An unmanned aerial vehicle intelligent emergency rescue method and system in a complex disaster environment

Abstract

The application discloses an unmanned aerial vehicle intelligent emergency rescue method and system in a complex disaster environment, and relates to the fields of unmanned aerial vehicle technology and public safety emergency rescue. The method comprises the following steps: firstly, constructing a temporary ad hoc network communication link to guarantee data transmission; secondly, utilizing multi-source sensor fusion technology to position trapped personnel and detect environmental risks; then, generating a rescue environment evaluation map including risk partitions based on real-time data; finally, dynamically deciding to execute material delivery or wounded personnel transfer tasks through a three-level decision process. The application effectively solves the problems of response lag and blind decision of traditional rescue in the case of communication interruption, complex environment and limited resources, significantly improves the rescue efficiency and safety, and is suitable for various major disaster scenes such as earthquakes, floods, landslides and dangerous goods leakage.

Description

Technical Field

[0001] This invention belongs to the field of unmanned aerial vehicle (UAV) application technology, specifically relating to an intelligent emergency rescue method and system for UAVs in complex disaster environments, which is particularly suitable for ground traffic disruptions, communication paralysis, and high-risk environmental scenarios caused by natural disasters, accidents, and other disasters. Background Technology

[0002] In the event of major disasters such as earthquakes, floods, landslides, or hazardous chemical leaks, traditional ground rescue often faces challenges such as road blockages, communication disruptions, unknown environments, and high risks of secondary injuries, resulting in wasted "golden rescue time." Drones, with their advantages of maneuverability, lack of terrain limitations, and remote operation capabilities, have gradually become an important technological means for disaster emergency rescue, and related technological solutions are continuously developing.

[0003] One existing technical solution focuses on optimizing the allocation of UAV monitoring resources in geological disaster scenarios. For example, patent application CN120765010A proposes a UAV radar monitoring method for geological disaster emergency response. This method first divides the geological disaster monitoring area into risk levels based on disaster type and frequency, extracting risk areas spatially associated with disaster-causing factors. Then, by comparing historical disaster data with historical UAV radar monitoring data, it identifies high-risk target areas with insufficient monitoring capabilities and adjacent associated areas with relatively redundant monitoring resources. Finally, it combines real-time terrain, meteorological, and other monitoring features with preset scheduling rules, dynamically allocating UAV radar monitoring resources through a resource scheduling model to improve the efficiency of monitoring resource utilization. However, this type of solution is limited to resource scheduling in the disaster monitoring stage and does not extend to core rescue execution scenarios such as aerial evacuation and material delivery, thus failing to form a complete emergency rescue closed loop.

[0004] Another type of technical solution revolves around the triggering of disaster events and the scheduling of drone missions. For example, patent application CN122451341A proposes a disaster emergency drone scheduling method based on hotlines. It uses citizen hotline data as the trigger source, extracts keywords, determines the urgency level, identifies core demands, and labels scenes on unstructured request texts. At the same time, it performs deduplication, completion, and format standardization on structured work order data. After automatically identifying disaster emergency events, it associates external multi-source data such as meteorology, traffic, holidays, and policies, automatically extracts the coordinates of the incident location, plans rescue routes, matches target drones from the equipment resource pool, and generates standardized scheduling and control commands. It is also equipped with execution capabilities such as drone autonomous cruise, intelligent identification of trapped personnel, hovering fine observation, pixel-level environmental analysis, and generation of on-site rescue maps, realizing the process linkage from event perception to drone operation. However, the scheduling logic of this type of solution uses event reporting as the single trigger source and lacks in-depth quantitative judgment of rescue priorities. When faced with a large number of trapped people and limited drone resources, it has not established a multi-dimensional comprehensive decision-making model based on the severity of injuries, physiological tolerance time windows, aircraft range and load limits. It is unable to scientifically weigh the resource allocation strategy of "prioritizing the transfer of the wounded" and "prioritizing the delivery of emergency supplies", which can easily lead to deviations in rescue decisions or waste of resources.

[0005] Overall, existing drone rescue applications are mostly focused on single searches or simple material drops, lacking systematic, end-to-end solutions. Especially when facing a large number of trapped people, how to scientifically allocate limited drone resources (prioritizing medicine delivery or rescue) remains a challenge. Current technology lacks comprehensive quantitative judgment standards based on injury severity, timeliness, and the physical limits of the aircraft, easily leading to rescue decision-making errors or resource waste. Furthermore, communication support and high-precision, all-domain environmental perception in complex disaster environments remain key bottlenecks restricting rescue efficiency, failing to meet the practical needs of major disaster emergency rescue. Summary of the Invention

[0006] The main objective of this invention is to overcome the shortcomings and deficiencies of the prior art and provide an intelligent emergency rescue method and system for unmanned aerial vehicles in complex disaster environments. This effectively solves the problems of delayed response and blind decision-making in traditional rescue under conditions of communication interruption, complex environment and limited resources, and significantly improves rescue efficiency and safety.

