Emergency response decision-making method and system based on air-ground cooperation

Abstract

The application discloses a kind of based on air-ground cooperation's emergency response decision-making method and system of sudden event, the method includes: in response to sudden event alarm, based on GIS and Internet of Things technology collection field perception data and constructs dynamic task situation map;Acquire the state information of ground rescue unit and unmanned aerial vehicle platform, respectively generate the candidate task set of air-ground both sides;Calculate the task support degree of air task to ground task and the task dependency of ground task to air task, based on this joint screening first target task and second target task;For ground rescue unit planning first travel path, and for unmanned aerial vehicle platform planning second detection path along the continuous detection of first travel path front;Whether first travel path is safe according to the front situation information of real-time feedback, if not satisfied, then re-planning and synchronously updating second detection path.The application realizes the task coupling of air-ground cooperation, improves the timeliness and accuracy of emergency response.

Description

Technical Field

[0001] This application relates to the field of emergency response technology, specifically to an emergency response decision-making method and system based on air-ground coordination. Background Technology

[0002] Currently, many companies are dedicated to developing integrated emergency response and drill solutions based on GIS (Geographic Information System), GPS (Global Positioning System), and IoT technologies, including integrated emergency network platforms, earthquake monitoring systems, and intelligent multi-rotor UAV platforms. In practical applications, while these technologies have played significant roles in geographic information monitoring, earthquake early warning, and aerial patrols, they often operate relatively independently when facing complex emergencies, making it difficult to form an effective collaborative response.

[0003] For example, after an earthquake, earthquake monitoring systems can output damage assessment data for buildings in the affected area. However, this data is usually presented in the form of static reports or background layers, making it difficult to transmit it to ground rescue units rushing to the scene in a timely manner. Although multi-rotor drone platforms have flexible aerial patrol capabilities, their flight paths are often manually controlled by flight operators. After the collected on-site video data is transmitted back to the command center, it still requires manual analysis by command personnel before issuing instructions to ground rescue teams. This serial information flow link of data collection-central analysis-ground execution has significant information delays and decision-making lags when facing complex scenarios with frequent secondary disasters such as earthquakes. The main reason for this problem is that existing technologies have failed to deeply couple earthquake monitoring and early warning, aerial mobile patrols, and ground rescue operations at the task level, lacking a decision-making mechanism that can coordinate and plan the action paths and task loads of both air and ground forces in real time based on dynamic on-site data.

[0004] Therefore, there is an urgent need for a solution for emergency response decision-making based on air-ground coordination. Summary of the Invention

[0005] This application provides an emergency response decision-making method and system based on air-ground coordination, which at least addresses the problems existing in the prior art.

[0006] A first aspect of this application provides an emergency response decision-making method for sudden events based on air-ground coordination, comprising the following steps: S1, based on geographic information system and Internet of Things technology, responds to emergency alarms, collects on-site perception data of the incident area, and constructs a dynamic mission situation map in combination with the incident location; S2, acquire the first status information of at least one ground rescue unit within the response range, and the second status information of at least one multi-rotor drone platform; S3, based on the dynamic mission situation map, the first state information and the second state information, generate a first candidate task set for the ground rescue unit and a second candidate task set for the multi-rotor UAV platform respectively; S4, calculate the mission support degree of each air mission in the second candidate mission set to each ground mission in the first candidate mission set, and the mission dependence degree of each ground mission in the first candidate mission set to each air mission in the second candidate mission set. S5. Based on the task support degree and task dependency degree, jointly select the first target task and the second target task with the maximum coupling correlation from the first candidate task set and the second candidate task set. S6, according to the first objective task, a first travel path and a first disposal strategy are planned for the ground rescue unit, and according to the second objective task, a second detection path and a second detection strategy are planned for the multi-rotor UAV platform. The second detection path is used to enable the multi-rotor UAV platform to continuously detect along the area in front of the first travel path and generate forward situation information. S7, based on the real-time situational information transmitted back from the multi-rotor UAV platform, determines whether the first travel path meets the safe passage conditions. If not, the first travel path is replanned and the second detection path is updated simultaneously.

[0007] In this embodiment, the entire process—including constructing a dynamic mission situation map, generating candidate mission sets for both air and ground forces, calculating mission support and dependency, jointly screening coupled missions, planning collaborative paths, and dynamically replanning based on real-time situational awareness—solves the technical problems of independent operation of air and ground platforms and delayed information flow in existing emergency responses. This method achieves deep coupling of air and ground missions, enabling ground rescue units to dynamically avoid travel risks based on real-time aerial detection information, significantly improving the timeliness and accuracy of emergency responses.

[0008] In some embodiments of this application, the construction of a dynamic task situation map in step S1 further includes: integrating the building damage assessment data of the earthquake zone output by the earthquake monitoring system, the environmental parameters collected by ground fixed sensors, and the crowdsourced information reported by mobile terminals around the incident area, dynamically marking the road network accessibility and building collapse risk areas in the ground traffic environment, and dynamically marking the airspace control area and airflow disturbance risk area in the air flight environment.

[0009] In this embodiment, by integrating building damage assessment data from the earthquake monitoring system with crowdsourced information reported by ground-based fixed sensors and mobile terminals, the technical problems of traditional situation maps—such as limited information sources and insufficient dynamic update capabilities—are solved. This solution incorporates the unique building collapse risk of earthquake disasters into the ground traffic environment assessment, while dynamically marking airflow disturbance risks and airspace control areas in the airspace. This makes the constructed dynamic mission situation map more closely resemble real disaster scenarios, providing a more accurate and reliable data foundation for subsequent air-ground collaborative decision-making.

[0010] In some embodiments of this application, the first state information obtained in step S2 further includes: real-time collection of the current location of the ground rescue unit, the remaining fuel on the vehicle, and the type and remaining quantity of the vehicle-mounted rescue supplies through the vehicle-mounted IoT terminal; the acquisition of the second state information further includes: real-time transmission of the current location, flight speed, remaining battery power, and the type and working mode of the currently mounted functional modules of the multi-rotor UAV platform through the airborne flight control system and mission payload system of the multi-rotor UAV platform, wherein the type of the functional module is at least one of visible light imaging module, infrared thermal imaging module, gas detection module, or airborne broadcasting module; the generation of the first candidate task set and the second candidate task set in step S3 further includes: dynamically estimating the maximum driving range of the ground rescue unit and the maximum flight range of the multi-rotor UAV platform based on the remaining fuel on the ground rescue unit and the remaining battery power of the multi-rotor UAV platform, and using the estimation results as the endurance constraints for generating the first candidate task set and the second candidate task set.

[0011] In this embodiment, by refining the status information collected from the ground rescue unit and the UAV platform, and dynamically estimating the maximum driving range and flight range based on remaining fuel and battery power as endurance constraints for mission generation, the technical problem of neglecting the real-time status of the execution unit during mission planning, which leads to the inability to actually execute the planned mission, is solved. This solution improves the executability and reliability of the candidate mission set.

[0012] In some embodiments of this application, calculating the mission support degree in step S4 specifically includes: extracting the detection coverage and data backhaul bandwidth of the airborne detection module for each air mission in the second candidate mission set; extracting the expected travel path and spatial range of the on-site handling area for each ground mission in the first candidate mission set; and generating a quantitative value of the mission support degree based on the spatial overlap between the detection coverage and the expected travel path, as well as the matching degree between the data backhaul bandwidth and the real-time data transmission volume required by the on-site handling area.

[0013] In this embodiment, the spatial overlap between the detection coverage area and the expected ground travel path, as well as the matching degree between the data return bandwidth and the data transmission volume required by the on-site handling area, are used as quantitative indicators. This solves the technical problem of relying on manual experience and lacking objective quantitative basis in the air-to-ground mission matching process. This scheme gives the calculation of mission support a clear geometric and physical meaning, providing an objective decision-making basis for subsequent joint screening.

[0014] In some embodiments of this application, step S5, which involves jointly selecting the first and second target tasks with the maximum coupling correlation, includes: constructing a bipartite graph model with a ground rescue unit as the first node and a multi-rotor UAV platform as the second node, where the first node corresponds to ground tasks in the first candidate task set and the second node corresponds to air tasks in the second candidate task set; using the weighted sum of the calculated task support and task dependency as the weight value of the connecting edge; and employing the Hungarian algorithm to solve the bipartite graph model for maximum weight matching to determine the optimal combination of the first and second target tasks.

[0015] In this embodiment, a bipartite graph model is constructed and the Hungarian algorithm is used to solve the maximum weight matching problem, thus addressing the technical challenge of achieving globally optimal matching of air-to-ground tasks in scenarios with multiple ground units and multiple UAV platforms. This solution selects the target task combination with the highest overall coupling correlation from the candidate task set, avoiding resource waste caused by local optima or random matching.

[0016] In some embodiments of this application, step S6, which plans a second detection path for the multi-rotor UAV platform according to the second objective mission, includes: determining at least one key node location as a key monitoring waypoint based on the first travel path of the ground rescue unit; and planning a second detection path that can sequentially fly over the key monitoring waypoints and meets the endurance constraints, by combining the remaining endurance of the multi-rotor UAV platform with the no-fly zone information in the dynamic mission situation map.

