset inspection unit can be determined (e.g., the higher the historical change rate, the shorter the inspection cycle). Thus, a daily inspection task list is generated according to the inspection unit sorting results and the inspection cycle of the daily inspection task of each preset inspection unit, etc.; this embodiment of the invention does not limit this. Optionally, the priority of a preset inspection unit can be used as the priority of the daily inspection task of the corresponding preset inspection unit. Therefore, the daily inspection task list may include, but is not limited to, the priority and / or inspection cycle of each daily inspection task in at least one daily inspection task, etc.

[0023] It should be noted that the weights of the various evaluation indicators in this embodiment of the invention are not limited, and can be set according to experience or actual needs. Optionally, when generating a daily inspection task list based on the total priority score of each preset inspection unit, the inspection cycle of the daily inspection task of any preset inspection unit can be determined according to a specified score range in which the total priority score of any preset inspection unit falls (e.g., a specified score range can correspond to a preset inspection cycle). Thus, a daily inspection task list can be generated based on the inspection cycle of the daily inspection tasks of each preset inspection unit and the total priority score of each preset inspection unit. The higher the total priority score of a preset inspection unit, the higher the priority of the daily inspection task of that preset inspection unit. Alternatively, the inspection cycle of the daily inspection task of each preset inspection unit can be determined based on the inspection object attribute indication information of each preset inspection unit, thereby generating a daily inspection task list according to the inspection cycle of the daily inspection tasks of each preset inspection unit and the total priority score of each preset inspection unit, and so on. This embodiment of the invention does not limit this.

[0024] In summary, the embodiments of the present invention do not limit the specific implementation of generating a daily inspection task list based on guidance data generated from daily inspection tasks; based on this, the embodiments of the present invention can automatically calculate the priority of each preset inspection unit based on multi-dimensional information, thereby generating a periodic daily inspection task list (also known as a daily inspection task checklist).

[0025] Optionally, the task processing device can also update the daily inspection task list at preset daily inspection update intervals. For example, it can re-execute the process of obtaining daily inspection task generation guidance data to generate a daily inspection task list, etc. Optionally, the preset daily inspection update interval can be set based on experience or actual needs, and this embodiment of the invention does not limit this. Optionally, the task processing device can update the daily inspection task list according to an improved daily inspection strategy; optionally, the improved daily inspection strategy can be set based on experience or actual needs, and this embodiment of the invention does not limit this. For example, assuming the risk value of the pre-defined inspection unit G001 in the risk layer increases, its inspection priority can be automatically increased by the low-altitude inspection processing system; and / or, the inspection cycle of the grid can be adjusted from "once a week" to "once a day", and the camera script in the inspection plan can be set to a version that pays more attention to "construction behavior" and "road occupation" (such as replacing the camera script with a construction-specific inspection script and / or a road occupation-specific inspection script, both of which can be set according to experience or actual needs), and so on.

[0026] Optionally, the task processing device may include an inspection task generation module. In this case, the task processing device can generate a daily inspection task list through the inspection task generation module. Optionally, the task processing device may also include a spatiotemporal information base module (also referred to as a spatiotemporal information base database, spatiotemporal information base, or base, etc.). In this case, the daily inspection task generation guidance data can be obtained from the spatiotemporal information base module; or, the task processing device may also obtain a download link for the daily inspection task generation guidance data to download the daily inspection task generation guidance data, thereby achieving the acquisition of the daily inspection task generation guidance data, etc.; this embodiment of the invention does not limit this. Based on this, the inspection task generation module can generate a daily inspection task list based on multi-dimensional information such as risk indication information, coverage indication information, and inspected object attribute indication information in the base.

[0027] Optionally, the triggering conditions for a daily inspection task can be set based on experience or actual needs, and this embodiment of the invention does not limit this. For example, the triggering conditions for a daily inspection task may include, but are not limited to, at least one of the following: the current time is within the execution period of the corresponding daily inspection task; there are no unscheduled daily inspection tasks with higher priority than the corresponding daily inspection task (i.e., unscheduled daily inspection tasks) within the execution period of the corresponding daily inspection task, etc., and this embodiment of the invention does not limit this. Optionally, an unscheduled daily inspection task may refer to a daily inspection task for which flight instructions have not been issued based on the corresponding task execution data, or it may refer to a daily inspection task for which a corresponding handling conclusion report has not been generated, etc.; this embodiment of the invention does not limit this.

[0028] In one implementation, the task processing device can designate any daily inspection task in the daily inspection task list as a task to be executed when it detects that the daily task triggering condition has been met, thereby confirming that a task to be executed has been detected. In another implementation, the task processing device can determine the alarm event handling task corresponding to the event alarm information upon receiving the event alarm information, and designate the alarm event handling task corresponding to the event alarm information as a task to be executed, thereby confirming that a task to be executed has been detected, and so on. Optionally, the event alarm information can be received through any channel, and this embodiment of the invention does not limit this. For example, it can be received through a citizen reporting program, such as receiving an event alarm information reported by a citizen that "there is unapproved road construction on the northeast side of the intersection of XX Road and YY Road, causing traffic congestion." Alternatively, it can be received through AI (Artificial Intelligence) monitoring, such as a fixed camera deployed at a key intersection, which, through video AI analysis algorithms (such as object detection algorithms), detects that the traffic density at the intersection of XX Road and YY Road has abnormally increased by 30% within 10 minutes, and identifies construction barriers and engineering vehicles occupying the original non-motorized vehicle lane, initially judging it as a "road construction" event and reporting the event alarm information, etc. Based on this, this embodiment of the invention can not only conduct daily inspections through the low-altitude inspection and processing system, but also monitor the city's operational status in real time, so as to handle alarm events when receiving event alarm information reported through any channel. Optionally, the event alarm information may include, but is not limited to, at least one of the following: alarm event type (such as road construction, vehicle collision, traffic congestion, road damage, etc.), event location (such as grid coordinates or road name), alarm timestamp, current status information of the inspected change target (such as vehicle speed, etc.), etc., and the embodiments of the present invention do not limit this.

[0029] Optionally, when determining the alarm event handling task corresponding to the alarm information, the alarm event type can be used as the task type of the alarm event handling task corresponding to the alarm information; and / or, the event location can be used as the task inspection location of the alarm event handling task corresponding to the alarm information, etc.; this embodiment of the invention does not limit this. Optionally, the task processing device may also include a plan arrangement module (also called a plan library and plan arrangement module), and the alarm information can be received by the task processing device through the plan arrangement module; and / or, the plan arrangement module can send a retrieval request to the spatiotemporal information base module, or the task processing device can also send a retrieval request through other modules, etc.; this embodiment of the invention does not limit this.

[0030] In this embodiment of the invention, the methods for obtaining task retrieval data for the task to be executed may include, but are not limited to, the following: The first method of acquisition: The task processing device can obtain the task retrieval data download link and use the task retrieval data download link to download the task retrieval data of the task to be executed, so as to obtain the task retrieval data of the task to be executed.

[0031] The second method of acquisition: The task processing device may include a spatiotemporal information base module. In this case, the task processing device may send a retrieval request for the task to be executed to the spatiotemporal information base module, so that the spatiotemporal information base module returns the task retrieval data of the task to be executed, thereby realizing the acquisition of the task retrieval data of the task to be executed, and so on.

[0032] Optionally, the spatiotemporal information base module may include, but is not limited to, at least one of the following: a static base map layer, a low-altitude compliance layer, a resource layer, a multi-temporal change layer, and a risk layer, etc., which are not limited in this embodiment of the invention. The static base map layer, also known as the static base layer, may include basic geospatial data, such as vector polygon / line data or raster tiles, and may contain static elements such as roads, water systems, building outlines, and critical infrastructure. Its key technology lies in establishing an efficient spatial index (such as a GeoHash grid or quadtree) to provide spatial anchors for upper-layer data. The low-altitude compliance layer, also known as the low-altitude compliance layer, may include a digital representation of airspace constraint rules; each constraint element includes at least a geographical boundary, an effective time window, a constraint type (such as no-fly zone, altitude restriction, speed limit), and an approval status; this layer is the mandatory basis for generating "compliant flight routes," ensuring flight legality. The resource layer, also known as the resource stack, enables dynamic management of execution resources such as flight equipment, flight equipment parking areas (e.g., airports / nails), and payloads. Each resource element may include, but is not limited to, at least one of the following: the real-time location of each flight equipment, its availability status (e.g., idle, on a mission, under maintenance), capability parameters (e.g., endurance, payload type (e.g., type of mission equipment, type of sensors), communication modules (e.g., communication method, type)). This layer is the core input for resource optimization and scheduling. The multi-temporal change layer, also known as the multi-temporal change stack, serves as a "long-term memory" carrier. It can use "spatial index + timestamp" as the key to record snapshots or changes in the state of the same location at different times. Each record may include, but is not limited to, at least one of the following: state value / change, confidence level, and data source (e.g., mission ID). For example, it can record "grid G001 had no water accumulation at time T1, and the water accumulation area was 30% at time T2." This layer makes it possible to calculate the "degree of change (Δ)" of the mission inspection location. The risk layer, also known as the risk profile, can be a derived layer generated by aggregating data such as historical event frequency and multi-temporal change rate (e.g., number of congestion occurrences / time, construction quantity / time, and whether changes have occurred in the inspected object). It can identify hotspots such as accident-prone areas and high-frequency violations, and can be used to guide the priority calculation of inspection tasks. Optionally, the spatiotemporal information base module can serve as a unified spatiotemporal data and rule hub for the entire mission of low-altitude flight equipment. This module can adopt a layered architecture, with each layer updated independently and linked through a unified spatiotemporal index, effectively avoiding data redundancy and coupling.Based on this, embodiments of the present invention can integrate multi-source geographic and business data by constructing and initializing a spatiotemporal information base module, and construct core layers such as static base map layer, compliance layer, resource layer, and multi-temporal change layer. Accordingly, embodiments of the present invention can use the spatiotemporal information base module as the "long-term memory" and "rule engine" of the low-altitude inspection and processing system, so as to manage static geographic, dynamic compliance, available resources and multi-temporal information in layers through the spatiotemporal information base module, and provide a unified spatiotemporal index interface.

