Unmanned aerial vehicle rescue system

Abstract

The invention discloses an unmanned aerial vehicle rescue system, and relates to the technical field of unmanned aerial vehicle control. The unmanned aerial vehicle rescue system comprises a rescue unmanned aerial vehicle, and a system controller, a camera device and a rope supply device which are arranged at the rescue unmanned aerial vehicle; the camera device is used for collecting a visual image below the rescue unmanned aerial vehicle and sending the visual image to the system controller; the system controller is configured to execute the following processing: acquiring a visual image, identifying a target image of a to-be-rescued target in the visual image, and determining a target space coordinate of the to-be-rescued target based on the visual image containing the target image and a space coordinate of the rescue unmanned aerial vehicle; and based on the target space coordinates, the rescue unmanned aerial vehicle is controlled to fly to the position above the to-be-rescued target, and a rope supply device is controlled to release a rescue rope to the to-be-rescued target. According to the technical scheme, the rope supply rescue efficiency of to-be-rescued personnel trapped on power equipment such as a power tower and the like can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of unmanned aerial vehicle (UAV) control, and in particular to a UAV rescue system. Background Art

[0002] As society's demand for electricity continues to grow, so too does the need for stable power system operation. To ensure this, power workers are required to perform high-altitude operations on power equipment, such as power poles, transmission lines, or distribution lines. However, these workers may become trapped on these equipment, requiring external rescue.

[0003] Currently, most existing rescue methods involve rope-feeding rescue, where rescuers on the ground use high-pressure gas or other methods to launch a rescue rope to a trapped person at a high altitude, allowing the trapped person to retrieve the rescue rope and use it to descend to the ground. However, this method of providing a rescue rope to a trapped person places high demands on the rescuer. There is a possibility that the rescuer may not be able to accurately launch the rescue rope to the trapped person, requiring multiple launches, which affects the rescue efficiency. Therefore, the existing rope-feeding rescue method has a low efficiency. Summary of the Invention

[0004] In view of this, the present application provides a drone rescue system, the main purpose of which is to solve the technical problem of low efficiency in providing rope rescue for people to be rescued who are trapped on power equipment such as power poles.

[0005] According to a first aspect of the present invention, there is provided a drone rescue system, comprising a rescue drone, and a system controller, a camera device, and a rope supply device provided at the rescue drone;

[0006] The camera device is used to collect visual images below the rescue drone and send the visual images to the system controller;

[0007] The system controller is configured to perform the following processing:

[0008] The system controller acquires the visual image, identifies a target image of a target to be rescued in the visual image, and determines the target spatial coordinates of the target to be rescued based on the visual image including the target image and the spatial coordinates of the rescue drone;

[0009] Based on the target space coordinates, the rescue drone is controlled to fly above the target to be rescued, and the rope supply device is controlled to lower the rescue rope toward the target to be rescued.

[0010] In an optional embodiment, the lower end of the rescue rope is provided with an electromagnetic lock buckle.

[0011] In an optional embodiment, the camera device has an active infrared camera unit and a natural light camera unit; the camera device collects visual images below the rescue drone, including: the camera device collects infrared images and natural light images below the rescue drone through the active infrared camera unit and the natural light camera unit respectively; determines the infrared image edge contour line in the infrared image, and obtains an infrared edge contour image based on the infrared image edge contour line; determines the natural light edge contour line in the natural light image, and obtains a natural light edge contour image based on the natural light edge contour line; superimposes the infrared edge contour image and the natural light edge contour image to obtain the visual image.

[0012] In an optional embodiment, the system controller identifies a target image of a target to be rescued in the visual image, comprising: the system controller sending the visual image to a pre-trained convolutional neural network model so that the convolutional neural network model identifies the target image in the visual image.

[0013] In an optional embodiment, a marker is provided at the lower end of the rescue rope, and the marker includes a position sensor, which is used to determine the position coordinates of the marker and send the position coordinates to the system controller; the system controller is also configured to perform the following processing: the system controller determines in real time the deviation direction of the position coordinates and the spatial coordinates in the horizontal direction, as well as the distance value in the horizontal direction; and controls the rescue drone to move the distance value in the opposite direction of the deviation direction.

