Intelligent monitoring system and method for low-altitude unattended service station

Abstract

This invention discloses an intelligent monitoring system and method for low-altitude unmanned service stations, relating to the field of low-altitude intelligent logistics and transportation monitoring. The system includes: an intelligent control center for global task scheduling, resource coordination, status monitoring, and remote management; multiple service stations, each including a takeoff and landing platform, a door control device, a charging / battery swapping interface, and an identification module; and an intelligent aircraft docking module for enabling automatic landing, identification, charging, and task data interaction for UAVs. This intelligent monitoring system and method for low-altitude unmanned service stations achieves "unmanned operation, remote monitoring, and self-recovery from anomalies" through the establishment of unmanned service stations. Furthermore, by establishing a dynamic adaptation model between flight tasks and ground resources, it supports high-frequency and diverse task operations. Based on the collaboration of edge intelligence and central intelligence, it ensures stable station operation and high task completion rates.

Description

Technical Field

[0001] This invention relates to low-altitude intelligent logistics traffic monitoring technology, specifically to an intelligent monitoring system and method for low-altitude unmanned service stations. Background Technology

[0002] With the rapid development of the urban low-altitude economy, the number of low-altitude aircraft for logistics delivery, traffic management, inspection, and security has surged. Traditional ground service nodes, which rely on manual operation, are struggling to meet the high-frequency, dynamic, and high-safety requirements of these operations. Current station operations largely depend on static configurations, lacking intelligent sensing and dynamic response mechanisms for mission status, airspace scheduling, and equipment health. As a key component of future low-altitude infrastructure construction, unmanned service stations necessitate the development of an unmanned service station system with intelligent monitoring capabilities and the ability to achieve closed-loop management of multi-source, heterogeneous missions throughout the entire process. Summary of the Invention

[0003] The purpose of this invention is to provide an intelligent monitoring system and method for low-altitude unattended service stations to address the aforementioned shortcomings in the prior art.

[0004] To achieve the above objectives, the present invention provides the following technical solution: an intelligent monitoring system for low-altitude unmanned service stations, comprising:

[0005] The intelligent control center is used for global task scheduling, resource coordination, status monitoring, and remote management.

[0006] Multiple service stations, each of which includes a take-off and landing platform, a door control device, a charging and swapping interface, and an identification module;

[0007] The intelligent docking module for aircraft is used to enable the automatic landing, identification, charging, and mission data interaction of drones;

[0008] The perception and safety system integrates video surveillance, anomaly detection, and AI recognition units, and is used to monitor the status of sites and aircraft.

[0009] The operation management platform is used for task decomposition, execution monitoring, anomaly warning, and maintenance management.

[0010] Among them, multiple service stations can be interconnected through a network and can perform task takeover and fault-tolerant switching.

[0011] Furthermore, the intelligent control center includes:

[0012] The task lifecycle management module is used to achieve closed-loop control of the entire process of a task from creation to completion;

[0013] The status awareness and anomaly response module is used to sense the status of devices and tasks in real time and automatically trigger notifications or recovery processes.

[0014] The scheduling and resource matching module is used to dynamically match site and aircraft resources based on task characteristics and airspace constraints;

[0015] Edge intelligent recognition module, which uses edge AI camera to identify aircraft and assist in landing guidance;

[0016] The site self-organization and collaboration module enables multiple sites to perform task takeover and backup scheduling within the network.

[0017] Furthermore, the scheduling and resource matching module employs a multi-dimensional matching scheduling algorithm, which performs dynamic optimal matching by constructing a task-device compatibility scoring function. The scoring function is as follows:

[0018] ;

[0019] in, Indicates the first The first task and the first Overall compatibility score between devices;

[0020] Indicates the first The urgency of each task;

[0021] Indicates the first Resource availability index of each device;

[0022] Indicates the first The device and the first The spatial matching degree between tasks is an inverse proportional function calculated based on the Euclidean distance between the current location of the device and the location or path of the task target. The closer the distance, the higher the value, and it is normalized to the interval [0, 1].

[0023] Indicates the first The health status score of each device is a normalized value calculated based on its historical failure rate, current self-inspection status, and life prediction of key components. The higher the value, the better the health status.

[0024] These are the weighting coefficients for the corresponding indicators, and they satisfy... .

[0025] Furthermore, the intelligent docking module for aircraft supports automatic identification of various drone models and dynamically adapts to drones of different sizes through an adjustable mechanical interface.

[0026] Furthermore, the service station achieves local decision-making through edge AI units, supporting aircraft self-identification, anomaly warning, and remote intervention control.

[0027] Furthermore, the system supports automatically selecting a backup site for task migration and restart based on the "N+1" redundancy mechanism when a failure occurs during task execution.

[0028] Furthermore, the intelligent control center adopts a distributed deployment architecture, supporting centralized scheduling and partitioned management of service sites in multiple regions.

[0029] A method for the full-process operation and management of low-altitude unmanned service stations, applied to the intelligent monitoring system of the low-altitude unmanned service stations, the method specifically includes the following steps:

[0030] System initialization steps: Complete service site registration, device status detection, and task readiness assessment;

[0031] Task reception and decomposition steps: Receive task requests and decompose tasks and plan flight segments according to task type and resource status;

[0032] Dynamic matching and scheduling steps: Based on a multi-dimensional adaptation scoring model, available sites and aircraft resources are dynamically matched, and scheduling instructions are generated;

[0033] Aircraft docking and execution steps: After the aircraft arrives at the station, it completes automatic landing, energy replenishment or material loading and unloading operations through edge intelligent recognition and guidance;

[0034] Status monitoring and feedback steps: Real-time monitoring of task execution status and equipment parameters, uploading to the control center and dynamically adjusting scheduling strategies;

[0035] Anomaly handling and recovery steps: When an anomaly is detected, the fault tolerance mechanism is automatically triggered, the task is migrated to a backup site and execution is resumed.

