Passive positioning system and method for trapped persons based on airship digital marks

Abstract

The invention discloses a trapped person passive positioning system and method based on airship digital marker.The system comprises an air reference end, a search verification end and a ground command end, the air reference end comprises an airship, a direction sensor and an airship terminal are arranged in the airship, and the direction sensor is connected with the airship terminal; annular areas with different colors are arranged on the outer wall of the airship, the ground command end comprises a ground resolving terminal and a communication receiving terminal, the airship terminal is in signal connection with the ground resolving terminal, the search verification end comprises an unmanned aerial vehicle, and a multispectral imaging device, an unmanned aerial vehicle control terminal and a remote data transmission device are carried on the unmanned aerial vehicle. The multi-spectral imaging device is connected with the unmanned aerial vehicle control terminal, the unmanned aerial vehicle control terminal is in signal connection with the ground resolving terminal through the remote data transmission device, and in the three-break (open circuit, power failure and network disconnection) disaster environment, through the airship digital marking and computer vision fusion positioning technology, trapped persons can be rapidly and accurately positioned.

Description

Technical Field

[0001] The present invention belongs to the field of intelligent emergency rescue technology, and specifically relates to a passive positioning system and method for trapped persons based on tethered airship digital tags, integrating tethered airship digital tags, drone swarm collaboration, and the YOLOv8 deep learning model. The system is particularly suitable for extreme environments with "three disconnections" (circuit, power, and network outages) caused by disasters such as earthquakes and floods. Background Art

[0002] With the development of human society and the intensification of climate change, natural disasters (such as earthquakes, floods, and landslides) are becoming more frequent and more destructive, placing higher demands on emergency rescue technologies. Existing passive positioning technologies have serious limitations, especially in extreme disaster environments such as "three disconnections" (power outages, network outages, and power outages):

[0003] 1. Satellite remote sensing technology relies on periodic transits, has long revisit cycles, and is susceptible to cloud cover. Its positioning accuracy is only 5-10 meters, making it unable to meet the timeliness requirements of the "golden 72 hours" rescue effort. For example, during the 2023 Türkiye earthquake, cloud cover prevented satellite imagery from assessing 30% of the affected area in a timely manner.

[0004] 2. The full-area scanning of drones requires relying on preset routes or pilot operation. The flight time of a single drone is insufficient. It takes 45-90 minutes to search 1 square kilometer, and the recognition rate of trapped people actively calling for help is low.

[0005] 3. Existing passive positioning technologies such as GNSS, UWB, and cellular positioning technologies all require the trapped person to carry terminal equipment or pre-deploy base station networks.

[0006] 4. Shortcomings in multi-platform collaboration: The existing data of airships and drone systems are isolated. For example, airships only serve as communication relays and fail to take advantage of their long flight time and strong visibility. Drone clusters lack efficient energy supply and are unable to operate continuously.

[0007] After searching:

[0008] Chinese invention patent publication number CN109451580B proposes a method for locating people trapped after an earthquake in a three-dimensional scene. This method addresses situations where people are trapped in buildings or other structures after an earthquake and their mobile phones are unable to receive GPS signals for positioning. Mobile communication base stations provide communication signals to the trapped person's mobile phone. Using a built-in mobile phone application and signal strength data, the trapped person's spatial location is determined. The method uses the signal strength collected from the trapped person's mobile phone and converts it into the distance between the mobile communication base station and the phone using a lognormal shadow model. The position of the mobile communication base station is regularly changed to obtain multiple sets of data, and the three-dimensional position of the trapped person within the measured site is calculated using spatial geometry formulas. The advantages of this patented positioning method include its adaptability to sudden events such as earthquakes, the need for the person being located to carry additional equipment, minimal equipment requirements, and simple computational complexity, meeting the needs for rapid and accurate post-earthquake positioning.

[0009] Chinese utility model patent publication number CN207488769U proposes a trapped person information collection aircraft. The aircraft comprises an aircraft housing equipped with a coreless motor module, a motor drive module, and an ultrasonic ranging module. Inside the housing are an STM32 microcontroller module, a sound collection module, a speech synthesis module, a speech recognition module, a satellite positioning module, a communication module, a nine-axis motion sensor module, a power supply module, and a clock module. This utility model combines wireless sensor network and sound processing technologies, taking advantage of the golden rescue vacuum period when rescuers are unable to reach, to automatically collect information and assess the health of trapped individuals. This provides relevant data for future rescuers entering the disaster site, thereby addressing the shortcomings of existing auxiliary rescue systems, maximizing rescue efficiency, reducing the mortality rate of trapped individuals, and preventing further family tragedies.