[0014] In an optional embodiment, the system controller controls the rope supply device to lower the rescue rope to the target to be rescued, including: the system controller controls the camera device to continuously capture the visual images, and identifies the target image of the target to be rescued in each visual image; performs gesture recognition on the target image of each visual image, and when a preset rope supply completion gesture is recognized in the target image, controls the rope supply device to release the rescue rope, so that the rescue rope falls off from the rope supply device.

[0015] In an optional embodiment, the drone rescue system further includes a wireless communication unit, which is used to establish a wireless communication connection between the system controller and a remote host computer.

[0016] In an optional embodiment, the system controller is further configured to send the visual image to the host computer.

[0017] In an optional embodiment, the drone rescue system also includes a magnetic field sensor, which is used to collect magnetic field strength information of the location of the rescue drone and send the magnetic field strength information to the system controller; the system controller controls the rescue drone to fly above the target to be rescued based on the target space coordinates, including: the system controller controls the rescue drone to fly above the target to be rescued and lowers the flight altitude at a preset speed; obtains the magnetic field strength information in real time, and compares the magnetic field strength information with a preset magnetic field threshold; when the magnetic field strength information exceeds the magnetic field threshold, controls the rescue drone to hover, and controls the rope supply device to lower the rescue rope to the target to be rescued.

[0018] In an optional embodiment, the marker includes an audible and visual alarm device; the rescue rope has multiple magnetic field detectors, each of which is arranged at a different position of the rescue rope, for collecting magnetic field magnitude information and magnetic field direction information of the magnetic field at the magnetic field detector, and sending the magnetic field magnitude information and the magnetic field direction information to the system controller; the system controller is also used to obtain the magnetic field magnitude information sent by each magnetic field detector in real time when controlling the rope supply device to lower the rescue rope to the target to be rescued, and compare the magnetic field magnitude information with a preset magnetic field magnitude threshold. When the magnetic field magnitude information sent by the magnetic field detector is greater than the magnetic field magnitude threshold, the magnetic field detector is determined as a target detector; the system controller is also used to obtain the magnetic field direction information sent by each target detector, and determine whether the magnetic field direction information sent by each target detector is the same. When the magnetic field direction information sent by any two target detectors is different, the audible and visual alarm device is controlled to issue an electric shock warning message.

[0019] The present invention provides a drone rescue system. First, a drone can fly near a trapped person, capture a visual image containing the trapped person, and identify the trapped person in the visual image. Then, based on the visual image containing the trapped person and the spatial coordinates of the drone, the target spatial coordinates of the trapped person's location are determined. Then, based on the target spatial coordinates, the drone is controlled to fly above the trapped person. Finally, a rope supply device is controlled to lower a rescue rope so that the trapped person can receive the rescue rope, completing the rope supply rescue, and allowing the trapped person to lower themselves to the ground using the rescue rope. The technical solution provided by this application can be based on a drone flying above a trapped person trapped on power equipment such as power poles, transmission lines, or distribution lines, and releasing the rescue rope from top to bottom, allowing the trapped person to easily obtain the rescue rope for self-rescue. Compared to rope supply rescue methods that use high-pressure gas to launch the rescue rope to the trapped person at high altitude, this application can improve the accuracy of rescue rope release and can quickly provide the rescue rope to the trapped person, thereby improving the efficiency of rope supply rescue for trapped people trapped on power equipment such as power poles.

[0020] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of this application. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0022] Figure 1 A schematic structural diagram of a drone rescue system provided by an embodiment of the present invention is shown;

[0023] Figure 2 A schematic diagram showing a flow chart of a rope supply operation performed by a system controller provided by an embodiment of the present invention;

[0024] Figure 3 A schematic structural diagram of a rescue rope and a marker provided by an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0025] The present invention will be described in detail below with reference to the accompanying drawings and in combination with embodiments. It should be noted that, unless there is a conflict, the embodiments and features in the embodiments of the present application can be combined with each other.