[0036] Furthermore, the dynamic matching and scheduling steps are constructed as a multi-objective optimization problem, whose objective function and constraints are defined as follows:

[0037] Objective 1: Maximize the overall compatibility between all assigned tasks and devices within the system. The specific expression for this is as follows:

[0038] ;

[0039] Objective 2: Minimize the average response latency of all tasks in the system, as expressed in the following expression:

[0040] ;

[0041] The constraints are as follows:

[0042] Unique allocation constraint: Each task must be assigned to one and only one available device, as expressed below:

[0043] ;

[0044] Capacity load constraint: The total demand of all tasks assigned to each device must not exceed the maximum service capacity of that device, as expressed below:

[0045] ;

[0046] in:

[0047] This indicates the total number of tasks to be assigned.

[0048] Indicates the total number of available devices;

[0049] It is a binary decision variable, when the... The task is assigned to the first The value is 1 if the device is specific, otherwise it is 0.

[0050] Indicates the first The time delay from when a task is ready to when it begins execution;

[0051] Indicates the first The resource requirement vector for each task includes the expected flight duration, computational resource consumption, and data throughput requirements.

[0052] Indicates the first The maximum service capacity vector of a device is the upper limit of the flight mission load, computing load, and communication load that it can withstand.

[0053] Furthermore, the method also includes data interaction with civil aviation regulatory platforms or urban low-altitude airspace management systems to ensure flight compliance and airspace coordination; and the operation management platform can provide user interface, API interface, data report and log export functions.

[0054] Compared with existing technologies, the intelligent monitoring system and method for low-altitude unmanned service stations provided by this invention achieves "unmanned operation, remote monitoring, and self-recovery from anomalies" by setting up unmanned service stations. Furthermore, by establishing a dynamic adaptation model between flight missions and ground resources, it supports high-frequency and diverse mission operations. Based on the collaboration of edge intelligence and central intelligence, it ensures stable station operation and mission completion rate. At the same time, it proposes a closed-loop management method for the entire mission process applicable to low-altitude logistics and urban air traffic operations. Attached Figure Description

[0055] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0056] Figure 1 A schematic diagram of the overall structure of the intelligent monitoring system for low-altitude unmanned service stations provided in an embodiment of the present invention;

[0057] Figure 2 A functional flowchart of the intelligent monitoring system for low-altitude unattended service stations provided in an embodiment of the present invention;

[0058] Figure 3 This is a schematic diagram of the structure of a low-altitude unattended service station provided in an embodiment of the present invention;

[0059] Figure 4 This is a schematic diagram of the dynamic matching and scheduling algorithm provided in an embodiment of the present invention. Detailed Implementation

[0060] To enable those skilled in the art to better understand the technical solution of the present invention, the present invention will be further described in detail below with reference to the accompanying drawings.

[0061] First, a brief introduction to the terms used in the embodiments of this application will be given.

[0062] Low-altitude unmanned service stations refer to automated infrastructure nodes deployed on urban rooftops, the ground, or other open airspace, capable of drone take-off and landing, energy replenishment, data exchange, and limited material storage. These stations require no on-site human supervision, receiving instructions and transmitting status data via remote communication links, forming the physical foundation for networked low-altitude operations.

[0063] The multidimensional adaptation scoring function is a mathematical model used to quantify the matching degree between task requirements and site / drone resources. Its input variables include the spatiotemporal attributes of the task, resource requirements, and the real-time status and health of the equipment. A comprehensive score is output through a weighted summation. This function is the core decision-making basis for achieving dynamic and optimal scheduling, and its formal expression is as follows: .

[0064] Closed-loop monitoring of the entire mission lifecycle refers to continuous, data-driven tracking and management of the entire process, from mission creation, resource matching, command issuance, aircraft docking and execution, to status feedback, anomaly handling, and final archiving. This mechanism ensures the visibility, controllability, and traceability of each mission stage, and is key to achieving high reliability and autonomy for the system.

[0065] The edge intelligent recognition module is an embedded computing unit deployed at the service site, integrating AI inference capabilities. By analyzing local camera video streams, this module performs real-time model identification, identity authentication, landing attitude analysis, and preliminary health status assessment of approaching drones, providing low-latency, highly reliable perception input for core scheduling decisions.

[0066] Based on the above definitions, the implementation environment of the intelligent monitoring system for low-altitude unmanned service stations provided in this application embodiment will be described. Indicatively, this implementation environment includes:

[0067] Intelligent sensing and execution terminals include edge AI cameras, millimeter-wave radars, intelligent door controllers, automatic charging and swapping devices, environmental sensors (such as weather stations) deployed at service sites, as well as the flight control system, GNSS module, and status sensors carried by the drone itself. These terminals are responsible for interacting with the physical world, collecting data, and performing specific operations.

[0068] Communication networks: including 5G / 5G-A networks, low-power wide area networks or satellite communication links, used to build high-bandwidth, low-latency and highly reliable two-way data channels between terminals, sites and centers.

[0069] Computing and processing units include a cloud-based intelligent control center (composed of a server cluster, responsible for global scheduling and optimization), edge computing nodes (deployed in regions or sites, responsible for local real-time decision-making and data processing), and access terminals for the operation and management platform. Processors can be central processing units, graph processors, multi-core processors, or artificial intelligence chips, etc.

[0070] Data storage and management facilities: These can be distributed storage devices or centralized storage, used to store task logs, device status history, map data, AI models, and system configuration parameters.

[0071] Application Scenario Description

[0072] Based on the above definitions and implementation environment, the application scenarios of the embodiments of this application are described. The intelligent monitoring system and method for low-altitude unmanned service stations provided in the embodiments of this application can be applied to scenarios including but not limited to the following:

[0073] In routine urban logistics and delivery scenarios, especially in response to the growing demand for instant retail and same-city express delivery, this solution provides transit, charging, and cargo handover services for logistics drones through a network of unmanned stations distributed throughout the city. The system utilizes a multi-dimensional adaptation scoring function to dynamically plan the most efficient delivery routes and station sequences, effectively avoiding traffic congestion areas and achieving 24 / 7 automated, grid-based logistics operations, significantly reducing labor costs and improving last-mile delivery efficiency.