[0017] In this embodiment, by determining key node locations as key monitoring waypoints based on the first travel path of the ground rescue unit, and combining the UAV's endurance and no-fly zone information to plan a second detection path, the technical problem of how to achieve accurate and efficient forward detection under limited UAV detection resources is solved. This solution allows the UAV to focus on the key nodes on the ground rescue path that require real-time perception, balancing detection efficiency with endurance constraints.

[0018] In some embodiments of this application, step S7, determining whether the first travel path meets the safe passage conditions, includes: parsing the situational information ahead and extracting the locations of newly added obstacles, the boundaries of secondary disaster spread trends, or the hotspots of trapped personnel distribution; performing spatial overlay analysis on the extracted information and the first travel path; if the locations of newly added obstacles or the boundaries of secondary disaster spread trends intersect with the first travel path, or if the hotspots of trapped personnel distribution deviate from the first travel path by more than a preset deviation threshold, then it is determined that the safe passage conditions are not met.

[0019] In this embodiment, by spatially overlaying and analyzing the locations of obstacles, the boundaries of secondary disaster spread trends, and the hotspots of trapped personnel distribution in the forward situation information with the first travel path, the technical problem of how to objectively and quantitatively determine whether a travel path is safe and feasible is solved. This solution transforms ambiguous safety hazards into clear spatial intersection judgments or deviation threshold judgments, providing a clear technical boundary for triggering path replanning.

[0020] In some embodiments of this application, after replanning the first travel path in step S7, the method further includes: generating a path change instruction based on the replanned first travel path; sending the path change instruction to the multi-rotor UAV platform and driving the multi-rotor UAV platform to adjust the second detection path according to the path change instruction, so as to maintain continuous detection of the area ahead of the replanned first travel path.

[0021] In this embodiment, by generating a path change command after replanning the first travel path and driving the UAV to synchronously update the second detection path, the technical problem of the break in air-ground coordination and the disconnect between UAV detection and ground rescue after path change is solved. This solution ensures that no matter how the ground path is dynamically adjusted, the UAV can always maintain continuous detection coverage of the area in front of the ground rescue unit.

[0022] In some embodiments of this application, during the process of the ground rescue unit moving along the first travel path, the method further includes: dynamically adjusting the personnel evacuation guidance range or the activation timing of rescue equipment in the first disposal strategy based on the forward situation information transmitted back in real time by the multi-rotor UAV platform; the dynamic adjustment specifically includes: when the forward situation information is parsed to indicate that there is a newly added area of ​​trapped personnel gathering ahead of the first travel path, triggering an adaptive expansion of the boundary of the personnel evacuation guidance range, so that the expanded evacuation guidance range includes the newly added area of ​​trapped personnel gathering, and broadcasting a temporary guidance path to the newly added area of ​​trapped personnel gathering through the air broadcast module carried by the multi-rotor UAV platform; when the forward situation information is parsed to indicate that there is a secondary disaster spread risk area ahead of the first travel path, triggering an advance adjustment of the activation timing of rescue equipment, so that the ground rescue unit can pre-activate the corresponding protective equipment or demolition equipment before arriving at the secondary disaster spread risk area; wherein, the dynamic adjustment result of the first disposal strategy is fed back to step S7, and if the dynamically adjusted personnel evacuation guidance range or the activation timing of rescue equipment conflicts with the first travel path in terms of spatial constraints, it is used as an additional judgment condition for triggering the replanning of the first travel path.

[0023] In this embodiment, by dynamically adjusting the personnel evacuation guidance range or the activation timing of rescue equipment based on forward situation information, and feeding the adjustment results back to the path replanning judgment, the technical problems of static plans being unable to adapt to dynamic changes on-site and the disconnect between the response strategy and the on-site situation are solved. This solution achieves dynamic collaborative optimization of the response strategy and the travel path, enabling the ground rescue unit's response measures to adaptively adjust based on real-time perceived information.

[0024] In some embodiments of this application, step S5, which involves jointly selecting the first and second target tasks with the highest coupling correlation, further includes: acquiring real-time building damage assessment data of the seismic zone output by the earthquake monitoring system, wherein the building damage assessment data includes the damage level and collapse risk probability of each building; superimposing the building damage assessment data as a spatial weighting factor onto the dynamic task situation map, and applying risk weighting to the on-site handling areas involved in each ground task in the first candidate task set; recalculating the task dependence of each ground task in the first candidate task set on each aerial task in the second candidate task set based on the risk-weighted on-site handling areas, so that ground tasks located in areas with high damage levels or high collapse risk probabilities receive higher task dependence weights; and re-performing the joint screening based on the updated task dependence and task support, so that aerial tasks with high collapse risk area detection capabilities are preferentially coupled and matched with ground tasks handling high-risk areas.

[0025] In this embodiment, by using the building damage assessment data output by the earthquake monitoring system as a spatial weighting factor, ground tasks located in areas with high damage levels or high collapse risk are assigned a higher task dependence weight. This solves the technical problem of prioritizing the matching of aerial resources with corresponding detection capabilities to high-risk areas. This solution enables UAVs with high collapse risk detection capabilities to be preferentially coupled and matched with ground rescue units handling high-risk areas, achieving precise resource alignment of "high risk - high capability," avoiding mismatch of detection resources in key areas, and significantly improving the reliability of emergency response in high-risk scenarios.

[0026] In some embodiments of this application, step S6, which plans a second detection path for the multi-rotor UAV platform according to the second target task, further includes: generating at least one aerial hovering broadcast waypoint in the second detection path according to the personnel evacuation guidance range in the first disposal strategy; when the multi-rotor UAV platform arrives at the aerial hovering broadcast waypoint, triggering the airborne aerial broadcast module to switch to directional broadcast mode, and automatically adjusting the sound wave radiation pointing angle of the aerial broadcast module according to the spatial orientation of the hotspot area of ​​the trapped personnel extracted from the dynamic task situation map, so that the main lobe direction of the broadcast sound wave is aligned with the hotspot area of ​​the trapped personnel; at the same time, dynamically calculating the continuous hovering duration of the aerial hovering broadcast waypoint according to the real-time travel speed and expected arrival time of the ground rescue unit, so that the broadcast content of the aerial broadcast module covers the complete time window from the current moment to the expected arrival time of the ground rescue unit, and triggering the multi-rotor UAV platform to exit the hovering state and continue to perform the detection task along the second detection path when a preset time threshold is set before the ground rescue unit arrives at the hotspot area of ​​the trapped personnel.

[0027] In this embodiment, by generating hovering broadcast waypoints in the second detection path of the UAV and automatically adjusting the broadcast sound wave radiation direction angle according to the hotspot areas of the trapped personnel distribution, while dynamically calculating the hovering duration based on the estimated arrival time of the ground rescue unit, the technical problems of inaccurate coverage and the disconnect between broadcast timing and ground rescue progress in traditional aerial broadcasting are solved. This solution enables the UAV to accurately cover areas where trapped personnel are gathered in a directional broadcast mode at key locations, and the hovering duration is coordinated with the arrival time of the ground rescue unit, achieving seamless connection from detection and discovery to evacuation guidance, significantly improving the accuracy and effectiveness of personnel evacuation guidance.

[0028] A second aspect of this application provides an emergency response decision-making system based on air-ground coordination, which is used to implement the above-described emergency response decision-making method based on air-ground coordination, including: The situation construction module is used to collect on-site perception data of the incident area in response to emergency alarms based on geographic information system and Internet of Things technology, and to build a dynamic task situation map in combination with the incident location. The status acquisition module is used to acquire the first status information of at least one ground rescue unit within the response range, and the second status information of at least one multi-rotor drone platform; The task generation module is used to generate a first candidate task set for ground rescue units and a second candidate task set for multi-rotor UAV platforms based on the dynamic task situation map, first state information and second state information. The coupled calculation module is used to calculate the mission support degree of each air mission in the second candidate mission set to each ground mission in the first candidate mission set, and the mission dependency degree of each ground mission in the first candidate mission set to each air mission in the second candidate mission set. The collaborative decision-making module is used to jointly select the first and second target tasks with the greatest coupling correlation from the first and second candidate task sets based on task support and task dependency. The path planning module is used to plan a first travel path and a first disposal strategy for the ground rescue unit according to the first objective task, and to plan a second detection path and a second detection strategy for the multi-rotor UAV platform according to the second objective task. The second detection path is used to enable the multi-rotor UAV platform to continuously detect the area in front of the first travel path and generate forward situation information. The dynamic adjustment module is used to determine whether the first travel path meets the safe passage conditions based on the real-time situation information transmitted back by the multi-rotor UAV platform. If it does not meet the conditions, the first travel path is replanned and the second detection path is updated simultaneously.