[0033] Optionally, the retrieval request for the task to be executed may carry the task inspection location of the task to be executed or the target spatial range of the task inspection location (such as a circular area within a preset retrieval radius centered on the task inspection location (such as an alarm point)). Optionally, the preset retrieval radius may be set according to experience or actual needs, and this embodiment of the invention does not limit this. Optionally, the task retrieval data may include, but is not limited to, at least one of the following: the target change degree corresponding to the task to be executed, flight equipment resource indication data corresponding to the task inspection location of the task to be executed (which can be used to indicate all available flight equipment resources within the target spatial range of the task inspection location, i.e., can be used to indicate the equipment information of each selected flight equipment in the set of selected flight equipment within the target spatial range of the task inspection location), airspace constraint rules corresponding to the task to be executed (such as airspace constraint rules within the target spatial range of the task inspection location), and compliance approval query results of the task to be executed, etc.; this embodiment of the invention does not limit this. Optionally, one task may correspond to one inspected object.

[0034] In one implementation, the task retrieval data includes the target change degree corresponding to the task to be executed. When acquiring the task retrieval data for the task to be executed, the task processing device can determine the task inspection location of the task to be executed and retrieve the historical baseline status information of the monitored object corresponding to the task at the task inspection location and at least one historical time (e.g., retrieved from a multi-temporal change layer). Based on the historical baseline status information, the target change degree can be determined and added to the task retrieval data. The at least one historical time can include any historical period or any historical point in time; this embodiment of the invention does not limit this. For example, the at least one historical time can include the same period within the past 7 days (e.g., a weekday afternoon). Optionally, the historical baseline status information (also referred to as the historical baseline) can be the historical baseline status information (such as the mean or median of the inspection change target) corresponding to the inspected object (i.e., the inspection change target corresponding to the corresponding task type); optionally, the inspection change target corresponding to a task type can be set according to experience or actual needs, and this embodiment of the invention does not limit this; optionally, the inspection change targets corresponding to different task types can be the same or different, and this embodiment of the invention does not limit this; for example, the inspection change target corresponding to the task type "road construction" can be vehicle speed, or it can be the construction status of road construction (such as under construction or not under construction, such as the historical baseline status information can be not under construction, etc.), and so on. Optionally, when determining the degree of target change based on historical baseline status information, the degree of target change can be determined based on the current status information of the inspected changed target (such as the current vehicle speed or the current construction status (such as 1 for construction, 0 for no construction, etc.)) and historical baseline status information. For example, the degree of target change can be: current status information - historical baseline status information, or the degree of target change can be: (current status information - historical baseline status information) / max(current status information, historical baseline status information), etc. This embodiment of the invention does not limit this. Optionally, if the task to be performed is an alarm event handling task, the current status information can be determined from the event alarm information; if the task to be performed is a daily inspection task, a command to obtain the status information of the inspected changed target can be sent to the monitoring equipment (such as a camera) at the task inspection location to obtain the current status information, etc. Optionally, the number of task inspection locations for the task to be executed can be one or more, and this embodiment of the invention does not limit this; Optionally, when the number of task inspection locations for the task to be executed is multiple, the degree of target change of the inspected change target at each of the multiple task inspection locations can be determined separately (at this time, the data collection method at different task inspection locations can be different), or the mean or median of the degree of target change of the inspected change target at each task inspection location can be used as the degree of target change, etc.; This embodiment of the invention does not limit this.

[0035] For example, the "traffic flow" status snapshots of the inspection location for the task within the past 7 days (such as a weekday afternoon) can be retrieved from the multi-temporal change layer to calculate the target change degree. For example, if the historical baseline shows that the average traffic speed at the intersection is 40 km / h (meaning the target change at this time can be vehicle speed), and the current status information (i.e., the current traffic speed) has dropped to 5 km / h, then the target change degree can be (5-40) / 40=-87.5%, indicating a significant abnormality, and so on.

[0036] In another implementation, when the task to be performed is an alarm event handling task, the task retrieval data may include the target change degree corresponding to the task to be performed; when the task to be performed is a daily inspection task, the task retrieval data may not include the target change degree, and in this case, no adjustment to the execution plan is required, and so on.

[0037] In another embodiment, the task processing device can also retrieve flight equipment resource indication data corresponding to the task inspection location and add the flight equipment resource indication data to the task retrieval data. For example, taking a drone as an example, it can retrieve all drone nests / airports within a radius of 3 kilometers (i.e., the target space range) centered on the task inspection location from the resource layer. For example, the query result may be: the nearest "XX Street nest" has a drone equipped with a visible light / thermal imaging dual-light pod that is in an "idle" state, with a flight endurance of 25 minutes and an estimated flight arrival time of 3 minutes.

[0038] In another embodiment, the airspace constraint rules corresponding to the task to be executed and / or the compliance approval query results of the task to be executed can also be retrieved, and the airspace constraint rules (such as identifying no-fly zones) and / or compliance approval query results can be added to the task retrieval data, etc.; for example, the compliance layer can be used to search for whether there is an approval record for "road occupation construction" within the target space range at the current timestamp. If the query result is "no valid approval record", it can be preliminarily judged as "illegal construction", etc. Optionally, when the task to be executed is a daily inspection task, the compliance approval query results may not be included in the task retrieval data, etc.; the embodiments of the present invention do not limit this.

[0039] As can be seen, this invention systematically constructs, for the first time, a low-altitude spatiotemporal information base that includes static base maps, compliance constraints, resource status, and multi-temporal change layers, forming a "long-term memory" that supports rapid retrieval and analysis. This provides a unified and computable source of data and constraints for daily inspections and alarm event handling, solving the core pain points of task silos and inconsistent data definitions. In other words, this invention can construct a layered spatiotemporal information base module with spatiotemporal indexes, uniformly managing geographical, compliance, resource, and historical change data, providing a unique and reliable source of data and rules for all tasks, thereby achieving an integrated data core.

[0040] S102, the task processing device determines the execution plan corresponding to the task to be executed, and determines the task execution data of the task to be executed based on the execution plan corresponding to the task to be executed and the task retrieval data.

[0041] Optionally, when the task processing equipment includes a playbook and a playbook arrangement module, the task processing equipment can determine the execution playbook corresponding to the task to be executed, and determine the task execution data, etc., through the playbook and playbook arrangement module. The playbook and playbook arrangement module can serve as the "intelligent scheduling hub" of the low-altitude inspection processing system, storing reusable task templates and automatically generating compliant flight paths, camera scripts, and data acquisition strategies based on specific task scenarios and real-time base information; that is, the playbook and playbook arrangement module can be the "intelligent operation manual" and "real-time scheduling engine" of the low-altitude inspection processing system. Optionally, a playbook (i.e., an execution playbook) can be a structured task template that solidifies best practices for a certain type of scenario. Optionally, the playbook can include at least one execution playbook. Optionally, the contingency plan library may include a set of execution contingency plans for each task type within at least one task type. The number of execution contingency plans in a set of execution contingency plans for a task type may be one or more, and this embodiment of the invention does not limit this. For example, for any task type, one preset inspection unit may correspond to one execution contingency plan (i.e., the set of execution contingency plans for any task type may include the execution contingency plans of each preset inspection unit for any task type, so a daily inspection task may correspond to one execution contingency plan, etc., i.e., the set of execution contingency plans for any task type may include the execution contingency plan corresponding to each daily inspection task for any task type), or each preset inspection unit may correspond to the same execution contingency plan (i.e., each daily inspection task for any task type may correspond to the same execution contingency plan, that is, each daily inspection task for the same task type may correspond to the same execution contingency plan, i.e., one task type may correspond to one execution contingency plan), and so on; this embodiment of the invention does not limit this. Optionally, the contingency plan library may be set according to experience or actual needs, and this embodiment of the invention does not limit this.

[0042] Optionally, an execution plan may include, but is not limited to, at least one of the following: plan triggering conditions (also known as plan matching indicators, such as task type identifiers or inspection cycles, etc.), flight path templates (such as, but not limited to, at least one of the following: flight path inspection information (such as waypoint sequences like "high-point hovering at intersections - low-altitude inspection along roads - circling the construction core area," etc., or when a preset inspection unit corresponds to an execution plan under a task type, it may include the waypoints involved in the corresponding preset inspection unit, etc.), planning parameters (cruising altitude, cruising speed, etc.), camera scripts (including but not limited to, at least one of the following: standardized gimbal and flight action sequences (such as first panoramic hovering, then corridor following, and finally circling the point of interest), triggering conditions for each action (such as reaching the inspection range of the task inspection location, detecting points of interest (such as road construction, illegal vehicles, etc.) etc.) and timeout rules (such as time limits for detecting the inspected object, etc.), data extraction strategies (also known as SoI (Segment of Interest (SoI) extraction strategies may include, but are not limited to, at least one of the following: fixed sampling frequency (e.g., 1 frame per second) and abnormal triggering conditions (e.g., a surge in inspected targets within the frame), data extraction threshold (also known as SoI extraction threshold), judgment list (which may include a list of elements that must be verified, such as confirming whether the construction has a permit), output format (e.g., field definitions for structured conclusions), KPI indicators (Key Performance Indicators, which can be used for subsequent improvement assessments, such as indicating the expected on-site duration), resource scheduling weights (e.g., endurance weights, distance weights, etc.); this embodiment of the invention does not limit these. The task type identifier (e.g., task type code or task type name, etc.) can be used to indicate the task type; for example, a task type may be any of the following: road construction, vehicle collision, traffic congestion, and road damage, etc., this embodiment of the invention does not limit these.

[0043] Optionally, when determining the execution plan corresponding to the task to be executed, the task type of the task to be executed can be determined, and the execution plan corresponding to the task type of the task to be executed can be determined from the plan library (i.e., the execution plan whose task type is indicated by the task type identifier in the plan triggering condition is the task type of the task to be executed; at this time, one task type can correspond to one execution plan). Then, the execution plan corresponding to the task type of the task to be executed can be used as the execution plan corresponding to the task to be executed, thereby realizing the selection of the optimal execution plan according to the task type to achieve plan matching; or, when the set of execution plans under a task type may include the execution plans of each preset inspection unit under the corresponding task type, the target preset inspection unit where the task inspection position of the task to be executed is located can be determined, and the execution plan of the target preset inspection unit under the task type of the task to be executed can be determined from the plan library, so as to use the execution plan of the target preset inspection unit under the task type of the task to be executed as the execution plan corresponding to the task to be executed, etc.; the embodiments of the present invention do not limit this. For example, let's take matching contingency plans based on task type as an example. When the task type to be executed is road construction, the corresponding "road construction execution plan" can be loaded from the contingency plan library (also known as the contingency plan arrangement library), and so on.