[0074] In the field of low-altitude emergency response and rescue, this technical solution, through its dynamic scheduling and fault-tolerant mechanism, can quickly establish a reliable "low-altitude lifeline" when roads are disrupted due to natural disasters or public events. For example, when delivering emergency medicines, blood plasma, or rescue equipment to isolated areas, the system can automatically avoid risky airspace and seamlessly switch to a backup station if a primary station becomes unavailable, ensuring a high success rate and timeliness of rescue missions under extreme conditions.

[0075] This solution provides essential infrastructure support for passenger drones and vertical takeoff and landing (VTOL) aircraft in supporting future urban air mobility operations. Through high-precision edge intelligence recognition and guidance, it ensures that aircraft can safely and accurately land at designated stations in complex urban environments for rapid energy replenishment or passenger transfers. The operations management platform globally optimizes and prioritizes dense takeoff and landing requests and resolves airspace conflicts, serving as a core operational support system for building a large-scale, high-density urban low-altitude three-dimensional transportation network.

[0076] As an illustration, the intelligent monitoring system and method for low-altitude unmanned service stations provided in this application embodiment can also be applied to many low-altitude economic fields such as regional security patrol, agricultural and forestry plant protection operation transfer, and urban public services (such as fire observation and lighting inspection). This is only an example and does not limit the specific application scenarios.

[0077] In one exemplary embodiment, such as Figure 1 As shown, the intelligent monitoring system for low-altitude unmanned service stations realizes modularization, data flow, and control flow in low-altitude unmanned service scenarios. Through the collaborative work of the intelligent scheduling and control center (i.e., intelligent control center), unmanned service stations, aircraft, and edge intelligent sensing modules, it achieves closed-loop management of the entire mission process, ensuring operational efficiency, safety, and fault tolerance.

[0078] By setting:

[0079] Ⅰ. Task Management Platform

[0080] It includes: task reception, user interface, and API service.

[0081] As the interface between the system and external users or third-party systems, it is responsible for receiving task requests (such as logistics delivery, emergency response, etc.) and uploading tasks to the intelligent dispatch and control center via API services. The platform supports task decomposition, status query, and result feedback, and has a user interface and data report generation capabilities.

[0082] II. Intelligent Dispatch and Control Center

[0083] It includes: task scheduling, status monitoring, and exception management.

[0084] As the core of the system, it is responsible for global task scheduling, resource matching, status monitoring, and anomaly handling. After receiving a task request, the center generates scheduling instructions based on a multi-dimensional matching algorithm (such as task urgency, site availability, and spatial location) and issues them to the aircraft or service station. Simultaneously, it monitors the task execution status in real time and dynamically adjusts scheduling strategies to ensure efficient task completion.

[0085] III. Aircraft

[0086] By performing specific low-altitude missions (such as logistics transportation and emergency delivery), the aircraft receives instructions from the control center to perform takeoff, landing, flight, and mission operations, and provides real-time feedback data such as position, battery level, and health status to the control center. In abnormal situations, the aircraft can receive remote intervention or mission relocation commands.

[0087] IV. Unmanned service stations

[0088] It includes: take-off and landing platform, door module, charging and swapping equipment, and sensing module.

[0089] The stations serve as low-altitude infrastructure nodes, providing functions such as aircraft takeoff and landing, charging / battery swapping, and cargo loading and unloading. Sensing modules (such as sensors and cameras) monitor the station status and aircraft docking process in real time, reporting data to the control center. The stations feature an "N+1" redundancy design, supporting task takeover and fault-tolerant switching in case of failure.

[0090] V. Edge Intelligent Sensing Module

[0091] It includes: camera recognition, status awareness, and local judgment.

[0092] Deployed at service sites, this module leverages edge computing capabilities to enable local intelligent processing such as aircraft identification, health status detection, and anomaly alerts. Working in conjunction with the site's perception system, it reduces the central communication load and improves response speed. For example, it can guide aircraft to land precisely using AI cameras and trigger local recovery processes or report to the control center when an anomaly is detected.

[0093] Regarding the explanation of data flow and control flow:

[0094] Task upload stream: The task management platform receives tasks through the user interface or API and uploads them to the intelligent scheduling and control center.

[0095] Command delivery flow: The control center generates dispatch commands and sends them to aircraft or service stations to control aircraft take-off and landing, mission execution, and other operations.

[0096] Status feedback stream: The aircraft and service stations report status data (such as power, location, and abnormal signals) to the control center in real time for dynamic monitoring and scheduling adjustments.

[0097] Edge interaction flow: The service station calls the edge intelligent perception module to perform local identification and judgment, and the result is fed back to the station or directly reported to the control center.

[0098] Comprehensive implementation:

[0099] Unmanned operation and remote monitoring: Through the collaboration of the intelligent dispatch and control center and the edge sensing module, the site can be "unmanned operation and remote monitoring".

[0100] Dynamic resource matching: The control center dynamically schedules tasks and resources based on a multi-dimensional scoring model, supporting high-frequency and diverse task operation.

[0101] Full-process closed-loop management: From task reception, decomposition, execution to feedback, a complete closed loop is formed to ensure that tasks are traceable and controllable.

[0102] Fault tolerance and redundancy mechanisms: The system has scheduling and takeover capabilities between stations, and automatically migrates tasks in the event of a failure, thereby improving system reliability.

[0103] Through this architecture, the system can achieve efficient, safe, and adaptive low-altitude operation management, providing infrastructure support for the urban low-altitude three-dimensional transportation system.

[0104] In one exemplary embodiment, an intelligent monitoring system for low-altitude unmanned service stations is provided, comprising:

[0105] The intelligent control center is used for global task scheduling, resource coordination, status monitoring and analysis, and remote monitoring and management; it is responsible for global task scheduling, status analysis, resource coordination and remote supervision.