[0029] In the above embodiments, a complete air-ground collaborative emergency response decision-making system was constructed, enabling earthquake monitoring data, UAV perception data, and ground rescue unit status to be deeply integrated and collaboratively processed within the same system framework, providing an integrated decision-making platform with air-ground mission coupling capabilities for emergency command. Attached Figure Description

[0030] Figure 1 A flowchart illustrating an emergency response decision-making method based on air-ground coordination provided in an embodiment of this application; Figure 2 A schematic diagram of an emergency response decision-making system based on air-ground coordination provided in an embodiment of this application; Figure 3 This is a schematic diagram of an electronic device provided in an embodiment of this application. Detailed Implementation

[0031] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0032] To make the purpose, technical solution, and advantages of this application clearer, the following will be described in conjunction with the appendix. Figure 1-3 The following is an explanation using specific examples.

[0033] Currently, with the acceleration of urbanization and the frequent occurrence of extreme weather, various emergencies place higher demands on the efficiency of emergency response. In practical applications, Geographic Information System (GIS), Global Positioning System (GPS), and Internet of Things (IoT) technologies have been widely used in the field of emergency response. For example, GIS platforms are used to display geographic information of the affected area, and IoT sensors are used to collect on-site environmental data. However, in existing technologies, these systems often operate relatively independently, making it difficult to form an effective collaborative response force. For example, after an earthquake, the earthquake monitoring system can output damage assessment data of buildings in the affected area, but this data is usually presented in the form of static reports or background layers, making it difficult to transmit to ground rescue units rushing to the scene in a timely manner. Although multi-rotor drone platforms have flexible aerial patrol capabilities, their flight paths are often manually controlled by flight controllers. After the collected on-site video data is transmitted back to the command center, it still needs to be manually analyzed by the command personnel before issuing instructions to the ground rescue teams. This serial information flow link of data collection, central analysis, and ground execution has significant information delays and decision-making lags when facing complex scenarios with frequent secondary disasters such as earthquakes.

[0034] In-depth analysis reveals that the primary cause of this problem lies in the failure of existing technologies to deeply couple earthquake monitoring and early warning, aerial mobile reconnaissance, and ground rescue operations at the mission level. On one hand, ground rescue units often plan their routes based on static map information before deployment, lacking the ability to perceive real-time dynamic changes ahead, such as sudden road collapses, the spread of secondary disasters, or changes in the distribution of trapped personnel. On the other hand, the detection paths of aerial drone platforms and the movement paths of ground rescue units lack coordinated planning, resulting in a spatiotemporal misalignment between the on-site information collected from the air and the actual needs of ground rescue. This makes it difficult for aerial platforms to provide accurate and continuous situational support to ground units. This disconnect between air and ground tasks makes it difficult to adjust emergency response decisions in real time based on dynamic on-site data.

[0035] To address the aforementioned issues, this application proposes an emergency response decision-making method based on air-ground collaboration. The core concept of this method lies in constructing a collaborative decision-making framework that deeply couples ground rescue units with multi-rotor UAV platforms. When an emergency occurs, the system first collects on-site perception data based on GIS and IoT technologies to construct a dynamic mission situation map reflecting both the ground access environment and the air flight environment. Based on this, a coupling correlation model of air-ground tasks is established by calculating the mission support degree of air tasks to ground tasks and the mission dependence degree of ground tasks on air tasks. Based on this model, the combination of air-ground target tasks with the highest coupling correlation is jointly selected from the candidate task set, thereby planning a first travel path for the ground rescue unit and a second detection path for the multi-rotor UAV platform to continuously probe the area ahead of the first travel path. During the journey, the system dynamically judges the path safety based on the real-time situation information transmitted back by the UAV. If the safe passage conditions are not met, the path is replanned, and the UAV's detection path is updated synchronously. Through the above methods, a shift from data silos and manual analysis to an air-ground collaborative and task-coupled model has been achieved, which helps to improve the timeliness and accuracy of emergency response.

[0036] Please refer to Figure 1 , Figure 1 An emergency response decision-making method based on air-ground coordination, provided in one embodiment of this application, includes the following steps: S1, based on geographic information system and Internet of Things technology, responds to emergency alarms, collects on-site perception data of the incident area, and constructs a dynamic mission situation map in combination with the incident location; S2, acquire the first status information of at least one ground rescue unit within the response range, and the second status information of at least one multi-rotor drone platform; S3, based on the dynamic mission situation map, the first state information and the second state information, generate a first candidate task set for the ground rescue unit and a second candidate task set for the multi-rotor UAV platform respectively; S4, calculate the mission support degree of each air mission in the second candidate mission set to each ground mission in the first candidate mission set, and the mission dependence degree of each ground mission in the first candidate mission set to each air mission in the second candidate mission set. S5. Based on the task support degree and task dependency degree, jointly select the first target task and the second target task with the maximum coupling correlation from the first candidate task set and the second candidate task set. S6, according to the first objective task, a first travel path and a first disposal strategy are planned for the ground rescue unit, and according to the second objective task, a second detection path and a second detection strategy are planned for the multi-rotor UAV platform. The second detection path is used to enable the multi-rotor UAV platform to continuously detect along the area in front of the first travel path and generate forward situation information. S7, based on the real-time situational information transmitted back from the multi-rotor UAV platform, determines whether the first travel path meets the safe passage conditions. If not, the first travel path is replanned and the second detection path is updated simultaneously.

[0037] In the technical solution of this application embodiment, the dynamic mission situation map refers to a spatial information model constructed based on a Geographic Information System (GIS) that reflects the real-time dynamic changes in the incident area. This map not only includes basic geographic information data, such as road networks and building distribution, but also integrates on-site perception data collected through Internet of Things (IoT) technology. For example, it may include real-time accessibility information of road networks in the ground transportation environment, such as which road sections are impassable due to disasters; it may also include marking information of building collapse risk areas; and it may also include airspace control area information and airflow disturbance risk area information in the air flight environment. It is understood that this dynamic mission situation map can be dynamically adjusted as the on-site perception data is updated, providing real-time data support for subsequent decision-making. A ground rescue unit refers to a mobile device or personnel combination capable of performing emergency rescue tasks on the ground. Exemplarily, it may include various emergency rescue vehicles such as fire trucks, ambulances, and rescue vehicles, and may also include rescue personnel teams carrying rescue equipment. Ground rescue units possess autonomous movement capabilities and on-site handling capabilities.

[0038] The first state information refers to the data set reflecting the current status and mission execution capabilities of the ground rescue unit. Specifically, it may include real-time GPS coordinates collected via onboard IoT terminals, vehicle fuel levels, and a list of onboard rescue supplies and their remaining quantities. Additionally, it may include the rescue unit's speed and the operational status of onboard equipment. This information is used to assess the current available capabilities and mission scope of the ground rescue unit.

[0039] A multi-rotor unmanned aerial vehicle (UAV) platform refers to an unmanned aerial vehicle that uses multiple rotors for flight. This platform can be equipped with different functional modules depending on mission requirements, such as visible light imaging modules for on-site reconnaissance, infrared thermal imaging modules for nighttime or smoky environments, gas detection modules for detecting harmful gas concentrations, and aerial broadcasting modules for disseminating evacuation instructions to stranded personnel. Multi-rotor UAV platforms possess capabilities such as vertical takeoff and landing, hovering, and autonomous flight. Secondary status information refers to the data set reflecting the current flight status and mission payload capabilities of the multi-rotor UAV platform. Specifically, this includes real-time data transmitted back through the onboard flight control system, such as current position, flight speed, flight altitude, remaining battery power, and current available flight time. It also includes information transmitted back through the mission payload system regarding the type of currently equipped functional modules, such as whether a visible light imaging module or an infrared thermal imaging module is currently equipped, as well as the current operating mode and real-time data acquisition rate of these modules. This information is used to assess the UAV platform's current detection capabilities and the range of missions it can perform.

[0040] The first candidate task set refers to the set of theoretically executable task options generated for available ground rescue units based on the dynamic task situation map and the first-state information of the ground rescue units. Each candidate task may include information such as the target area, task type (e.g., personnel search and rescue or material transportation), and task priority. Similarly, the second candidate task set refers to the set of theoretically executable aerial task options generated for available multi-rotor UAV platforms. Each aerial task may include information such as the detection area, detection target, and the type of functional modules required. Task support is an indicator used to quantitatively evaluate the degree of support that aerial tasks can provide to ground tasks. The calculation of this indicator can consider multiple factors, such as the spatial overlap between the detection coverage of the airborne detection module of the aerial task and the expected path of the ground task. The higher the overlap, the longer the UAV can accompany the ground unit's movement, and the higher the support. Another example is the degree of matching between the data transmission bandwidth of the aerial task and the amount of real-time data transmission required by the ground task's on-site handling area. The higher the matching degree, the smoother the UAV can provide the ground unit with real-time video or data, and the higher the support. Mission dependency is an indicator used to quantitatively assess the degree to which ground missions require air mission support. Corresponding to mission support, mission dependency reflects the extent to which the execution of a ground mission requires the cooperation of a specific air mission. For example, if a ground mission requires entering an area with a high risk of building collapse, then this ground mission may have a high dependency on air missions with infrared detection or gas detection capabilities, because drones need to detect heat sources or harmful gas concentrations in the area beforehand to ensure the safety of ground personnel.