[0044] Optionally, the task processing device can also load the execution plan corresponding to the task to be executed. For example, taking the task type of road construction to be executed as an example, the flight path template can be: a preset waypoint sequence of "high-point hovering at the intersection - low-altitude inspection along the road - circling around the core construction area"; the camera script can be: [Action 1: high-altitude panoramic view, pitch angle -30°, lasting 30 seconds] - [Action 2: low-altitude inspection, gimbal tracking the boundary of the construction fence, speed 5m / s] - [Action 3: core circling, gimbal focusing on the license plate / construction permit of the engineering vehicle, radius 20m]; the SoI extraction strategy can be: [fixed sampling frequency: extract 1 frame every 5 seconds] + [abnormal trigger: when the AI ​​recognizes "no license plate" or "personnel not wearing safety helmets", automatically extract 15 seconds of video before and after]; the judgment list can be: [(1) verify whether the construction has a license plate], [(2) estimate the road area], [(3) record fence / vehicle characteristics], etc.

[0045] Based on this, the task processing equipment can determine the task execution data (also known as the task execution package or task package) of the task to be executed based on the execution plan and task retrieval data corresponding to the task to be executed. This enables knowledge-based deployment, meaning that before the flight equipment takes off, the low-altitude inspection and processing system has already grasped the historical background, compliance status, and available resources of the task to be executed, providing key inputs for subsequent adaptive task orchestration, and so on. Furthermore, embodiments of the present invention can combine task retrieval data to dynamically and adaptively orchestrate the execution plan, thereby generating the final task execution data.

[0046] Optionally, the task retrieval data may include the degree of change of the target corresponding to the task to be executed. Then, when determining the task execution data of the task to be executed based on the execution plan corresponding to the task to be executed and the task retrieval data, the parameters of the execution plan corresponding to the task to be executed may be adjusted based on the degree of change of the target to obtain the task execution adjustment plan for the task to be executed; and the task execution data of the task to be executed may be generated based on the task execution adjustment plan.

[0047] In one implementation, when adjusting the parameters of the execution plan corresponding to the task to be executed based on the degree of target change to obtain the task execution adjustment plan, the degree of target change can be used to adjust the inspection parameters of the execution plan corresponding to the task to be executed, resulting in the execution plan after the inspection parameters are adjusted. This adjusted execution plan is then used as the task execution adjustment plan for the task to be executed. Optionally, when adjusting the inspection parameters of the execution plan corresponding to the task to be executed based on the degree of target change to obtain the execution plan after the inspection parameters are adjusted, the degree of target change can be used to determine the target risk level (e.g., severe (i.e., high risk), medium risk, low risk, etc., where one risk level corresponds to a range of change, and the risk level corresponding to the range of change in the target change can be used as the target risk level). The inspection parameters (e.g., shot script, SoI extraction threshold, etc.) of the execution plan corresponding to the task to be executed can be adjusted according to a preset inspection parameter adjustment strategy and the target risk level to obtain the execution plan after the inspection parameters are adjusted. Optionally, the preset inspection parameter adjustment strategy can be set according to experience or actual needs, and this embodiment of the invention does not limit this.

[0048] For example, when the target change rate is -87.5% (severe congestion), the severity of the event can be judged to be high, and the contingency plan needs to be strengthened. This can be achieved by adjusting the sequence of shots, such as postponing the "high-altitude panoramic" shot (which has a lower priority in the original shot script) and advancing the "core surround" action (used for detailed evidence collection) to the second position, to ensure that key evidence (such as license plates, permits, etc.) is obtained in the shortest possible time. And / or, the SoI extraction threshold can be adjusted to improve the sensitivity of abnormal triggering, such as lowering the confidence threshold for AI to identify "no permit" from 0.85 to 0.7, in order to more actively capture potential violation evidence, thereby achieving the adjustment of inspection parameters, and so on.

[0049] Optionally, the task processing device can also update the waypoint sequence included in the route template of the execution plan after the inspection parameters are adjusted, according to the task inspection location of the task to be executed, so as to realize the inspection of the task inspection location; in other words, the inspection parameter adjustment process can also include route template adjustment to update the waypoint sequence to the waypoint sequence corresponding to the task inspection location, etc. For example, when the task to be executed is an alarm event handling task, or when different preset inspection units correspond to the same execution plan under a task type, the waypoint sequence can be updated, etc. Optionally, the number of task inspection locations of the task to be executed can be one or more, and this embodiment of the invention does not limit this; for example, when the task to be executed is an alarm event handling task, the number of task inspection locations can be one; when the task to be executed is a daily inspection task, the task inspection location of the task to be executed can include all inspection points in the corresponding preset inspection unit. The inspection points in a preset inspection unit can be set according to experience or actual needs, etc. For ease of explanation, the following will use one task inspection location as an example.

[0050] In another embodiment, the task retrieval data may further include flight equipment resource indication data corresponding to the task inspection location. The flight equipment resource indication data supports equipment information for each candidate flight device in the set of candidate flight devices within the target space range of the task inspection location. Optionally, the equipment information of a candidate flight device may include, but is not limited to, at least one of the following: the corresponding candidate flight device's equipment location, endurance (i.e., endurance capability, also known as endurance time), etc., which are not limited in this embodiment of the invention; and / or, a candidate flight device may be a flight device whose available state is idle, etc. Optionally, when adjusting the parameters of the execution plan corresponding to the task to be executed based on the degree of target change to obtain the task execution adjustment scheme, the degree of target change can also be used to adjust the inspection parameters of the execution plan corresponding to the task to be executed to obtain the execution plan after the inspection parameters are adjusted; and based on the flight equipment resource indication data, the resource scheduling weight of the execution plan after the inspection parameters are adjusted to obtain the task execution adjustment scheme, etc. Optionally, when adjusting the resource scheduling weights of the execution plan after the inspection parameters are adjusted based on the flight equipment resource indication data to obtain the task execution adjustment plan for the task to be executed, the candidate flight equipment closest to the task inspection position of the task to be executed can be determined from the flight equipment resource indication data, and it can be determined whether the range of the determined candidate flight equipment supports the completion of the execution plan after the inspection parameters are adjusted; if it does not support the completion of the execution plan after the inspection parameters are adjusted, the resource scheduling weights of the execution plan after the inspection parameters are adjusted according to the preset resource scheduling weight adjustment strategy to obtain the task execution adjustment plan for the task to be executed; if it supports the completion of the execution plan after the inspection parameters are adjusted, the execution plan after the inspection parameters are adjusted is used as the task execution adjustment plan for the task to be executed, that is, at this time the resource scheduling weight adjustment process may not perform any adjustment operation. Optionally, the preset resource scheduling weight adjustment strategy can be set according to experience or actual needs, and this embodiment of the invention does not limit this. For example, when the endurance is insufficient to complete the execution plan after the inspection parameter adjustment (i.e., insufficient to complete the shot script of the execution plan after the inspection parameter adjustment and return to home, etc.), the preset resource scheduling weight adjustment strategy can be used to indicate an increase in endurance weight, such as using the first weight as the endurance weight and the second weight as the distance weight (i.e., the weight used to measure the distance between the flight equipment and the mission inspection position), etc., which can prompt the subsequent selection of flight equipment that is slightly farther away but has a longer endurance, etc. Correspondingly, if the current resources meet the requirements (i.e., support the completion of the execution plan after the inspection parameter adjustment), the resource scheduling weight may not be adjusted, such as still using the nearest resource, i.e., the flight equipment to be selected that is closest to the mission inspection position to be executed, etc.Optionally, when determining whether the range of the selected flight equipment supports the execution plan after the inspection parameters are adjusted, the required flight time for completing the execution plan can be determined (i.e., the flight time required to fly and collect data according to the execution plan after the inspection parameters are adjusted until return). If the range of the selected flight equipment is greater than or equal to the required flight time, it can be determined that it supports the execution plan after the inspection parameters are adjusted; if the range of the selected flight equipment is less than the required flight time, it can be determined that it does not support the execution plan after the inspection parameters are adjusted, and so on. Optionally, the required flight time can be the flight time required to complete the execution plan after the inspection parameters are adjusted when flying around a no-fly zone, etc.

[0051] Optionally, when generating task execution data for a task to be executed based on a task execution adjustment scheme, a compliant flight path file can be generated based on the task execution adjustment scheme and added to the task execution data of the task to be executed; and / or, a camera script instruction set (which may include the sequence of camera actions, triggering conditions, etc., such as a camera script in the task execution adjustment scheme) can be generated based on the task execution adjustment scheme and added to the task execution data; and / or, data acquisition parameters (which may include, but are not limited to, at least one of the following: target sampling frequency, SoI extraction threshold, image size, resource scheduling weight, target flight equipment identifier, etc.) can be generated and added to the task execution data; and / or, flight equipment resource indication data can also be added to the task execution data, etc.; the embodiments of the present invention do not limit this. In other words, task execution data may include, but is not limited to, at least one of the following: compliant flight path file, camera script instruction set, data acquisition parameters, and flight equipment resource indication data, etc. Among them, a task identifier for a task can be used to indicate the corresponding task. Optionally, when the task retrieval data also includes airspace constraint rules, the compliant route file can be determined based on the task execution adjustment plan and the airspace constraint rules in the task retrieval data. In this case, the compliant route file may include a waypoint sequence that automatically avoids no-fly zones indicated by the space constraint rules. For example, the compliant route file may include, but is not limited to, at least one of the following: a waypoint sequence that avoids no-fly zones (which may point to the task inspection location, or may be called a bypass waypoint sequence), flight time (such as including start timestamp and / or end timestamp), task identifier of the task to be executed, route ID, etc. The embodiments of the present invention do not limit this.

[0052] Optionally, when the task retrieval data also includes compliance approval query results, these results can also be added to the task execution data to mark the task to be executed as a violation verification task. For example, the compliance approval query results can be added to the task execution data in the form of metadata, etc.; this embodiment of the invention does not limit this. For example, this embodiment of the invention can combine the dynamically adjusted flight path with the "no approval record" information obtained from the compliance layer to obtain task execution data, etc.

[0053] As can be seen, the embodiments of the present invention can calculate the degree of target change of the task inspection position relative to the historical baseline based on the multi-temporal change layer, and dynamically adjust the lens script order and SoI extraction threshold of the flight equipment inspection accordingly, realizing intelligent adaptation of "where to look and how detailed to look", which significantly improves the pertinence and efficiency of event interpretation, and realizes adaptive inspection driven by multi-temporal change.

[0054] S103, the mission processing device determines the target flight device from at least one flight device to perform the mission to be performed, and issues the target flight command to the target flight device based on the mission execution data.