[0106] Multiple service stations (i.e., multiple low-altitude unmanned service stations), each station including at least a take-off and landing platform, a door control device, a charging and swapping interface and an identification module;

[0107] The intelligent docking module for aircraft is used to enable drones to land automatically, identify, charge, and exchange mission data.

[0108] The perception and safety system integrates video surveillance, anomaly detection, and AI recognition units to monitor the status of sites and aircraft.

[0109] The operation management platform is used for task decomposition, execution monitoring, anomaly warning, and maintenance management.

[0110] Among them, multiple service stations are interconnected through the network, and have the ability to take over tasks and switch faults.

[0111] The system supports closed-loop monitoring of the entire process of tasks, from creation and assignment to execution and completion.

[0112] In one exemplary embodiment, such as Figure 2 As shown, the intelligent monitoring system for low-altitude unmanned service stations completes the entire lifecycle of a task from receipt to completion (or anomaly handling). This system embodies the core "full-process closed-loop management method" of this embodiment, integrating key mechanisms such as intelligent task decomposition, dynamic matching and scheduling, real-time status monitoring, and automatic fault tolerance and recovery. This ensures the efficient and reliable operation of the system in unmanned mode. Its specific functional flow is as follows:

[0113] Ⅰ. Receive task request

[0114] Specifically, the system receives task requests from users or third-party systems (such as logistics platforms or emergency command centers) through the user interface or API service of the task management platform.

[0115] II. Task decomposition into flight segments

[0116] Specifically, the intelligent dispatch and control center breaks down long-distance or complex tasks into multiple short-range segments that can be executed between adjacent service stations, based on the task's starting point, ending point, airspace constraints, and logistics rules.

[0117] III. Check the status of service stations

[0118] Specifically, the system obtains status parameters from each unattended service station in real time, including: power level, busy level, equipment health status, and accessibility, providing a data foundation for subsequent matching calculations.

[0119] IV. Calculate the matching score

[0120] Specifically, the multidimensional matching scheduling algorithm is invoked, based on the urgency of the task ( ), site resource availability ( Spatial matching degree ( ) and equipment health status ( Factors such as ) are used to calculate the suitability score between the task and the candidate site / aircraft. This allows for optimal resource allocation.

[0121] V. Determine: Are there any available sites?

[0122] Specifically, this is a key decision point in the process. The system determines whether there are available and suitable sites based on the matching score.

[0123] Yes: The process enters the normal task execution branch.

[0124] No: The process enters the exception handling branch, triggering an alarm or queuing.

[0125] VI. Issuing task instructions

[0126] Specifically, the intelligent dispatch and control center will issue instructions containing mission details, time windows, and target stations to the selected aircraft and their corresponding service stations.

[0127] VII. Aircraft performing missions

[0128] Specifically, the aircraft autonomously flies to the target service station according to instructions.

[0129] VIII. Automatic Site Connection and Charging

[0130] Specifically, after the aircraft arrives, the edge intelligent recognition module (such as an AI camera) at the service station guides it to land precisely, and the aircraft docking module automatically completes operations such as identity authentication, hatch opening, charging / battery swapping, or loading and unloading of supplies.

[0131] IX. Real-time status feedback

[0132] Specifically, throughout the entire mission execution and station operation process, the aircraft and service station continuously transmit status data (such as location, battery level, operation progress, and equipment parameters) back to the control center through the sensing system, achieving transparent monitoring of the entire process.

[0133] X. Judgment: Is it abnormal?

[0134] Specifically, the system continuously monitors the returned data to determine whether any abnormalities occur during task execution (such as equipment failure, identification failure, insufficient power, or severe timeout).

[0135] Yes: A fault-tolerant process that records exceptions and initiates task rescheduling upon process entry.

[0136] No: The process proceeded smoothly to completion.

[0137] XI. Task Rescheduling

[0138] Specifically, this is a key innovation of this embodiment. When a mission fails, the system does not simply end in failure, but automatically initiates a recovery mechanism. The control center recalculates the matching degree based on the current global state, migrates the mission to the nearest backup site or replaces the aircraft, and reissues mission instructions, forming a negative feedback loop to ensure the final completion of the mission.

[0139] XII. Mark task as completed

[0140] Specifically, once all segments of the mission have been successfully completed, the system marks the mission status as completed on the platform and generates relevant logs and reports for subsequent analysis and auditing.

[0141] In summary, the following was achieved:

[0142] Full-process closed-loop management: The process covers every step from task creation to completion / recovery, forming a complete lifecycle management.

[0143] Dynamic intelligent scheduling: The steps of "calculating matching score" and "task rescheduling" embody the dynamic resource matching and optimization capabilities based on a multi-dimensional scoring model.

[0144] Anomaly self-recovery mechanism: The clear anomaly detection and rescheduling branches in the diagram demonstrate the system's high fault tolerance and self-healing capabilities under "unattended" conditions.

[0145] Edge and center collaboration: "Automatic site docking" relies on edge intelligence, while global scheduling and rescheduling are handled by the center. The collaboration between the two ensures the efficiency and stability of the system.

[0146] This verifies that the system can achieve efficient, reliable, and fully automated operation without human intervention in practical applications (such as urban low-altitude logistics and emergency response), fully demonstrating the practicality, advancement, and industrial value of this embodiment.

[0147] In one exemplary embodiment, the intelligent control center further includes:

[0148] The task lifecycle management module is used to achieve closed-loop control of the entire process of a task from creation to completion;

[0149] The status awareness and anomaly response module is used to sense the status of devices and tasks in real time and automatically trigger notifications or recovery processes.

[0150] The scheduling and resource matching module is used to dynamically match site and aircraft resources based on mission characteristics and airspace constraints;

[0151] The edge intelligent recognition module is used to identify aircraft using edge AI cameras and assist in landing guidance;

[0152] The site self-organization and collaboration module enables multiple sites to perform task takeover and backup scheduling within the network.