[0041] Maximum coupling correlation refers to the objective pursued when jointly selecting the first and second target missions. This indicator reflects the overall matching degree between the selected ground mission and air mission combination. By maximizing this correlation, air missions can provide strong support for their paired ground missions, while ground missions can also fully utilize the capabilities of their paired air missions, achieving optimized allocation of air and ground resources. The first travel path refers to the planned route for the ground rescue unit from its current location to the incident area or mission target area. The planning of this path needs to consider factors such as road network accessibility, traffic conditions, and risk areas in the dynamic mission situation map, striving to reach the scene as quickly as possible while ensuring safety. The first response strategy refers to the specific action plan executed by the ground rescue unit after arriving at the mission target area, such as the scope of personnel evacuation guidance, the timing and method of activating rescue equipment, and the order of rescuing trapped personnel. The second detection path refers to the flight route planned for the multi-rotor UAV platform for performing aerial detection missions. In this patent, the planning of the second detection path is closely related to the first travel path, aiming to enable the UAV to continuously detect along the area ahead of the first travel path. Specifically, the second detection path can be a flight trajectory that extends along the first path and is a certain distance ahead of the ground rescue unit, so as to ensure that the drone can detect obstacles on the road ahead, the spread of secondary disasters, or newly trapped people in advance, and generate this information as forward situation information to be transmitted back.

[0042] Forward situation information refers to data about the state of the area ahead of the first travel path, generated in real time by the multi-rotor UAV platform through its onboard functional modules during the execution of the second reconnaissance path. This information may include the location information of newly added obstacles, dynamic boundary information of secondary disasters such as fire spread or harmful gas diffusion, and information on hotspots of trapped personnel identified through image recognition technology. Safe passage conditions are a standard used to determine whether the first travel path is suitable for the ground rescue unit to continue along its original route at the current moment. This condition is determined based on the spatial relationship analysis between the forward situation information and the first travel path. For example, if the forward situation information shows that there is a newly added obstacle ahead of the first travel path and that obstacle happens to be located on the path, causing a complete blockage, then it can be determined that the safe passage conditions are not met. If the forward situation information shows that the boundary of the spread trend of secondary disasters is approaching the first travel path and is expected to intersect the path before the ground unit arrives, it can also be determined that the safe passage conditions are not met. If the situational information ahead shows that the hotspots of trapped personnel are far from the first route, for example, exceeding the preset deviation threshold, although this does not directly affect the safety of the route, it may affect the effectiveness of the rescue mission. Therefore, it can also be used as a reference for judging whether the route needs to be adjusted.

[0043] The above technical solution constructs a complete closed loop from information collection, task generation, collaborative matching to dynamic adjustment through a series of interrelated steps.

[0044] Step S1 first constructs a dynamic mission situation map, providing a spatiotemporal benchmark and real-time data foundation for the entire decision-making process. It integrates basic geographic information from the Geographic Information System (GIS) with dynamic on-site perception data collected by IoT technology, ensuring that all subsequent decisions are based on an accurate understanding of the current situation in the incident area. Without this foundation, subsequent path planning and task matching would lack a realistic basis. Based on the dynamic mission situation map, Step S2 acquires real-time status information of the ground rescue unit and the multi-rotor UAV platform. This step clarifies available execution resources and their current capability boundaries, such as fuel reserves and remaining battery power, setting feasibility boundaries for subsequent task generation. Step S3 generates candidate task sets for both air and ground operations based on map information and resource status information. This process transforms "what might be done" into "specific task options," providing a pool of candidates for subsequent collaborative matching. Step S4 calculates task support and task dependency, a core quantitative step for achieving air-ground collaboration. It no longer treats air and ground tasks as two isolated categories but establishes a correlation between them through quantitative indicators. Support level measures a unit's contribution to the ground from an air perspective, while dependence level measures its demand for air support from a ground perspective. Together, they form a quantitative bridge for air-to-ground mission coupling. Step S5 uses these quantitative indicators to jointly select target mission combinations with the highest coupling correlation, ensuring that the selected air missions can support their paired ground missions to the greatest extent possible, while the ground missions can also fully utilize the capabilities of their paired air units, achieving optimized resource allocation. Step S6 plans ground travel paths and air reconnaissance paths according to the selected target missions, and clearly defines the collaborative relationship between the air reconnaissance paths and ground travel paths—that is, the air unit needs to continuously reconnoiter the area ahead of the ground path. This collaborative planning ensures that air reconnaissance is no longer aimless wide-area patrols, but precisely focuses on the area that the ground unit is about to pass through, ensuring that the ground unit can obtain real-time situational awareness. Step S7 constitutes a dynamic feedback adjustment link. During its journey, the ground unit continuously receives real-time situational information from the air unit and uses this information to determine whether the original path is still safe and feasible. Once a risk is detected ahead, the system can promptly replan the ground path and simultaneously update the aerial reconnaissance path, allowing the aerial units to continue providing forward reconnaissance for the new path. This closed-loop mechanism enables the entire air-ground collaborative system to adapt to dynamic changes on the ground.

[0045] Through the close coordination of the above steps, this technical solution achieves a transformation from static contingency planning to dynamic collaboration. A dynamic task situation map provides the foundation for perception; status information and candidate task sets clarify resources and options; the calculation of support and dependency establishes air-ground correlation; the selection of the maximum coupling correlation achieves optimized matching; collaborative path planning implements task allocation; and dynamic replanning based on real-time situational awareness endows the system with adaptability. These features are interdependent and progressive, collectively constituting a complete air-ground collaborative emergency response decision-making solution.

[0046] In some embodiments of this application, the construction of a dynamic task situation map in step S1 further includes: integrating the building damage assessment data of the earthquake zone output by the earthquake monitoring system, the environmental parameters collected by ground fixed sensors, and the crowdsourced information reported by mobile terminals around the incident area, dynamically marking the road network accessibility and building collapse risk areas in the ground traffic environment, and dynamically marking the airspace control area and airflow disturbance risk area in the air flight environment.

[0047] In constructing a dynamic task situation map, this application embodiment integrates multiple data sources to improve the map's accuracy and dynamism. The earthquake monitoring system outputs building damage assessment data for the earthquake zone, which refers to information about the degree of damage to buildings after an earthquake, generated through an earthquake monitoring network and building structure analysis model. This may include, for example, the damage level classification of each building, such as minor, moderate, or severe damage, as well as the probability value of building collapse. Environmental parameters collected by ground-based fixed sensors refer to real-time data transmitted from various IoT sensors pre-deployed around the incident area, such as traffic flow monitoring data on roads, pressure monitoring data of underground pipelines, or environmental temperature and humidity data. Crowdsourced information reported by mobile terminals around the incident area refers to on-site information voluntarily or automatically reported by people near the incident area using smartphones or other mobile devices with their built-in GPS modules and applications, such as photos of road flooding and descriptions of congested road sections. Based on the fusion of the above multi-source data, the accessibility of the road network in the ground traffic environment is dynamically marked. Road network accessibility refers to the actual passability of each road segment in the road network, such as being marked as unobstructed, slow-moving, or interrupted. Areas with building collapse risk are dynamically marked, that is, the areas of buildings with a high risk of collapse are marked on the map. Airspace control areas in the aerial flight environment are dynamically marked. Airspace control areas refer to the airspace where drone flights are prohibited or restricted for safety or management reasons, such as airport airspace protection zones and temporary control zones for major events. Areas with airflow disturbance risk are dynamically marked. Areas with airflow disturbance risk refer to areas where drone flights may be unstable due to terrain or meteorological conditions, such as strong wind areas around tall buildings or turbulent areas in valleys.

[0048] In some embodiments of this application, the first state information obtained in step S2 further includes: real-time collection of the current location of the ground rescue unit, the remaining fuel on the vehicle, and the type and remaining quantity of the vehicle-mounted rescue supplies through the vehicle-mounted IoT terminal; the acquisition of the second state information further includes: real-time transmission of the current location, flight speed, remaining battery power, and the type and working mode of the currently mounted functional modules of the multi-rotor UAV platform through the airborne flight control system and mission payload system of the multi-rotor UAV platform, wherein the type of the functional module is at least one of visible light imaging module, infrared thermal imaging module, gas detection module, or airborne broadcasting module; the generation of the first candidate task set and the second candidate task set in step S3 further includes: dynamically estimating the maximum driving range of the ground rescue unit and the maximum flight range of the multi-rotor UAV platform based on the remaining fuel on the ground rescue unit and the remaining battery power of the multi-rotor UAV platform, and using the estimation results as the endurance constraints for generating the first candidate task set and the second candidate task set.