[0026] Currently, existing rope-based high-altitude rescue methods often involve rescuers on the ground using methods such as high-pressure gas to launch a rescue rope to a trapped person at a high altitude, allowing the trapped person to retrieve the rescue rope and descend to the ground using the rescue rope. However, this method of providing rescue ropes to trapped people places high demands on the rescuers. There is a risk that the rescuers may not be able to accurately launch the rescue rope to the trapped person, requiring multiple launches, which affects the rescue efficiency. Therefore, existing rope-based rescue methods are inefficient.

[0027] To solve the above problem, in one embodiment, Figure 1 As shown, a drone rescue system is provided. This method is described using a rotary-wing drone as an example. The drone rescue system includes a rescue drone 100, a system controller 200, a camera 300, and a rope supply device 400, which are provided on the rescue drone 100. The system controller 200 can be a computer device such as a single-chip microcomputer or a digital signal processor, and is connected to the flight control system of the rescue drone 100 to control the flight and movement of the rescue drone 100. The camera 300 can be a camera with its lens facing directly below the rotary-wing drone. Furthermore, the rope supply device 400 can be a winch, used to wind a rescue rope and lower or retrieve it. Here, the rescue rope can be an insulated and elastic rope.

[0028] Specifically, the camera device 300 is used to capture visual images of the area below the rescue drone 100 and transmit the images to the system controller 200. Furthermore, the drone rescue system also includes a wireless communication unit (not shown), which may be a wireless signal transceiver. The wireless communication unit is used to establish a wireless communication connection between the system controller 200 and a remote host computer. Here, the host computer may be a computer device in a remote power system control center, so that relevant personnel can understand the rescue situation through the host computer. In actual use, the system controller 200 can establish a wireless communication connection with the remote host computer, receive the location information of the trapped person sent by the host computer, and navigate to the location. When the rescue drone 100 reaches the trapped person's location, the system controller 200 controls the camera device 300 to begin capturing visual images of the area below the rescue drone 100 and transmit the images to the host computer. Here, the images may be photos or videos. Furthermore, when the rescue drone 100 is heading for the approximate location of the trapped persons, its flight altitude is set to be higher than the height of power equipment such as iron towers to prevent the rescue drone 100 from colliding with the power equipment during flight.

[0029] Further, such as Figure 2 As shown, the system controller 200 is configured to perform the following processes:

[0030] 201. The system controller acquires the visual image, identifies a target image of the target to be rescued in the visual image, and determines the target spatial coordinates of the target to be rescued based on the visual image containing the target image and the spatial coordinates of the rescue drone.

[0031] Specifically, the system controller can send the visual image to a pre-trained convolutional neural network model so that the convolutional neural network model can recognize the target image in the visual image. Here, the convolutional neural network model can be trained by a large number of images marked with power workers at various angles. The image can be a contour image, an infrared image, or a natural light image, so that the convolutional neural network model can receive the visual image, and recognize the target image of the target to be rescued in the visual image, and determine the position of the target image in the visual image. Furthermore, a monocular ranging algorithm can be used to determine the distance between the target to be rescued and the camera device 300 based on the position and range of the target image in the visual image.

[0032] Furthermore, the solvePNP algorithm can be used to calculate the relative angle between the target to be rescued and the camera device 300 based on the distance between the target to be rescued and the camera device 300 and the size of the target to be rescued. In addition, the internal parameter matrix of the camera device 300 can also be obtained, including the x-axis focal length f of the camera device 300 on the x-axis. x , the focal length f of the camera device 300 in the y-axis direction y and the principal point coordinates of the camera device 300 in the visual image, including the principal coordinate point c in the horizontal direction x , and the vertical principal coordinate point c y Furthermore, the horizontal coordinate u and the vertical coordinate v of the image center point of the target image in the visual image are obtained. Here, the origin of the coordinate system where the horizontal coordinate u and the vertical coordinate v are located can be the center point of the visual image, and the horizontal coordinate u and the vertical coordinate v can be determined by the position of the pixel of the image center point of the target image in the visual image.

[0033] Furthermore, the horizontal offset and vertical offset of the image center point of the target image in the visual image can be calculated based on Formula 1 and Formula 2:

[0034] △x=uc x (1)

[0035] △y=vc y (2)

[0036] Among them, △x is the horizontal offset, △y is the vertical offset, u is the horizontal coordinate of the image center point of the target image in the visual image, v is the vertical coordinate of the image center point of the target image in the visual image, c x is the principal coordinate point in the horizontal direction, c y is the principal coordinate point in the vertical direction.