[0153] In one exemplary embodiment, the scheduling and resource matching module employs a multi-dimensional matching scheduling algorithm, which performs dynamic optimal matching by constructing a task-device fit scoring function. The scoring function is as follows:

[0154] ;

[0155] in, Indicates the first The first task and the first Overall compatibility score between devices;

[0156] Indicates the first The urgency of a task is a value normalized to the range [0, 1] by the system rule base. The higher the value, the higher the urgency.

[0157] Indicates the first The resource availability index of a device is a value normalized to the range [0, 1] after taking into account its remaining power, calculated load margin and communication bandwidth availability.

[0158] Indicates the first The device and the first The spatial matching degree between tasks is an inverse proportional function calculated based on the Euclidean distance between the current location of the device and the location or path of the task target. The closer the distance, the higher the value, and it is normalized to the interval [0, 1].

[0159] Indicates the first The health status score of each device is a normalized value calculated based on its historical failure rate, current self-inspection status, and life prediction of key components. The higher the value, the better the health status.

[0160] These are the weighting coefficients for the corresponding indicators, and they satisfy... Its specific value is dynamically adjusted by the system based on historical task completion data through a machine learning model.

[0161] The core logic of the multidimensional matching and scheduling algorithm is as follows:

[0162] (a) Status parameter collection: task urgency, target location, time window; site status, power consumption, busy level, and accessibility;

[0163] (b) Construct the scoring function Score(t,s) and perform maximum score matching;

[0164] (c) During execution, dynamically adjust the score weights and perform real-time rescheduling and task migration.

[0165] In one exemplary embodiment, the intelligent docking module supports automatic identification of various drone models and dynamically adapts to drones of different sizes through an adjustable mechanical interface.

[0166] In one exemplary embodiment, the service station enables local decision-making through an edge AI unit, supporting aircraft self-identification, anomaly warning, and remote intervention control.

[0167] In one exemplary embodiment, such as Figure 3 As shown, the low-altitude unmanned service station serves as the hardware foundation of a distributed, intelligent low-altitude infrastructure node. Its internal modules work together to enable autonomous takeoff and landing, energy replenishment, status awareness, material handover, and local intelligent control of aircraft, making it crucial for achieving the goal of "unmanned" operation.

[0168] It may also specifically include:

[0169] Ⅰ. Take-off and landing platform

[0170] Specifically, it provides standardized, safe, and reliable physical landing areas for aircraft. Its structure is specially designed to adapt to different aircraft types and can work with guidance systems (such as lights and markings) to ensure accurate landings in various environments.

[0171] II. Charging / Battery Swapping Module

[0172] Specifically, this is one of the core functional modules of this site. It supports automatic wired charging or automatic battery pack replacement (battery swapping) for drones, greatly improving the operational endurance of the aircraft and forming the basis for continuous, multi-segment mission execution. The automation of this module is the core element for achieving unmanned operation.

[0173] III. Edge Recognition Camera

[0174] Specifically, this is a concrete manifestation of "edge intelligence" in this embodiment. The camera is not a typical video surveillance system; rather, it integrates AI algorithms, enabling real-time local (edge-side) performance analysis of aircraft type, identity, preliminary health status (such as visual detection of propeller damage), and visual guidance during landing. This reduces the computational burden on the central platform and significantly improves the response speed and reliability of the docking process.

[0175] IV. Environmental monitoring equipment

[0176] Specifically, it is used to monitor micro-environmental parameters around the site in real time, such as wind speed, temperature, humidity, precipitation, and visibility. This data is used to assess takeoff and landing safety, providing a basis for decision-making by the dispatch center; on the other hand, it can be used for the site's own management (such as triggering protection mechanisms in advance of severe weather).

[0177] V. Automatic hatch system

[0178] Specifically, the system automatically opens when the aircraft completes identification and prepares to charge or handle supplies, providing a protective space for the aircraft; it automatically closes after the operation is completed, serving as a safety protection, anti-theft, and anti-environmental interference function. Its opening and closing are linked to commands from the control center and local sensing signals.

[0179] VI. Local Controller

[0180] Specifically, the "brain" of the service station is the main carrier of edge computing capabilities. It is responsible for coordinating and controlling the collaborative work of all modules within the station (such as doors, charging equipment, and cameras), executing instructions from the central platform, processing data generated by edge recognition cameras and other sensing devices, and making rapid local decisions (such as immediately terminating unsafe landing procedures).

[0181] VII. Material Interface

[0182] Specifically, it features a standardized physical interface designed for logistics tasks. This interface enables automatic locking, unlocking, and transfer of cargo containers, facilitating fully automated cargo transfer between drones and service stations. It is a key piece of equipment in the "last link" of the logistics closed loop.

[0183] Module Collaboration Workflow (Example):

[0184] Taking a complete logistics task transfer as an example, the collaborative work follows the process below:

[0185] 1. Once the aircraft arrives over the station, the edge recognition camera begins to work, identifies the aircraft, and guides it to land.

[0186] II. After the aircraft lands stably on the landing platform, the local controller confirms safety and triggers the automatic door system to close, forming a closed work cabin.

[0187] III. The local controller coordinates the charging and swapping modules to replenish the aircraft's energy according to the mission instructions, and at the same time directs the material interface to complete the unloading or loading of cargo containers.

[0188] IV. Throughout the process, the environmental monitor continuously monitors the data to ensure the stability of the cabin environment.

[0189] V. After all operations are completed, the automatic door system opens, and the aircraft takes off to continue the next leg of its journey.

[0190] In summary, the following was achieved:

[0191] Unmanned operation and remote monitoring: Through highly automated modular combinations, no manual intervention is required at the site.

[0192] Edge intelligence and central intelligence work together: Edge recognition cameras and local controllers constitute edge intelligence nodes, which handle high-frequency, real-time local tasks and form an efficient division of labor with central intelligence.