[0010] A Chinese invention patent, publication number CN108169816A, proposes a post-disaster rescue and trapped person positioning system, comprising a rescue positioning controller, a disaster relief platform main controller connected to at least one rescuer positioning device, the rescue positioning controller connected to at least one life detection device, and the rescue positioning controller connected to at least one display device. The rescue positioning controller is configured to receive positioning signals from the rescuer positioning device and the life detection device, process the signals, convert different types of signals into location information for the rescuer and trapped persons, and transmit the location information for the rescuer and trapped persons to the display device. When a disaster occurs, the system switches to an emergency state and simultaneously activates the personnel positioning and life detection systems, which can display the current location of rescuers entering the scene, as well as the location of trapped persons who have no time or are unable to escape, thereby increasing the chances of a successful rescue.

[0011] The technical comparison between this application and the above patent is as follows:

[0012] Patent publication number CN109451580B proposes a method for locating trapped individuals after an earthquake in a three-dimensional scenario. This method utilizes mobile communication base stations moving along a right-angle path around the edge of the disaster area. By collecting signal strength from the trapped individuals' phones and converting it into distance data, the method, combined with a lognormal shadow model and spatial geometry algorithms, accurately locates the trapped individuals in three dimensions. This method does not rely on GPS signals and requires only a small amount of equipment for rapid positioning. It is suitable for sudden disasters such as earthquakes, boasts simple computation and rapid response, significantly improving rescue efficiency.

[0013] Utility model patent publication number CN207488769U discloses a trapped person information collection aircraft. This aircraft integrates a coreless motor module, an ultrasonic ranging module, an STM32 microcontroller control system, and multimodal sensors. Through sound acquisition and voice interaction, it rapidly locates and retrieves information about trapped individuals after a disaster. Its innovation lies in its integration of nine-axis motion sensor precision navigation, ZigBee wireless networking, and satellite positioning technology. It can autonomously perform three-dimensional spatial searches in complex environments inaccessible to rescuers, transmitting real-time data on the trapped individual's location, physiological status, and environment. Compared to traditional life detection equipment, this design offers a breakthrough in contactless human-computer interaction. By integrating the LD3320 voice recognition chip and the XFS5152 voice synthesis chip, it significantly improves the proactive response rate for trapped individuals in confined spaces, providing an intelligent solution for the golden 72-hour rescue effort.

[0014] Patent publication number CN108169816A proposes a post-disaster collaborative positioning system for rescuers and trapped individuals. This system integrates multi-source positioning data through a rescue positioning controller, enabling two-way visualization of personnel at the rescue site. Its innovation lies in combining rescuer positioning technology using RFID / 2.4G wireless inter-screen communication with infrared / radar life detection technology. This system utilizes an RS-485 / TCP-IP / ZigBee multimode communication architecture to build a real-time dynamic positioning network. The system features a three-level fault tolerance mechanism: a power management module ensures continuous power supply, a fault detection module enables device self-diagnosis, and a main control module employs redundant data processing algorithms to ensure stable output of rescuer movement trajectories and trapped individuals' vital sign coordinates, even in complex environments such as building collapse. Compared to traditional single-positioning solutions, this system achieves a breakthrough in spatial and temporal correlation between rescue forces and targets. Through battlefield situation visualization on the display device, this system improves rescue efficiency by over 40%, making it particularly suitable for time-critical disaster scenarios such as earthquakes and fires.

[0015] This invention addresses the limitations of traditional technologies for locating trapped individuals in disaster situations, which rely on devices such as GNSS terminals and UWB tags. Furthermore, these technologies struggle to operate effectively under extreme conditions such as "three disconnections" (circuit, power, and network outages). By doing so, it proposes a rapid and precise positioning solution that eliminates the need for trapped individuals to carry any electronic devices. This solution innovatively utilizes a tethered helium airship as a visual positioning reference, combined with a multi-level response mechanism between drones and ground stations, to accurately locate trapped individuals in any complex environment. In the event of a low-voltage fault or adverse weather conditions, the system dynamically adjusts the airship's altitude and brightness mode to maintain stable operation of the tagging system, significantly improving the reliability and efficiency of positioning during disaster relief efforts.

[0016] The invention patent with publication number CN109451580B focuses on three-dimensional scene positioning by measuring distance through signal strength, the utility model patent with publication number CN108169816A focuses on the contactless human-computer dialogue function of sound collection and voice interaction technology, and the invention patent with publication number CN108169816A focuses on building a real-time dynamic positioning network through a multi-mode communication architecture to achieve two-way personnel visualization. The present invention uses a tethered airship equipped with a large LED digital marker as an aerial reference, combined with a drone cluster and a YOLOv8 model to achieve passive positioning, which is suitable for extreme environments with "three cuts". The trapped person only needs to visually report the airship code, without the need for electronic equipment, filling the gap in existing technologies that rely on electronic equipment or infrastructure. At the same time, it achieves fast and accurate passive positioning through a collaborative architecture and efficient algorithms, which is an important innovation in the field of emergency rescue. Summary of the Invention

[0017] To address the above technical issues, the present invention proposes a passive positioning system and method for trapped persons based on airship digital markers. This application can achieve rapid and accurate positioning of trapped persons in a disaster environment characterized by "three cuts" (circuit, power, and network outages) by integrating airship digital markers with computer vision positioning technology.

