Efficient rescue system and method for unmanned aerial vehicle carrying emergency rescue cabin

Abstract

The invention discloses an efficient rescue system and method for an emergency rescue cabin carried by an unmanned aerial vehicle, and relates to the technical field of disaster rescue. Comprising the following steps: firstly, carrying out comprehensive inspection on equipment, and calculating the endurance of an unmanned aerial vehicle; disaster area static and dynamic environment data and resource deployment data are imported in a layered mode; and finally, establishing a three-layer communication network and executing a stability test. Setting a device data acquisition frequency parameter and a data priority parameter, and starting a multi-device cooperation mode to complete multi-source data acquisition; according to the technical key points, full-state perception and high-reliability communication are integrated, a full-link guarantee system is constructed, the problems of insufficient rescue pertinence and easy process interruption caused by information fault, easy communication interruption and the like existing in traditional rescue are solved, and the technology realizes continuous monitoring of vital signs of the wounded personnel in the whole process from rescue to transfer and in-cabin environment, so that the rescue efficiency is improved. Key data support is provided for medical assistance, it is ensured that the whole rescue process is not interrupted, and the method has good use prospects.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of disaster rescue, in particular to an efficient rescue system and method of an unmanned aerial vehicle carrying an emergency rescue cabin. BACKGROUND

[0002] Disaster rescue is a multi-element coordinated emergency action to cope with natural disasters and man-made disasters, with the core of protecting life safety and reducing property loss. It runs through the whole process of disaster. Before the disaster, risk investigation, early warning and material reserve are carried out to build a defense line. During the disaster, personnel search and rescue, medical rescue and transfer of trapped people are focused on, and the emergency repair of infrastructure such as transportation, communication and power is simultaneously promoted, while secondary disasters such as aftershocks and epidemics are strictly prevented. After the disaster, the management of temporary resettlement sites and the basic life security of disaster victims are focused on, and the reconstruction of homes and the restoration of production and life order are gradually promoted.

[0003] Traditional disaster rescue (such as earthquake, flood, landslide) faces many challenges: road damage makes it difficult for rescue forces to quickly enter; the complex environment of the disaster area leads to low search and rescue efficiency; the condition of the wounded is unknown, making it difficult to implement precise rescue; the communication network is damaged, requiring the use of temporary emergency communication. Although unmanned aerial vehicle technology has been applied to material delivery, existing solutions have obvious limitations.

[0004] However, the existing disaster rescue scheme has the following defects,

[0005] Insufficient communication capability: the existing network is prone to interruption in disasters, with high latency and limited bandwidth, making it difficult to support real-time return of multi-channel high-definition video, sensor data and high-precision real-time control;

[0006] Low degree of intelligence: the flight path of the unmanned aerial vehicle is mostly preset or manually controlled, and cannot be dynamically and intelligently planned according to real-time environmental and wounded information;

[0007] Poor delivery accuracy: traditional air delivery methods have low accuracy and can easily cause secondary injury to the wounded, and cannot achieve safe and precise delivery of materials in complex terrain (such as cliffs and debris gaps);

[0008] Weak coordination: multiple unmanned aerial vehicles and multiple rescue units lack efficient coordination, making it difficult to form systematic rescue capabilities;

[0009] In summary, the existing disaster rescue scheme does not meet the needs of society, and therefore we propose an efficient rescue system and method of an unmanned aerial vehicle carrying an emergency rescue cabin. SUMMARY

[0010] To achieve the above purpose, the present application is implemented by the following technical scheme:

[0011] An efficient rescue method of an unmanned aerial vehicle carrying an emergency rescue cabin, comprising the following steps:

[0012] First, the device is comprehensively tested, and the endurance of the unmanned aerial vehicle is calculated; then, static and dynamic environment data and resource deployment data in the disaster area are introduced in layers; finally, a three-layer communication network is built and stability testing is performed;

[0013] Set the device data collection frequency parameter and data priority parameter, start the multi-device collaboration mode to complete multi-source data collection; perform filtering, fusion and compression processing on the collected data, and transmit the data to the ground command center system after marking the data priority; transmit the data in layers, perform data integrity verification synchronously, plan the path according to the data priority, and generate the optimal rescue scheme;

[0014] Execute the optimal rescue scheme, and after the unmanned aerial vehicle arrives at the lowering point, perform the wounded rescue operation according to the lowering scheme; transport the wounded to the designated terminal, and then return to the supply point to complete the wounded transportation task.

[0015] Preferably, the step of comprehensively testing the device is as follows:

[0016] The scope of the devices to be tested is determined, including unmanned aerial vehicle devices, intelligent rescue cabin bodies and communication terminals;

[0017] Perform the unmanned aerial vehicle device testing process: check the appearance state of the propeller, confirm whether there are deformation, crack defects; test the rotation function of the propeller, and determine whether there are jamming phenomena;

[0018] Carry out the detection of the endurance of the unmanned aerial vehicle, and simultaneously detect the sensing devices carried by the unmanned aerial vehicle, to confirm that the unmanned aerial vehicle body and the sensing devices meet the normal operation requirements;

[0019] Start the first round of inspection of the intelligent rescue cabin body: detect the protection performance of the intelligent rescue cabin body, confirm that the protection effect meets the standard; check the internal electronic instruments and various sensing devices of the intelligent rescue cabin body, and verify their running state;

[0020] Perform the second inspection of the intelligent rescue cabin body: check the internal drug reserve amount and instrument oxygen supply amount of the intelligent rescue cabin body, and confirm that the supply is sufficient; detect the power-on function of the intelligent rescue cabin body equipment, and determine whether it can be normally powered on; check the connection state of the equipment interface, and confirm that there is no looseness;

[0021] Test the communication terminal, and check whether the communication equipment of the ground command center system and the handheld communication terminal can be normally used.

[0022] Preferably, the endurance of the unmanned aerial vehicle is calculated as the estimated flight distance, and the specific calculation method is as follows:

[0023]

[0024] In the formula, is the estimated distance that can be flown, For standard battery capacity, This refers to the annual battery degradation rate. For battery lifespan, This represents the percentage of remaining battery power. For the no-load power consumption of the drone, For the power consumption of the intelligent rescue cabin under heavy load, This is the actual load capacity. For the drone's flight speed, The remaining percentage after deducting the reserved percentage for safety reasons.