[0193] The foundation of the dynamic adaptation model: standardized take-off and landing platforms, charging and swapping modules, and material interfaces are prerequisites for supporting the rapid access and service of various drone models, providing physical possibilities for dynamic resource matching.

[0194] The establishment of low-altitude unmanned service stations ensures that the service stations can independently, safely, and reliably complete all necessary interactions with the aircraft under unmanned conditions, and is the physical foundation for the entire intelligent monitoring system to be implemented and operated.

[0195] In one exemplary embodiment, the system supports automatically selecting a backup site for task migration and restart based on an "N+1" redundancy mechanism when a failure occurs during task execution.

[0196] Multi-site redundancy and fault tolerance mechanisms are achieved through:

[0197] (a) Each site is equipped with “N+1” redundant replaceable nodes;

[0198] (b) Supports loosely coupled task-site binding relationship, and the system can replace it when the operation is interrupted.

[0199] In one exemplary embodiment, the intelligent control center adopts a distributed deployment architecture to support centralized scheduling and partitioned management of service sites in multiple regions.

[0200] In one exemplary embodiment, a method for the full-process operation management of low-altitude unmanned service stations is applied to an intelligent monitoring system for low-altitude unmanned service stations. The method includes:

[0201] System initialization steps: Complete service site registration, device status detection, and task readiness assessment;

[0202] Task reception and decomposition steps: Receive task requests and decompose tasks and plan flight segments according to task type and resource status;

[0203] Dynamic matching and scheduling steps: Based on a multi-dimensional adaptation scoring model, available sites and aircraft resources are dynamically matched, and scheduling instructions are generated;

[0204] Aircraft docking and execution steps: After the aircraft arrives at the station, it completes automatic landing, energy replenishment or material loading and unloading operations through edge intelligent recognition and guidance;

[0205] Status monitoring and feedback steps: Real-time monitoring of task execution status and equipment parameters, uploading to the control center and dynamically adjusting scheduling strategies;

[0206] Anomaly handling and recovery steps: When an anomaly is detected, the fault tolerance mechanism is automatically triggered, the task is migrated to a backup site and execution is resumed.

[0207] In an exemplary embodiment, the dynamic matching and scheduling step is constructed as a multi-objective optimization problem, whose objective function and constraints are defined as follows:

[0208] Objective 1: Maximize the overall compatibility between all assigned tasks and devices within the system. The specific expression for this is as follows:

[0209] ;

[0210] Objective 2: Minimize the average response latency of all tasks in the system, as expressed in the following expression:

[0211] ;

[0212] The constraints are as follows:

[0213] Unique allocation constraint: Each task must be assigned to one and only one available device, as expressed below:

[0214] ;

[0215] Capacity load constraint: The total demand of all tasks assigned to each device must not exceed the maximum service capacity of that device, as expressed below:

[0216] ;

[0217] in:

[0218] This indicates the total number of tasks to be assigned.

[0219] This indicates the total number of available devices (including drones and service site resources);

[0220] It is a binary decision variable, when the... The task is assigned to the first The value is 1 if the device is specific, otherwise it is 0.

[0221] Indicates the first The time delay from when a task is ready to when it begins execution;

[0222] Indicates the first The resource requirement vector for each task may include the expected flight duration, computational resource consumption, and data throughput requirements.

[0223] Indicates the first The maximum service capacity vector of a device is the upper limit of the flight mission load, computing load, and communication load that it can withstand.

[0224] In one exemplary embodiment, such as Figure 4 As shown, the dynamic matching and scheduling algorithm in the full-process operation management method of low-altitude unmanned service stations is the core decision engine of the intelligent scheduling and control center. It realizes real-time, adaptive, and optimal matching and scheduling of multiple tasks and multiple station resources. It embodies a complete closed loop based on priority scheduling, multi-dimensional adaptability calculation, state-driven execution, and anomaly self-recovery, and is the core key to achieving efficient full-process operation management in this embodiment.

[0225] The overall process is as follows:

[0226] Ⅰ. Initialize task queue T

[0227] Specifically, when the system starts or runs, a global task queue T is initialized to store all received task requests.

[0228] II. Obtain the state of site set S

[0229] Specifically, the latest status data is obtained in real time or periodically from all unattended service sites (set S) to provide dynamic and accurate input for matching calculations.

[0230] III. Refresh Task Priority

[0231] Specifically, the urgency level of each task in task queue T is dynamically adjusted based on the real-time status of the task (such as approaching deadlines, user-manually increased priority, triggering of emergency events, etc.). ).

[0232] IV. Sort by priority T

[0233] Specifically, queue T is reordered according to the refreshed task priority to ensure that the system prioritizes the most urgent tasks and meets the needs of high timeliness scenarios.

[0234] V. Calculate the appropriate site for each task

[0235] Specifically, this is the core step of the algorithm. For each task (t) in the queue... i ), call the fit score function ( ), calculate its relationship with each available site (s j The function takes into account the task urgency (match score). ), site resource availability ( Spatial matching degree ( ) and equipment health status ( Factors from multiple dimensions, such as ( ).

[0236] VI. Judgment: Find available sites

[0237] Specifically, based on the adaptation calculation results, it is determined whether there are any usable sites with scores higher than the threshold that meet the basic requirements of the task.

[0238] Yes: The process enters the normal execution branch of the assigned task.

[0239] No: Mark the task as pending assignment, which may result in it entering a queue or triggering an alarm mechanism (such as notifying operations personnel to intervene).

[0240] VII. Assign tasks & update site status

[0241] Specifically, tasks are formally assigned to the best site (and associated aircraft) with the highest suitability score, and the status of that site is immediately updated in the system (such as being marked as busy, having its remaining power or load capacity reduced) to prevent resource conflicts.

[0242] VIII. Execute the task and listen for feedback.

[0243] Specifically, after the mission command is issued, the system continuously monitors the real-time feedback data from the aircraft and the station, including flight trajectory, docking progress, and operational status.