[0018] To achieve the above object, the technical solution adopted by the present invention is:

[0019] The passive positioning system for trapped persons based on airship digital markings is characterized by comprising an aerial reference end, a search and verification end and a ground command end, wherein the aerial reference end comprises an airship, a direction sensor and an airship terminal are arranged in the airship, the direction sensor is connected to the airship terminal and transmits the real-time position and posture data of the airship to the airship terminal, the outer wall of the airship is provided with annular areas of different colors, each of the annular areas is provided with three layers, the outer layer represents the partition, the middle layer represents the height, and the inner layer is a QR code check bit, the ground command end comprises a ground solution terminal and a communication receiving terminal for receiving calls from trapped persons, the airship terminal is connected to the ground solution terminal signal, and uploads the position and posture data of the airship to the ground solution terminal in real time, the search and verification end comprises a drone, and the drone is equipped with a multi-spectral component. The multispectral imaging device is connected to the drone control terminal, and the drone control terminal is connected to the ground solution terminal signal through the remote data transmission device. The drone control terminal has built-in water search and rescue models and land search and rescue models. When the trapped person reports the airship information seen to the communication receiving terminal, the ground solution terminal calculates the initial positioning area of ​​the trapped person based on the reported information and the position data of the spacecraft and plans the drone flight route, and then uploads the drone flight route to the drone control terminal. When the initial positioning area is located in the water area, the drone control terminal activates the water search and rescue model and heads there. When the initial positioning area is located in the land area, the drone control terminal activates the land search and rescue model and heads there.

[0020] As a preferred technical solution of the present invention: the aerial reference end also includes a winch device, the winch device includes a mooring cable and a fixing seat, the fixing seat is fixed on the ground, the mooring cable is stored on the fixing seat, and the airship is pulled up to a predetermined height by the mooring cable.

[0021] As a preferred technical solution of the present invention: there are eight annular areas, each of which is provided with three layers of digital rings, the outer layer is letters AH, the middle layer is numbers 1-8, and the inner layer is a QR code check digit. The digital ring is a large LED digital code of 2m×2m.

[0022] As a preferred technical solution of the present invention: the ground solution terminal is provided with a position solution engine. When the trapped personnel report the airship data they have seen, the airship terminal uploads the real-time yaw angle θ of the airship to the position solution engine. The position solution engine calculates the azimuth annular area using a geometric projection algorithm: α = θ + (S_id × 45°) ± 22.5°.

[0023] As a preferred technical solution of the present invention: the water search and rescue model has a built-in feature library of people who fall into the water. When the drone arrives at the target water area, it uses a multispectral imaging device to perform identification and search based on the data in the feature library of people who fall into the water. The land search and rescue model has a built-in feature library of human activities. When the drone arrives at the target land area, it uses a multispectral imaging device to perform identification and search based on the data in the feature library of human activities.

[0024] As a preferred technical solution of the present invention: when the UAV searches on the flight route, when it goes to search in the water area, it adopts the water mode, that is, spiral expansion search; when it goes to search in the landing area, it adopts the land mode, that is, flying along the street topology path; when there is land in the water area or water in the land area, it adopts the mixed mode, that is, preferentially scanning the abnormal reflective area on the water surface.

[0025] As a preferred technical solution of the present invention: the multispectral imaging device includes a visible infrared detector, an infrared thermal imager and a laser radar rangefinder for identifying human body features and distress movements.

[0026] As a preferred technical solution of the present invention: the remote data transmission device carried by the drone is a local area network, a hotspot network or a LoRa gateway.

[0027] In the above structure: the passive positioning system for trapped persons based on airship digital marking proposed by the present invention includes an aerial reference terminal, a search and verification terminal and a ground command terminal. The aerial reference terminal is used for patrol and as a positioning reference for trapped persons. The search and verification terminal is used for accurate search and location confirmation of trapped persons. The ground command terminal is used for information processing and rescue plan formulation.

[0028] The aerial reference end includes an airship, which is equipped with a direction sensor and an airship terminal. The direction sensor is connected to the airship terminal and transmits the real-time position data of the airship to the airship terminal. The outer wall of the airship is provided with annular areas of different colors. Each of the annular areas is provided with three layers, namely, the outer layer represents the partition, the middle layer represents the height, and the inner layer is the QR code check bit.

[0029] The airship is a tethered helium-filled blimp, lifted to a predetermined altitude of 300 meters by a ground-mounted winch. The mooring cable also serves as a power transmission and communications relay. Once airborne, the airship activates a self-stabilizing system to maintain stability in the air. A MEMS inertial measurement unit (IMU) uses a directional sensor to monitor the airship's attitude in real time, maintaining hovering stability in wind speeds ≤ Level 8. The ducted thrusters automatically adjust thrust according to wind conditions to combat sudden gusts.