[0055] Optionally, the mission processing device may also include a mission execution control module. In this case, the pre-plan library and pre-plan arrangement module can send mission execution data to the mission execution control module. Based on this, the mission processing device can use the mission execution control module to determine the target flight device for executing the mission from at least one flight device, and send target flight commands to the target flight device based on the mission execution data, and so on.

[0056] Optionally, the target flight device can be determined based on resource scheduling weights and flight device resource indication data. For example, when the distance weight is 1, the target flight device can be the idle flight device closest to the task inspection location indicated by the flight device resource indication data among at least one flight device. Alternatively, based on the distance weight and endurance weight, a weighted sum of the distance score and endurance score of each candidate flight device indicated by the flight device resource indication data can be calculated, and the flight device with the largest weighted sum score can be used as the target flight device, etc. This embodiment of the invention does not limit this. Optionally, the distance score corresponding to a distance and the endurance score corresponding to an endurance can be determined according to a preset score rule. Optionally, the preset score rule can be set according to experience or actual needs, and this embodiment of the invention does not limit this. At least one flight device may include each candidate flight device indicated by the flight device resource indication data.

[0057] Optionally, in other embodiments, the target flight device can also be determined by a plan library and a plan arrangement module. In this case, the task execution data may include a target flight device identifier (which can be used to indicate the target flight device). Then, the target flight device for performing the task to be performed can be determined from at least one flight device based on the target flight device identifier, and so on.

[0058] Optionally, when issuing target flight commands to the target flight equipment based on task execution data, task acquisition indication data can be determined based on the task execution data, and then the target flight commands can be sent to the target flight equipment using the task acquisition indication data, so that the target flight commands carry the task acquisition indication data. Optionally, the task acquisition indication data may include, but is not limited to, at least one of the following: target compliant flight path files (which may be determined from compliant flight path files, such as, but not limited to, at least one of the following: waypoint sequence for avoiding no-fly zones, task identifier for the task to be executed, flight path ID, etc.), target camera script (such as camera script instruction set), target generated data acquisition parameters (which may be determined from generated data acquisition parameters, such as, but not limited to, at least one of the following: target sampling frequency, SoI extraction threshold, image size, etc.), and compliance approval query results, etc., which are not limited in this embodiment of the present invention.

[0059] S104, the target flight equipment collects data according to the target flight command, obtains target acquisition data, and generates target acquisition evidence data based on the target acquisition data, and then returns the target acquisition evidence data to the mission processing equipment.

[0060] Optionally, a flight device may include a data extraction and evidence packet generation module (also known as a SoI extraction and evidence packet generation module). In this case, the target flight device can use the SoI extraction and evidence packet generation module to collect data according to the target flight instructions, thereby obtaining the target collected data, and so on. Among them, the SoI extraction and evidence packet generation module can extract the "essence of information" from massive video streams, effectively solving the bottleneck of "fast arrival but slow interpretation".

[0061] S105, the task processing equipment generates a disposal conclusion report based on the evidence data collected from the target.

[0062] Optionally, the mission processing equipment may also include a fusion interpretation and structured output module. In this case, the mission processing equipment can generate a disposal conclusion report, etc., based on the target acquisition evidence data, through the fusion interpretation and structured output module. The fusion interpretation and structured output module can receive the SoI evidence package (i.e., target acquisition evidence data, also simply referred to as the evidence package) transmitted back by the flight equipment and, through correlation with the data in the base station, produce in-depth analysis conclusions.

[0063] Optionally, the execution plan corresponding to the task to be executed can be one of at least one execution plan. Optionally, for any execution plan among the at least one execution plan, the task processing device can also determine the plan evaluation index of any execution plan; and based on the plan evaluation index of any execution plan, it can determine whether any execution plan has a plan optimization requirement; if any execution plan has the plan optimization requirement, then the execution plan is optimized to update the execution plan. Optionally, the plan evaluation index of any execution plan may include, but is not limited to, at least one of the following: KPI indicators and closed-loop indicators of any execution plan, etc., which are not limited in this embodiment of the present invention. Optionally, the closed-loop indicators for any execution plan may include, but are not limited to, at least one of the following: the probability of achieving the target completion time (e.g., the probability that the task completion time within the preset evaluation period is less than the task completion time threshold, such as the number of times any execution plan with a task completion time less than the task completion time threshold is used / the total number of times any execution plan is used) and the probability of achieving the KPI (i.e., the probability of achieving the KPI, such as the number of times any execution plan's KPI is achieved within the preset evaluation period / the total number of times any execution plan is used), etc.; this embodiment of the invention does not limit this. Optionally, the preset evaluation period (e.g., the past 7 days) and the task completion time threshold can be set according to experience or actual needs, and this embodiment of the invention does not limit this.

[0064] Optionally, the task processing equipment may trigger the execution of determining the evaluation index of any execution plan when any execution plan is used and target collection evidence data under any execution plan is received from the flight equipment; and / or, it may trigger the execution of determining the evaluation index of any execution plan after the disposal conclusion report under any execution plan is generated; and / or, it may trigger the execution of determining the evaluation index of any execution plan at preset optimization intervals, etc.; the embodiments of the present invention do not limit this. Optionally, the preset optimization interval can be set according to experience or actual needs, and the embodiments of the present invention do not limit this.

[0065] In one implementation, the evaluation index of any execution plan includes the KPI index of any execution plan; optionally, the KPI index of any execution plan may include the KPI index value of any execution plan in the last task (such as the time to arrive, such as the time to reach the inspection range (i.e., the task area of ​​the task to be executed) where the task inspection location is located), or include the KPI index value of any execution plan in each of the M tasks closest to the current time. Based on this, when determining whether any execution plan requires optimization based on its evaluation indicators, if any execution plan's KPIs are met (i.e., KPIs are achieved, such as arrival time being less than or equal to the expected arrival time, or the average of M arrival times being less than or equal to the expected arrival time, or one of the M arrival times being less than or equal to the expected arrival time, etc.), then it can be determined that any execution plan does not require optimization. Conversely, if any execution plan fails to meet its KPIs (e.g., arrival time exceeding the expected arrival time, or the average of M arrival times exceeding the expected arrival time, or all M arrival times exceeding the expected arrival time, etc.), then it can be determined that any execution plan requires optimization. Here, M is a positive integer. For example, taking the case where the arrival times of M drones all exceed the expected arrival times as an example to illustrate the need for plan optimization, the low-altitude inspection and processing system can periodically calculate the KPIs of each execution plan, such as "arrival time from event triggering to drone approach." If the KPI of a certain plan consistently fails to meet the target (e.g., continuous arrival timeouts), it can be determined that there is a need for plan optimization. The system can then automatically adjust the parameters of the execution plan through refinement algorithms (such as Bayesian optimization or reinforcement learning), such as increasing the distance weight in the resource scheduling formula or increasing the cruising speed, thereby achieving continuous optimization of the execution plan strategy and making the low-altitude inspection and processing system increasingly "intelligent" over time. Optionally, the expected arrival time of an execution plan can be set based on experience or actual needs; this embodiment of the invention does not limit this.

[0066] In another implementation, the evaluation index of any execution plan includes the closed-loop index of any execution plan. Based on this, when determining whether any execution plan has optimization needs based on the evaluation index of any execution plan, if the closed-loop index of any execution plan is less than the closed-loop index threshold (e.g., the probability of achieving the target in the completion time is less than the probability threshold and / or the probability of achieving the KPI indicator is less than the probability threshold), then it can be determined that any execution plan has optimization needs. If the closed-loop index of any execution plan is greater than or equal to the closed-loop index threshold (e.g., each indicator in the closed-loop index is greater than or equal to the corresponding indicator threshold, such as the probability of achieving the target in the completion time is greater than or equal to the probability threshold, and the probability of achieving the KPI indicator is greater than or equal to the probability threshold), then it can be determined that any execution plan does not have optimization needs, and so on. Optionally, the probability threshold for achieving the target in the completion time and the probability threshold for achieving the KPI indicator can be set according to experience or actual needs, and this embodiment of the invention does not limit this.

[0067] Optionally, when optimizing any execution plan, the parameters to be optimized (also called plan parameters) in any execution plan can be optimized using a refinement algorithm; or, the parameters to be optimized in any execution plan can be optimized according to a preset parameter optimization strategy. Optionally, the parameters to be optimized may include, but are not limited to, at least one of the following: SoI extraction threshold, action order in the shot script, resource scheduling weight, etc.; this embodiment of the invention does not limit this. Optionally, the refinement algorithm may be a Bayesian optimization algorithm or a reinforcement learning algorithm, etc.; this embodiment of the invention does not limit this. Optionally, the preset parameter optimization strategy may be set according to experience or actual needs; this embodiment of the invention does not limit this. Based on this, this embodiment of the invention can effectively improve the closure rate of subsequent tasks, such as increasing the probability that the task completion time is less than the task completion time threshold.

[0068] Optionally, the task processing equipment may also include a write-back and refinement module. In this case, the task processing equipment can use the write-back and refinement module to determine the evaluation index of any execution plan for any of the at least one execution plan, and so on. The write-back and refinement module can achieve closed-loop and automatic optimization of the low-altitude inspection processing system, and so on.

[0069] Optionally, the task processing device can also use the aforementioned disposal conclusion report to generate a base write-back record, thereby writing the base write-back record back to the multi-temporal change layer. For example, the disposal conclusion report can be used as the base write-back record, or the base write-back record can be generated using the disposal conclusion report according to a preset write-back format, etc. This embodiment of the invention does not limit this. Optionally, the preset write-back format can be set according to experience or actual needs, and this embodiment of the invention does not limit this. Optionally, when writing the base write-back record back to the multi-temporal change layer, the index of the base write-back record can be determined so that the base write-back record is written back to the multi-temporal change layer according to the index. For example, the index of the base write-back record may include, but is not limited to, a spatial index (such as the area identifier of the grid area where the task inspection location is located (such as grid G001)) and data recording time (such as video shooting time or current timestamp), etc. For example, the spatial index + data recording time can be used as the key to write the base write-back record back to the multi-temporal change layer, etc. For example, taking road occupation construction as an example, when the handling conclusion report indicates the event type as illegal road occupation construction, the base write-back record can be: {"Status": "Illegal Road Occupation Construction", "Confidence Level": 0.98, "Data Source": "EVT_20240401_001"}, thus becoming the future historical baseline for that location; where EVT can represent Engineering Verification Test, engineering verification test data, and the content following EVT can respectively represent the data recording time and spatial index, etc. Optionally, the event type in the handling conclusion report can be used to indicate the inspection and handling conclusion of the task to be performed, such as illegal road occupation construction or non-illegal road occupation construction, etc.