[0244] IX. Check: Task completed & Check: Exception

[0245] Specifically, this is a key decision point to ensure closed-loop processes and system robustness.

[0246] If the task is completed, it is removed from the queue.

[0247] If an anomaly is detected (such as equipment failure or task execution timeout), the task process is rescheduled. This demonstrates the loose coupling of the task-site binding relationship and the fault-tolerant switching capability in this embodiment.

[0248] X. Reschedule tasks

[0249] Specifically, the abnormal task is reinserted into task queue T and given a new priority (usually increased), waiting for the next scheduling cycle to rematch it with an available backup site.

[0250] XI. Update adaptation weights

[0251] Specifically, this is a step involving machine learning and adaptive optimization. The system periodically adjusts the fitness score function based on the execution results of a large number of historical tasks (such as success rate and time taken) and outlier data. The weighting coefficients (w) in ) k This allows scheduling strategies to continuously evolve and better adapt to the actual operating environment and business needs.

[0252] XII. Check: Task queue T is not empty

[0253] Specifically, the main loop controls the algorithm. As long as the task queue is not empty, the algorithm will continue to run, achieving 24 / 7 uninterrupted intelligent scheduling service.

[0254] In summary, the following was achieved:

[0255] Dynamic resource matching model: Through steps such as "calculating adapted sites" and "updating adapted weights", the theoretical model is transformed into executable dynamic adaptation logic.

[0256] End-to-end closed-loop management and anomaly self-recovery: The process, from task enqueuing to completion or rescheduling, forms a complete closed loop. A clearly defined "anomaly-rescheduling" path is the concrete manifestation of the "anomaly self-recovery" capability.

[0257] Centralized intelligent collaboration: This algorithm runs in the intelligent scheduling and control center and collaborates with edge intelligence (site-specific decision-making) to jointly ensure the stable operation of the system.

[0258] The dynamic matching and scheduling algorithm is not only a concretization of the scheduling method in the embodiment, but also a highly feasible technical solution. It clearly outlines the working mechanism of an intelligent decision-making system capable of coping with complex and ever-changing low-altitude operating environments. Through continuous perception, real-time calculation, iterative optimization, and proactive fault tolerance, it ensures the efficient and reliable operation of the low-altitude unmanned service station network.

[0259] In one exemplary embodiment, the method further includes data interaction with a civil aviation regulatory platform or an urban low-altitude airspace management system to ensure flight compliance and airspace coordination; and the operation management platform provides a user interface, API interface, data reports, and log export functions.

[0260] As an example, the specific diagnostic application scenarios in this embodiment can be as follows:

[0261] I. System Deployment and Initialization

[0262] Within the urban area, several unmanned service stations (NUPS) are pre-deployed. Each station is equipped with takeoff and landing capabilities, a power interface, identification functions, and edge intelligent processing capabilities. All stations are connected to the intelligent dispatch and control center via 5G / IoT. The dispatch center is deployed on the city's operation management platform or within the aviation management unit, possessing cross-regional flight mission dispatch and supervision capabilities. Upon system startup, it automatically completes resource registration, health checks, and mission readiness status assessments.

[0263] II. Flight Mission Execution Procedure (Taking Urban Low-Altitude Logistics as an Example)

[0264] II. I. Task Reception and Allocation

[0265] (a) The platform user or a third-party system submits a task request (such as express delivery or emergency supplies transfer).

[0266] (b) Based on mission timeliness, route planning and logistics rules, the control center breaks down the mission into multiple flight segments, each of which is executed by the aircraft between two stations.

[0267] II.2 Site Matching and Scheduling

[0268] (a) The system obtains the idle status, power consumption, current load, and historical health records of each site in real time;

[0269] (b) Use the fit rating model to perform optimal matching of candidate sites;

[0270] (c) The dispatching platform sends task instructions to the designated station. The instructions include the task number, expected aircraft ID, time window, and operation details.

[0271] II. III. Spacecraft Docking and Mission Execution

[0272] (a) When the drone arrives above the service station, the edge camera identifies the aircraft and guides it to land precisely;

[0273] (b) The hatch opens automatically, triggering the charging / battery swapping or material loading / unloading module to work;

[0274] (c) After the operation is completed, the hatch will close automatically, the aircraft will take off and continue to perform the next mission.

[0275] II. IV. Condition Monitoring and Feedback

[0276] (a) During flight and on-site operations, mission execution status, equipment parameters, environmental indicators, etc. are uploaded to the platform in real time;

[0277] (b) The control center automatically adjusts subsequent scheduling strategies based on feedback;

[0278] (c) All operations are recorded in the log database for easy review and analysis later.

[0279] III. Exception Handling and Task Recovery Mechanism

[0280] (a) If a station detects problems such as aircraft identification failure, abnormal power, or stuck hatch, the edge processing module will be automatically triggered to perform preliminary troubleshooting.

[0281] (b) Simultaneously, an abnormal signal is sent to the central platform, and the system schedules the nearest backup site to take over the current task;

[0282] (c) If the mission is interrupted, the system will readjust the flight path, update the station allocation, and coordinate airspace resources to ensure flight safety.

[0283] IV. Application Scenarios Examples

[0284] IV. I. Emergency Medical Supplies Distribution

[0285] In the event of a public health emergency, hospitals issue emergency delivery orders through the platform. The system then dispatches unmanned stations as transfer nodes based on the shortest route between the target hospital and the drug reserve center. The aircraft completes multiple delivery tasks without human intervention, significantly improving response speed.

[0286] IV. II. Drone parking and dispatching during peak urban traffic hours

[0287] During peak hours, due to airspace congestion or overlapping tasks, some aircraft need to temporarily dock, buffer, or standby. The system can dynamically allocate nearby vacant sites as temporary docking points for aircraft, which will then re-enter the work queue after dispatch instructions are issued.