[0049] In acquiring the status information of the ground rescue unit and the drone platform, this application embodiment employs a specific data acquisition method. The vehicle-mounted IoT terminal refers to a device installed on the ground rescue unit, possessing wireless communication and data acquisition capabilities. It can collect and upload the rescue unit's own status parameters in real time. The current location collected by this terminal is the GPS coordinate of the rescue unit; the vehicle fuel remaining amount refers to the amount of fuel remaining in the vehicle's fuel tank; the type and remaining quantity of vehicle-mounted rescue supplies refer to the specific types and remaining quantities of various rescue items carried by the vehicle, such as demolition tools, first-aid medicines, and emergency food. For the multi-rotor drone platform, the airborne flight control system refers to the drone's flight control core, responsible for collecting data such as flight attitude, speed, and position, and transmitting it back in real time via a communication link; the mission payload system refers to the mission execution module carried by the drone, such as a visible light imaging module, an infrared thermal imaging module, a gas detection module, or an airborne broadcasting module. The mission payload system is responsible for collecting detection data and transmitting its own working status, such as whether the current working mode is photo mode or video recording mode, and the data acquisition rate. When generating the first and second candidate task sets, the maximum driving range of the ground rescue unit can be dynamically estimated based on its remaining onboard fuel and parameters such as fuel consumption per 100 kilometers. Similarly, the maximum flight range of the multi-rotor UAV platform can be dynamically estimated based on its remaining battery power and flight power consumption. These two estimates serve as endurance constraints, meaning that any task exceeding the unit's current energy replenishment capability will be automatically excluded from the candidate task set, ensuring that all generated tasks are within the unit's capability to execute.

[0050] In some embodiments of this application, calculating the mission support degree in step S4 specifically includes: extracting the detection coverage and data backhaul bandwidth of the airborne detection module for each air mission in the second candidate mission set; extracting the expected travel path and spatial range of the on-site handling area for each ground mission in the first candidate mission set; and generating a quantitative value of the mission support degree based on the spatial overlap between the detection coverage and the expected travel path, as well as the matching degree between the data backhaul bandwidth and the real-time data transmission volume required by the on-site handling area.

[0051] When calculating the quantitative indicator of mission support, this application embodiment selects two key evaluation dimensions. The first dimension is spatial overlap, which is calculated based on the detection coverage of the airborne detection module of the aerial mission and the expected travel path of the ground mission. The detection coverage refers to the ground area that the detection equipment carried by the UAV can effectively perceive at a specific flight altitude, such as the ground swath of a visible light camera or the scanning width of an infrared thermal imager. The expected travel path is the planned route from the starting point to the target point included in the ground mission. By superimposing these two in space, the proportion of the path length or area falling within the detection coverage can be calculated. The higher the proportion, the higher the spatial overlap, and the longer the UAV can conduct accompanying detection of ground units. The second dimension is data bandwidth matching, which is calculated based on the data return bandwidth of the aerial mission and the amount of real-time data transmission required by the on-site handling area of ​​the ground mission. The data return bandwidth refers to the transmission rate of the wireless data link between the UAV and the ground control center or ground rescue unit. The real-time data transmission volume required for on-site response refers to the amount of data that ground units need to acquire from the UAV in real time to obtain effective on-site perception during mission execution. For example, responding to a fire may require high-definition, high-frame-rate video streams, resulting in a large data requirement; while responding to the confirmation of a static obstacle may only require low-resolution images, resulting in a smaller data requirement. By analyzing the matching degree between data transmission bandwidth and the required data transmission volume, it can be determined whether the UAV's communication capabilities can meet the real-time data requirements of the ground mission. Finally, by weighting or comprehensively calculating the spatial overlap and data bandwidth matching degree, a quantitative value of mission support is generated.

[0052] In some embodiments of this application, step S5, which involves jointly selecting the first and second target tasks with the maximum coupling correlation, includes: constructing a bipartite graph model with a ground rescue unit as the first node and a multi-rotor UAV platform as the second node, where the first node corresponds to ground tasks in the first candidate task set and the second node corresponds to air tasks in the second candidate task set; using the weighted sum of the calculated task support and task dependency as the weight value of the connecting edge; and employing the Hungarian algorithm to solve the bipartite graph model for maximum weight matching to determine the optimal combination of the first and second target tasks.

[0053] When jointly selecting air-to-ground target tasks with the maximum coupling correlation, this embodiment of the application uses a bipartite graph model from graph theory for mathematical solution. A bipartite graph is a special graph model whose vertices can be divided into two disjoint subsets, and the two vertices associated with each edge in the graph belong to these two different subsets respectively. Here, the constructed bipartite graph model uses ground rescue units as the first set of nodes, with each first node corresponding to a specific ground task in the first candidate task set; and multi-rotor UAV platforms as the second set of nodes, with each second node corresponding to a specific air task in the second candidate task set. The weight value of the connecting edge is obtained by weighted summing the task support and task dependency calculated for the specific ground task and the specific air task associated with the edge. The weighting coefficient can be set according to the actual application scenario, for example, emphasizing support or dependency. The Hungarian algorithm is a classic combinatorial optimization algorithm used to solve assignment problems or bipartite graph maximum matching problems. This embodiment uses the existing Hungarian algorithm for implementation. The Hungarian algorithm is used to solve the constructed weighted bipartite graph model. The goal is to find a matching such that the sum of the weights of all matching edges is maximized. This matching corresponds to the optimal combination of the first and second target tasks selected from the candidate task set.

[0054] In some embodiments of this application, step S6, which plans a second detection path for the multi-rotor UAV platform according to the second objective mission, includes: determining at least one key node location as a key monitoring waypoint based on the first travel path of the ground rescue unit; and planning a second detection path that can sequentially fly over the key monitoring waypoints and meets the endurance constraints, by combining the remaining endurance of the multi-rotor UAV platform with the no-fly zone information in the dynamic mission situation map.

[0055] When planning a second detection path for a multi-rotor UAV platform, this embodiment uses the characteristics of the first travel path as the core basis for path planning. First, based on the first travel path of the ground rescue unit, at least one key node location is identified as a key monitoring waypoint. Key node locations can be understood as points on the first travel path that are of significant importance to the rescue operation. Examples include turning points or intersections on the path, where it is necessary to confirm the correct direction; they can also include entrances to high-risk areas such as bridges, tunnels, or densely built-up areas; and they can also include accident-prone areas identified based on historical data or experience. After identifying these points as key monitoring waypoints, a comprehensive path planning is performed by combining the remaining endurance of the multi-rotor UAV platform (estimated flight time or distance based on remaining battery power) and information on no-fly zones in the dynamic mission situation map (airspaces where UAVs are not permitted to fly). The goal of the plan is to generate a second detection path that can fly over the identified key monitoring waypoints in a certain order, while meeting the endurance constraints throughout the flight, namely, the total flight distance or total flight time does not exceed the current capability range of the UAV, and avoids no-fly zones throughout the entire flight.

[0056] In some embodiments of this application, step S7, determining whether the first travel path meets the safe passage conditions, includes: parsing the situational information ahead and extracting the locations of newly added obstacles, the boundaries of secondary disaster spread trends, or the hotspots of trapped personnel distribution; performing spatial overlay analysis on the extracted information and the first travel path; if the locations of newly added obstacles or the boundaries of secondary disaster spread trends intersect with the first travel path, or if the hotspots of trapped personnel distribution deviate from the first travel path by more than a preset deviation threshold, then it is determined that the safe passage conditions are not met.

[0057] When determining whether the first travel path meets the safe passage conditions, this application embodiment makes a decision based on the analysis and spatial analysis of the situational information ahead. First, the situational information ahead transmitted in real time by the multi-rotor UAV platform is analyzed to extract key dynamic change elements. New obstacle locations refer to those locations that were not originally present in the base map but are obstacles created by sudden events, such as the location of boulders rolling down due to an earthquake or the location of vehicles stranded due to traffic accidents. The secondary disaster spread trend boundary refers to the leading edge boundary line of secondary disasters such as fires, hazardous gas leaks, and floods that are spreading; this boundary line is dynamically changing. Hotspot areas of trapped personnel refer to the areas where trapped personnel are relatively concentrated, identified through analysis using the infrared imaging module or visible light image recognition algorithm onboard the UAV. These extracted elements are then spatially overlaid with the first travel path; that is, the path layer is overlaid with these element layers in the geographic information system. If the new obstacle location or the secondary disaster spread trend boundary spatially intersects with the first travel path—that is, the obstacle is located exactly on the path, or the disaster boundary has spread to intersect with the path—then the path is clearly no longer safe. If the hotspot area where trapped people are located deviates from the first travel path by more than a preset deviation threshold, for example, if the rescue personnel are pre-set to be no more than 200 meters away from the trapped people, but the center of the hotspot area actually deviates from the path by 300 meters, although the path itself may still be safe, it can be determined that the safe passage conditions are not met in order to carry out the rescue mission more effectively, thereby triggering a path adjustment so that the rescue unit can get closer to the trapped people.

[0058] In some embodiments of this application, after replanning the first travel path in step S7, the method further includes: generating a path change instruction based on the replanned first travel path; sending the path change instruction to the multi-rotor UAV platform and driving the multi-rotor UAV platform to adjust the second detection path according to the path change instruction, so as to maintain continuous detection of the area ahead of the replanned first travel path.