[0070] Optionally, when the handling conclusion report indicates an increase in risk in the area where the task inspection location is located (e.g., a grid area, i.e., the inspection area itself) (e.g., the event type is determined to be illegal road construction, or the risk level is high risk), the risk value of the target event type in the area where the task inspection location is located can be increased according to a preset risk increment. The target event type can be the event type indicated in the handling conclusion report (i.e., the event type in the handling conclusion report). And / or, when the handling conclusion report indicates a decrease in risk in the area where the task inspection location is located (e.g., the event type is determined to be non-illegal road construction, or the risk level is low risk), the risk value of the target event type in the area where the task inspection location is located can be decreased according to a preset risk reduction amount, thereby updating the risk layer. In other words, this embodiment of the invention can update the risk value of the corresponding area (i.e., the area where the task inspection location is located) in the risk layer according to the event type in the handling conclusion report. Optionally, both the preset risk increment and the preset risk reduction amount can be set according to experience or actual needs, and this embodiment of the invention does not limit this. For example, both the preset risk increment and the preset risk reduction amount can be 10, etc. Based on this, the risk priority of this area can be changed when the daily inspection task list is generated in the future, and so on.

[0071] Optionally, when the handling conclusion report indicates the existence of a new obstruction (such as at least one of the following: temporary high-risk sources (such as gas leaks, temporary obstacles, etc.), the task processing equipment may also generate a temporary no-fly rule with an effective time window (such as the next 24 hours) and write the temporary no-fly rule back to the compliance layer; optionally, the temporary no-fly rule may also include the no-fly zone corresponding to the new obstruction, such as the no-fly zone may include the area where the new obstruction is located, etc.

[0072] For example, the content written back for this task to be executed can be captured by the inspection task generation module. Therefore, in the next round of daily inspection task planning (i.e., after a preset daily inspection update interval, such as the next day), the inspection task generation module will recalculate the priority of all preset inspection units, and so on. Based on this, the embodiments of the present invention can complete a complete, traceable, and self-optimizing closed loop of "inspection accumulation - disposal and utilization - result writing back - strategy refinement" from a specific task execution to the final inspection strategy adjustment. Correspondingly, the low-altitude inspection processing system, through this task execution, not only solves immediate problems but also optimizes its long-term operating strategy, demonstrating the core value of the embodiments of the present invention.

[0073] In summary, the embodiments of the present invention provide an integrated closed-loop method and system for daily inspection and alarm event handling of low-altitude flight equipment, effectively addressing the core issues of insufficient systemicity, intelligence, and sustainability in existing technologies. Specifically, the embodiments of the present invention can introduce structured contingency plans as the core process-driven middleware, solidifying the automated link of "task generation - base retrieval - contingency plan arrangement - compliance verification - camera script - effective information extraction - fusion interpretation - write-back update - contingency plan optimization," breaking down task silos and enabling bidirectional empowerment and collaborative optimization of daily inspection and alarm event handling, thus constructing a complete business closed loop. Secondly, the embodiments of the present invention can construct a low-altitude spatiotemporal information base, integrating historical state memory and real-time compliance rules, providing unified spatiotemporal retrieval and dynamic compliance verification for tasks, thereby achieving traceability of historical states and immediate response to constraint rules. In addition, the embodiments of the present invention can deeply embed compliance layers. The system is integrated into the base and plays a core constraining and verification role in the entire process of contingency planning, real-time route planning, dynamic obstacle avoidance, and post-event auditing. This ensures that every step from task generation to completion meets compliance requirements, enabling dynamic, end-to-end, and traceable safety management. Furthermore, this invention establishes a continuous evolution mechanism based on effect feedback. The inspection and handling results can be written back to the base to update the spatiotemporal memory. Based on key performance indicators (KPIs), the system can automatically analyze and iteratively optimize contingency plan parameters, enabling the system to learn from historical execution and optimize strategies. This achieves a leap from "static execution" to "dynamic refinement," obtaining true self-adaptive and continuous evolution characteristics.

[0074] Based on this, the embodiments of the present invention realize efficient collaboration and data closure between daily inspection and alarm event handling tasks. While improving task execution efficiency and compliance, it provides reliable support for refined urban governance and intelligent decision-making by relying on continuous data accumulation and algorithm self-optimization, and has significant practical value and good scalability.

[0075] In this embodiment of the invention, the task processing device can acquire task retrieval data for a task to be executed upon detection of such a task. The task to be executed may be a routine inspection task or an alarm event handling task. Then, the task processing device can determine the execution plan corresponding to the task to be executed, and based on the execution plan and the task retrieval data, determine the task execution data. Further, the task processing device can identify a target flight device from at least one flight device to execute the task to be executed, and issue a target flight command to the target flight device based on the task execution data. Correspondingly, the target flight device can collect data according to the target flight command, obtaining target collected data; and generate target collected evidence data based on the target collected data, thereby returning the target collected evidence data to the task processing device. Based on this, the task processing device can generate a handling conclusion report based on the target collected evidence data. As can be seen, the embodiments of the present invention can treat daily inspection tasks and / or alarm event handling tasks as tasks to be executed, and can automatically determine task execution data through the execution plan corresponding to the tasks to be executed. In this way, the target flight equipment is driven to collect data through the task execution data, and a handling conclusion report of the tasks to be executed can be easily generated. In other words, the embodiments of the present invention can integrate daily inspection and alarm event handling, and can determine the task execution data of the tasks to be executed without relying on manual planning and intervention, thus enabling convenient inspection processing.

[0076] Based on the above description, this embodiment of the invention also proposes a more specific low-altitude inspection processing method. Accordingly, this low-altitude inspection processing method can be executed by the aforementioned low-altitude inspection processing system, which may include mission processing equipment and at least one flight device, etc. Please refer to... Figure 2 The low-altitude inspection and processing method may include the following steps S201-S208: S201, when the task processing device detects a task to be executed, it obtains the task retrieval data of the task to be executed; wherein, the task to be executed is a daily inspection task or an alarm event handling task.

[0077] S202, the task processing device determines the execution plan corresponding to the task to be executed, and determines the task execution data of the task to be executed based on the execution plan corresponding to the task to be executed and the task retrieval data.

[0078] S203, the mission processing device determines the target flight device from at least one flight device to perform the mission to be performed, and issues a target flight command to the target flight device based on the mission execution data.

[0079] The target flight command carries mission acquisition instruction data from the mission execution data. The mission acquisition instruction data includes the target camera script, and a camera script supports the use of flight equipment to fly and acquire video streams.

[0080] S204: After arriving at the mission area to be executed, the target flight equipment flies according to the target camera script and collects the target video stream.

[0081] Correspondingly, the target flight equipment can fly and collect target video streams according to the mission acquisition instruction data, such as the target shot script, cruise speed, cruise altitude, waypoint sequence for avoiding no-fly zones, etc. in the mission acquisition instruction data.

[0082] Based on this, the target flight equipment can fly to the mission inspection location to be performed according to the route indicated by the mission acquisition instruction data. Optionally, the target flight equipment can also transmit flight status information in real time through an onboard communication module (such as a 4G / 5G module, i.e., a fourth-generation / fifth-generation mobile communication module), that is, transmit flight status information back to the mission processing equipment. For example, flight status information can be transmitted once every preset flight status transmission interval. The preset flight status transmission interval can be set according to experience or actual needs, and this embodiment of the invention does not limit it. Optionally, the flight status information may include, but is not limited to, at least one of the following: the target flight equipment's position, altitude, and battery level at the current time, etc., and this embodiment of the invention does not limit it.

[0083] Optionally, the task area to be executed can refer to the area where the distance between the task inspection position and the task inspection position is less than or equal to a preset inspection distance threshold. For example, the task area to be executed can refer to the area with the task inspection position as the center and the preset inspection distance threshold as the radius, etc. Optionally, the preset inspection distance threshold can be set according to experience or actual needs, and this embodiment of the present invention does not limit it.

[0084] Optionally, the target video stream may include all video streams captured during flight according to the target shot script, or it may only include video streams captured under specified actions, etc.; this embodiment of the invention does not limit this. Optionally, the specified actions may be set according to experience or actual needs, and this embodiment of the invention does not limit this; for example, the specified actions may include core orbit actions (i.e., orbit actions around points of interest), in which case the target video stream may include video streams captured during the core orbit script execution phase, etc. Based on this, the target flight device can automatically execute gimbal and flight actions according to the target shot script after arriving at the mission area, and capture high-definition video streams.

[0085] S205: During the acquisition of the target video stream, the target flight equipment continuously extracts images to be analyzed from the target video stream according to the target sampling frequency, and performs anomaly detection on the currently extracted images to be analyzed, and obtains the current anomaly detection result.

[0086] In this embodiment of the invention, the target flight device can continuously run the SoI extraction and evidence packet generation module during the execution of a task (i.e., during the acquisition of the target video stream) to process the target video stream. The target sampling frequency can be determined from the task acquisition instruction data; that is, the task acquisition instruction data may also include the target sampling frequency. Optionally, in other embodiments, the target flight device can also obtain the target sampling frequency from its own storage space. In this case, the target sampling frequency can be a preset sampling frequency in the target flight device, and so on. For example, assuming the target sampling frequency is 1 frame every 5 seconds, the target flight device can extract one keyframe from the target video stream every 5 seconds as a basic record; for example, the extracted frame can be used as the image to be analyzed every 5 seconds.

[0087] In one implementation, when performing anomaly detection on the currently extracted image to be analyzed and obtaining the current anomaly detection result, a target image recognition model can be invoked. Based on the currently extracted image to be analyzed, the target image recognition result can be determined. For example, the currently extracted image to be analyzed can be input into the target image recognition model, and the target image recognition model can output the target image recognition result. When the target image recognition result includes a target anomaly trigger indicator, the current anomaly detection result can be determined to be anomaly, that is, the current anomaly detection result satisfies the anomaly triggering condition in the task acquisition instruction data. When the target image recognition result does not include the target anomaly trigger indicator, the current anomaly detection result can be determined to be non-anomaly, that is, the current anomaly detection result does not satisfy the anomaly triggering condition in the task acquisition instruction data. Meeting the anomaly triggering condition indicates the triggering of an anomaly event.