[0007] To achieve the above objectives, the present invention adopts the following technical solution:

[0008] In a first aspect, the present invention provides an intelligent emergency rescue method using unmanned aerial vehicles (UAVs) in complex disaster environments, comprising the following steps:

[0009] Deploy a cluster of communication relay drones to build a highly resilient mobile ad hoc network and establish emergency communication links;

[0010] The reconnaissance drone is equipped with multiple sensors to search for the target area and acquire multi-sensor data, including visible light images, infrared thermal imaging data, three-dimensional terrain data, and environmental hazard factors.

[0011] Multi-sensor data is fused and processed to generate a rescue environment risk assessment map, which includes the location of trapped personnel, travel routes, and dangerous areas.

[0012] Based on the rescue environment risk assessment map, rescue tasks are allocated and decisions are made to obtain the results.

[0013] Based on the decision-making results, control the operation drones to perform material delivery or casualty evacuation tasks.

[0014] As a preferred technical solution, the highly resilient mobile ad hoc network adopts a distributed mesh topology, which allows nodes to dynamically join and leave.

[0015] As a preferred technical solution, the multi-source sensor adopts one or more of the following: visible light camera, infrared thermal imager, lidar, ion mobility spectrometer, and gamma ray detector; the environmental hazard factors include toxic gas concentration, radiation intensity, and building structure stability index.

[0016] As a preferred technical solution, a pre-defined three-level decision-making process is adopted to allocate rescue missions, specifically as follows:

[0017] Level 1 assessment: Based on the severity score of the trapped person's injuries, determine whether to carry out a rescue operation;

[0018] The second level of assessment: compare the estimated time of physiological tolerance of the trapped personnel with the round-trip flight time of the drone to determine the feasibility of the timeliness;

[0019] The third level of judgment: Based on the current payload, remaining energy and target distance of the UAV, calculate the maximum range and determine the physical feasibility.

[0020] As a preferred technical solution, when the injury severity score is lower than a set threshold, the location and status information of the injured person are retained, and instructions for remote guidance, self-rescue prompts, or waiting for subsequent rescue are output; otherwise, the second-level judgment is executed.

[0021] If the timeliness and feasibility are met, the material delivery task shall be carried out first; otherwise, the third-level judgment shall be executed.

[0022] If the physical feasibility is met, the casualty evacuation mission shall be carried out; otherwise, the material delivery mission shall be carried out first.

[0023] As a preferred technical solution, the third-level determination specifically includes:

[0024] Using the Breguet range relationship, we analyze the relationship between UAV range and propulsion efficiency, power consumption coefficient, lift-to-drag ratio, takeoff weight and final weight.

[0025] The impact of payload and environmental drag on maximum range is calculated using the following formula:

[0026] ,

[0027] in, This represents the maximum usable range of the drone under a given payload. This represents a constant determined by the aerodynamic characteristics of the aircraft model. Indicates the empty weight of the drone. Indicates the effective payload weight of the drone. Indicates the environmental resistance coefficient;

[0028] The decision variables for calculating the flight range are ,but:

[0029] ,

[0030] in, Time-based decision variables representing the feasibility of a timeframe. Indicates the drone's take-off and landing point to the number The one-way distance from the location of the injured person. Indicates the weight of the injured person. Indicates the weight of the airborne medical equipment;

[0031] when At that time, the drone will be carrying the first After transporting the wounded and onboard medical equipment, its maximum available range is not less than the round-trip distance of the mission, possessing the physical conditions to complete the air evacuation and return; when When the mission's round-trip distance exceeds the maximum available range under the current configuration, the mission is switched to a supply delivery mission.

[0032] As a preferred technical solution, if the decision is to carry out a material delivery mission, the drone will use a temperature-controlled storage and transportation cabin to load the materials and deliver them by rappelling, airdropping, or in collaboration with ground robots; if the decision is to carry out a casualty evacuation mission, the drone will use an active stabilization stretcher system based on an inertial measurement unit and transmit the casualty's vital signs data back to the rear medical center in real time.

[0033] Secondly, the present invention also provides an intelligent emergency rescue system for unmanned aerial vehicles (UAVs) in complex disaster environments, applied to the aforementioned intelligent emergency rescue method for UAVs in complex disaster environments, comprising:

[0034] The communication support unit is used to temporarily establish a highly resilient, self-organizing emergency communication network.

[0035] The perception and detection unit is a reconnaissance drone equipped with multi-source sensors, used for disaster environmental data collection;

[0036] The data processing unit is used for multi-source data fusion, environmental modeling, and risk assessment.

[0037] The decision control unit has a built-in three-level judgment algorithm model for generating rescue mission instructions; the three-level judgment algorithm model includes injury severity score, golden time window, and physical feasibility.

[0038] The execution unit is used to carry out rescue missions of the decision control unit, including material delivery drones and casualty evacuation drones.