[0059] After replanning the first travel path, this embodiment establishes a linkage update mechanism for air-to-ground paths. When step S7 determines that the original path does not meet the safe passage conditions and completes the replanning, the system generates a path change instruction based on the replanned first travel path. The path change instruction is a formatted control message containing the geometric coordinate sequence of the new first travel path, key node information, and related attribute descriptions. This instruction is sent to the multi-rotor UAV platform performing the detection mission via a wireless communication network. After receiving the instruction, the UAV platform's onboard flight control system parses the instruction content and automatically adjusts the currently executing second detection path according to the new path information described in the path change instruction. The goal of the adjustment is to enable the UAV's detection path to re-establish a cooperative relationship with the newly planned first travel path. Specifically, this involves recalculating and planning a detection route that can cover the area ahead of the new path to maintain continuous detection of the area ahead of the replanned first travel path and ensure that the continuity of air-to-ground cooperation is not interrupted.

[0060] In some embodiments of this application, during the process of the ground rescue unit moving along the first travel path, the method further includes: dynamically adjusting the personnel evacuation guidance range or the activation timing of rescue equipment in the first disposal strategy based on the forward situation information transmitted back in real time by the multi-rotor UAV platform; the dynamic adjustment specifically includes: when the forward situation information is parsed to indicate that there is a newly added area of ​​trapped personnel gathering ahead of the first travel path, triggering an adaptive expansion of the boundary of the personnel evacuation guidance range, so that the expanded evacuation guidance range includes the newly added area of ​​trapped personnel gathering, and broadcasting a temporary guidance path to the newly added area of ​​trapped personnel gathering through the air broadcast module carried by the multi-rotor UAV platform; when the forward situation information is parsed to indicate that there is a secondary disaster spread risk area ahead of the first travel path, triggering an advance adjustment of the activation timing of rescue equipment, so that the ground rescue unit can pre-activate the corresponding protective equipment or demolition equipment before arriving at the secondary disaster spread risk area; wherein, the dynamic adjustment result of the first disposal strategy is fed back to step S7, and if the dynamically adjusted personnel evacuation guidance range or the activation timing of rescue equipment conflicts with the first travel path in terms of spatial constraints, it is used as an additional judgment condition for triggering the replanning of the first travel path.

[0061] During the movement of the ground rescue unit, this embodiment of the application realizes dynamic adjustment of the response strategy based on real-time situational awareness and establishes a feedback mechanism between strategy adjustment and path planning. Based on the real-time situational information transmitted back by the UAV, the specific action parameters in the first response strategy can be dynamically adjusted. When a newly added area of ​​trapped personnel is identified ahead of the first movement path, an adaptive expansion of the boundary of the personnel evacuation guidance range is triggered. The personnel evacuation guidance range is the area boundary pre-set in the first response strategy for planned evacuation guidance of trapped personnel. Adaptive expansion means automatically extending the original boundary outward according to the spatial location and extent of the newly added area, so that the expanded evacuation guidance range can completely include the newly added area of ​​trapped personnel. Simultaneously, a temporary guidance path is broadcast to the newly added area via the UAV's airborne broadcast module, guiding trapped personnel to evacuate along a safe route. When a secondary disaster risk area is identified ahead of the first movement path, an adjustment to the timing of rescue equipment activation is triggered. The timing of rescue equipment activation is a pre-set plan in the strategy to activate certain equipment at a specific time or location, such as activating a fire pump upon reaching the edge of a fire. Forward adjustment refers to advancing the activation time, allowing ground rescue units to pre-activate corresponding protective equipment, such as turning on the vehicle's NBC protection system or pre-activating demolition equipment, before arriving at areas at risk of secondary disaster spread, so that they can immediately begin operations upon arrival. Furthermore, the dynamic adjustment results of the first response strategy are fed back into the judgment process in step S7. If the dynamically adjusted personnel evacuation guidance range becomes too large spatially, causing a spatial constraint conflict with the first travel path itself—for example, the evacuation range covers the originally planned path, requiring the path to avoid evacuated populations; or the forward timing of rescue equipment activation necessitates stopping at an earlier location on the path, which may not be suitable for stopping—these situations will serve as additional judgment conditions triggering the replanning of the first travel path, ensuring that path planning takes into account the new needs brought about by the strategy adjustment.

[0062] In some embodiments of this application, step S5, which involves jointly selecting the first and second target tasks with the highest coupling correlation, further includes: acquiring real-time building damage assessment data of the seismic zone output by the earthquake monitoring system, wherein the building damage assessment data includes the damage level and collapse risk probability of each building; superimposing the building damage assessment data as a spatial weighting factor onto the dynamic task situation map, and applying risk weighting to the on-site handling areas involved in each ground task in the first candidate task set; recalculating the task dependence of each ground task in the first candidate task set on each aerial task in the second candidate task set based on the risk-weighted on-site handling areas, so that ground tasks located in areas with high damage levels or high collapse risk probabilities receive higher task dependence weights; and re-performing the joint screening based on the updated task dependence and task support, so that aerial tasks with high collapse risk area detection capabilities are preferentially coupled and matched with ground tasks handling high-risk areas.

[0063] Building upon the joint screening in step S5, building damage assessment data from the earthquake monitoring system was further introduced as a spatial weighting factor. This data includes the damage level and collapse risk probability of each building, providing crucial information reflecting the distribution of ground risks in the incident area. Overlaying this data onto the dynamic mission situation map allows for risk weighting of the on-site response areas involved in each ground mission in the first candidate mission set, assigning different risk weight values ​​based on the degree of building damage within each area. Based on the risk-weighted on-site response areas, the mission dependency of each ground mission in the first candidate mission set on each aerial mission in the second candidate mission set is recalculated. This results in ground missions located in areas with high damage levels or high collapse risk probabilities receiving higher mission dependency weights. For example, ground missions requiring search and rescue operations within severely damaged building complexes will have a significantly increased dependency on aerial missions with infrared detection or structural scanning capabilities. Based on the updated mission dependencies and the original mission support levels, the joint screening process is re-executed. This allows aerial missions with high collapse risk detection capabilities, such as UAVs equipped with specialized payloads capable of detecting internal heat sources or structural cracks, to be prioritized for matching with ground missions that need to address high-risk areas. This risk-weighted secondary screening mechanism helps achieve precise coordination in prioritizing high-risk areas with high-capacity resources, enabling limited aerial detection resources to be more focused on the critical areas that most need their support.

[0064] In some embodiments of this application, step S6, which plans a second detection path for the multi-rotor UAV platform according to the second target task, further includes: generating at least one aerial hovering broadcast waypoint in the second detection path according to the personnel evacuation guidance range in the first disposal strategy; when the multi-rotor UAV platform arrives at the aerial hovering broadcast waypoint, triggering the airborne aerial broadcast module to switch to directional broadcast mode, and automatically adjusting the sound wave radiation pointing angle of the aerial broadcast module according to the spatial orientation of the hotspot area of ​​the trapped personnel extracted from the dynamic task situation map, so that the main lobe direction of the broadcast sound wave is aligned with the hotspot area of ​​the trapped personnel; at the same time, dynamically calculating the continuous hovering duration of the aerial hovering broadcast waypoint according to the real-time travel speed and expected arrival time of the ground rescue unit, so that the broadcast content of the aerial broadcast module covers the complete time window from the current moment to the expected arrival time of the ground rescue unit, and triggering the multi-rotor UAV platform to exit the hovering state and continue to perform the detection task along the second detection path when a preset time threshold is set before the ground rescue unit arrives at the hotspot area of ​​the trapped personnel.

[0065] Building upon the second detection path planned in step S6, refined control functions related to personnel evacuation guidance are further introduced. Based on the personnel evacuation guidance range in the first response strategy, at least one aerial hovering broadcast waypoint is generated within the second detection path. This means that the UAV's detection path includes not only patrol monitoring points but also hovering points specifically for performing evacuation broadcasting tasks. When the multi-rotor UAV platform reaches the aerial hovering broadcast waypoint, the onboard aerial broadcast module switches to directional broadcast mode and automatically adjusts the sound wave radiation pointing angle of the aerial broadcast module according to the spatial orientation of the hotspot areas of trapped personnel extracted from the dynamic mission situation map, aligning the main lobe of the broadcast sound wave with the hotspot areas of trapped personnel. Compared to traditional omnidirectional broadcasting, this directional broadcasting can project sound energy more concentratedly onto the target area, helping to improve the clarity and effective coverage of the broadcast content while reducing noise interference to non-target areas. Simultaneously, based on the real-time travel speed and estimated arrival time of the ground rescue unit, the continuous hovering duration of the aerial hovering broadcast waypoint is dynamically calculated, ensuring that the broadcast content of the aerial broadcast module covers the complete time window from the current moment to the estimated arrival time of the ground rescue unit. This means that the drone can continuously broadcast guidance over the area where people are stranded until the ground unit is about to arrive, avoiding information interruption caused by prematurely ending the broadcast. When the ground rescue unit arrives at the hotspot area where people are stranded, a preset time threshold is triggered, and the multi-rotor drone platform exits the hovering state and continues to perform the detection mission along the second detection path, achieving a seamless connection from aerial guidance to ground handover.