[0037] Furthermore, the horizontal angle θ between the image center of the target image and the camera device 300 can be calculated based on Formula 3 and Formula 4: x and vertical angle θ y :

[0038]

[0039] Wherein, △x is the horizontal offset, △y is the vertical offset, and Z is the distance between the target to be rescued and the camera device 300. Further, based on the horizontal angle θ between the image center of the target image and the camera device 300 x and vertical angle θ y The relative angle between the target to be rescued and the camera device 300 can be determined.

[0040] Furthermore, after determining the relative angle between the target to be rescued and the camera device 300, as well as the distance between the target to be rescued and the camera device 300, the relative position between the target to be rescued and the camera device 300 can be determined. Subsequently, the three-dimensional spatial coordinates of the location of the rescue drone 100 can be determined based on a positioning device installed on the rescue drone 100. The spatial coordinates may include latitude, longitude, and elevation. In this case, the positioning device may be a Beidou satellite positioning device. Furthermore, based on the three-dimensional spatial coordinates of the rescue drone 100 and the relative position between the target to be rescued and the camera device 300, the three-dimensional spatial coordinates of the location of the target to be rescued can be determined and used as the target spatial coordinates. In this case, the target spatial coordinates also include latitude, longitude, and elevation.

[0041] 202. Based on the target space coordinates, control the rescue drone to fly above the target to be rescued, and control the rope supply device to lower the rescue rope toward the target to be rescued.

[0042] Specifically, the rescue drone 100 can be controlled to fly to the location indicated by the latitude and longitude of the target spatial coordinates, and the flight altitude of the rescue drone 100 can be controlled to be higher than the altitude of the target spatial coordinates. Furthermore, the rope supply device 400 can be controlled to lower the rescue rope so that the trapped person can receive the rescue rope and use it to rescue themselves. Furthermore, the lower end of the rescue rope can be equipped with an electromagnetic lock buckle for automatically connecting to the trapped person's metal belt / safety hook to ensure that the trapped person can be successfully extricated.

[0043] The present embodiment provides a drone rescue system. First, a drone can fly near a trapped person, capture a visual image containing the trapped person, and identify the person to be rescued in the visual image. Then, based on the visual image containing the trapped person and the spatial coordinates of the drone, the target spatial coordinates of the trapped person are determined. Then, based on the target spatial coordinates, the drone is controlled to fly above the trapped person. Finally, the rope supply device is controlled to lower the rescue rope so that the trapped person can receive the rescue rope and lower himself to the ground using the rescue rope, completing the high-altitude rescue operation. The technical solution provided by the present application can be based on the drone flying above the trapped person on power equipment such as power towers, transmission lines, or distribution lines, and releasing the rescue rope from top to bottom, so that the trapped person can easily obtain the rescue rope for self-rescue. Compared with the method of launching the rescue rope to the trapped person at high altitude by means of high-pressure gas, the present application can improve the accuracy of the rescue rope release, can quickly provide the rescue rope to the trapped person, and thus improve the efficiency of the rope supply rescue for trapped people on power equipment such as power towers.

[0044] In an optional embodiment, if Figure 3 As shown, a marker 411 is provided at the lower end of the rescue rope 410. Here, a flashlight and a speaker can be provided at the marker 411 to facilitate the drone to carry out rescue at night or in low light. Furthermore, the marker 411 includes a position sensor 420, which is used to collect the position coordinates of the marker 411 in real time. The position coordinates are in the same coordinate system as the spatial coordinates, and send the position coordinates of the marker 411 to the system controller (not shown in the figure). Here, the flashlight, the speaker and the position sensor 420 can be powered by the battery in the marker 411, and are connected to the system controller through the wires in the rescue rope 410 for data communication.