[0039] As a preferred technical solution, the casualty transport drone is equipped with an active stabilization stretcher system based on an inertial measurement unit to isolate high-frequency vibrations and low-frequency swaying during flight.

[0040] As a preferred technical solution, the system is equipped with a modular load interface, which allows switching between the material delivery and casualty transfer modules under different conditions.

[0041] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0042] 1. Scientific Decision-Making: By combining medical indicators, time windows, and engineering limits through a three-level judgment model, optimal allocation of rescue tasks is achieved, avoiding resource misallocation. 2. Multi-Dimensional Perception: Integrating data from multiple sensors, it can not only locate trapped personnel but also identify hidden dangers such as toxic gases, radiation, and collapses, ensuring the safety of rescue operations. 3. Reliable Communication: Utilizing self-organizing network technology, an independent communication link is built in environments without a public network, ensuring smooth command and control and real-time data transmission. 4. Strong Adaptability: Modular design and autonomous obstacle avoidance technology enable it to cope with various complex terrains and environments such as urban ruins, deep mountains and forests, and chemical plants. Attached Figure Description

[0043] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0044] Figure 1 This is a flowchart of an embodiment of the intelligent emergency rescue method for unmanned aerial vehicles (UAVs) in complex disaster environments according to an embodiment of the present invention;

[0045] Figure 2This is a structural diagram of the unmanned aerial vehicle (UAV) intelligent emergency rescue system under complex disaster environments, according to an embodiment of the present invention. Detailed Implementation

[0046] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are merely some embodiments of the present application, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present application without creative effort are within the scope of protection of the present application.

[0047] In this application, the reference to "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a mutually exclusive, independent, or alternative embodiment. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described in this application can be combined with other embodiments.

[0048] Please see Figure 1 This embodiment provides an intelligent emergency rescue method using unmanned aerial vehicles (UAVs) in complex disaster environments, including the following steps:

[0049] S1. Deploy a cluster of communication relay drones to build a highly resilient mobile ad hoc network and establish emergency communication links.

[0050] Deploying a swarm of communication relay drones to construct a highly resilient mobile ad hoc network establishes emergency communication links. This highly resilient mobile ad hoc network adopts a distributed mesh topology, supports dynamic node joining and leaving, and features multi-band fusion and adaptive anti-interference capabilities, enabling independent operation without ground infrastructure support. Specifically, when a relay drone is damaged or unable to transmit communication due to interference, the ad hoc network can automatically reconstruct its topology, and data packets will automatically select alternative transmission paths to ensure uninterrupted communication. Furthermore, the incorporation of frequency hopping and spread spectrum technologies further enhances the network's resilience against environmental electromagnetic interference, providing stable and reliable information transmission for the entire rescue operation.

[0051] S2. Control the reconnaissance drone equipped with multi-source sensors to search the target area and acquire multi-sensor data, including visible light images, infrared thermal imaging data, three-dimensional terrain data, and environmental hazard factor data.

[0052] In this embodiment, step S2 controls a reconnaissance drone equipped with multi-source sensors to search the target area and acquire visible light images, infrared thermal imaging data, three-dimensional terrain data, and environmental hazard factor data. The multi-source sensors include a visible light camera, an infrared thermal imager, a lidar, an ion mobility spectrometer, and a gamma-ray detector; the environmental hazard factors include toxic gas concentrations, radiation intensity, and building structural stability indices. In complex disaster environments, such as those with smoke obscuration, dark nighttime conditions, and the identification of concealed targets, a single sensor is insufficient for accurate identification and detection.

[0053] S3. The multi-sensor data is fused and processed to generate a rescue environment risk assessment map, which includes the location of the trapped personnel, the passage route, and the dangerous area.

[0054] Next, in step S3, the multi-sensor data is fused and processed to generate a rescue environment risk assessment map that includes the location of trapped personnel, travel routes, and dangerous areas. It should be explained that multi-sensor data fusion processing technology can comprehensively process data from visible light cameras, infrared thermal imagers, lidar, ion mobility spectrometers, and gamma-ray detectors. Essentially, it's about complementary enhancement and collaborative decision-making of multi-source information: by performing spatiotemporal registration, correlation integration, and hierarchical processing on sensor data from different physical principles and sensing dimensions, it ultimately obtains a more comprehensive, reliable, and accurate perception result than a single sensor, while also compensating for the inherent blind spots and limitations of a single sensor.

[0055] Specifically, it typically covers a three-tiered integration process from the bottom layer to the top layer:

[0056] First, spatiotemporal unification. All sensor data is given a unified timestamp, ensuring that multi-source data at the same moment corresponds to the same scene state. Simultaneously, calibration algorithms unify the coordinate systems of all sensors to the same world coordinate system. For example, the 3D point cloud of the LiDAR is mapped one-to-one with the pixels of the visible light image, and the temperature value of the infrared thermal image is mapped to the corresponding pixel position in the visible light image. This forms a "high-definition fused image with depth and temperature information." Furthermore, the detection results of ion mobility spectrometry and gamma detectors are bound to corresponding spatial coordinate points, solving the pain point of chemical / radiation sensors "measuring but not being able to pinpoint the exact location."