[0066] Please refer to Figure 2A second aspect of this application provides an emergency response decision-making system based on air-ground coordination, which is used to implement the above-described emergency response decision-making method based on air-ground coordination, including: The situation construction module 21 is used to collect on-site perception data of the incident area in response to emergency alarms based on geographic information system and Internet of Things technology, and to construct a dynamic task situation map in combination with the incident location. The status acquisition module 22 is used to acquire the first status information of at least one ground rescue unit within the response range, and the second status information of at least one multi-rotor drone platform; The task generation module 23 is used to generate a first candidate task set for the ground rescue unit and a second candidate task set for the multi-rotor UAV platform based on the dynamic task situation map, the first state information and the second state information. The coupled calculation module 24 is used to calculate the mission support degree of each air mission in the second candidate mission set to each ground mission in the first candidate mission set, and the mission dependency degree of each ground mission in the first candidate mission set to each air mission in the second candidate mission set. The collaborative decision-making module 25 is used to jointly select the first target task and the second target task with the maximum coupling correlation from the first candidate task set and the second candidate task set based on the task support degree and the task dependency degree. The path planning module 26 is used to plan a first travel path and a first disposal strategy for the ground rescue unit according to the first objective task, and to plan a second detection path and a second detection strategy for the multi-rotor UAV platform according to the second objective task. The second detection path is used to enable the multi-rotor UAV platform to continuously detect along the area in front of the first travel path and generate forward situation information. The dynamic adjustment module 27 is used to determine whether the first travel path meets the safe passage conditions based on the forward situation information transmitted back in real time by the multi-rotor UAV platform. If it does not meet the conditions, the first travel path is replanned and the second detection path is updated simultaneously.

[0067] In the above embodiments, a complete air-ground collaborative emergency response decision-making system was constructed, enabling earthquake monitoring data, UAV perception data, and ground rescue unit status to be deeply integrated and collaboratively processed within the same system framework, providing an integrated decision-making platform with air-ground mission coupling capabilities for emergency command.

[0068] See Figure 3 , Figure 3 This is a schematic block diagram of an electronic device provided according to an embodiment of this application. Figure 3The electronic device 300 in this embodiment may include one or more processors 301, one or more input devices 302, one or more output devices 303, and one or more memories 304. The processors 301, input devices 302, output devices 303, and memories 304 communicate with each other via a communication bus 305. The memories 304 store computer programs, including program instructions. The processors 301 execute the program instructions stored in the memories 304. Specifically, the processors 301 are configured to invoke the program instructions to execute the aforementioned air-ground coordinated emergency response decision-making method.

[0069] It should be understood that, in the embodiments of this application, the processor 301 may be a central processing unit (CPU), or it may be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor.

[0070] Input device 302 may include a touchpad, a fingerprint sensor (for collecting the user's fingerprint information and fingerprint orientation information), a microphone, etc., and output device 303 may include a display (LCD, etc.), a speaker, etc.

[0071] The memory 304 may include read-only memory and random access memory, and provides instructions and data to the processor 301. A portion of the memory 304 may also include non-volatile random access memory. For example, the memory 304 may also store device type information.

[0072] In specific implementations, the processor 301, input device 302, and output device 303 described in the embodiments of this application can execute the implementation methods described in any embodiment of the emergency response decision-making method based on air-ground coordination provided in the embodiments of this application, or they can execute the implementation methods of the electronic devices described in the embodiments of this application, which will not be repeated here.

[0073] In another embodiment of this application, an electronic device is provided. The electronic device stores a computer program, which includes program instructions. When executed by a processor, the program instructions implement all or part of the processes in the above-described air-ground coordinated emergency response decision-making method for sudden events. Alternatively, the computer program can instruct related hardware to complete the process. The computer program can be stored in an electronic device, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. Computer-readable media can include any entity or device capable of carrying computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc.

[0074] The computer-readable storage medium can be an internal storage unit of the electronic device in any of the foregoing embodiments, such as a hard disk or memory of the electronic device. The computer-readable storage medium can also be an external storage device of the electronic device, such as a plug-in hard disk, smart media card (SMC), secure digital card (SD), flash card, etc., equipped on the electronic device. Furthermore, the computer-readable storage medium can include both internal and external storage units of the electronic device. The computer-readable storage medium is used to store computer programs and other programs and data required by the electronic device. The computer-readable storage medium can also be used to temporarily store data that has been output or will be output.

[0075] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this application.

[0076] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the electronic devices and units described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0077] In the several embodiments provided in this application, it should be understood that the disclosed electronic devices and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed may be indirect coupling or communication connection through some interfaces or units, or it may be an electrical, mechanical, or other form of connection.

[0078] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of the embodiments of this application, depending on actual needs.

[0079] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0080] The above are merely specific embodiments of this application, but the scope of protection of this application is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in this application, and these modifications or substitutions should all be covered within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A decision-making method for emergency response to sudden events based on air-ground coordination, characterized in that, The sudden event is an earthquake sudden event, and the method includes the following steps: S1, based on geographic information system and Internet of Things technology, responds to earthquake emergency alarms by collecting on-site perception data of the incident area and constructing a dynamic mission situation map in combination with the incident location; S2, acquire the first status information of at least one ground rescue unit within the response range, and the second status information of at least one multi-rotor drone platform, specifically: The vehicle-mounted IoT terminal collects the current location of the ground rescue unit, the remaining fuel level, and the types and quantities of rescue supplies in real time. The acquisition of the second status information further includes: transmitting the current location, flight speed, remaining battery power, and the type and working mode of the currently equipped functional modules of the multi-rotor UAV platform in real time through the airborne flight control system and mission payload system of the multi-rotor UAV platform. S3, based on the dynamic mission situation map, the first state information, and the second state information, generate a first candidate task set for the ground rescue unit and a second candidate task set for the multi-rotor UAV platform, respectively. Specifically: Based on the remaining fuel of the ground rescue unit and the remaining battery power of the multi-rotor UAV platform, the maximum driving range of the ground rescue unit and the maximum flight range of the multi-rotor UAV platform are dynamically estimated, and the estimation results are used as the endurance constraints for generating the first candidate task set and the second candidate task set. S4, calculate the mission support degree of each air mission in the second candidate mission set to each ground mission in the first candidate mission set, and the mission dependence degree of each ground mission in the first candidate mission set to each air mission in the second candidate mission set. S5, based on the task support degree and the task dependency degree, jointly select the first target task and the second target task with the maximum coupling correlation from the first candidate task set and the second candidate task set; the joint selection process further includes: The system acquires real-time building damage assessment data from the earthquake monitoring system, which includes the damage level and collapse risk probability of each building. This data is then overlaid as a spatial weighting factor onto the dynamic task situation map to apply risk weighting to the on-site handling areas involved in each ground task in the first candidate task set. Based on the risk-weighted on-site handling areas, the task dependency of each ground task in the first candidate task set on each aerial task in the second candidate task set is recalculated, giving higher task dependency weights to ground tasks located in areas with high damage levels or high collapse risk probabilities. Based on the updated task dependency and task support, the joint screening is re-executed to prioritize the coupling and matching of aerial tasks with high collapse risk area detection capabilities with ground tasks handling high-risk areas. The aerial tasks with high collapse risk area detection capabilities are performed by multi-rotor UAV platforms carrying dedicated payloads capable of detecting internal heat sources or structural cracks in buildings. S6, based on the first objective task, a first travel path and a first response strategy are planned for the ground rescue unit, and based on the second objective task, a second detection path and a second detection strategy are planned for the multi-rotor UAV platform. Specifically, the second detection path enables the multi-rotor UAV platform to continuously detect and generate forward situational information along the area ahead of the first travel path. Based on the personnel evacuation guidance range in the first response strategy, at least one aerial hovering broadcast waypoint is generated in the second detection path. When the multi-rotor UAV platform arrives at the aerial hovering broadcast waypoint, the airborne aerial broadcast module is triggered to switch to directional broadcast mode. Based on the spatial orientation of the hotspot area of ​​trapped personnel extracted from the dynamic mission situation map, the sound wave radiation pointing angle of the aerial broadcast module is automatically adjusted so that the main lobe of the broadcast sound wave is aligned with the hotspot area of ​​trapped personnel. At the same time, based on the real-time travel speed and expected arrival time of the ground rescue unit, the continuous hovering duration of the aerial hovering broadcast waypoint is dynamically calculated so that the broadcast content of the aerial broadcast module covers the complete time window from the current moment to the expected arrival time of the ground rescue unit. When a preset time threshold is reached before the ground rescue unit arrives at the hotspot area of ​​trapped personnel, the multi-rotor UAV platform is triggered to exit the hovering state and continue to perform the detection mission along the second detection path. S7. Based on the forward situation information transmitted back in real time by the multi-rotor UAV platform, determine whether the first travel path meets the safe passage conditions. If not, replan the first travel path and update the second detection path simultaneously.