[0030] The airship unfolds into eight circular sections, each displaying a 2m x 2m LED code (e.g., "A1" - "D2"), with brightness automatically adjusted based on ambient light:

[0031] Day mode: 1000cd / m² (visible in strong light environment)

[0032] Night mode: 500cd / m² (with orange-red backlight turned on)

[0033] After the airship is launched, the direction sensor is used to complete the calibration as follows:

[0034] Fiber optic gyroscope initialization

[0035] Synchronize with BeiDou differential positioning data from the ground station to eliminate accumulated errors.

[0036] When using:

[0037] When the trapped people see the airship in the sky, the airship guidance communication is established:

[0038] The airship displays the rescue phone number through LED coding (such as flashing "Dial 189-XXXX-099")

[0039] Synchronous broadcast voice prompt (5G broadcast channel, content: "Please dial the number displayed on the screen")

[0040] The ground command terminal includes a ground solution terminal and a communication receiving terminal for receiving calls from trapped personnel. The airship terminal is connected to the ground solution terminal signal and uploads the airship's position data to the ground solution terminal in real time.

[0041] Simple interactions for trapped persons are as follows:

[0042] After the trapped person dials the rescue number on the airship, the ground control terminal communicates with him through the communication receiving terminal. The content of the call is as follows:

[0043] Rescuer: "Please describe the airship code you saw, such as the brightest letter and number combination"

[0044] Trapped person: "I see B and 3 glowing right in front of me, B is brighter than 3" → Determined to be area B3

[0045] Position determination logic:

[0046] If the trapped person says "B is on the left" → starboard side of the airship (needs to be calculated based on the airship's real-time yaw angle θ)

[0047] If the number is above the light source, the elevation angle is φ≈20°-30° (experienced value)

[0048] Rescuers use the ground solution terminal to combine the trapped person's description with the position and posture transmitted in real time by the airship, and convert the airship code and observation angle reported by the trapped person into specific coordinates through the geometric projection model. They also generate an EPSO optimized path, which is transmitted to the drone control terminal to control the drone to go to the specific coordinates for precise search.

[0049] The search verification end includes a drone, which is equipped with a multispectral imaging device, a drone control terminal and a remote data transmission device. The multispectral imaging device is connected to the drone control terminal, and the drone control terminal is connected to the ground solution terminal signal through the remote data transmission device.

[0050] The UAV conducts detection according to the flight route and feeds back information to the ground solution terminal in real time through the remote data transmission device.

[0051] The ground control terminal generates the UAV search path:

[0052] The drone is loaded with a land search and rescue model, adopts a spiral expansion search mode, and avoids known obstacles (collapsed buildings are marked as no-fly zones).

[0053] The drone takes off with its payload:

[0054] Flight speed 8m / s, search altitude 30 meters (taking into account both field of view and resolution)

[0055] The 1080p video stream is transmitted back to the ground control terminal in real time through the remote data transmission device.

[0056] When the drone reaches the designated area, it begins precise search and positioning through the multispectral imaging device:

[0057] Visible infrared detector:

[0058] Based on improved YOLOv8 model

[0059] Identify human features (head, limb outlines) and distress gestures (waving, flashing lights)

[0060] Infrared thermal imager:

[0061] Thermal imaging detects body temperature and eliminates false alarms from debris

[0062] Dynamic threshold segmentation technology eliminates sunlight reflection interference

[0063] LiDAR rangefinder:

[0064] After the trapped person is detected, the spatial position is confirmed by the lidar rangefinder

[0065] Geographic coordinate system conversion (WGS84 to UTM grid)

[0066] And push the geographic coordinate position to the ground solution terminal:

[0067] Three-dimensional coordinates: X=125.7m, Y=88.3m, Z=1.2m (relative to the ground station)

[0068] Confidence level: 98.4% (visible light + infrared dual verification).

[0069] The present invention can also be used for nighttime flood disaster location search, specifically as follows:

[0070] Special environment adaptation measures:

[0071] Airship Adjustment

[0072] LED backlight switches to night mode (strong ability to penetrate rain and fog)

[0073] The airship descended to 200 meters (to improve the visibility of the marker and compensate for night visibility)

[0074] Drone sensor configuration:

[0075] Infrared thermal imager set to high sensitivity mode

[0076] Load the water search and rescue model:

[0077] Feature database of people falling into the water (head exposure, floating object correlation detection)

[0078] Positioning process optimization:

[0079] After the trapped person calls the rescue hotline, the rescue team receives the signal through the LoRa gateway (a frequency band with strong water penetration ability).

[0080] Multi-source data fusion positioning:

[0081] Airship position calculation and UAV thermal imaging detection collaboration:

[0082] Heat source cluster analysis (excluding areas with abnormal water temperature)

[0083] Motion trajectory tracking (continuous frame displacement vector verification)

[0084] The final positioning error is less than 1.5 meters (slightly higher than the land scene due to the influence of water surface fluctuations).