[0045] Furthermore, the system controller is also configured to perform the following processing: first, determine in real time the deviation direction of the position coordinates and the spatial coordinates in the horizontal direction, as well as the distance value of the deviation in the horizontal direction; here, the deviation direction and deviation value of the point where the position coordinates are located and the point where the spatial coordinates are located can be compared in the coordinate system. Then, control the rescue drone to move the distance value in the opposite direction of the deviation direction. As an example, if the deviation direction is due north and the distance value is 2 meters, the system controller can control the rescue drone to fly 2 meters to due south. The technical solution provided in the present application can make timely wind deviation adjustments when the rescue rope is blown by the wind and is not lowered vertically downward, so that the rescue rope can be accurately delivered to the trapped person.

[0046] In an optional embodiment, the camera device includes an active infrared camera unit and a natural light camera unit; wherein, the active infrared camera unit can be an infrared camera, and the natural light camera unit can be a video camera, and the lens of the infrared camera and the lens of the video camera are arranged side by side and adjacent to each other and the lenses are oriented in the same direction to achieve the same image acquisition range of the infrared camera and the image acquisition range of the video camera.

[0047] Furthermore, the camera device collects visual images below the rescue drone, including:

[0048] First, the camera device collects infrared images and natural light images of the area below the rescue drone through an active infrared camera unit and a natural light camera unit, respectively. Specifically, the active infrared camera unit collects infrared images, and the natural light camera unit collects natural light images.

[0049] Then, the infrared image edge contour lines are determined in the infrared image to obtain an infrared edge contour image, and the natural light edge contour lines are determined in the natural light image to obtain a natural light edge contour image. Specifically, the contour lines in the infrared image can be determined based on edge detection operators such as Sobel and Canny, and the contour lines are extracted from the infrared image. The extracted contour lines are then set in a blank image base of the same size as the visual image to obtain an infrared edge contour image. Similarly, the contour lines in the natural light image can be determined based on edge detection operators such as Sobel and Canny, and the contour lines are extracted from the natural light image. The extracted contour lines are then set in a blank image base of the same size as the visual image to obtain a natural light edge contour image, so that the natural light edge contour image and the natural light edge contour image have the same image size.

[0050] Finally, the infrared edge profile image is superimposed on the natural light edge profile image to obtain the visual image. Specifically, the image center point of the infrared edge profile image and the image center point of the natural light edge profile image can be determined, and the infrared edge profile image and the natural light edge profile image are superimposed so that the center points of the two images overlap to obtain the visual image.

[0051] The embodiment provided in the present application can superimpose infrared images with natural light images to obtain images of trapped persons under different light intensities, and perform subsequent image recognition processing, thereby improving the success rate of delivering rescue ropes to trapped persons.

[0052] In an optional embodiment, the system controller controls the rope supply device to lower the rescue rope toward the target to be rescued, including:

[0053] First, during the lowering of the rescue rope, the system controller controls the camera device to continuously capture the visual images, and identifies the target image of the target to be rescued in each visual image; specifically, the system controller can control the camera device to shoot and generate a visual image at preset intervals, and identify the target image of the target to be rescued in each visual image, and separate the target image from the visual images.

[0054] Then, gesture recognition is performed on the target image of each visual image, and when a preset rope supply completion gesture is recognized in the target image, the rope supply device is controlled to release the rescue rope so that the rescue rope falls off from the rope supply device.

[0055] Specifically, each target image can be input into a pre-trained gesture recognition model to determine whether a preset rope supply completion gesture appears in the target image; wherein the preset rope supply completion gesture is a gesture used to determine that the trapped person has obtained the rescue rope, such as a V-gesture with the palm facing outward. The specific gesture style can be determined according to actual conditions and notified to each power system operator in advance.

[0056] Here, the gesture recognition model can also be a convolutional neural network model, which can be trained by multiple gesture photos marked with preset rope feeding completion gestures to realize recognition of the preset rope feeding completion gestures.

[0057] Furthermore, when the system controller recognizes a preset rope supply completion gesture from a target image, it controls the rope supply device to disconnect the rescue rope, so that the rescue rope falls off from the rope supply device and falls freely downward, so that the trapped person can obtain the rescue rope and use the rescue rope to save himself.

[0058] The embodiment provided in the present application can perform gesture recognition on the trapped person. When it is recognized that the trapped person makes a preset gesture, it is determined that the trapped person has obtained the rescue rope, and the rescue rope is separated from the rescue drone, so that the trapped person can freely use the rescue rope to save himself and complete the escape operation.