[0288] V. System Security Design

[0289] (a) Communication uses TLS encrypted transmission;

[0290] (b) All stations support remote power-off, cabin locking, alarm and other mechanisms;

[0291] (c) The control center has a manual intervention interface for key processes;

[0292] (d) Support data interconnection with civil aviation regulatory and urban air traffic control systems to ensure legal and compliant flights.

[0293] The foregoing has only described certain exemplary embodiments of the present invention by way of illustration. Undoubtedly, those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the foregoing drawings and descriptions are illustrative in nature and should not be construed as limiting the scope of protection of the claims of the present invention.

Claims

1. An intelligent monitoring system for low-altitude unmanned service stations, characterized in that: include: The intelligent control center is used for global task scheduling, resource coordination, status monitoring, and remote management. Multiple service stations, each of which includes a take-off and landing platform, a door control device, a charging and swapping interface, and an identification module; The intelligent docking module for aircraft is used to enable the automatic landing, identification, charging, and mission data interaction of drones; The perception and safety system integrates video surveillance, anomaly detection, and AI recognition units, and is used to monitor the status of sites and aircraft. The operation management platform is used for task decomposition, execution monitoring, anomaly warning, and maintenance management. Among them, multiple service stations can be interconnected through a network and can perform task takeover and fault-tolerant switching.

2. The intelligent monitoring system for low-altitude unmanned service stations according to claim 1, characterized in that, The intelligent control center includes: The task lifecycle management module is used to achieve closed-loop control of the entire process of a task from creation to completion; The status awareness and anomaly response module is used to sense the status of devices and tasks in real time and automatically trigger notifications or recovery processes. The scheduling and resource matching module is used to dynamically match site and aircraft resources based on task characteristics and airspace constraints; Edge intelligent recognition module, which uses edge AI camera to identify aircraft and assist in landing guidance; The site self-organization and collaboration module enables multiple sites to perform task takeover and backup scheduling within the network.

3. The intelligent monitoring system for low-altitude unmanned service stations according to claim 2, characterized in that, The scheduling and resource matching module employs a multi-dimensional matching scheduling algorithm, which performs dynamic optimal matching by constructing a task-device compatibility scoring function. The scoring function is as follows: ； in, Indicates the first The first task and the first Overall compatibility score between devices; Indicates the first The urgency of each task; Indicates the first Resource availability index of each device; Indicates the first The device and the first The spatial matching degree between tasks is an inverse proportional function calculated based on the Euclidean distance between the current location of the device and the location or path of the task target. The closer the distance, the higher the value, and it is normalized to the interval [0, 1]. Indicates the first The health status score of each device is a normalized value calculated based on its historical failure rate, current self-inspection status, and life prediction of key components. The higher the value, the better the health status. These are the weighting coefficients for the corresponding indicators, and they satisfy... .

4. The intelligent monitoring system for low-altitude unmanned service stations according to claim 1, characterized in that, The intelligent docking module for aircraft supports automatic identification of various drone models and dynamically adapts to drones of different sizes through an adjustable mechanical interface.

5. The intelligent monitoring system for low-altitude unmanned service stations according to claim 1, characterized in that, The service station achieves local decision-making through edge AI units, supporting aircraft self-identification, anomaly warning, and remote intervention control.

6. The intelligent monitoring system for low-altitude unmanned service stations according to claim 1, characterized in that, The system supports automatically selecting a backup site for task migration and restart based on the "N+1" redundancy mechanism when a failure occurs during task execution.

7. The intelligent monitoring system for low-altitude unmanned service stations according to claim 1, characterized in that, The intelligent control center adopts a distributed deployment architecture, supporting centralized scheduling and partitioned management of service sites in multiple regions.

8. A method for the full-process operation and management of low-altitude unattended service stations, applied to the intelligent monitoring system of low-altitude unattended service stations as described in any one of claims 1-7, characterized in that, The method specifically includes the following steps: System initialization steps: Complete service site registration, device status detection, and task readiness assessment; Task reception and decomposition steps: Receive task requests and decompose tasks and plan flight segments according to task type and resource status; Dynamic matching and scheduling steps: Based on a multi-dimensional adaptation scoring model, available sites and aircraft resources are dynamically matched, and scheduling instructions are generated; Aircraft docking and execution steps: After the aircraft arrives at the station, it completes automatic landing, energy replenishment or material loading and unloading operations through edge intelligent recognition and guidance; Status monitoring and feedback steps: Real-time monitoring of task execution status and equipment parameters, uploading to the control center and dynamically adjusting scheduling strategies; Anomaly handling and recovery steps: When an anomaly is detected, the fault tolerance mechanism is automatically triggered, the task is migrated to a backup site and execution is resumed.

9. The full-process operation management method for low-altitude unmanned service stations according to claim 8, characterized in that, The dynamic matching and scheduling steps are constructed as a multi-objective optimization problem, whose objective function and constraints are defined as follows: Objective 1: Maximize the overall compatibility between all assigned tasks and devices within the system. The specific expression for this is as follows: ； Objective 2: Minimize the average response latency of all tasks in the system, as expressed in the following expression: ； The constraints are as follows: Unique allocation constraint: Each task must be assigned to one and only one available device, as expressed below: ； Capacity load constraint: The total demand of all tasks assigned to each device must not exceed the maximum service capacity of that device, as expressed below: ； in: This indicates the total number of tasks to be assigned. Indicates the total number of available devices; It is a binary decision variable, when the... The task is assigned to the first The value is 1 if the device is specific, otherwise it is 0. Indicates the first The time delay from when a task is ready to when it begins execution; Indicates the first The resource requirement vector for each task includes the expected flight duration, computational resource consumption, and data throughput requirements. Indicates the first The maximum service capacity vector of a device is the upper limit of the flight mission load, computing load, and communication load that it can withstand.

10. The full-process operation management method for low-altitude unmanned service stations according to claim 8, characterized in that, The method also includes data interaction with civil aviation regulatory platforms or urban low-altitude airspace management systems to ensure flight compliance and airspace coordination; and the operation management platform can provide user interface, API interface, data report and log export functions.

Images