[0085] UAV adaptive scanning strategy

[0086] Water Mode: Spiral Expansion Search

[0087] City Mode: Flying along street topology paths

[0088] Hybrid mode: Prioritize scanning of abnormal reflective areas on the water surface

[0089] Multispectral verification

[0090] Thermal imaging detection

[0091] LiDAR water surface fluctuation compensation

[0092] Recognition of drowning people using visible light + YOLOv8.

[0093] The method for passively locating trapped persons based on airship digital markings is characterized by comprising the following steps:

[0094] S1. The trapped person observes the airship code and elevation angle φ, and calls the rescue phone through the communication equipment to report to the ground command end; S2. The ground command end calculates the azimuth sector α=θ+(S_id×45°)±22.5° based on the reported information and the real-time yaw angle θ of the airship, determines the area where the trapped person is located, generates the EPSO optimized path based on the area where the trapped person is located, and transmits the optimization to the UAV control terminal; S3. The UAV control terminal controls the UAV to approach the area where the trapped person is located according to the EPSO optimized path and conducts a detailed search through the multi-spectral imaging device to determine the trapped state and precise trapped location of the trapped person, and transmits the video and the trapped location of the trapped person to the ground command end in real time through the remote data transmission device; S4. After receiving the video and trapped location transmitted by the UAV, the ground command end generates a rescue plan according to the situation.

[0095] Compared with the prior art, the present invention has the following beneficial effects:

[0096] 1. Passive positioning without any equipment dependency: Through the large visual digital marker (2m×2m LED code) on the airship, trapped people can report their location independently, completely eliminating the dependence on GNSS terminals, UWB tags and other equipment, and can still work stably in extreme environments with "three disconnections".

[0097] 2. Rapid response capability: Adopting a two-level model of "airship coarse positioning + drone precise search", the positioning process, which traditional technology requires 45-90 minutes per square kilometer, is shortened to within 10 minutes.

[0098] 3. High-precision positioning: By integrating a geometric projection algorithm (error <25°) with multispectral computer vision verification (visible light + infrared + lidar), the final positioning accuracy reaches 0.5 meters.

[0099] 4. Excellent environmental adaptability: The airship marking system supports an operating temperature range of -30°C to 60°C, and the LED backlight ensures a visual distance of 500 meters at night / in rainy and foggy environments; the drone is equipped with IP67 protection and can operate normally in level 7 winds and moderate rain conditions.

[0100] 5. Long-term continuous operation capability: The airship achieves energy self-sufficiency through flexible photovoltaics (conversion efficiency 28%) and ducted fans (300W), and cooperates with drone wireless charging (efficiency >85%) to support the system's uninterrupted operation for 30 days.

[0101] 6. Intelligent Collaborative Architecture: Establish an integrated air-space-ground network of "airship-UAV-ground station", achieve multi-device data synchronization through the TDMA communication protocol, and achieve a positioning success rate of 98%.

[0102] 7. Versatility in multiple disaster scenarios: This technology has been verified in earthquake ruins, floods, mountainous areas, and other scenarios. It is particularly suitable for complex environments such as collapsed buildings and flooded water bodies. The technical adaptability is more than three times that of a single sensor solution. BRIEF DESCRIPTION OF THE DRAWINGS

[0103] Figure 1 This is a flow chart of a method for passively locating trapped persons based on digital markings on airships;

[0104] Figure 2 It is a flow chart of the airship hull color and number marking system;

[0105] Figure 3 It is the positioning process in earthquake debris environment;

[0106] Figure 4 It is an identification mark map in the earthquake ruins environment;

[0107] Figure 5 This is a flowchart of the application of nighttime flood disaster positioning. DETAILED DESCRIPTION

[0108] The present invention is further described in detail below with reference to the accompanying drawings and specific embodiments:

[0109] like Figure 1-2As shown, the passive positioning system for trapped persons based on airship digital marking proposed by the present invention is characterized in that it includes an aerial reference end, a search and verification end, and a ground command end. The aerial reference end includes an airship, and a direction sensor and an airship terminal are provided in the airship. The direction sensor is connected to the airship terminal and transmits the real-time posture data of the airship to the airship terminal. The outer wall of the airship is provided with annular areas of different colors, and each of the annular areas is provided with three layers, namely, the outer layer represents the partition, the middle layer represents the height, and the inner layer is a QR code check bit. The ground command end includes a ground solution terminal and a communication receiving terminal for receiving calls from trapped persons. The airship terminal is connected to the ground solution terminal signal and uploads the posture data of the airship to the ground solution terminal in real time. The search and verification end includes a drone, which is equipped with The invention comprises a multispectral imaging device, a drone control terminal, and a remote data transmission device. The multispectral imaging device is connected to the drone control terminal, which is connected to the ground solution terminal signal via the remote data transmission device. The drone control terminal has built-in water search and rescue models and land search and rescue models. When the trapped person reports the airship information he sees to the communication receiving terminal, the ground solution terminal calculates the trapped person's initial positioning area based on the reported information combined with the airship's position data and plans the drone's flight route. The drone's flight route is then uploaded to the drone control terminal. When the initial positioning area is within the water area, the drone control terminal activates the water search and rescue model and proceeds. When the initial positioning area is within the land area, the drone control terminal activates the land search and rescue model and proceeds. The aerial reference end also includes a winch device, which includes a mooring cable and a fixing seat. The fixing seat is fixed to the ground, and the mooring cable is stored in the fixing seat. The airship is pulled to a predetermined height by the mooring cable. There are eight annular zones, each with three layers of digital rings: the outer layer is the letters AH, the middle layer is the numbers 1-8, and the inner layer is a QR code check digit. The digital rings are large 2m×2m LED digital codes. The ground-based solution terminal is equipped with a position solution engine. When the trapped person reports the airship data they have seen, the airship terminal uploads the airship's real-time yaw angle θ to the position solution engine. The position solution engine calculates the azimuth annular zone using a geometric projection algorithm: α = θ + (S_id × 45°) ± 22.5°. The water search and rescue model has a built-in human feature library. When the drone arrives at the target water area, it uses the data in the human feature library to perform identification and search using a multispectral imaging device. The land search and rescue model has a built-in human feature activity library. When the drone arrives at the target land area, it uses the data in the human feature activity library to perform identification and search using a multispectral imaging device.When searching along its flight path, the drone uses a water mode (i.e., a spiral expansion search) when searching in water areas. When searching in landing areas, it uses a land mode (i.e., flying along a street topology path). When searching in water areas with land or land areas with water, it uses a hybrid mode (i.e., prioritizing areas with abnormally reflective water surfaces). The multispectral imaging device includes a visible infrared detector, an infrared thermal imager, and a lidar rangefinder for identifying human features and distress signals. The drone's onboard remote data transmission device is a local area network, hotspot network, or LoRa gateway.

[0110] The passive positioning method for trapped persons based on airship digital marking proposed in the present invention comprises the following steps:

[0111] S1. The trapped person observes the airship code and elevation angle φ and reports the situation to the ground control center by dialing a rescue line via communication equipment.

[0112] S2. The ground control terminal uses the reported information and the airship's real-time yaw angle θ to calculate the azimuth sector α = θ + (S_id × 45°) ± 22.5°. This determines the area where the trapped person is located, generates an optimized EPSO path based on the area, and transmits the optimized path to the UAV control terminal.

[0113] S3. The UAV control terminal controls the UAV along the EPSO-optimized path to approach the area where the trapped individuals are located and conducts a detailed search using a multispectral imaging device to determine the trapped individuals' condition and precise location. The UAV then transmits the video and the trapped individuals' location to the ground control terminal in real time via a remote data transmission device.

[0114] S4. After receiving the video and trapped location transmitted by the drone, the ground control terminal generates a rescue plan based on the situation.

[0115] Example 1:

[0116] like Figure 3-4 As shown, the passive positioning system for trapped persons based on airship digital marking proposed by the present invention includes an aerial reference terminal, a search and verification terminal, and a ground command terminal. The aerial reference terminal is used for patrol and as a positioning reference for trapped persons, the search and verification terminal is used for accurate search and location confirmation of trapped persons, and the ground command terminal is used for information processing and rescue plan formulation.

[0117] The aerial reference end includes an airship, which is equipped with a direction sensor and an airship terminal. The direction sensor is connected to the airship terminal and transmits the real-time position data of the airship to the airship terminal. The outer wall of the airship is provided with annular areas of different colors. Each of the annular areas is provided with three layers, namely, the outer layer represents the partition, the middle layer represents the height, and the inner layer is the QR code check bit.

[0118] The airship is a tethered helium-filled blimp, lifted to a predetermined altitude of 300 meters by a ground-mounted winch. The mooring cable also serves as a power transmission and communications relay. Once airborne, the airship activates a self-stabilizing system to maintain stability in the air. A MEMS inertial measurement unit (IMU) uses a directional sensor to monitor the airship's attitude in real time, maintaining hovering stability in wind speeds ≤ Level 8. The ducted thrusters automatically adjust thrust according to wind conditions to combat sudden gusts.

[0119] The airship unfolds into eight circular sections, each displaying a 2m x 2m LED code (e.g., "A1" - "D2"), with brightness automatically adjusted based on ambient light:

[0120] Day mode: 1000cd / m² (visible in strong light environment)

[0121] Night mode: 500cd / m² (with orange-red backlight turned on)

[0122] After the airship is launched, the direction sensor is used to complete the calibration as follows:

[0123] Fiber optic gyroscope initialization

[0124] Synchronize with BeiDou differential positioning data from the ground station to eliminate accumulated errors.