[0057] Second, feature-level fusion. First, specific features are extracted from the raw data of each sensor, and then cross-dimensional fusion analysis is performed on these features, including:

[0058] (1) Visible light extraction: visual features such as edges, textures, colors, and target outlines;

[0059] (2) LiDAR extraction: geometric features such as three-dimensional dimensions, spatial position, and surface normals;

[0060] (3) Infrared extraction: thermal characteristics such as temperature gradient and hotspot distribution;

[0061] (4) Ion mobility spectrum / gamma detector extraction: material composition, radiation dose, nuclide type and other property characteristics.

[0062] (5) Fusion logic: Through data association algorithms, multi-dimensional features of the same target are bound together. For example, for a suspicious package, its appearance shape, volume estimation, internal heat source status, surface chemical composition, and radiation level are integrated to significantly improve the accuracy of target identification and risk assessment.

[0063] Third, decision-level fusion. Each sensor first independently outputs its own judgment result, and then a fusion algorithm comprehensively arbitrates all decisions to output the final conclusion. Example scenario: Visible light determines "the target is a suitcase", infrared determines "no abnormal heat source", lidar determines "the size conforms to a standard suitcase", ion mobility spectrometry determines "toxic and hazardous components are detected", and gamma detector determines "no radioactive anomalies". The fusion algorithm (such as DS evidence theory and Bayesian inference) integrates the confidence levels of each sensor and finally outputs a comprehensive decision: "the suitcase poses a risk of explosives but no risk of nuclear radiation". Its core advantage is strong fault tolerance; if one sensor fails or malfunctions, the remaining sensors can still support the decision, resulting in higher overall system robustness.

[0064] This sensor suite covers five dimensions: visual perception, spatial geometry, thermal state, chemical composition, and nuclear radiation. When fused, it can achieve all-round, blind-spot-free target detection, and is typically used in scenarios such as terrain inspection and disaster emergency response.

[0065] The flight path is constructed by building a local 3D map based on environmental perception data from the drone's onboard sensors, and then using a path planning algorithm to comprehensively calculate the drone's dynamic limitations and external environmental constraints, thereby planning a basic flight path. Hazardous areas are identified by fusing multi-sensor data and marking them on the 3D map as areas of collapse risk and toxic gas diffusion.

[0066] S4. Make decisions and allocate rescue tasks based on the rescue environment risk assessment map, and obtain the decision results.

[0067] In step S4, the preset three-level judgment process has the following criteria:

[0068] Level 1 assessment: Based on the severity score of the trapped person's injuries, determine whether to carry out a rescue operation;

[0069] The second level of assessment: compare the estimated time of physiological tolerance of the trapped personnel with the round-trip flight time of the drone to determine the feasibility of the timeliness;

[0070] The third level of judgment: Based on the current payload, remaining energy and target distance of the UAV, calculate the maximum range and determine the physical feasibility.

[0071] The first level of assessment determines whether the injured person has priority for air evacuation. The Injury Severity Score (ISS) ranges from 1 to 75 points; this embodiment uses... This serves as a preset threshold for entering the subsequent air evacuation feasibility assessment. This threshold is used to distinguish seriously injured patients who require priority assessment for air evacuation from those who do not yet have the priority for air evacuation.

[0072] Let the first The decision variables for the injury status of each wounded person are: ,but:

[0073] ,

[0074] when At that time, the system determines the first The injured person is not currently prioritized for air evacuation and will be transferred to the awaiting rescue / self-rescue or remote guidance queue; when When the injury reaches the preset severe injury threshold, the system enters the second-level physiological tolerance time window assessment. The first-level assessment is only used to determine whether to continue the UAV evacuation feasibility assessment and does not directly replace the subsequent assessment of time and range conditions.

[0075] The second-level assessment determines whether the drone can complete the rescue loop within the injured person's physiological tolerance time, thus avoiding aerial evacuation missions that cannot be completed within the effective time window. This includes the mission loop time for the drone to complete the round-trip flight and on-site handling. It consists of round-trip flight time and on-site handling time.

[0076] ,

[0077] in, Indicates the drone's take-off and landing point to the number The one-way distance from the location of the injured person. This represents the theoretical round-trip flight time required for the drone to fly from the take-off and landing point to the location of the injured person and complete its return journey. This indicates the time required for the drone to land at the location of the injured person, load the injured person, connect onboard medical equipment, provide necessary initial first aid, and take off again.