[0088] Optionally, the target image recognition model may include any target detection model, any semantic segmentation model, etc.; this embodiment of the invention does not limit this. Optionally, the target anomaly triggering index may be determined based on the anomaly triggering conditions in the task acquisition instruction data; this embodiment of the invention does not limit the specific content of the target anomaly triggering index. Optionally, the target image recognition result including the target anomaly triggering index may also be expressed as the target image recognition result indicating the existence of the target anomaly triggering index. For example, when the anomaly triggering condition includes detecting a construction vehicle without a construction permit, the target anomaly triggering indicator may include the construction vehicle without a construction permit; and / or, when the anomaly triggering condition includes a sudden change in the density of a target being inspected (such as a density greater than a preset density threshold, which may be set according to experience or actual needs), the target anomaly triggering indicator may include a sudden change in the density of a target being inspected (i.e., a target being inspected that includes a density change), such as road construction or water accumulation exceeding a preset water accumulation threshold (i.e., the preset density threshold may be the preset water accumulation threshold at this time); and / or, when the anomaly triggering condition includes abnormal dwelling of aerial equipment, the target anomaly triggering indicator may include abnormal dwelling of aerial equipment, such as no detection of a target being inspected for an extended period of time; and / or, when the anomaly triggering condition includes the appearance of an abnormal hot zone in thermal imaging, the target anomaly triggering indicator may include the abnormal hot zone, etc.; the embodiments of the present invention do not limit this. For example, taking the construction vehicle without a construction permit as the target anomaly triggering indicator as an example, when a construction vehicle is identified in the current image to be analyzed and it is detected that the construction vehicle does not have a construction permit, the target image recognition result may include the target anomaly triggering indicator, etc.

[0089] In another implementation, the task acquisition indication data also includes an anomaly confidence threshold. When performing anomaly detection on the currently extracted image to be analyzed and obtaining the current anomaly detection result, a target image recognition model can be invoked to determine the target image recognition result based on the currently extracted image to be analyzed. When the target image recognition result includes a target anomaly triggering indicator, the anomaly detection confidence of the target anomaly triggering indicator can be determined. If the target image recognition result also includes the anomaly detection confidence of the target anomaly triggering indicator, then the anomaly detection confidence of the target anomaly triggering indicator can be determined from the target image recognition result. If the anomaly detection confidence is greater than the anomaly confidence threshold, then the current anomaly detection result can be determined to be anomaly. If the anomaly detection confidence is less than or equal to the anomaly confidence threshold, then the current anomaly detection result can be determined to be non-anomaly, and so on.

[0090] Optionally, in other embodiments, the low-altitude inspection and processing system may further include an edge computing node. In this case, the target flight device can send the target video stream to the edge computing node, so that the edge computing node continuously extracts the image to be analyzed from the target video stream according to the target sampling frequency, performs anomaly detection on the currently extracted image to be analyzed, obtains the current anomaly detection result, and then determines the target collection evidence data, etc.; the present invention does not limit this.

[0091] S206, if the current anomaly detection result indicates an anomaly, the target flight device obtains the abnormal image data corresponding to the currently extracted image to be analyzed from the target video stream and adds the abnormal image data to the target acquisition data.

[0092] In this embodiment of the invention, when the current anomaly detection result indicates an anomaly, an anomaly trigger can be implemented to obtain abnormal image data.

[0093] Optionally, when obtaining abnormal image data corresponding to the currently extracted image to be analyzed from the target video stream, the position of the currently extracted image to be analyzed in the target video stream can be used as the abnormal trigger point, and abnormal image data can be extracted from the target video stream according to the abnormal trigger point, such as extracting video segments within a preset extraction time before and after the abnormal trigger point; or, abnormal image data can be extracted from the target video stream according to the abnormal trigger point and the target sampling frequency. In this case, the abnormal image data may include all keyframes within a preset extraction time before and after the abnormal trigger point (i.e., all frames extracted from the video stream within the preset extraction time before and after the abnormal trigger point according to the target sampling frequency, and the abnormal image data may be a list of keyframes), etc.; the embodiments of the present invention do not limit this. Optionally, the preset extraction time can be set according to experience or actual needs, and the embodiments of the present invention do not limit this; for example, the preset extraction time can be 15 seconds, such as extracting video segments 15 seconds before and after the abnormal trigger point.

[0094] Therefore, the target acquisition data may include abnormal image data.

[0095] S207, the target flight equipment generates target acquisition evidence data based on the target acquisition data, and then returns the target acquisition evidence data to the mission processing equipment.

[0096] Optionally, the target-collected evidence data can also be referred to as a target evidence package or evidence package, etc. Optionally, the target flight equipment can package the abnormal image data and the corresponding full metadata to form target-collected evidence data, thereby achieving structured encapsulation of the evidence package.

[0097] Optionally, the target acquisition evidence data may include, but is not limited to, at least one of the following: target acquisition data (also known as SoI content), spatiotemporal reference information (such as the start and end timestamps of the target acquisition data and / or the timestamps of each frame of the target acquisition data), attitude trajectory (such as the latitude, longitude, altitude, and gimbal angle corresponding to each time point within the time period of the target acquisition data (such as the time point of each frame of the image), task context (such as the task ID and the plan version number), script execution log (such as the currently executing core orbiting action, which can record the execution status (such as normal or abnormal) and time of each step of the shot script), and integrity verification (such as adding hash values ​​or digital signatures to ensure that the evidence chain is tamper-proof during transmission and storage), etc.; the embodiments of the present invention do not limit this. Based on this, the embodiments of the present invention can align and package the target acquisition data and the corresponding full metadata (such as attitude trajectory, task context, etc.) to obtain target acquisition evidence data, thereby generating a structured SoI evidence package containing all the above information; and / or, its hash value can also be calculated to ensure integrity, etc.

[0098] Based on this, embodiments of the present invention can prioritize transmitting small, high-value SoI evidence packets back to the task processing device (such as the command center in the backhaul value task processing device, such as the fusion interpretation and structured output module) via 4G / 5G networks, instead of waiting for the full video to be transmitted back. It is evident that embodiments of the present invention propose an "event-driven effective fragment (SoI) extraction" and "evidence packetization" mechanism, replacing the full video transmission with "key fragments + aligned metadata," which can reduce data transmission bandwidth and backend analysis computing power consumption by more than 90%. Simultaneously, the evidence packet (i.e., target-collected evidence data) structurally integrates geographical, spatiotemporal, and task information, providing a solid foundation for subsequent multi-source fusion interpretation and reliable tracing.

[0099] S208, the task processing equipment generates a disposal conclusion report based on the evidence data collected from the target.

[0100] In one implementation, the task processing device can invoke a target analysis model to generate a conclusion analysis result based on the target-collected evidence data. For example, the evidence analysis content from the target-collected evidence data can be input into the target analysis model to output the conclusion analysis result, and a disposal conclusion report (also known as a disposal conclusion report for the task to be executed) can be generated based on the conclusion analysis result. Alternatively, the disposal conclusion report can be directly output through the target analysis model, and so on. Optionally, the evidence analysis content may include, but is not limited to, at least one of the following: target collection data, spatiotemporal reference information, attitude trajectory, script execution logs, etc., from the target collection evidence data. This embodiment of the invention does not limit this. For example, the evidence analysis content may include target collection data (such as SoI video content) from the target collection evidence data, in which case the target collection data can be input into the target analysis model for analysis. For example, the evidence analysis content and the interpretation list from the task execution data can also be input into the target analysis model to output the conclusion analysis result or disposal conclusion report, etc., through the target analysis model.

[0101] Optionally, the target analysis model can be any target detection model, any semantic segmentation model, or any multimodal large model (in which case a disposal conclusion report can be directly output), etc.; this embodiment of the invention does not limit this. Optionally, when generating a disposal conclusion report based on the conclusion analysis results, the disposal conclusion report can be generated using the conclusion analysis results according to a preset report template; optionally, the preset report template can be set according to experience or actual needs, this embodiment of the invention does not limit this.

[0102] In another implementation, the task processing device can also determine the base layer indication information corresponding to the task to be executed, which includes the basic geospatial data corresponding to the task; and can call the target analysis model to generate a disposal conclusion report based on the target collection evidence data and the base layer indication information. Optionally, the basic geospatial data corresponding to the task to be executed can be determined from the static base map layer according to the spatial location indicated by the task collection indication data (such as the location indicated by latitude and longitude in the attitude trajectory, etc.). For example, the basic geospatial data of the grid area where the spatial location is located can be used as the basic geospatial data corresponding to the task to be executed, etc.

[0103] Optionally, the data format of the basic geospatial data corresponding to the task to be performed can be vector polygon / line data or raster tiles; optionally, the basic geospatial data can include, but is not limited to, at least one of the following: static geographic features such as roads, water systems, building outlines, and critical infrastructure, etc., which are not limited in this embodiment of the invention. Based on this, the basic geospatial data can serve as the spatial reference for all upper-level data, providing an "anchoring" function to confirm the planned use of the land parcel, such as "urban road" etc.

[0104] Optionally, the base layer indication information may also include, but is not limited to, at least one of the following: compliance approval query results and historical baseline status information corresponding to the task to be executed, etc., which are not limited in this embodiment of the invention; for example, it can be determined from the base (e.g., the compliance layer can be retrieved again to determine whether there is construction approval at the corresponding time and location, etc.), or it can be determined from the task retrieval data, etc. Based on this, this embodiment of the invention can associate the spatial location in the evidence package with at least one of the static base map layer, compliance layer, and multi-temporal change layer in the base to achieve multi-source information alignment.

[0105] Optionally, when calling the target analysis model to generate a disposal conclusion report based on the target-collected evidence data and the base layer indication information, the evidence analysis content and the base layer indication information can be input into the target analysis model (e.g., the target analysis model can be a multimodal large model) to output the conclusion analysis results through the target analysis model, and generate a disposal conclusion report based on the conclusion analysis results; or, the evidence analysis content and the base layer indication information can be input into the target analysis model to output the disposal conclusion report through the target analysis model; or, the evidence analysis content can be input into the target analysis model to output the conclusion analysis results through the target analysis model, and generate a disposal conclusion report based on the conclusion analysis results and the base layer indication information, etc.; this embodiment of the invention does not limit this. Optionally, when inputting the evidence analysis content into the target analysis model to output the conclusion analysis results through the target analysis model, the evidence analysis content can be input into the target analysis model all at once, or each image frame in the evidence analysis content can be input into the target analysis model separately (in this case, the conclusion analysis results may include the image analysis results of each image frame), etc.; this embodiment of the invention does not limit this.