[0025] The preferred steps for importing static and dynamic environmental data and resource deployment data in the disaster area in a layered manner are as follows:

[0026] A hierarchical data import method was adopted to import static and dynamic data of the disaster area according to the type of disaster.

[0027] Initiate the import operation of static data for the disaster area: The data type to be imported is high-definition map static data of the disaster area; it is clear that the high-definition map static data includes terrain slope and building information, and the data source is satellite information and information collected by surveying equipment.

[0028] Perform the disaster area dynamic data import operation, and import the data type as dynamic environmental data;

[0029] Initiate the data fusion and generation operation: combine the imported static and dynamic data to generate visualized environmental data; overlay the drone flight distance parameters onto the visualized environmental data to generate a drone rescue map;

[0030] Perform the resource deployment data import operation, and specify that the imported data includes the number of drones, the number of backup drones, the safe areas in the disaster area, temporary charging and maintenance points, and personnel responsibilities.

[0031] Preferably, the multi-device collaborative acquisition of multi-source data includes: lidar deployed on drones to collect 3D data of disaster area terrain and obstacles, generating 3D point cloud models; optical cameras deployed on drones to collect disaster area terrain image data; infrared thermal imagers deployed on drones to collect data on the location of the injured and preliminary heat source identification; vital sign sensors deployed inside the intelligent rescue cabin to collect vital sign data of the injured; environmental sensors deployed on drones and environmental sensors deployed on handheld terminals of ground rescue personnel to collect environmental data; drone status sensors deployed on drones to collect drone's own status data; and resource monitoring sensors deployed on the intelligent rescue cabin to collect data on the remaining resources of the intelligent rescue cabin.

[0032] The preferred steps for multi-device collaborative acquisition of multi-source data are as follows:

[0033] Set the data collection frequency of the drones and the equipment on the intelligent rescue cabin, and clarify the transmission cycle and priority of different types of data;

[0034] Set the data acquisition model and accuracy parameters for the drone and the equipment on the intelligent rescue cabin;

[0035] The multi-device collaborative acquisition mode is activated, controlling optical equipment, infrared equipment, and radar equipment to operate synchronously and collect disaster area terrain data and injured status data;

[0036] The lidar scanning function is activated to scan the terrain of the disaster area and generate a 3D point cloud model; the optical camera is controlled to perform image capture operations simultaneously, and the captured images are superimposed with the 3D point cloud model to achieve data visualization;

[0037] The edge computing unit on the drone is used to filter the image data collected by the optical camera, removing invalid and abnormal data with a resolution of less than 80%.

[0038] The system performs fusion processing on the collected data of the same type and source; for large data volumes greater than 10Mbps, it uses efficient video coding technology for compression; and it marks the priority of the pre-processed data before transmitting it to the ground command center system.

[0039] Preferably, the steps for generating the optimal rescue plan are as follows:

[0040] Perform layered data backhaul: the core layer backhauls high-volume, high-priority data, the collaboration layer backhauls shared data from multiple drones, and the terminal layer backhauls simplified task data.

[0041] The ground command center system receives layered data transmission and initiates the data integrity verification process.

[0042] Define the path planning constraints, covering the drone's flight distance and casualty priority;

[0043] Sorting operations are performed according to the priority of the wounded from high to low; the real-time location of the drone, the location of the wounded, and the location of the transport destination are obtained, and an initial rescue path is generated;

[0044] The real-time obstacle monitoring mechanism is activated. When a sudden obstacle is detected, the obstacle location data is added and a new rescue path is generated. The newly generated rescue path is then simulated and verified to confirm its feasibility.

[0045] When there are multiple drones and multiple casualties, calculate the straight-line distance between each drone and each casualty, the drone's operating energy consumption, and the operating time.

[0046] A multi-objective optimization model was constructed to calculate the evaluation value of each drone-wound matching scheme. The matching scheme with the highest total evaluation value was selected as the optimal rescue scheme.

[0047] The preferred steps for rescuing the wounded according to the rappelling plan are as follows:

[0048] Receive the coordinates of the injured, verify whether there are any sudden obstacles in the target area, and determine whether it is suitable for rappelling.

[0049] The UAV system receives a hovering command, flies to the target area, keeps the lidar continuously running, and scans the surrounding environment in real time.

[0050] The UAV generates a 3D environment model of the target area based on real-time scanning data from LiDAR; it then marks the location information of obstacles in the model, triggering flight attitude adjustments.

[0051] Adjust the drone so that it is directly below the cabin and coincides with the coordinates of the injured person. Retrieve the superimposed image of the drone's hovering position and the target point to verify and confirm that there is no positional deviation.

[0052] After the drone sends a signal to the ground command system indicating that it is ready to rappel, it performs the rope release operation according to the preset rappelling plan.

[0053] Collect real-time wind field data and UAV acceleration data, calculate rope tension and wind field component based on the collected data, and start the rope descent speed adjustment program when the data exceeds the preset threshold.

[0054] The system monitors and determines in real time whether the swing amplitude of the intelligent rescue cabin exceeds a preset angle threshold; if it exceeds the threshold, it sends a rappelling pause command and restarts the rappelling process after the wind is detected to have weakened.

[0055] The preferred approach after rappelling to the ground is as follows: After the intelligent rescue cabin touches the ground, the drone system controls the cabin door to open, and ground personnel assist the injured person into the intelligent rescue cabin. The intelligent rescue cabin uses the matched resources to provide initial first aid to the injured person. At the same time, the vital signs sensors inside the intelligent rescue cabin are activated to collect the injured person's data and transmit it back to the ground command center system in real time for the command personnel to monitor the status. The intelligent rescue cabin is then shut down, and the intelligent rescue cabin sends a shutdown signal to the drone system. The drone then initiates the rope retrieval function. After the rope is retrieved, the injured person is transported according to the planned path in the optimal rescue plan, and the drone transports the injured person to the final delivery location.