[0078] To make the calculation results applicable to the low-altitude environment of the disaster area, the average cruising speed The effective cruising speed of the drone under the current mission environment is used, rather than the nominal speed under ideal windless conditions. Indicates the first The physiological tolerance time for each injured person can be given by on-site rescuers based on the injured person's condition, or it can be determined by vital sign sensor data and statistical predictions based on ISS and injury type. According to the golden hour principle, effective rescue must meet the following conditions:

[0079] ,

[0080] To allow for a safety margin due to weather changes, takeoff and landing delays, or fluctuations in on-site handling time, this invention introduces a safety factor. and adopt This serves as the actual time window for judgment.

[0081] Let the time decision variable be... ,but:

[0082] ,

[0083] when When this indicates that the drone, after considering safety margins, may still be able to complete the aerial evacuation within the patient's physiological tolerance time, the system enters the third level of physical feasibility assessment; when If the system indicates that the time required for air evacuation has reached or exceeded the safe time window, it will switch to the material delivery mode and prioritize delivering first aid kits, hemostatic supplies, oxygen supply devices, thermal insulation supplies or other life support equipment to the location of the injured to extend the waiting time for the injured.

[0084] The third level of assessment verifies whether the drone, after carrying casualties and onboard medical equipment, possesses the range to complete a round trip and return safely. As the drone's payload increases, its total takeoff weight increases, propulsion system energy consumption rises, and the maximum usable range decreases accordingly. Therefore, range determination should be based on the maximum usable range under payload conditions, rather than directly using the drone's nominal range when unloaded.

[0085] Based on the Breguet range relationship, the range of the UAV The relationship between propulsion efficiency, power consumption coefficient, lift-to-drag ratio, takeoff gross weight, and final weight can be expressed as:

[0086] ,

[0087] in, Indicates the efficiency of advancement. Indicates the power consumption coefficient. Indicates lift. Indicates resistance. Indicates the lift-to-drag ratio; Indicates the total weight at takeoff. The final weight after fuel or electricity consumption is expressed, respectively, and satisfies:

[0088] ,

[0089] in, Indicates the weight of the drone's fuel or equivalent battery. Indicates the empty weight of the drone. Indicates the effective payload weight of the drone. Indicates the weight of the injured person. This indicates the weight of the airborne medical equipment.

[0090] In the rescue decision model of this embodiment, for ease of real-time calculation, the impact of payload and environmental resistance on the maximum range is approximated as follows:

[0091] ,

[0092] in, This indicates the maximum usable range of the drone under given payload conditions. This represents a constant determined by the aerodynamic characteristics of the aircraft model. This represents the environmental resistance coefficient.

[0093] As can be seen from the above relationships, given a fixed UAV model and energy status, the payload... The larger the size, the greater the maximum usable range of the drone. The smaller the value; the lower the environmental resistance coefficient. The larger the value, the more significant the adverse environmental effects such as low-altitude wind resistance, and the smaller the maximum range that the drone can use for rescue missions.

[0094] Let the flight range decision variable be... ,but:

[0095] ,

[0096] when At that time, it was explained that the drone was carrying the first After transporting the wounded and onboard medical equipment, its maximum available range is not less than the round-trip distance of the mission, possessing the physical conditions to complete the air evacuation and return; when If the mission's round-trip distance exceeds the maximum available range in the loaded state, continuing to perform air delivery may cause energy depletion or prevent return. The system then switches to material delivery mode.

[0097] Based on the comprehensive assessment of the first layer of injury severity, the second layer of physiological tolerance time window, and the third layer of physical feasibility, the following measures are taken regarding the first layer of injury severity. Output the final rescue decision command to the injured. The judgment method is as follows.

[0098] ,

[0099] The decision results include three modes: Mode 0 (waiting for rescue / self-rescue), Mode 1 (supply delivery), and Mode 2 (airborne delivery), as shown in Table 1 below.

[0100] Table 1. Rescue modes and their output conditions.

[0101]

[0102] In actual operation, the system performs judgments in the order of severity of injury, physiological tolerance time, and physical feasibility. If the conditions for air evacuation are not met at any level, the subsequent evacuation judgment is terminated in a timely manner, and the corresponding alternative rescue mode is directly output. This hierarchical judgment method reduces invalid calculations and unsafe mission instructions, and can form a clear decision relationship between the constraints of low-altitude environment in the disaster area, the life time constraints of the injured, and the range constraints of UAV payload.

[0103] When the vital signs and location of the injured, the drone's cruising speed, the on-site response time, the payload, or the environmental drag coefficient change, the above model should be recalculated. and and update , , and final decision This allows for real-time adjustments to the rescue approach. The system ultimately outputs rescue instructions that match the current condition of the injured and the capabilities of the drone mission.

[0104] Furthermore, when In this case, the system retains the location and status information of the injured person and outputs instructions for remote guidance, self-rescue prompts, or waiting for subsequent rescue.