[0059] In an optional embodiment, the drone rescue system further includes a magnetic field sensor, which is disposed within the rescue drone and configured to collect magnetic field information at the rescue drone's location and transmit the magnetic field information to the system controller. The magnetic field sensor may be a magnetometer or a magnetic sensor.

[0060] Furthermore, the system controller controls the rescue drone to fly above the target to be rescued based on the target space coordinates, including:

[0061] First, the system controller controls the rescue drone to fly above the target and then descend at a preset speed. Before the rope is supplied, the drone's altitude can be set higher than that of power equipment, such as a tower, to prevent collisions with the equipment. Furthermore, after the drone reaches the target, the system controller can control the drone to slowly descend vertically at a preset speed.

[0062] Then, the magnetic field strength information is obtained in real time and compared with a preset magnetic field threshold. When the magnetic field strength information exceeds the magnetic field threshold, the rescue drone is controlled to hover and the rope supply device is controlled to lower the rescue rope to the target to be rescued.

[0063] Specifically, when a drone rescues a person trapped on a live distribution line or power transmission line, a magnetic field is generated around the distribution line or power transmission line. The strength of the magnetic field is shown in Formula 5:

[0064]

[0065] Where B(d) is the strength of the magnetic field, μ0 is the vacuum permeability, I is the instantaneous current in the distribution or transmission line, and d is the distance to the distribution or transmission line. As can be seen, the closer the rescue drone is to the live distribution or transmission line, the higher the magnetic field strength at the rescue drone.

[0066] Furthermore, because the current in power equipment such as distribution lines or transmission lines is pre-set by the power system, the magnetic field strength at a specific distance from the power equipment can also be calculated through the current in the power equipment and the length of the specific distance; therefore, a safety distance of a specific length can be set in advance. When the distance between the rescue drone and the power equipment exceeds or is equal to the safety distance, the power equipment will not affect the normal flight of the rescue drone; further, through formula 5, the safety distance and the current in the power equipment, the magnetic field strength at the position at the safety distance from the power equipment can be calculated, and the difference after subtracting a smaller buffer value from the magnetic field strength is determined as the magnetic field threshold. Here, the buffer value and the value of the safety distance can be determined according to actual conditions.

[0067] During the descent of a rescue drone, if the magnetic field strength information collected by the drone's magnetic field sensor exceeds a magnetic field threshold, it can be determined that the rescue drone is close to power equipment. If it continues to descend, it may pose a safety hazard. At this time, the control system controller controls the rescue drone to hover and controls the rope supply device to lower the rescue rope. The technical solution provided by this application can control the rescue drone to get as close as possible to the trapped person without colliding with power transmission and distribution lines, thereby improving the success rate of rope supply.

[0068] In an optional embodiment, if Figure 3 As shown, the marker 411 is provided with an audible and visual alarm device (not shown in the figure), and the audible and visual alarm device is connected to the system controller through a wire provided in the rescue rope 410 for data communication. The audible and visual alarm device can be a buzzer or an alarm light; further, the rescue rope 410 has a plurality of magnetic field detectors 430, and each of the magnetic field detectors 430 is provided at a different position of the rescue rope 410. Specifically, a magnetic field detector 430 can be provided at every preset length on the rescue rope 410; further, the magnetic field detector 430 is used to collect the magnetic field magnitude information and the magnetic field direction information of the alternating magnetic field at the magnetic field detector 430, and send the magnetic field magnitude information and the magnetic field direction information to the system controller (not shown in the figure).

[0069] Here, if a person is trapped in a transmission or distribution line carrying three-phase AC power, the lowered rescue rope may approach multiple conductors simultaneously, each of which may be carrying a different phase of AC power. Furthermore, if the rescue rope is too close to or touches different conductors, and each conductor carries a different phase of AC power, the rescue rope may cause two-phase electric shock. This is especially likely to occur in extremely humid environments or during rainfall, making interphase electric shock more likely. In this case, if the trapped person touches the rescue rope, there is a risk of electric shock.