[0056] A highly efficient rescue system for drones equipped with an emergency rescue cabin includes a network construction module, a scheme formulation module, and a rescue control module.

[0057] Network construction module: First, conduct a comprehensive inspection of the equipment and calculate the drone's battery life; then, import static and dynamic environmental data and resource deployment data of the disaster area in layers; finally, build a three-layer communication network and perform stability tests.

[0058] Solution Development Module: Sets equipment data acquisition frequency parameters and data priority parameters, initiates multi-device collaborative mode to complete multi-source data acquisition; performs filtering, fusion and compression processing on the acquired data, marks data priority and transmits it to the ground command center system; transmits data in layers, performs data integrity verification synchronously, plans paths based on data priority, and generates the optimal rescue plan;

[0059] Rescue and Control Module: Executes the optimal rescue plan. After the drone arrives at the rappelling location, it performs the rescue operation for the wounded according to the rappelling plan; transports the wounded to the designated destination, and then returns to the supply point to complete the wounded transportation mission.

[0060] This invention provides a highly efficient rescue system and method for an unmanned aerial vehicle (UAV) equipped with an emergency rescue cabin, which has the following beneficial effects:

[0061] This invention employs UAV-based technology to establish independent relay station networks and gigabit emergency networks. By constructing low-latency transmission channels through three-layer communication links, and using UAVs to establish independent relay station networks and gigabit emergency networks, ultra-low air interface latency is achieved, providing core support for accurate docking with the wounded.

[0062] This invention controls the rappelling of the intelligent rescue cabin, achieving real-time closed-loop control of the drone and the rappelling mechanism. This ensures extremely high precision and stability of the cabin rappelling in complex airflow environments. Even in environments such as fires, dense smoke, floods, and waves, it can still achieve a smooth rappelling, avoiding secondary injuries to the wounded. It can greatly improve the efficiency and safety of rescue operations, has good performance, and has promising application prospects.

[0063] This invention integrates multi-source data through a ground edge center algorithm, enabling multi-source information fusion and intelligent decision-making. This allows multiple drones to work collaboratively like a swarm, maximizing rescue efficiency and prioritizing the treatment of the most critically injured. It effectively solves the problems of low efficiency caused by task conflicts and prioritizing the treatment of minor injuries in traditional multi-drone rescue operations. It can maximize the rescue coverage area, has good performance, and has promising application prospects.

[0064] This invention integrates full-state perception and high-reliability communication to construct a full-link support system, solving the problems of insufficient targeted rescue and easy interruption in traditional rescue, such as information gaps and easy communication interruptions. This technology realizes continuous monitoring of the vital signs and cabin environment of the injured throughout the entire process from rescue to transfer, providing key data support for medical rescue and ensuring uninterrupted rescue. This integrated technology lays the foundation for intelligent and precise disaster emergency rescue, has good results, and has good application prospects. Attached Figure Description

[0065] Figure 1 This is a flowchart illustrating an efficient rescue method using an unmanned aerial vehicle (UAV) equipped with an emergency rescue cabin, according to the present invention.

[0066] Figure 2 This is a structural block diagram of an efficient rescue system for an unmanned aerial vehicle (UAV) equipped with an emergency rescue cabin, according to the present invention. Detailed Implementation

[0067] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0068] This invention targets disaster scenarios such as earthquakes, floods, and forest fires. It utilizes drones to carry emergency rescue cabins, combined with the ultra-low latency, ultra-high reliability, and massive connectivity of networks, as well as the speed of artificial intelligence edge computing, to achieve precise, automatic, and coordinated rappelling and delivery of rescue cabins, enabling rapid rescue of patients.

[0069] Example 1:

[0070] Please see Figure 1 This embodiment provides an efficient rescue method for drones equipped with emergency rescue cabins, including the following steps:

[0071] S1. After equipment inspection, import environmental data and resource deployment data, and build a three-layer communication network;

[0072] S1.1 Equipment inspection;

[0073] The equipment includes drones, intelligent rescue cabins, and communication terminals. Before building the communication network, the equipment needs to be comprehensively tested to ensure it can be used normally. When testing drones, check whether the drone propellers are deformed, have cracks, or have any jamming during rotation. Also check the drone's endurance and whether its sensing devices, such as lidar and infrared thermal imagers, can operate normally. When testing intelligent rescue cabins, check their protective effect, whether the internal electronic instruments can work properly, and whether the sensing devices, such as hazardous gas sensors, can operate normally. The communication terminals include the communication equipment of the ground command center system and handheld communication terminals, and their functionality must be tested.

[0074] Drone equipment needs to have its remaining range calculated in advance during inspection to facilitate subsequent rescue operations;

[0075] When estimating range, the estimated flight distance is determined using the following calculation method:

[0076]

[0077] In the formula, To estimate the possible flight distance, For standard battery capacity, This refers to the annual battery degradation rate. For battery lifespan, This represents the percentage of remaining battery power. For the no-load power consumption of the drone, Power consumption for cabin load capacity, This is the actual load capacity. For the drone's flight speed, The remaining percentage after deducting the reserve ratio for safety reasons, for example, if the reserve ratio is 20%, then... It is 80%.

[0078] During the inspection of the intelligent rescue cabin, it is also necessary to further check whether the oxygen supply to the medicines and instruments inside the rescue cabin is sufficient, whether the equipment can be powered on normally, and whether there are any loose interfaces.

[0079] S1.2, Import data;

[0080] When importing static and dynamic data from disaster areas, a hierarchical data import method is used, and data is imported according to disaster type.

[0081] The steps for importing static and dynamic data from the disaster area are as follows:

[0082] Import static data from high-resolution maps of the disaster area, including terrain slopes and buildings, specifically importing data collected in real time by satellites and surveying equipment.

[0083] Importing disaster dynamic data involves importing dynamic environmental data, such as fires, typhoons, and floods. For fires, infrared monitoring data is used to mark temperature and visibility to divide areas. For floods, satellite remote sensing data is used to mark landing sites and water flow speed.

[0084] The imported data is combined to generate a visual representation of the environment, which is then overlaid with the drone's flight distance to create a drone rescue map.