[0105] Furthermore, the material delivery mission will be carried out in the following steps:

[0106] (1) The admission criteria are met, and the serious injury determination is satisfied (ISS≥16), but any of the following restrictions apply:

[0107] Time Limit Exceeded: The injured person's physiological tolerance time is shorter than the round-trip time of the drone, and direct evacuation cannot be completed within the golden time.

[0108] Flight or environment unreachable: The target distance exceeds the drone's maximum return radius, or the ground roads are completely damaged, or high-risk areas are impassable, making direct transfer impossible.

[0109] Core objective: To rapidly deliver emergency medical supplies, extend the survival window for the wounded, and create conditions for subsequent evacuation.

[0110] (2) The equipment and supplies system consists of "base terminal + delivery drones + control platform", of which:

[0111] Base end: The rear supply base is responsible for the standardized storage, buffer packaging, and rapid installation of medical supplies;

[0112] Delivery equipment: Medium-sized cargo drones with parachute or precision drop capabilities, supporting modular medical supply cargo compartments;

[0113] Management and control platform: A global monitoring and scheduling system that integrates AI dynamic path planning algorithms to achieve visualized management and control of the entire delivery process;

[0114] Material configuration: Standardized modules to meet the needs of critical injury emergency care, including hemostatic consumables, emergency medicines, fluid resuscitation devices, ventilation tools, fracture fixation equipment, etc., and adopt shock-resistant cushioning packaging to adapt to airdrop scenarios.

[0115] (3) Step-by-step execution process.

[0116] ① Material matching and route planning.

[0117] Based on the injured person's ISS score and injury type, the command center intelligently matches the corresponding emergency medical supplies module and completes rapid packaging; the AI ​​system combines the ground road damage, the distribution of high-risk disaster areas, and weather conditions to dynamically plan the optimal flight path and proactively avoid risky areas.

[0118] ② Drone takeoff and full-process monitoring.

[0119] The cargo drones take off from the rear supply base and fly towards the target area along the planned route; the control platform monitors the flight status, route deviation and surrounding environment throughout the process, and can adjust the route in real time in case of sudden risks.

[0120] ③ Precise airdrop and landing point guidance.

[0121] After the drone arrives over the target area, it selects either a parachute drop or a low-altitude drop mode based on the terrain to ensure accurate landing of supplies and prevent secondary damage. After the drop is completed, the landing coordinates and instructions for using the supplies are simultaneously sent to the on-site rescue personnel.

[0122] ④ Confirmation of results and follow-up procedures.

[0123] Through reconnaissance drones or feedback from on-site personnel, confirm that supplies have been delivered intact and that the wounded have received first aid; continuously assess the on-site environment and the condition of the wounded, and once time and flight conditions are met, initiate the air evacuation process.

[0124] Furthermore, the specific plan for the air evacuation mission is as follows:

[0125] (1) To trigger the admission conditions, three prerequisite judgments must be met simultaneously:

[0126] Injury criteria met: ISS score ≥16, indicating life-threatening serious injury, meeting the criteria for emergency evacuation;

[0127] Timeframe met: The injured person's physiological tolerance time exceeds the round-trip flight time of the drone, and the transfer can be completed within the golden window for treatment;

[0128] Range meets requirements: The maximum return radius of the drone is greater than or equal to the target distance, and the battery power and range support a complete round trip.

[0129] (2) Equipment and personnel system.

[0130] The elements in the diagram adopt a three-tiered architecture: "command center + air formation + ground nodes".

[0131] Command end: The rear medical command center is equipped with a team of telemedicine experts and a drone dispatch platform, relying on satellite communication and wireless data transmission links to achieve full flight control and remote medical guidance;

[0132] Aerial formation: The formation consists of one dedicated rescue drone and N reconnaissance and support drones.

[0133] Specialized medical drones: equipped with lightweight medical stretchers, casualty restraint systems, onboard vital signs monitors, emergency oxygen supply or hemostasis and first aid kits, undertaking core transport functions;

[0134] Reconnaissance and collaborative UAVs: forward escort, responsible for real-time optical image transmission, terrain obstacle reconnaissance, risk area early warning, and ensuring the safety of transfer routes;

[0135] Ground nodes: On-site emergency personnel in high-risk areas and forward transfer stations are responsible for the pre-treatment of the injured and the handover upon arrival.

[0136] (3) Step-by-step execution process.

[0137] ①Task assignment and preliminary preparation.

[0138] After the command center confirms that the three assessments have been passed, it issues a mission to the nearest rescue team, pre-loading the target coordinates, the injured person's condition, and the basic flight path; the on-site rescue personnel simultaneously perform pre-treatment on the injured person, such as stopping bleeding, opening the airway, and temporarily fixing fractures, and place the injured person on a lightweight stretcher and take protective measures.

[0139] ② The formation approaches and connects with the site.

[0140] The formation takes off from a rear base or transit station. Reconnaissance drones lead the way and transmit real-time risk information such as terrain, fire, and combat situation along the route. AI dynamically adjusts the flight path to avoid obstacles. After the rescue drone arrives at the target point, it hovers or lands precisely, and on-site personnel complete the connection and fixation of the stretcher to the drone.