[0070] Based on this, the system controller is also used to obtain the magnetic field magnitude information sent by each magnetic field detector 430 in real time when controlling the rope supply device to lower the rescue rope 410 to the target to be rescued, and compare the magnetic field magnitude information with the preset magnetic field magnitude threshold. When the magnetic field magnitude information sent by the magnetic field detector 430 is greater than the magnetic field magnitude threshold, the magnetic field detector 430 is determined as a target detector.

[0071] Here, the magnetic field magnitude threshold can be calculated based on Formula 5. Specifically, an experiment can be conducted to determine when the distance between a wet rescue rope 410 and a power transmission line or distribution line is less than or equal to a certain value, indicating that the rescue rope 410 poses an electric shock risk. Subsequently, based on the current in the power transmission line or distribution line and the distance value, the magnetic field strength is calculated and determined as the magnetic field magnitude threshold. Based on this, in actual operation, if the system controller detects that the magnetic field magnitude information collected by a magnetic field detector 430 on the rescue rope 410 is greater than the magnetic field magnitude threshold, it can be determined that the magnetic field detector 430 may pose an electric shock risk and be designated as a target detector. Furthermore, if the magnetic field magnitude information collected by multiple magnetic field detectors 430 on the rescue rope 410 is greater than the magnetic field magnitude threshold, each of the multiple magnetic field detectors 430 can be designated as a target detector.

[0072] Furthermore, the system controller is also used to obtain the magnetic field direction information sent by each of the target detectors, and determine whether the magnetic field direction information sent by each of the target detectors is the same. When the magnetic field direction information sent by any two of the target detectors is different, the sound and light alarm device is controlled to issue an electric shock warning message. Here, because the phase difference between the three-phase currents of three-phase alternating current is 120 degrees, the magnetic field directions generated by the transmission lines or distribution lines that transmit each phase of alternating current are also different from each other. Based on this, if the magnetic field direction information sent by any two target detectors that may cause electric shock is different, the rescue rope 410 may have come into contact with the wires transmitting alternating current of different phases. At this time, if the trapped person touches the rescue rope 410 without any precautions, the trapped person may be in danger of electric shock. At this time, the system controller controls the sound and light alarm device to issue an electric shock warning message to remind the trapped person to pay attention to the risk of electric shock. In addition, the system controller can also control the rope supply device to retract the rescue rope 410 to abandon the rope supply rescue to avoid electric shock to the trapped person, and send a rescue stop message to the host computer so that the rescue personnel at the host computer can know the on-site situation and take corresponding handling measures.

[0073] The technical solution provided in this application can determine whether the rescue rope has the possibility of contacting the wire transmitting alternating current of different phases through the magnetic field size information and the magnetic field direction information, determine whether the trapped person who receives the rescue rope is at risk of electric shock, and send an alarm to the trapped person in a timely manner when the trapped person is at risk of electric shock, so as to improve the safety of the rescue work.

[0074] The drone rescue system provided in this embodiment can accurately identify the specific location of a trapped person, control the drone to approach the trapped person as closely as possible, and provide a rescue rope to them. At the same time, it can correct any deviations that may occur during the descent of the rescue rope, ensuring that the trapped person can successfully obtain the rescue rope. Furthermore, it can determine whether there is a risk of electric shock during the rope supply process and issue a timely warning if there is a risk of electric shock. Compared to methods that use high-pressure gas or other methods to launch the rescue rope to the trapped person at high altitude, the technical solution of this application can significantly improve the efficiency and safety of the rope supply process.

[0075] The serial numbers of the above application are for descriptive purposes only and do not represent the advantages or disadvantages of the implementation scenarios. The above disclosure only discloses several specific implementation scenarios of the present application, but the present application is not limited thereto. Any changes that can be conceived by those skilled in the art should fall within the scope of protection of the present application.

Claims

1. A drone rescue system, characterized in that: The UAV rescue system includes a rescue UAV, and a system controller, a camera device, and a rope supply device arranged at the rescue UAV; The camera device is used to collect visual images below the rescue drone and send the visual images to the system controller; The system controller is configured to perform the following processing: The system controller acquires the visual image, identifies a target image of a target to be rescued in the visual image, and determines the target spatial coordinates of the target to be rescued based on the visual image including the target image and the spatial coordinates of the rescue drone; Based on the target space coordinates, the rescue drone is controlled to fly above the target to be rescued, and the rope supply device is controlled to lower the rescue rope toward the target to be rescued.