[0106] Optionally, when inputting the evidence analysis content and base layer indication information into the target analysis model (such as a multimodal large model), the judgment list in the task execution data can also be input into the target analysis model (that is, the evidence analysis content, base layer indication information and judgment list can be input into the target analysis model) so as to output the conclusion analysis results or disposal conclusion report through the target analysis model, etc.

[0107] Optionally, the conclusion analysis results may include, but are not limited to, the image recognition interpretation results under each interpretation item in the interpretation list in the task execution data. Optionally, when generating a disposal conclusion report based on the conclusion analysis results and the base layer indication information, the image recognition interpretation results under each incomplete interpretation item (i.e., interpretation items that need to be interpreted in conjunction with the base layer indication information) in the interpretation list in the task execution data and the base layer indication information can be analyzed to obtain at least one interpretation result, thereby generating a disposal conclusion report based on the conclusion analysis results and at least one interpretation result. For example, taking road construction as an example, assuming the judgment checklist includes judgment items such as verifying whether the construction has a permit, estimating the road area occupied, and feature recording, then through image recognition using the target analysis model, if no valid permit is found on the construction vehicles and fences in the evidence analysis, the image recognition judgment result under the judgment item "verifying whether the construction has a permit" can be determined as "no permit is held." Furthermore, the road area occupied in the image can be determined through image recognition, thus determining the image recognition judgment result under the judgment item "estimating the road area" as the road area occupied. Additionally, if the license plate number of the construction vehicle and the presence of the construction unit's name on the fence (but without a permit number) can be determined through image recognition, then the image recognition judgment result under the judgment item "feature recording" can be "features recorded." Correspondingly, the incomplete judgment item at this time may include estimating the road area, which can be based on the road area occupied, the altitude of the flight equipment, the gimbal angle data, and the base layer indication information (such as GIS). Using a Geographic Information System (GIS) base map, the area of ​​the fence encroaching on the non-motorized vehicle lane and part of the motorized vehicle lane can be calculated. If the area is approximately 150 square meters, the judgment result can be determined as serious road occupation. Furthermore, based on the judgment results such as serious road occupation, recorded features, and lack of permits, a disposal conclusion report can be generated, and so on. It is evident that the embodiments of this invention can achieve machine-assisted judgment, that is, by combining the judgment list, guiding analysts or using AI algorithms to conduct specialized analysis of evidence analysis content (i.e., SoI content). For example, when verifying illegal dumping of garbage, the system will automatically retrieve the planned land use nature of the plot for comparison, and so on.

[0108] Based on this, embodiments of the present invention can integrate all interpretation results to generate a field-based, structured disposal conclusion report. For example, the disposal conclusion report may include, but is not limited to, at least one of the following: event ID (e.g., task ID, which can be constructed using spatial index + data collection time), event type (e.g., determined to be illegal road occupation construction based on the interpretation result of not holding a permit), location (e.g., including latitude and longitude and / or location description (e.g., northeast side of the intersection of XX Road and YY Road)), confidence level (e.g., can be determined from the target-collected evidence data, or the target analysis model can also output the confidence level that no valid permit signs were found on the construction vehicles and the fence), degree of target change, verification content (e.g., may include, but is not limited to, at least one of the following: whether a permit is held: false (meaning no permit is held), road occupation area: 150 square meters, vehicle characteristics: license plate and construction unit, risk level: high, etc.), disposal recommendations (e.g., immediately notify department A to manage traffic on the road section, department B to go to the site to order the work to stop and verify qualifications, and simultaneously upload the evidence package as a basis for enforcement, etc.); embodiments of the present invention do not limit this. Therefore, the format of the incident response report can directly drive subsequent work order systems or command decisions, etc. Optionally, the incident response recommendations can be determined based on risk level and / or event type, etc.

[0109] Optionally, the task processing device can also send the disposal conclusion report to the target disposal object (such as the corresponding regulatory department, etc.). The target disposal object can be the regulatory department or management personnel corresponding to the task type to be executed. Optionally, the disposal conclusion report of each task can be sent to the corresponding target disposal object separately, or it can be sent to the target disposal object when the risk level is medium risk and / or high risk, etc. The embodiments of the present invention do not limit this.

[0110] In summary, the embodiments of this invention can reposition inspections as a continuous feeding process of multi-temporal change layers and risk layers in the "base," allowing the accumulated data to be directly transformed into "historical baselines" and "risk prophets" that can be invoked when performing tasks (such as alarm event handling tasks). Thus, through pre-planning, historical temporal phases, compliance constraints, and available resources can be retrieved from the base the moment a task is triggered, and the degree of target change can be calculated, enabling flight equipment to "go into action with knowledge," achieving a qualitative leap from "blind inspection" to "precise verification." This connects daily inspections with alarm event handling, constructing a flywheel of "inspection data accumulation - data service handling - handling results feeding back into inspection strategies." This is not only a technological integration innovation but also a practical need to solve the efficiency and cost problems of low-altitude operations, achieving a closed-loop technology for integrated daily inspections and alarm event handling. Furthermore, the base in this embodiment of the invention does not rely on expensive full-scale real-scene 3D modeling. Instead, it adopts a lightweight structure of "static base map + multi-layer thematic information (compliance, resources, and time phase)". Each layer can be built and updated at low cost by connecting to existing GIS systems, airspace approval platforms, airport management systems, and historical task logs, which has a high degree of engineering feasibility. In addition, it can abstract complex route planning, camera control, and evidence collection logic into configurable and reusable "contingency plans", which greatly reduces the cost of customizing and developing intelligent flight applications for each new scenario. Operation and maintenance personnel can adapt to business changes by adjusting the parameters of the contingency plans instead of modifying the code. Furthermore, embodiments of the present invention can align and bind video clips, precise spatiotemporal stamps, flight attitude, camera parameters, and mission context through SoI evidence packages, forming an immutable and traceable complete chain of evidence that meets the audit and compliance requirements in urban governance. Additionally, it can combine the "what the flight equipment sees" (SoI) with the "what the base knows" (such as facility attributes, historical status, and surrounding risks) to output a structured conclusion such as "a certain intersection is congested due to a construction barrier blocking the lane in the northeast direction (95% confidence level)," far exceeding the crude judgment of "congestion detected" derived from simple video analysis. This effectively improves the accuracy of inspection and handling, and makes the handling conclusion report more detailed.

[0111] Based on this, embodiments of the present invention can be designed as a closed-loop system integrating daily inspection and alarm event handling, centered on the "base," driven by the "contingency plan," and carried by the "SoI evidence package" (e.g., Figure 3 As shown, it not only theoretically constructs a complete low-altitude data value cycle, but also reduces the implementation complexity through modular and template-based design in engineering. Furthermore, it ensures the long-term vitality and practicality of the system through a self-evolution mechanism, providing a reliable and efficient solution for city-level low-altitude intelligent operation and maintenance.

[0112] In this embodiment of the invention, the task processing device can acquire task retrieval data for a task to be executed when it detects such a task. The device can also determine the execution plan corresponding to the task and, based on the execution plan and task retrieval data, determine the task execution data. Then, the device identifies a target flight device from at least one flight device to execute the task and issues a target flight command to the target flight device based on the task execution data. The target flight command carries task acquisition instruction data from the task execution data, including a target shot script. A shot script can guide the flight of the flight device and the acquisition of video streams. Based on this, the target flight device can fly according to the target shot script and acquire the target video stream after reaching the task area of ​​the task to be executed. Furthermore, during the acquisition of the target video stream, it can continuously extract images to be analyzed from the target video stream according to the target sampling frequency and perform anomaly detection on the currently extracted images to be analyzed, obtaining the current anomaly detection result. Correspondingly, if the current anomaly detection result indicates an anomaly, the target flight device can obtain the anomaly image data corresponding to the currently extracted image to be analyzed from the target video stream and add the anomaly image data to the target acquisition data. Furthermore, the target flight device can generate target acquisition evidence data based on the target acquisition data, and then return the target acquisition evidence data to the task processing device. Based on this, the task processing device can generate a handling conclusion report based on the target acquisition evidence data. Therefore, this embodiment of the invention can only send target acquisition evidence data back to the task processing device, achieving bandwidth savings, structured output, and reliable traceability, thereby effectively improving analysis efficiency and evidence collection capabilities.

[0113] Based on the description of the relevant embodiments of the low-altitude inspection processing method above, this invention also proposes a low-altitude inspection processing system; such as Figure 4 As shown, the low-altitude inspection and processing system may include mission processing equipment 401 and at least one flight device 402. The low-altitude inspection and processing system can perform... Figure 1 or Figure 2 The low-altitude inspection processing method shown, that is, the low-altitude inspection processing system can operate the above-mentioned units: The task processing device 401 is used to acquire task retrieval data of the task to be executed when a task to be executed is detected; wherein the task to be executed is a daily inspection task or an alarm event handling task. The task processing device 401 is further configured to determine the execution plan corresponding to the task to be executed, and to determine the task execution data of the task to be executed based on the execution plan corresponding to the task to be executed and the task retrieval data; The task processing device 401 is further configured to determine, from the at least one flight device 402, a target flight device 4021 for performing the task to be performed, and issue a target flight command to the target flight device 4021 based on the task execution data. The target flight device 4021 is used to collect data according to the target flight command to obtain target collection data; and to generate target collection evidence data based on the target collection data, thereby returning the target collection evidence data to the task processing device 401; The task processing device 401 is also used to generate a disposal conclusion report based on the evidence data collected from the target.

[0114] In one embodiment, the task retrieval data includes the degree of target change corresponding to the task to be executed; when acquiring the task retrieval data of the task to be executed, the task processing device 401 may specifically be used for: Determine the task inspection location of the task to be executed, and retrieve the historical baseline status information of the monitored object corresponding to the task to be executed at the task inspection location and at least one historical time. Based on the historical baseline status information, the degree of change of the target is determined, and the degree of change of the target is added to the task retrieval data; When the task processing device 401 determines the task execution data of the task to be executed based on the execution plan corresponding to the task to be executed and the task retrieval data, it can be specifically used for: Based on the degree of change of the target, the parameters of the execution plan corresponding to the task to be executed are adjusted to obtain the task execution adjustment plan for the task to be executed; and based on the task execution adjustment plan, the task execution data for the task to be executed is generated.