[0141] ③Transfer and remote monitoring during transit.

[0142] The formation returned along the optimal route, with the onboard vital signs module collecting data such as heart rate, blood pressure, and blood oxygen in real time and transmitting it back to the command center. Remote medical experts provided full-time monitoring and guided the onboard system or on-site personnel to carry out emergency response when vital signs fluctuated.

[0143] ④ Arrival handover and task closure.

[0144] After the drone arrives at the forward transfer station or the rear medical center, the ground medical team completes the handover of the wounded and continues to carry out specialized treatment; the flight data and vital sign data throughout the mission are archived simultaneously to complete the rescue loop.

[0145] The two modules mentioned above work together, with material delivery serving as a prerequisite for air evacuation: when direct evacuation conditions are not met, material delivery is used first to stabilize the wounded's condition and extend their survival time; once the on-site risks are reduced and transfer conditions are met, the air evacuation process is then initiated, forming a tiered rescue chain of "first extending life, then transferring," maximizing the success rate of treating seriously injured patients.

[0146] S5. Based on the decision-making results, control the operation drone to perform material delivery or casualty transfer tasks.

[0147] It should be noted that, for the sake of simplicity, the aforementioned method embodiments are all described as a series of actions. However, those skilled in the art should understand that the present invention is not limited to the described order of actions, because according to the present invention, some steps can be performed in other orders or simultaneously.

[0148] Based on the same concept as the UAV intelligent emergency rescue method in complex disaster environments described in the above embodiments, this invention also provides a UAV intelligent emergency rescue system for complex disaster environments. This system can be used to execute the aforementioned UAV intelligent emergency rescue method for complex disaster environments. For ease of explanation, the structural diagram of the UAV intelligent emergency rescue system embodiment in complex disaster environments only shows the parts related to the embodiments of this invention. Those skilled in the art will understand that the illustrated structure does not constitute a limitation on the device, and may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0149] Please see Figure 2 In another embodiment of this application, an intelligent emergency rescue system 10 for unmanned aerial vehicles (UAVs) in complex disaster environments is provided. This system includes:

[0150] Communication support unit 11 is used to temporarily establish an emergency, highly resilient, self-organizing communication network.

[0151] The sensing and detection unit 12 is a reconnaissance drone equipped with multi-source sensors, used for disaster environmental data collection;

[0152] Data processing unit 13 is used for multi-source data fusion, environmental modeling, and risk assessment;

[0153] The decision control unit 14 has a built-in three-level judgment algorithm model for generating rescue mission instructions; the three-level judgment algorithm model includes injury severity score, golden time window and physical feasibility.

[0154] The execution unit 15 is used to perform rescue missions of the decision control unit, including material delivery drones and casualty transfer drones.

[0155] The aforementioned casualty transport drone is equipped with an active stabilization stretcher system based on an inertial measurement unit, which is used to isolate high-frequency vibrations and low-frequency swaying during flight and ensure the physiological stability of the casualty during transport.

[0156] It's worth noting that the active stabilization stretcher system here aims to address the technical problem that traditional passive stretchers cannot counteract the multi-dimensional tilting and impact vibrations in complex transport scenarios, easily causing secondary injuries to critically ill patients. During actual transport, when the external transport vehicle experiences conditions such as pitching uphill / downhill, tilting during turns, road bumps, ship swaying, or takeoff / landing impacts, the attitude sensing unit captures the attitude shift and vibration impact of the base with a millisecond-level response; the central control unit calculates compensation parameters in real time and drives the actuators to maintain a horizontal and stable posture on the patient's support bed, while effectively filtering out vibration impacts across the entire frequency band, ensuring that the patient's position is not affected by transport disturbances.

[0157] The system supports a modular payload interface, allowing for rapid switching between material delivery and casualty transport modules under different conditions, enabling multiple uses from a single machine.

[0158] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0159] The above embodiments are preferred embodiments of the present invention, but the embodiments of the present invention are not limited to the above embodiments. Any changes, modifications, substitutions, combinations, or simplifications made without departing from the spirit and principle of the present invention shall be considered equivalent substitutions and shall be included within the protection scope of the present invention.

Claims

1. A method for intelligent emergency rescue using unmanned aerial vehicles (UAVs) in complex disaster environments, characterized in that: Includes the following steps: Deploy a cluster of communication relay drones to build a highly resilient mobile ad hoc network and establish emergency communication links; The reconnaissance drone is equipped with multiple sensors to search for the target area and acquire multi-sensor data, including visible light images, infrared thermal imaging data, three-dimensional terrain data, and environmental hazard factors. Multi-sensor data is fused and processed to generate a rescue environment risk assessment map, which includes the location of trapped personnel, travel routes, and dangerous areas. Based on the rescue environment risk assessment map, rescue tasks are allocated and decisions are made to obtain the results. Based on the decision-making results, control the operation drones to perform material delivery tasks or casualty evacuation tasks.