2. The emergency response decision-making method for sudden events based on air-ground coordination according to claim 1, characterized in that, The construction of a dynamic task situation map in step S1 further includes: integrating the building damage assessment data of the earthquake zone output by the earthquake monitoring system, the environmental parameters collected by ground fixed sensors, and the crowdsourced information reported by mobile terminals around the incident area, dynamically marking the accessibility of the road network and the risk areas of building collapse in the ground traffic environment, and dynamically marking the airspace control area and the airflow disturbance risk area in the air flight environment.

3. The emergency response decision-making method for sudden events based on air-ground coordination according to claim 1, characterized in that, The functional module is at least one of the following: visible light imaging module, infrared thermal imaging module, gas detection module, or air broadcasting module.

4. The emergency response decision-making method for sudden events based on air-ground coordination according to claim 1, characterized in that, The calculation of the mission support degree in step S4 specifically includes: extracting the detection coverage and data backhaul bandwidth of the airborne detection module for each air mission in the second candidate mission set; extracting the expected travel path and spatial range of the on-site handling area for each ground mission in the first candidate mission set; and generating a quantitative value of the mission support degree based on the spatial overlap between the detection coverage and the expected travel path, and the matching degree between the data backhaul bandwidth and the real-time data transmission volume required by the on-site handling area.

5. The emergency response decision-making method for sudden events based on air-ground coordination according to claim 1, characterized in that, Step S5 involves re-performing the joint screening based on the updated task dependency and task support, including: constructing a bipartite graph model with the ground rescue unit as the first node and the multi-rotor UAV platform as the second node, where the first node corresponds to ground tasks in the first candidate task set and the second node corresponds to air tasks in the second candidate task set; using the weighted sum of the task support and the updated task dependency calculated for the corresponding ground and air tasks as the weight values ​​of the connecting edges; and employing the Hungarian algorithm to perform maximum weight matching on the bipartite graph model to determine the optimal combination of the first target task and the second target task.

6. The emergency response decision-making method for sudden events based on air-ground coordination according to claim 1, characterized in that, In step S6, planning a second detection path for the multi-rotor UAV platform according to the second target mission includes: determining at least one key node location as a key monitoring waypoint based on the first travel path of the ground rescue unit; and planning a second detection path that can sequentially fly over the key monitoring waypoints and meets the endurance constraints, in combination with the remaining endurance of the multi-rotor UAV platform and the no-fly zone information in the dynamic mission situation map.

7. The emergency response decision-making method for sudden events based on air-ground coordination according to claim 1, characterized in that, Step S7, determining whether the first travel path meets the safe passage conditions, includes: parsing the forward situation information and extracting the locations of newly added obstacles, the boundaries of secondary disaster spread trends, or hotspots of trapped personnel distribution; performing spatial overlay analysis with the extracted information and the first travel path; if the locations of newly added obstacles or the boundaries of secondary disaster spread trends intersect with the first travel path, or if the hotspots of trapped personnel distribution deviate from the first travel path by more than a preset deviation threshold, then it is determined that the safe passage conditions are not met.

8. The emergency response decision-making method for sudden events based on air-ground coordination according to claim 1, characterized in that, After replanning the first travel path in step S7, the method further includes: generating a path change command based on the replanned first travel path; sending the path change command to the multi-rotor UAV platform and driving the multi-rotor UAV platform to adjust the second detection path according to the path change command, so as to maintain continuous detection of the area ahead of the replanned first travel path.

9. The emergency response decision-making method for sudden events based on air-ground coordination according to claim 1, characterized in that, During the process of the ground rescue unit moving along the first travel path, the method further includes: dynamically adjusting the personnel evacuation guidance range or the activation timing of rescue equipment in the first response strategy based on the forward situation information transmitted back in real time by the multi-rotor UAV platform; the dynamic adjustment specifically includes: when the forward situation information reveals that there is a newly added area of ​​trapped personnel gathering ahead of the first travel path, triggering an adaptive expansion of the boundary of the personnel evacuation guidance range, so that the expanded evacuation guidance range includes the newly added area of ​​trapped personnel gathering, and broadcasting the information to the designated area via the airborne broadcast module carried by the multi-rotor UAV platform. The newly added areas where trapped personnel are gathered will be broadcast with temporary guidance routes; when the situation information ahead is analyzed to show that there is a risk of secondary disaster spread ahead of the first travel route, the timing of the activation of the rescue equipment will be adjusted in advance, so that the ground rescue unit can activate the corresponding protective equipment or demolition equipment in advance before reaching the risk area of ​​secondary disaster spread; wherein, the dynamic adjustment result of the first disposal strategy is fed back to step S7, and if the dynamically adjusted personnel evacuation guidance range or the timing of the activation of rescue equipment conflicts with the first travel route in terms of spatial constraints, it will be used as an additional judgment condition to trigger the replanning of the first travel route.

10. An emergency response decision-making system for sudden events based on air-ground coordination, characterized in that, The emergency is an earthquake emergency, and the system includes: The situation construction module is used to collect on-site perception data of the incident area in response to earthquake emergency alarms based on geographic information system and Internet of Things technology, and to build a dynamic task situation map in combination with the incident location. The status acquisition module is used to acquire the first status information of at least one ground rescue unit within the response range, and the second status information of at least one multi-rotor drone platform. Specifically, it collects the current location, remaining fuel, and types and quantities of onboard rescue supplies of the ground rescue unit in real time via an onboard IoT terminal, and transmits back the current location, flight speed, remaining battery power, and types and operating modes of the currently mounted functional modules of the multi-rotor drone platform in real time via the onboard flight control system and mission payload system. The task generation module is used to generate a first candidate task set for the ground rescue unit and a second candidate task set for the multi-rotor UAV platform based on the dynamic task situation map, the first state information, and the second state information; and to dynamically estimate the maximum driving range of the ground rescue unit and the maximum flight range of the multi-rotor UAV platform based on the remaining fuel of the ground rescue unit and the remaining battery power of the multi-rotor UAV platform, respectively, and use the estimation results as the endurance constraints for generating the first candidate task set and the second candidate task set. The coupled calculation module is used to calculate the mission support degree of each air mission in the second candidate mission set to each ground mission in the first candidate mission set, and the mission dependency degree of each ground mission in the first candidate mission set to each air mission in the second candidate mission set. The collaborative decision-making module is used to jointly filter out the first target task and the second target task with the highest coupling correlation from the first candidate task set and the second candidate task set based on the task support degree and the task dependency degree. During the joint filtering process, the collaborative decision-making module is also used to acquire real-time building damage assessment data of the earthquake zone output by the earthquake monitoring system. The building damage assessment data includes the damage level and collapse risk probability of each building. The building damage assessment data is superimposed as a spatial weighting factor onto the dynamic task situation map to perform risk weighting on the on-site handling areas involved in each ground task in the first candidate task set. Based on the risk weighting... After weighting the on-site disposal area, the task dependence of each ground task in the first candidate task set on each aerial task in the second candidate task set is recalculated, so that ground tasks located in areas with high damage levels or high collapse risk probability receive higher task dependence weights; based on the updated task dependence and the task support, the joint screening is re-executed so that aerial tasks with high collapse risk area detection capabilities are preferentially coupled and matched with ground tasks for handling high-risk areas; wherein, the aerial tasks with high collapse risk area detection capabilities are aerial tasks performed by multi-rotor UAV platforms carrying dedicated payloads capable of detecting heat sources or structural cracks inside buildings; The path planning module is used to plan a first travel path and a first response strategy for the ground rescue unit according to the first objective task, and to plan a second detection path and a second detection strategy for the multi-rotor UAV platform according to the second objective task. The second detection path enables the multi-rotor UAV platform to continuously detect and generate forward situational information along the area ahead of the first travel path. The path planning module is also used to generate at least one hovering broadcast waypoint in the second detection path based on the personnel evacuation guidance range in the first response strategy. When the multi-rotor UAV platform reaches the hovering broadcast waypoint, it triggers the airborne airborne broadcast module to switch to directional broadcast mode and... Based on the spatial orientation of the hotspots of trapped personnel extracted from the dynamic mission situation map, the sound wave radiation pointing angle of the aerial broadcast module is automatically adjusted so that the main lobe of the broadcast sound wave is aligned with the hotspots of trapped personnel. At the same time, based on the real-time travel speed and estimated arrival time of the ground rescue unit, the continuous hovering duration of the aerial hovering broadcast waypoint is dynamically calculated so that the broadcast content of the aerial broadcast module covers the complete time window from the current moment to the estimated arrival time of the ground rescue unit. When the ground rescue unit arrives at the hotspots of trapped personnel before the preset time threshold, the multi-rotor UAV platform is triggered to exit the hovering state and continue to perform the detection mission along the second detection path. The dynamic adjustment module is used to determine whether the first travel path meets the safe passage conditions based on the forward situation information transmitted back in real time by the multi-rotor UAV platform. If it does not meet the conditions, the first travel path is replanned and the second detection path is updated synchronously.

Images