[0125] When using:

[0126] When the trapped people see the airship in the sky, the airship guidance communication is established:

[0127] The airship displays the rescue phone number through LED coding (such as flashing "Dial 189-XXXX-099")

[0128] Synchronous broadcast voice prompt (5G broadcast channel, content: "Please dial the number displayed on the screen")

[0129] The ground command terminal includes a ground solution terminal and a communication receiving terminal for receiving calls from trapped personnel. The airship terminal is connected to the ground solution terminal signal and uploads the airship's position data to the ground solution terminal in real time.

[0130] Simple interactions for trapped persons are as follows:

[0131] After the trapped person dials the rescue number on the airship, the ground control terminal communicates with him through the communication receiving terminal. The content of the call is as follows:

[0132] Rescuer: "Please describe the airship code you saw, such as the brightest letter and number combination"

[0133] Trapped person: "I see B and 3 glowing right in front of me, B is brighter than 3" → Determined to be area B3

[0134] Position determination logic:

[0135] If the trapped person says "B is on the left" → starboard side of the airship (needs to be calculated based on the airship's real-time yaw angle θ)

[0136] If the number is above the light source, the elevation angle is φ≈20°-30° (experienced value)

[0137] Rescuers use the ground solution terminal to combine the trapped person's description with the position and posture transmitted in real time by the airship, and convert the airship code and observation angle reported by the trapped person into specific coordinates through the geometric projection model. They also generate an EPSO optimized path, which is transmitted to the drone control terminal to control the drone to go to the specific coordinates for precise search.

[0138] The search verification end includes a drone, which is equipped with a multispectral imaging device, a drone control terminal and a remote data transmission device. The multispectral imaging device is connected to the drone control terminal, and the drone control terminal is connected to the ground solution terminal signal through the remote data transmission device.

[0139] The UAV conducts detection according to the flight route and feeds back information to the ground solution terminal in real time through the remote data transmission device.

[0140] The ground control terminal generates the UAV search path:

[0141] The drone is loaded with a land search and rescue model, adopts a spiral expansion search mode, and avoids known obstacles (collapsed buildings are marked as no-fly zones).

[0142] The drone takes off with its payload:

[0143] Flight speed 8m / s, search altitude 30 meters (taking into account both field of view and resolution)

[0144] The 1080p video stream is transmitted back to the ground control terminal in real time through the remote data transmission device.

[0145] When the drone reaches the designated area, it begins precise search and positioning through the multispectral imaging device:

[0146] Visible infrared detector:

[0147] Based on improved YOLOv8 model

[0148] Identify human features (head, limb outlines) and distress gestures (waving, flashing lights)

[0149] Infrared thermal imager:

[0150] Thermal imaging detects body temperature and eliminates false alarms from debris

[0151] Dynamic threshold segmentation technology eliminates sunlight reflection interference

[0152] LiDAR rangefinder:

[0153] After the trapped person is detected, the spatial position is confirmed by the lidar rangefinder

[0154] Geographic coordinate system conversion (WGS84 to UTM grid)

[0155] And push the geographic coordinate position to the ground solution terminal:

[0156] Three-dimensional coordinates: X=125.7m, Y=88.3m, Z=1.2m (relative to the ground station)

[0157] Confidence level: 98.4% (visible light + infrared dual verification).

[0158] Example 2:

[0159] like Figure 5 As shown, the present invention can also be used for nighttime flood disaster location search, specifically as follows:

[0160] Special environment adaptation measures:

[0161] Airship Adjustment

[0162] LED backlight switches to night mode (strong ability to penetrate rain and fog)

[0163] The airship descended to 200 meters (to improve the visibility of the marker and compensate for night visibility)

[0164] Drone sensor configuration:

[0165] Infrared thermal imager set to high sensitivity mode

[0166] Load the water search and rescue model:

[0167] Feature database of people falling into the water (head exposure, floating object correlation detection)

[0168] Positioning process optimization:

[0169] After the trapped person calls the rescue hotline, the rescue team receives the signal through the LoRa gateway (a frequency band with strong water penetration ability).

[0170] Multi-source data fusion positioning:

[0171] Airship position calculation and UAV thermal imaging detection collaboration:

[0172] Heat source cluster analysis (excluding areas with abnormal water temperature)

[0173] Motion trajectory tracking (continuous frame displacement vector verification)

[0174] The final positioning error is less than 1.5 meters (slightly higher than the land scene due to the influence of water surface fluctuations).

[0175] UAV adaptive scanning strategy

[0176] Water Mode: Spiral Expansion Search

[0177] City Mode: Flying along street topology paths

[0178] Hybrid mode: Prioritize scanning of abnormal reflective areas on the water surface

[0179] Multispectral verification

[0180] Thermal imaging detection

[0181] LiDAR water surface fluctuation compensation

[0182] Recognition of drowning people using visible light + YOLOv8.