[0085] The imported data also includes resource deployment data, which includes the number of drones, the number of backup drones, the safe areas in the disaster area obtained based on the analysis of static and dynamic environmental data in the disaster area, the temporary charging and maintenance points deployed in the disaster area, and the responsibilities of relevant personnel in the disaster area.

[0086] S1.3, Set up the network;

[0087] The steps to build a three-layer communication network are as follows:

[0088] The core communication layer is established by activating the gigabit emergency private network transmitter at the ground command and control center, launching the UAV relay station, establishing a private network connection with the ground, calculating the total bandwidth requirement of the core layer, verifying transmission stability, and having the UAV relay station enter hovering mode to maintain the core layer connection.

[0089] When technicians turn on the gigabit private network transmitter at the ground command and control center, they usually set the frequency to 5.8GHz to avoid interference from civilian frequency bands. The drone relay station selects a drone with a payload of 50 kg, which flies to a height of 50-100 meters, avoiding trees and buildings, thus ensuring the signal range. The drone relay station can automatically search for ground private network signals and complete the connection.

[0090] Calculate the total bandwidth requirement of the core layer, select the bandwidth based on the total bandwidth requirement of the core layer, use the bandwidth redundancy calculation method to calculate the bandwidth requirement of a single drone, and then summarize them. In order to prevent unexpected events, the corresponding proportion is increased. 100Mbps bandwidth is sufficient for small disasters, and gigabit private network bandwidth is used for large disasters.

[0091] To test transmission stability, the ground command and control center sends standard video and a standard number of simulated control commands to the UAV relay station. The data transmission status is monitored to determine if there are any issues such as lag, command response exceeding the preset time, or data packet loss, ensuring that the core layer can stably transmit critical data and commands.

[0092] The system controls the drone relay station to hover at a preset altitude, continuously transmitting its own position and signal strength data back to the ground command center system, maintaining stable coverage of the core layer, and providing a data forwarding foundation for the coordination layer and terminal layer.

[0093] A communication collaboration layer was established, specifically by activating the self-organizing network module on the drone performing the mission. The self-organizing network module automatically searches for and joins the collaboration layer network. The network setup time and node switching stability were tested to verify the data sharing function of multiple drones.

[0094] Specifically, the ground command and control center sends network formation instructions to all drones that need to perform missions. After receiving the instructions, the drones activate the self-organizing network module, which automatically searches for and joins the collaborative layer network to achieve decentralized communication between drones.

[0095] When in use, the first drone in the core layer serves as the first node. Subsequent drones automatically scan the surrounding area and connect to the network when they discover the first node, thus forming a collaborative layer network.

[0096] After the collaborative layer network is constructed, performance testing is conducted to test the time taken for the drone to start the self-organizing network module and join the network, as well as the time taken to automatically join the network after disconnecting from it, in order to test whether it is suitable for use by drones in rescue operations.

[0097] Establish a communication terminal layer, activate the LoRa links of all access devices, complete device pairing, and verify data transmission functionality.

[0098] For example, when rescuers activate the communication mode on their handheld terminals, the drone and the intelligent rescue cabin receive instructions from the ground command center system, activate their LoRa transmission modules, and both enter a signal search state, enabling low-power, long-distance communication to facilitate search and rescue operations.

[0099] Communication data can be encrypted using the AES-256 algorithm to ensure the security of data transmission. Ordinary rescue information can also be transmitted directly without encryption.

[0100] Before communication, data transmission tests are conducted, such as handheld terminals sending coordinates, intelligent rescue cabins sending vital sign data of rescued persons, drones transmitting flight data, and real-time collected data.

[0101] Real-time monitoring of communication link stability includes bandwidth utilization, data packet loss rate, signal strength, and command response time. The monitored data is displayed in charts and graphs, with annotations indicating any anomalies, to help relevant personnel understand the situation.

[0102] To determine if there is an anomaly, simply compare the monitored data with the corresponding set threshold.

[0103] For example, it can determine whether the bandwidth utilization exceeds the set 80%. If it does, it can query the duration of the excess. If the excess duration is less than 10 seconds, it is marked as an instantaneous anomaly. If it exceeds 10 seconds, it is marked as an anomaly, and the ground command center system issues an audible and visual warning to facilitate relevant technical personnel to view and handle the situation.

[0104] A link monitoring log is generated every hour. The log includes fields such as time, monitoring indicators, values, whether it is abnormal, handling plan, and handling result, and is stored in the corresponding database for easy subsequent analysis.

[0105] This invention employs UAV-based technology to establish independent relay station networks and gigabit emergency networks. By constructing low-latency transmission channels through three-layer communication links, and using UAVs to establish independent relay station networks and gigabit emergency networks, ultra-low air interface latency is achieved, providing core support for accurate docking with the wounded.

[0106] S2. Collect data, preprocess the collected data, analyze the preprocessed data, and generate a rescue plan;

[0107] S21, Data Acquisition;

[0108] Collect multi-source data such as disaster area terrain, injured status, and equipment parameters.

[0109] Before data collection, it is necessary to set the collection frequency of the drone and the equipment on the intelligent rescue cabin, and determine the transmission cycle and priority of different types of data.

[0110] For example, vital sign data collected by the intelligent rescue cabin is prioritized over the operational data of the drone, facilitating better rescue efforts.

[0111] Setting the acquisition frequency of the drone and the equipment on the intelligent rescue cabin includes setting the acquisition model and acquisition accuracy of the optical camera, the acquisition frequency and acquisition accuracy of the infrared thermal imager, the acquisition frequency and acquisition accuracy of the lidar, the acquisition frequency and acquisition accuracy of the vital signs sensor, the acquisition frequency and acquisition accuracy of the environmental sensor, and the acquisition frequency and acquisition accuracy of the drone status sensor.

[0112] The equipment on the drones and intelligent rescue cabins needs to be calibrated regularly to ensure the accuracy of the collected data.

[0113] The highest priority is the vital signs of critically injured patients, the real-time location of the injured, and the alarm data of drones and intelligent rescue cabins. The medium priority is the status of ordinary injured patients, terrain and obstacle data, and the remaining resources of intelligent rescue cabins. The lowest priority is the ambient temperature and non-critical parameters of drones and intelligent rescue cabins.