2. The drone rescue system according to claim 1, characterized in that: The lower end of the rescue rope is provided with an electromagnetic lock buckle.

3. The drone rescue system according to claim 1, characterized in that: The camera device has an active infrared camera unit and a natural light camera unit; The camera device collects visual images below the rescue drone, including: The camera device collects an infrared image and a natural light image below the rescue drone through the active infrared camera unit and the natural light camera unit respectively; Determining an infrared image edge contour line in the infrared image, and obtaining an infrared edge contour image based on the infrared image edge contour line; Determining a natural light edge contour line in the natural light image, and obtaining a natural light edge contour image based on the natural light edge contour line; The infrared edge contour image and the natural light edge contour image are superimposed to obtain the visual image.

4. The drone rescue system according to claim 1, characterized in that: The system controller identifies a target image of a target to be rescued in the visual image, comprising: The system controller sends the visual image to a pre-trained convolutional neural network model so that the convolutional neural network model recognizes the target image in the visual image.

5. The drone rescue system according to claim 1, characterized in that: A marker is provided at the lower end of the rescue rope, wherein the marker includes a position sensor, and the position sensor is used to determine the position coordinates of the position of the marker and send the position coordinates to the system controller; The system controller is further configured to perform the following processing: The system controller determines in real time the deviation direction of the position coordinates from the space coordinates in the horizontal direction, as well as the distance value in the horizontal direction; The rescue drone is controlled to move the distance value in the opposite direction of the deviation direction.

6. The drone rescue system according to claim 3, characterized in that: The system controller controls the rope supply device to lower the rescue rope toward the target to be rescued, including: The system controller controls the camera device to continuously capture the visual images and identify the target image of the target to be rescued in each visual image; Gesture recognition is performed on the target image of each visual image, and when a preset rope supply completion gesture is recognized in the target image, the rope supply device is controlled to release the rescue rope so that the rescue rope falls off from the rope supply device.

7. The drone rescue system according to claim 3, characterized in that: The drone rescue system further includes a wireless communication unit, which is used to establish a wireless communication connection between the system controller and a remote host computer.

8. The drone rescue system according to claim 7, characterized in that: The system controller is further configured to send the visual image to the host computer.

9. The drone rescue system according to claim 1, characterized in that: The drone rescue system further includes a magnetic field sensor, which is used to collect magnetic field strength information at the rescue drone and send the magnetic field strength information to the system controller; The system controller controls the rescue drone to fly above the target to be rescued based on the target space coordinates, including: The system controller controls the rescue drone to fly above the target to be rescued and to reduce the flight altitude at a preset speed; The magnetic field strength information is obtained in real time, and the magnetic field strength information is compared with a preset magnetic field threshold. When the magnetic field strength information exceeds the magnetic field threshold, the rescue drone is controlled to hover, and the rope supply device is controlled to lower the rescue rope to the target to be rescued.

10. The drone rescue system according to claim 5, characterized in that: The marker includes an audible and visual alarm device; the rescue rope has a plurality of magnetic field detectors, each of which is arranged at a different position of the rescue rope and is used to collect magnetic field magnitude information and magnetic field direction information of the magnetic field at the magnetic field detector, and send the magnetic field magnitude information and the magnetic field direction information to the system controller; The system controller is further configured to obtain, in real time, magnetic field magnitude information sent by each magnetic field detector when controlling the rope supply device to lower the rescue rope toward the target to be rescued, and compare the magnetic field magnitude information with a preset magnetic field magnitude threshold, and determine the magnetic field detector as a target detector when the magnetic field magnitude information sent by the magnetic field detector is greater than the magnetic field magnitude threshold; The system controller is also used to obtain the magnetic field direction information sent by each of the target detectors, and determine whether the magnetic field direction information sent by each of the target detectors is the same. When the magnetic field direction information sent by any two of the target detectors is different, the sound and light alarm device is controlled to issue an electric shock warning message.

Images