[0183] The above description is merely a preferred embodiment of the present invention and does not constitute any other form of limitation to the present invention. Any modification or equivalent variation based on the technical essence of the present invention shall still fall within the scope of protection claimed by the present invention.

Claims

1. A passive positioning system for trapped persons based on digital airship markers, characterized by: It includes an aerial reference end, a search and verification end and a ground command end. The aerial reference end includes an airship, which is equipped with a direction sensor and an airship terminal. The direction sensor is connected to the airship terminal and transmits the real-time posture data of the airship to the airship terminal. The outer wall of the airship is provided with annular areas of different colors. Each of the annular areas is provided with three layers, the outer layer represents the partition, the middle layer represents the height, and the inner layer is a QR code check bit. The ground command end includes a ground solution terminal and a communication receiving terminal for receiving calls from trapped personnel. The airship terminal is connected to the ground solution terminal signal and uploads the airship's posture data to the ground solution terminal in real time. The search and verification end includes a drone, which is equipped with a multispectral imaging device, a drone control terminal and A remote data transmission device, the multispectral imaging device is connected to the drone control terminal, and the drone control terminal is connected to the ground solution terminal signal through the remote data transmission device. The drone control terminal has built-in water search and rescue models and land search and rescue models. When the trapped person reports the airship information he sees to the communication receiving terminal, the ground solution terminal calculates the initial positioning area of ​​the trapped person based on the reported information and the position data of the spacecraft and plans the drone flight route, and then uploads the drone flight route to the drone control terminal. When the initial positioning area is located in the water area, the drone control terminal activates the water search and rescue model and heads there. When the initial positioning area is located in the land area, the drone control terminal activates the land search and rescue model and heads there.

2. The passive positioning system for trapped persons based on airship digital marking according to claim 1, characterized in that: The aerial reference end further comprises a winch device, which comprises a mooring cable and a fixing seat, wherein the fixing seat is fixed on the ground and the mooring cable is stored on the fixing seat. The airship is pulled up to a predetermined height by the mooring cable.

3. The passive positioning system for trapped persons based on airship digital marking according to claim 1, characterized in that: There are eight annular areas, each of which has three layers of digital rings. The outer layer is letters AH, the middle layer is numbers 1-8, and the inner layer is a QR code check digit. The digital ring is a large LED digital code of 2m×2m.

4. The passive positioning system for trapped persons based on airship digital marking according to claim 1, characterized in that: The ground-based solution terminal is provided with a position solution engine. When the trapped personnel report the airship data they have seen, the airship terminal uploads the real-time yaw angle θ of the airship to the position solution engine. The position solution engine calculates the azimuth annular area using a geometric projection algorithm: α = θ + (S_id × 45°) ± 22.5°.

5. The passive positioning system for trapped persons based on airship digital marking according to claim 1, characterized in that: The water search and rescue model has a built-in feature library of people who have fallen into the water. When the drone arrives at the target water area, it uses a multispectral imaging device to perform identification and search based on the data in the feature library of people who have fallen into the water. The land search and rescue model has a built-in feature library of human activities. When the drone arrives at the target land area, it uses a multispectral imaging device to perform identification and search based on the data in the feature library of human activities.

6. The passive positioning system for trapped persons based on airship digital marking according to claim 5, characterized in that: When the drone searches on the flight route, when it goes to search in a water area, it adopts the water mode, i.e., spiral expansion search; when it goes to search in a landing area, it adopts the land mode, i.e., flying along the street topology path; when there is land in the water area or water in the land area, it adopts the mixed mode, i.e., preferentially scanning the abnormal reflective area on the water surface.

7. The passive positioning system for trapped persons based on airship digital marking according to claim 5, characterized in that: The multispectral imaging device includes a visible infrared detector, an infrared thermal imager and a laser radar rangefinder for identifying human features and distress movements.

8. The passive positioning system for trapped persons based on airship digital marking according to claim 5, characterized in that: The remote data transmission device carried by the drone is a local area network, a hotspot network or a LoRa gateway.

9. The method for passively locating trapped persons based on airship digital marking according to any one of claims 1 to 8, characterized in that: The steps include: S1. The trapped person observes the airship code and elevation angle φ, and reports the result to the ground control terminal by dialing a rescue line through communication equipment. S2. The ground control terminal combines the reported information with the airship's real-time yaw angle θ to calculate the azimuth sector α=θ+(S_id×45°)±22.5°, determines the area where the trapped person is located, generates an EPSO optimized path based on the area, and transmits the optimization to the UAV control terminal. S3. The UAV control terminal controls the UAV to approach the area where the trapped person is located along the EPSO optimized path and conducts a detailed search using a multispectral imaging device to determine the trapped person's trapped state and precise trapped location. The video and the trapped person's trapped location are then transmitted back to the ground control terminal in real time via a remote data transmission device. S4. After receiving the video and trapped location transmitted by the drone, the ground control terminal generates a rescue plan based on the situation.

Images