[0114] The data collected by the drone includes three-dimensional data of the disaster area terrain and obstacles, data on the location and preliminary condition of the injured, data on the disaster area environment and the drone's own status, and the collected data is transmitted to the edge computing unit carried by the drone in real time.

[0115] Preliminary status data was collected through the onboard intelligent rescue cabin. Since the intelligent rescue cabin is mounted on a drone, it is categorized as drone-collected for ease of analysis and processing.

[0116] When collecting data, the drones use a multi-device collaborative acquisition method, which means that optical, infrared, radar and other devices work synchronously to complement each other's data and achieve comprehensive acquisition of information about the disaster area.

[0117] When collecting data, the lidar scans the disaster area terrain 10-15 times per second to generate a 3D point cloud model, marking the location of various obstacles, such as ruins, trees, buildings, and fire zones, and recording the coordinates of the obstacles.

[0118] Furthermore, the optical camera simultaneously captures terrain images, which are then overlaid with the point cloud model to achieve visualization.

[0119] The initial condition data of the injured is obtained by identifying the heat source of the human body through an infrared thermal imager. When the rescued person is inside the intelligent rescue cabin, the initial condition data of the injured person is obtained by collecting the vital signs data of the rescued person through the instruments inside the intelligent rescue cabin.

[0120] Data on the disaster area environment and the drone's own status were collected through sensors installed on the drone. All collected data were named using timestamp sets for the drone and the data collection device, which facilitates subsequent statistical archiving.

[0121] S22, Data preprocessing;

[0122] Data preprocessing involves filtering, fusing, and compressing data using edge computing units mounted on drones.

[0123] The steps of data preprocessing are as follows:

[0124] The collected data is filtered to remove invalid and abnormal data. Invalid and abnormal data are data that do not meet the standards, such as data with an image clarity of less than 80% from optical cameras or data with a point cloud missing rate of more than 10% from lidar.

[0125] By integrating data from multiple devices that share the same source, data accuracy can be improved. For example, by merging similar types of data, such as by weighted averaging the coordinates of the injured marked by the infrared thermal imager and the terrain coordinates marked by the lidar, the fused coordinates can be obtained. This method can effectively improve accuracy. By combining the environmental data from the drone environmental sensor with the environmental data fed back by the handheld terminal of the ground rescue personnel, the risk level of different areas can be marked.

[0126] Large-volume data is compressed to reduce transmission bandwidth usage; data larger than 10Mbps is compressed using an efficient video encoding compression algorithm with a compression ratio of 10:3 to ensure no significant loss of image quality after compression; the 3D point cloud model generated by LiDAR is downsampled to retain only key feature points, thereby reducing the amount of data processing.

[0127] The preprocessed data is marked with priority and transmitted to the ground command center. Priority is added to the preprocessed data according to the previously set standards. For example, high, medium and low Chinese characters can be added, or the serial number 123 or the letters ABC can be used instead.

[0128] S23, Data Analysis;

[0129] The data analysis process involves the core layer transmitting high-volume, high-priority data, the collaboration layer transmitting shared data from multiple UAVs, the terminal layer transmitting simplified mission data, and the ground command center system receiving and verifying the data integrity.

[0130] By using a layered approach to transmission, congestion on a single link can be avoided, ensuring efficient transmission of different types of data.

[0131] The core layer, coordination layer, and terminal layer transmit different types of data, and transmit them sequentially according to priority.

[0132] For example, in the data transmitted in the core layer, the corresponding high priority is the vital signs data and real-time location data of critically injured patients, the corresponding medium priority is the status data of ordinary patients and the pre-processed 3D point cloud model data, and the corresponding low priority is the environmental data and non-critical data of UAVs.

[0133] In the data transmitted by the collaboration layer, the high priority corresponds to the status data and cabin resource data of the UAV for critically injured patient rescue, the medium priority corresponds to the status data and cabin resource data of the UAV for ordinary patient rescue, and the low priority corresponds to the status key data of the backup UAV.

[0134] In the data transmitted at the terminal layer, the high priority corresponds to the coordinates of critically injured patients and rescue precautions, the medium priority corresponds to the coordinates of ordinary injured patients and rescue precautions, and the low priority corresponds to the safety tips for the disaster area environment.

[0135] When the ground command center system receives and verifies data integrity, it mainly verifies the packet loss rate. For high-priority data, the packet loss rate should be below 1%. If it reaches 1%, immediate transmission is triggered until the data is completely transmitted. For medium-priority data, the packet loss rate should be below 3%. If it reaches 3%, reordering and transmission is triggered until the data is completely transmitted, with a maximum of 3 retransmissions. For low-priority data, the packet loss rate should be below 8%. If it reaches 8%, on-demand retransmission is triggered.

[0136] S24, Path planning;

[0137] The input path planning has multiple constraints, which are conditions set in the previous steps, such as the distance that the drone can fly, the priority of the wounded, and the requirement to rescue one person at a time. Critically injured patients have a higher priority than ordinary patients. This step follows the order of high priority, medium priority, and low priority set above.

[0138] Prioritize the injured and plan routes accordingly; ensure that critically injured patients receive timely treatment.

[0139] Dynamically adjust the path to sudden obstacles, verify the feasibility of the path, and generate a path plan when the feasibility is verified.

[0140] The location of the drone, the location of the wounded, and the destination of the transport are obtained. The RRT algorithm is used to generate an initial path, that is, to fly from the current location to the location of the wounded, and then to the destination of the transport.

[0141] When a sudden obstacle is detected that affects takeoff, flight, or landing, the obstacle data is added and a new path is generated. This is mainly to deal with sudden situations in disaster areas. After software simulation verifies that the drone can carry out rescue along the path, the path data is sent to the corresponding drone, which then carries out the rescue. At the same time, the ground command center system sends relevant information to staff around the route to assist in the transfer of the injured or supplies.

[0142] During a rescue operation, if none of the drones are assigned a rescue mission or the drone closest to the injured person is assigned a rescue mission, the route will be assigned to the drone closest to the injured person if the drone's remaining range is sufficient to cover the path.