2. The intelligent emergency rescue method for unmanned aerial vehicles (UAVs) in complex disaster environments according to claim 1, characterized in that, The highly resilient mobile ad hoc network adopts a distributed mesh topology, allowing nodes to dynamically join and leave.

3. The intelligent emergency rescue method for unmanned aerial vehicles (UAVs) in complex disaster environments according to claim 1, characterized in that, The multi-source sensor employs one or more of the following: visible light camera, infrared thermal imager, lidar, ion mobility spectrometer, and gamma ray detector; the environmental hazard factors include toxic gas concentration, radiation intensity, and building structural stability index.

4. The intelligent emergency rescue method for unmanned aerial vehicles (UAVs) in complex disaster environments according to claim 1, characterized in that, A pre-defined three-level decision-making process is adopted to allocate rescue missions, specifically as follows: Level 1 assessment: Based on the severity score of the trapped person's injuries, determine whether to carry out a rescue operation; The second level of assessment: compare the estimated time of physiological tolerance of the trapped personnel with the round-trip flight time of the drone to determine the feasibility of the timeliness; The third level of judgment: Based on the current payload, remaining energy and target distance of the UAV, calculate the maximum range and determine the physical feasibility.

5. The intelligent emergency rescue method for unmanned aerial vehicles in complex disaster environments according to claim 4, characterized in that, When the injury severity score is lower than the set threshold, the location and status information of the injured person are retained, and instructions for remote guidance, self-rescue prompts or waiting for subsequent rescue are output; otherwise, the second level of judgment is executed. If the timeliness and feasibility are met, the material delivery task shall be carried out first; otherwise, the third-level judgment shall be executed. If the physical feasibility is met, the casualty evacuation mission shall be carried out; otherwise, the material delivery mission shall be carried out first.

6. The intelligent emergency rescue method for unmanned aerial vehicles in complex disaster environments according to claim 4, characterized in that, The third level of determination is specifically as follows: Using the Breguet range relationship, we analyze the relationship between UAV range and propulsion efficiency, power consumption coefficient, lift-to-drag ratio, takeoff weight and final weight. The impact of payload and environmental drag on maximum range is calculated using the following formula: ， in, This represents the maximum usable range of the drone under a given payload. This represents a constant determined by the aerodynamic characteristics of the aircraft model. Indicates the empty weight of the drone. Indicates the effective payload weight of the drone. Indicates the environmental resistance coefficient; The decision variables for calculating the flight range are ,but: ， in, Time-based decision variables representing the feasibility of a timeframe. Indicates the drone's take-off and landing point to the number The one-way distance from the location of the injured person. Indicates the weight of the injured person. Indicates the weight of the airborne medical equipment; when At that time, the drone will be carrying the first After transporting the wounded and onboard medical equipment, its maximum available range is not less than the round-trip distance of the mission, possessing the physical conditions to complete the air evacuation and return; when When the mission's round-trip distance exceeds the maximum available range under the current configuration, the mission is switched to a supply delivery mission.

7. The intelligent emergency rescue method for unmanned aerial vehicles in complex disaster environments according to claim 1, characterized in that, If the decision is to carry out a material delivery mission, the drone will use a temperature-controlled storage and transportation cabin to load the materials and deliver them by rappelling, airdropping, or in collaboration with ground robots. If the decision is to carry out a casualty evacuation mission, the drone will activate an active stabilization stretcher system based on an inertial measurement unit and transmit the casualty's vital signs data back to the rear medical center in real time.

8. A drone-based intelligent emergency rescue system for complex disaster environments, characterized in that: The intelligent emergency rescue method for unmanned aerial vehicles (UAVs) applied to any one of claims 1-7 in a complex disaster environment includes: The communication support unit is used to temporarily establish a highly resilient, self-organizing emergency communication network. The perception and detection unit is a reconnaissance drone equipped with multi-source sensors, used for disaster environmental data collection; The data processing unit is used for multi-source data fusion, environmental modeling, and risk assessment. The decision control unit has a built-in three-level judgment algorithm model for generating rescue mission instructions; the three-level judgment algorithm model includes injury severity score, golden time window, and physical feasibility. The execution unit is used to carry out rescue missions of the decision control unit, including material delivery drones and casualty evacuation drones.

9. The intelligent emergency rescue system for unmanned aerial vehicles (UAVs) under complex disaster environments according to claim 8, characterized in that, The casualty transport drone is equipped with an active stabilization stretcher system based on an inertial measurement unit to isolate high-frequency vibrations and low-frequency swaying during flight.

10. The intelligent emergency rescue system for unmanned aerial vehicles (UAVs) under complex disaster environments according to claim 8, characterized in that, The system is equipped with a modular load interface, which allows switching between the material delivery and casualty evacuation modules under different conditions.

Images