[0115] In another embodiment, the task retrieval data further includes flight equipment resource indication data corresponding to the task inspection location. The flight equipment resource indication data supports equipment information for indicating each selectable flight device in the set of selectable flight devices within the target space range of the task inspection location. When the task processing device 401 adjusts the parameters of the execution plan corresponding to the task to be executed based on the degree of target change to obtain the task execution adjustment plan, it can be specifically used for: Based on the degree of change of the target, the inspection parameters of the execution plan corresponding to the task to be executed are adjusted to obtain the execution plan after the inspection parameters are adjusted. Based on the flight equipment resource indication data, the resource scheduling weight of the execution plan after the inspection parameters are adjusted is adjusted to obtain the task execution adjustment plan for the task to be executed.

[0116] In another embodiment, the target flight command carries task acquisition instruction data from the task execution data. The task acquisition instruction data includes a target shot script, which supports guiding the flight equipment's flight and video stream acquisition. When the target flight equipment 4021 acquires data according to the target flight command and obtains the target acquisition data, it can specifically be used for: After arriving at the task area to be performed, fly according to the target camera script and acquire the target video stream; During the acquisition of the target video stream, images to be analyzed are continuously extracted from the target video stream according to the target sampling frequency, and anomaly detection is performed on the currently extracted images to be analyzed to obtain the current anomaly detection result. If the current anomaly detection result indicates an anomaly, then the abnormal image data corresponding to the currently extracted image to be analyzed is obtained from the target video stream, and the abnormal image data is added to the target acquisition data.

[0117] In another implementation, the mission acquisition indication data also includes an anomaly confidence threshold; when the target flight device 4021 performs anomaly detection on the currently extracted image to be analyzed and obtains the current anomaly detection result, it can be specifically used for: The target image recognition model is invoked, and the target image recognition result is determined based on the currently extracted image to be analyzed; When the target image recognition result includes a target anomaly triggering indicator, the anomaly detection confidence level of the target anomaly triggering indicator is determined; If the anomaly detection confidence level is greater than the anomaly confidence level threshold, the current anomaly detection result is determined to be anomaly; if the anomaly detection confidence level is less than or equal to the anomaly confidence level threshold, the current anomaly detection result is determined to be non-anomaly.

[0118] In another implementation, when the task processing device 401 generates a disposal conclusion report based on the evidence data collected from the target, it can specifically be used for: Determine the base layer indication information corresponding to the task to be executed, wherein the base layer indication information includes the basic geospatial data corresponding to the task to be executed; The target analysis model is invoked, and a disposal conclusion report is generated based on the evidence data collected from the target and the indication information of the base layer.

[0119] In another embodiment, the execution plan corresponding to the task to be executed is one of at least one execution plan; the task processing device 401 can also be used for: For any one of the at least one execution plan, determine the plan evaluation index for that execution plan; Based on the evaluation indicators of any of the execution plans, determine whether any of the execution plans require optimization. If any of the execution plans requires optimization, then the execution plan is optimized to update the execution plan.

[0120] In another embodiment, the task processing device 401 can also be used for: Obtain daily inspection task generation guidance data, which includes at least one of the following: risk indication information, coverage indication information and inspection object attribute indication information of each preset inspection unit in at least one preset inspection unit; Based on the guidance data generated from the daily inspection tasks, a list of daily inspection tasks is generated; wherein, any daily inspection task in the list can be used as the task to be executed when the daily task triggering condition is met.

[0121] According to one embodiment of the present invention, Figure 4 Each device in the illustrated low-altitude inspection and processing system can be individually or entirely combined into one or more other units, or one or more of the devices can be further divided into multiple functionally smaller devices. This achieves the same operation without affecting the technical effects of the embodiments of the present invention. In practical applications, the function of one device can also be implemented by multiple devices, or the function of multiple devices can be implemented by one device. In other embodiments of the present invention, any low-altitude inspection and processing system may also include other devices. In practical applications, these functions can also be implemented with the assistance of other devices, and can be implemented collaboratively by multiple devices.

[0122] An exemplary embodiment of the present invention also provides a non-transitory computer-readable storage medium storing a computer program, wherein the computer program, when executed by a computer's processor, is used to cause the computer to perform a method according to an embodiment of the present invention.

[0123] An exemplary embodiment of the present invention also provides a computer program product, including a computer program, wherein, when executed by a computer's processor, the computer program is used to cause the computer to perform a method according to an embodiment of the present invention.

[0124] Furthermore, it should be understood that the above-disclosed embodiments are merely preferred embodiments of the present invention and should not be construed as limiting the scope of the present invention. Therefore, any equivalent variations made in accordance with the claims of the present invention are still within the scope of the present invention.

Claims

1. A low-altitude inspection and processing method, characterized in that, The low-altitude inspection and processing method is applied to a low-altitude inspection and processing system, which includes mission processing equipment and at least one flight device. The method includes: When the task processing device detects a task to be executed, it acquires the task retrieval data of the task to be executed; wherein, the task to be executed is a daily inspection task or an alarm event handling task. The task processing device determines the execution plan corresponding to the task to be executed, and determines the task execution data of the task to be executed based on the execution plan corresponding to the task to be executed and the task retrieval data; The task processing device determines the target flight device from the at least one flight device to perform the task to be performed, and issues a target flight command to the target flight device based on the task execution data; The target flight device collects data according to the target flight command to obtain target collection data; and generates target collection evidence data based on the target collection data, thereby returning the target collection evidence data to the task processing device; The task processing device generates a disposal conclusion report based on the evidence data collected from the target.

2. The method according to claim 1, characterized in that, The task retrieval data includes the degree of target change corresponding to the task to be executed; obtaining the task retrieval data of the task to be executed includes: Determine the task inspection location of the task to be executed, and retrieve the historical baseline status information of the monitored object corresponding to the task to be executed at the task inspection location and at least one historical time. Based on the historical baseline status information, the degree of change of the target is determined, and the degree of change of the target is added to the task retrieval data; The step of determining the task execution data of the task to be executed based on the execution plan corresponding to the task to be executed and the task retrieval data includes: Based on the degree of change of the target, the parameters of the execution plan corresponding to the task to be executed are adjusted to obtain the task execution adjustment plan for the task to be executed; and based on the task execution adjustment plan, the task execution data for the task to be executed is generated.

3. The method according to claim 2, characterized in that, The task retrieval data also includes flight equipment resource indication data corresponding to the task inspection location. This flight equipment resource indication data supports equipment information for each selectable flight device in the set of selectable flight devices within the target spatial range of the task inspection location. The step of adjusting the parameters of the execution plan corresponding to the task to be executed based on the degree of target change to obtain a task execution adjustment scheme for the task to be executed includes: Based on the degree of change of the target, the inspection parameters of the execution plan corresponding to the task to be executed are adjusted to obtain the execution plan after the inspection parameters are adjusted. Based on the flight equipment resource indication data, the resource scheduling weight of the execution plan after the inspection parameters are adjusted is adjusted to obtain the task execution adjustment plan for the task to be executed.

4. The method according to any one of claims 1-3, characterized in that, The target flight command carries the mission acquisition instruction data from the mission execution data. The mission acquisition instruction data includes a target shot script, and a shot script supports the use of guiding the flight equipment to fly and acquire video streams. The target flight equipment collects data according to the target flight command to obtain target collected data, including: After arriving at the mission area of ​​the task to be performed, the target flight device flies according to the target camera script and acquires the target video stream; During the acquisition of the target video stream, images to be analyzed are continuously extracted from the target video stream according to the target sampling frequency, and anomaly detection is performed on the currently extracted images to be analyzed to obtain the current anomaly detection result. If the current anomaly detection result indicates an anomaly, then the abnormal image data corresponding to the currently extracted image to be analyzed is obtained from the target video stream, and the abnormal image data is added to the target acquisition data.

5. The method according to claim 4, characterized in that, The task acquisition instruction data also includes an anomaly confidence threshold; the anomaly detection of the currently extracted image to be analyzed, to obtain the current anomaly detection result, includes: The target image recognition model is invoked, and the target image recognition result is determined based on the currently extracted image to be analyzed; When the target image recognition result includes a target anomaly triggering indicator, the anomaly detection confidence level of the target anomaly triggering indicator is determined; If the anomaly detection confidence level is greater than the anomaly confidence level threshold, the current anomaly detection result is determined to be anomaly; if the anomaly detection confidence level is less than or equal to the anomaly confidence level threshold, the current anomaly detection result is determined to be non-anomaly.

6. The method according to any one of claims 1-3, characterized in that, The task processing device generates a handling conclusion report based on the evidence data collected from the target, including: Determine the base layer indication information corresponding to the task to be executed, wherein the base layer indication information includes the basic geospatial data corresponding to the task to be executed; The target analysis model is invoked, and a disposal conclusion report is generated based on the evidence data collected from the target and the indication information of the base layer.

7. The method according to any one of claims 1-3, characterized in that, The execution plan corresponding to the task to be executed is one of at least one execution plan; the method further includes: The task processing device determines the plan evaluation index for any one of the at least one execution plan; Based on the evaluation indicators of any of the execution plans, determine whether any of the execution plans require optimization. If any of the execution plans requires optimization, then the execution plan is optimized to update the execution plan.

8. The method according to any one of claims 1-3, characterized in that, The method further includes: The task processing device acquires daily inspection task generation guidance data, which includes at least one of the following: risk indication information, coverage indication information and inspection object attribute indication information of each preset inspection unit in at least one preset inspection unit. Based on the guidance data generated from the daily inspection tasks, a list of daily inspection tasks is generated; wherein, any daily inspection task in the list can be used as the task to be executed when the daily task triggering condition is met.

9. A low-altitude inspection and processing system, characterized in that, The low-altitude inspection and processing system includes task processing equipment and at least one flight device; wherein... The task processing device is used to acquire task retrieval data of the task to be executed when a task to be executed is detected; wherein the task to be executed is a daily inspection task or an alarm event handling task. The task processing device is further configured to determine the execution plan corresponding to the task to be executed, and to determine the task execution data of the task to be executed based on the execution plan corresponding to the task to be executed and the task retrieval data; The task processing device is further configured to determine, from the at least one flight device, a target flight device for performing the task to be performed, and to issue a target flight command to the target flight device based on the task execution data; The target flight device is used to collect data according to the target flight command to obtain target collection data; and to generate target collection evidence data based on the target collection data, thereby returning the target collection evidence data to the task processing device; The task processing device is also used to generate a disposal conclusion report based on the evidence data collected from the target.

10. A non-transitory computer-readable storage medium storing computer instructions, characterized in that, The computer instructions are used to cause the computer to perform the method according to any one of claims 1-8.

Images