[0143] When there are multiple drones and multiple casualties, first organize the relevant data of the casualties and drones, calculate the correlation parameters between drones and casualties, and construct the objective function and constraints of the multi-objective optimization model.

[0144] The casualty data includes injury assessment results, and the drone data includes drone parameters. The correlation parameters between the drone and the casualty are calculated, including the straight-line distance between the drone and the casualty, which is directly calculated using coordinates. The energy consumption required to run along the path, the time required to run, and the path length are also calculated. The path length is directly obtained during path planning. The energy consumption required to run the path = path length × energy consumption per unit path + additional energy consumption; the additional energy consumption is the energy consumption for takeoff and landing; the time required to run = path length / drone flight speed + takeoff and landing time.

[0145] The rescue benefits for different types of casualties are set, and the rescue benefits are positively correlated with the severity of the casualties. The benefits of rescuing critically injured casualties are much higher than those of minor casualties. A matching scheme is generated by randomly assigning drones and casualties. Then, the evaluation value of each scheme is calculated. The evaluation value is the sum of the evaluation values ​​generated by all drones rescuing casualties in the scheme. Evaluation value = rescue benefit - rescue loss. Rescue loss = 0.6 × time required for operation + 0.4 × energy consumption required for path operation / standard battery capacity × 100.

[0146] The ground command center system selects the option with the highest total evaluation value by comparing all options as the optimal rescue option, and then executes the rescue according to the optimal option.

[0147] During the rescue operation, the ground command center system categorizes and organizes the received data according to data type, time, and disaster scenario, stores the data in a local database and an off-site backup center, sets a data retention period, and ensures traceability.

[0148] This invention controls the rappelling of the intelligent rescue cabin, achieving real-time closed-loop control of the drone and the rappelling mechanism. This ensures extremely high precision and stability of the cabin rappelling in complex airflow environments. Even in environments such as fires, dense smoke, floods, and waves, it can still achieve a smooth rappelling, avoiding secondary injuries to the wounded. It can greatly improve the efficiency and safety of rescue operations, has good performance, and has promising application prospects.

[0149] S3. Execute the optimal rescue plan. When the drone flies to the rappelling location, it will rescue the wounded according to the rappelling plan and complete the transport of the wounded.

[0150] S31. Rope descent confirmed;

[0151] Before rappelling, the rappelling location needs to be determined. At this time, ground personnel receive the coordinates of the injured person, verify the target environment on site, and determine whether it is suitable for rappelling. The drone system receives the hovering command, flies to the airspace above the target area, scans the environment with lidar, adjusts its attitude to achieve precise positioning, confirms the hovering position through a handheld terminal, and sends back a rappelling signal.

[0152] When verifying the target environment on-site, personnel in the vicinity confirm that there are no sudden obstacles at the target point. If there are obstacles, personnel in the vicinity report a new rappelling point.

[0153] While the drone is in flight, the lidar remains activated to scan the area around the target point, generate a 3D environment model, mark the location of obstacles, and automatically adjust the flight attitude so that the area directly below the drone coincides with the coordinates of the injured person. Relevant personnel can view the superimposed image of the drone's hovering position and the target point through a handheld terminal. After determining that there is no deviation, the drone will be rappelled down.

[0154] S31, rappelling control;

[0155] The rappelling plan is a standard rappelling plan developed based on the environment, which records the rappelling speed, etc. However, since the site is uncontrollable, it is necessary to collect data on the surrounding environment and make adjustments.

[0156] The rappelling process is as follows: The drone system releases the rope at a set speed. The drone system collects wind field and acceleration data in real time, calculates the tension and wind field component. If the tension exceeds the threshold, the drone automatically adjusts its speed. In case of an emergency, ground personnel send a pause command and observe the swaying amplitude of the cabin, and report environmental interference.

[0157] For example: Under normal circumstances, the drone system first accelerates from 0.2 m / s to 0.5 m / s, reaching its maximum speed of 0.5 m / s when the descent distance is 0.625 meters, and then descends at a constant speed of 0.5 m / s. The drone system collects wind speed every second. If the wind speed is lower than the standard value, the above plan remains unchanged. If the wind speed is higher than the standard value, the descent speed is reduced. The reduced speed is calculated based on the corresponding wind force. When ground personnel observe that the sway of the cabin exceeds the set angle, they can send a pause command through a handheld device, and the drone will pause the descent and restart after the wind weakens.

[0158] When the wind is weak, although you can rappel directly, you will be off-target by a certain distance. Therefore, you can adjust the hovering power output of the drone to counteract the horizontal pull of the wind on the cabin. Specifically, the drone system outputs power corresponding to the wind speed in the opposite direction of the wind direction to counteract the influence of the wind.

[0159] When the intelligent rescue cabin is lowered to the standard height, the drone automatically switches to deceleration mode, reducing the descent speed to 0.1 meters per second until the pressure sensor at the bottom of the intelligent rescue cabin detects a landing signal, at which point the drone stops lowering the rope.

[0160] S32, Wounded personnel reception;

[0161] After the intelligent rescue cabin touches the ground, the drone system controls the cabin door to open, and ground personnel assist the injured person into the intelligent rescue cabin to provide initial first aid using matching resources. The vital signs sensors in the intelligent rescue cabin collect the injured person's data and transmit it back to the ground command center system in real time. Ground personnel close the intelligent rescue cabin, and the intelligent rescue cabin feeds back data to the drone system. The drone system then activates the rope retrieval function, retrieving the rope at a set speed. The drone system transports the injured person to the destination location along the planned path. The drone system then initiates rope retrieval again, setting the retrieval speed to 1.0 m / s. When 1.5 meters of rope remain, the speed automatically decreases to 0.3 m / s.

[0162] The drone returned to the resupply point and landed smoothly. Personnel treated the injured, cleaned the fuselage, inspected the propellers, ropes and other parts for wear, collected battery power data again to determine whether to continue or recharge, and removed the intelligent rescue cabin from the drone, cleaned and disinfected the interior, and replenished the consumed emergency rescue resources.

[0163] All data generated during the rescue operation is compiled and stored for easy review and analysis later.

[0164] This invention integrates multi-source data through a ground edge center algorithm, enabling multi-source information fusion and intelligent decision-making. This allows multiple drones to work collaboratively like a swarm, maximizing rescue efficiency and prioritizing the treatment of the most critically injured. It effectively solves the problems of low efficiency caused by task conflicts and prioritizing the treatment of minor injuries in traditional multi-drone rescue operations. It can maximize the rescue coverage area, has good performance, and has promising application prospects.

[0165] Example 2:

[0166] Based on Example 1, such as Figure 2 As shown, this embodiment also provides a highly efficient rescue system for drones equipped with an emergency rescue cabin, including a network construction module, a scheme formulation module, and a rescue control module:

[0167] Network construction module: First, conduct a comprehensive inspection of the equipment and calculate the drone's battery life; then, import static and dynamic environmental data and resource deployment data of the disaster area in layers; finally, build a three-layer communication network and perform stability tests.

[0168] Solution Development Module: Sets equipment data acquisition frequency parameters and data priority parameters, initiates multi-device collaborative mode to complete multi-source data acquisition; performs filtering, fusion and compression processing on the acquired data, marks data priority and transmits it to the ground command center system; transmits data in layers, performs data integrity verification synchronously, plans paths based on data priority, and generates the optimal rescue plan;

[0169] Rescue and Control Module: Executes the optimal rescue plan. After the drone arrives at the rappelling location, it performs the rescue operation for the wounded according to the rappelling plan; transports the wounded to the designated destination, and then returns to the supply point to complete the wounded transportation mission.

[0170] This invention integrates full-state perception and high-reliability communication to construct a full-link support system, solving the problems of insufficient targeted rescue and easy interruption in traditional rescue, such as information gaps and easy communication interruptions. This technology realizes continuous monitoring of the vital signs and cabin environment of the injured throughout the entire process from rescue to transfer, providing key data support for medical rescue and ensuring uninterrupted rescue. This integrated technology lays the foundation for intelligent and precise disaster emergency rescue, has good results, and has good application prospects.

[0171] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented in software, the above embodiments can be implemented, in whole or in part, as a computer program product. 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, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution.

[0172] 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; 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 this embodiment, depending on actual needs.

[0173] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application.

Claims

1. A highly efficient rescue method for using an unmanned aerial vehicle (UAV) equipped with an emergency rescue cabin, characterized in that, Includes the following steps: First, a comprehensive inspection of the equipment was conducted, and the drone's battery life was calculated. Then, static and dynamic environmental data and resource deployment data of the disaster area were imported in layers. Finally, a three-layer communication network was built and stability tests were performed. Set the device data acquisition frequency parameters and data priority parameters, and start the multi-device collaborative mode to complete multi-source data acquisition; The collected data is filtered, fused, and compressed, and then transmitted to the ground command center system after being prioritized. Data is transmitted in layers, data integrity checks are performed synchronously, paths are planned based on data priority, and the optimal rescue plan is generated. The optimal rescue plan is executed. After the drone arrives at the rappelling location, it carries out the rescue operation for the wounded according to the rappelling plan; the wounded are transported to the designated destination, and then the drone returns to the supply point to complete the wounded transportation mission.

2. The efficient rescue method for an unmanned aerial vehicle (UAV) equipped with an emergency rescue cabin according to claim 1, characterized in that: The steps for conducting a comprehensive inspection of the equipment are as follows: The scope of equipment to be inspected should be clearly defined, including unmanned aerial vehicles (UAVs), intelligent rescue cabins, and communication terminals. Perform the drone equipment inspection process: Inspect the appearance of the propeller to confirm whether there are any defects such as deformation or cracks; Test the propeller rotation function to determine if there is any jamming. Conduct drone endurance testing and simultaneously test the drone's onboard sensors to confirm that both the drone itself and the sensors meet normal operating requirements. The first round of inspections of the intelligent rescue cabin was initiated: the protective performance of the intelligent rescue cabin was tested to confirm that the protective effect met the standards; the electronic instruments and various sensors inside the intelligent rescue cabin were inspected to verify their operational status. Perform a second inspection of the intelligent rescue cabin: check the internal drug reserves and oxygen supply of the instruments to confirm that the supply is sufficient; test the power-on function of the intelligent rescue cabin equipment to determine whether it can be powered on normally. Check the device interface connection status to ensure there are no loose connections. Inspect the communication terminals to check whether the communication equipment of the ground command center system and the handheld communication terminals are working properly.

3. The efficient rescue method for an unmanned aerial vehicle (UAV) equipped with an emergency rescue cabin according to claim 1, characterized in that: Calculating drone endurance involves estimating the drone's flight distance, and the specific calculation method is as follows: In the formula, To estimate the possible flight distance, For standard battery capacity, This refers to the annual battery degradation rate. For battery lifespan, This represents the percentage of remaining battery power. For the no-load power consumption of the drone, For the power consumption of the intelligent rescue cabin under heavy load, This is the actual load capacity. For the drone's flight speed, The remaining percentage after deducting the reserved percentage for safety reasons.

4. The efficient rescue method for an unmanned aerial vehicle (UAV) equipped with an emergency rescue cabin according to claim 1, characterized in that: The steps for importing static and dynamic environmental data and resource deployment data of the disaster area in layers are as follows: A hierarchical data import method was adopted to import static and dynamic data of the disaster area according to the type of disaster. Start the static data import operation for the disaster area: Import data type is high-definition map static data of the disaster area; It is clear that the static data of high-definition maps includes terrain slope and building information, and the data sources are satellite data and survey equipment data. Perform the disaster area dynamic data import operation, and import the data type as dynamic environmental data; Initiate the data fusion and generation operation: combine the imported static and dynamic data to generate visualized environmental data; overlay the drone flight distance parameters onto the visualized environmental data to generate a drone rescue map; Perform the resource deployment data import operation, and specify that the imported data includes the number of drones, the number of backup drones, the safe areas in the disaster area, temporary charging and maintenance points, and personnel responsibilities.

5. The efficient rescue method for an unmanned aerial vehicle (UAV) equipped with an emergency rescue cabin according to claim 4, characterized in that: Multi-device collaborative data collection from multiple sources includes: lidar deployed on drones to collect 3D data of disaster area terrain and obstacles, generating 3D point cloud models; optical cameras deployed on drones to collect disaster area terrain image data; infrared thermal imagers deployed on drones to collect data on the location of the injured and preliminary heat source identification; vital sign sensors deployed inside the intelligent rescue cabin to collect vital sign data of the injured; environmental sensors deployed on drones and environmental sensors deployed on handheld terminals of ground rescue personnel to collect environmental data; drone status sensors deployed on drones to collect drone's own status data; and resource monitoring sensors deployed on the intelligent rescue cabin to collect data on the remaining resources of the intelligent rescue cabin.

6. The efficient rescue method for an unmanned aerial vehicle (UAV) equipped with an emergency rescue cabin according to claim 5, characterized in that: The steps for collaborative data acquisition from multiple sources using multiple devices are as follows: Set the data collection frequency of the drones and the equipment on the intelligent rescue cabin, and clarify the transmission cycle and priority of different types of data; Set the data acquisition model and accuracy parameters for the drone and the equipment on the intelligent rescue cabin; The multi-device collaborative acquisition mode is activated, controlling optical equipment, infrared equipment, and radar equipment to operate synchronously and collect disaster area terrain data and injured status data; The lidar scanning function is activated to scan the terrain of the disaster area and generate a 3D point cloud model; the optical camera is controlled to perform image capture operations simultaneously, and the captured images are superimposed with the 3D point cloud model to achieve data visualization; The edge computing unit on the drone is used to filter the image data collected by the optical camera, removing invalid and abnormal data with a resolution of less than 80%. Perform fusion processing on collected data of the same type and source; for large data volumes greater than 10Mbps, use efficient video coding technology for compression; The pre-processed data is prioritized and then transmitted to the ground command center system.

7. The efficient rescue method for an unmanned aerial vehicle (UAV) equipped with an emergency rescue cabin according to claim 1, characterized in that: The steps to generate the optimal rescue plan are as follows: Perform layered data backhaul: the core layer backhauls high-volume, high-priority data, the collaboration layer backhauls shared data from multiple drones, and the terminal layer backhauls simplified task data. The ground command center system receives layered data transmission and initiates the data integrity verification process. Define the path planning constraints, covering the drone's flight distance and casualty priority; Sorting operations are performed according to the priority of the wounded from high to low; the real-time location of the drone, the location of the wounded, and the location of the transport destination are obtained, and an initial rescue path is generated; The real-time obstacle monitoring mechanism is activated. When a sudden obstacle is detected, the obstacle location data is added and a new rescue path is generated. The newly generated rescue path is then simulated and verified to confirm its feasibility. When there are multiple drones and multiple casualties, calculate the straight-line distance between each drone and each casualty, the drone's operating energy consumption, and the operating time. A multi-objective optimization model was constructed to calculate the evaluation value of each drone-wound matching scheme. The matching scheme with the highest total evaluation value was selected as the optimal rescue scheme.

8. The efficient rescue method for an unmanned aerial vehicle (UAV) equipped with an emergency rescue cabin according to claim 1, characterized in that: The steps for rescuing the wounded according to the rappelling plan are as follows: Receive the coordinates of the injured, verify whether there are any sudden obstacles in the target area, and determine whether it is suitable for rappelling. The UAV system receives a hovering command, flies to the target area, keeps the lidar continuously running, and scans the surrounding environment in real time. The UAV generates a 3D environment model of the target area based on real-time scanning data from LiDAR; it then marks the location information of obstacles in the model, triggering flight attitude adjustments. Adjust the drone so that it is directly below the cabin and coincides with the coordinates of the injured person. Retrieve the superimposed image of the drone's hovering position and the target point to verify and confirm that there is no positional deviation. After the drone sends a signal to the ground command system indicating that it is ready to rappel, it performs the rope release operation according to the preset rappelling plan. Collect real-time wind field data and UAV acceleration data, calculate rope tension and wind field component based on the collected data, and start the rope descent speed adjustment program when the data exceeds the preset threshold. Real-time monitoring and judgment of whether the swing amplitude of the intelligent rescue cabin exceeds the preset angle threshold; If the threshold is exceeded, a rappelling pause command is sent, and the rappelling process is restarted after the wind is detected to have weakened.

9. A highly efficient rescue method for an unmanned aerial vehicle (UAV) equipped with an emergency rescue cabin according to claim 8, characterized in that: The procedure after rappelling to the ground is as follows: After the intelligent rescue cabin touches the ground, the drone system controls the cabin door to open, and ground personnel assist the injured person to enter the intelligent rescue cabin. The intelligent rescue cabin uses the matched resources to provide initial first aid to the injured person. At the same time, the vital signs sensors inside the intelligent rescue cabin are activated to collect the injured person's data and transmit it back to the ground command center system in real time for the command personnel to monitor the status. The intelligent rescue cabin is then shut down, and the intelligent rescue cabin sends a shutdown signal to the drone system. The drone then initiates the rope retrieval function. After the rope is retrieved, the injured person is transported according to the planned path in the optimal rescue plan, and the drone transports the injured person to the final delivery location.

10. A highly efficient rescue system for drones equipped with an emergency rescue cabin, characterized in that, include: Network construction module: First, conduct a comprehensive test of the equipment and calculate the drone's battery life; Import the static and dynamic environmental data and resource deployment data of the disaster area in layers; Finally, a three-layer communication network was built and stability tests were performed. Solution formulation module: Set the device data acquisition frequency parameters and data priority parameters, and start the multi-device collaborative mode to complete multi-source data acquisition; The collected data is filtered, fused, and compressed, and then transmitted to the ground command center system after being prioritized. Data is transmitted in layers, data integrity checks are performed synchronously, paths are planned based on data priority, and the optimal rescue plan is generated. Rescue and Control Module: Executes the optimal rescue plan. After the drone arrives at the rappelling location, it performs the rescue operation for the wounded according to the rappelling plan; transports the wounded to the designated destination, and then returns to the supply point to complete the wounded transportation mission.

Images