CN211403566U - Children prevent drowned monitored control system and unmanned aerial vehicle based on unmanned aerial vehicle - Google Patents

Abstract

An unmanned aerial vehicle-based monitoring system for preventing drowning of children and an unmanned aerial vehicle, the monitoring system includes a casing; a navigation unit includes a processor and a navigation camera, a monitoring camera and an ultrasonic distance sensor connected to the processor; a buoy communication unit , including a communication buoy, the processor is connected to the server through the communication buoy; the flight control unit includes an underwater propeller set at the rear of the casing and a flying wing propeller set at the top of the casing, the underwater propeller is driven by a propulsion motor drive; the flying-wing propeller is driven by the rotor motor; the propulsion motor and the rotor motor are connected to the output end of the processor; the charging unit includes a rechargeable battery and a battery life state detection unit, and the battery life state monitoring unit is connected to the input end of the processor ; Airbag control unit, including inflation airbag and airbag valve controller. The utility model also includes an unmanned aerial vehicle. The utility model has the advantages of high rescue efficiency, fast rescue speed and the like.

Description

A drone-based child drowning prevention monitoring system and drone

technical field

The utility model relates to the technical field of intelligent rescue, in particular to a child drowning prevention monitoring system based on an unmanned aerial vehicle and an unmanned aerial vehicle.

Background technique

Outdoor drowning accidents often occur. Due to limited conditions, in many cases, the drowning person cannot be rescued in time because the drowning person is too far from the shore, or there is no person who can provide rescue on the shore, resulting in danger.

In 2016, the patent application No. [2016112684434] designed a drone rescue equipment that can be used for rescue. The equipment includes a wearable monitoring and rescue terminal and a drone monitoring terminal. The patent specifically designs a wearable device. The structure of the device, after the human body is drowning, on the one hand, the wearable device broadcasts the position information to the lifeguard, and releases the airbag at the same time, on the other hand, the drone flies to find the drowning person, and shoots the real-time situation of the drowning person as a video and transmits it to the lifeguard, Which supports communication methods 3G, 4G, WIFI, Bluetooth or infrared. In this patent, the role of drones is relatively limited, only for querying and tracking drowning people.

In 2017, the patent application number [201710784016X] used a rotary-wing drone to design a marine rescue system, combining four ordinary rotary-wing drones and a stage, enabling it to have two states of flight and sea navigation. After locating the drowning person, it falls vertically to the sea surface where the drowning person is located. The drowning person climbs down the ladder and gets it onto the platform carried by the drone. The drone pushes the drowning person to move on the sea surface. In the same year, the patent application number [201710497150.1] proposed a remote-controlled maritime rescue drone, including a fuselage, a flight control module, a flight mechanism and a delivery mechanism, which can be used in rescue scenarios where the drowning person does not wear a positioning device. The drone is searched under remote manual control. After the drowning person is found, the pull rope and lifebuoy are lowered, and a positioning device is set on the pull rope to inform others to help. In the same year, the patent with the application number [2017210238462] proposed a drone that can drop a lifebuoy. The main difference from the previous patent is that the mechanical claw for dropping the drone is designed in detail.

In 2018, the patent application No. [2018105911697] provided a design of a drone for first aid. The design is suitable for drowning people who have been rescued ashore. The drone is remotely controlled to locate the drowning who has been rescued ashore. personnel, and put oxygen masks and other relief supplies. In the same year, the patent application No. [2018110415697] disclosed a drowning prevention monitoring method based on a tethered drone. The drone was equipped with a camera, and the camera in the four directions of the drone was used to photograph the swimming pool. All the scenes, avoiding blind spots, arrange and combine the images in the four directions to form a new set of images in sequence, and analyze the changing rules of swimming motion with the previous set to find out the images that do not conform to the rules, and put them together. The image is compared with the drowning image, and finally the approximate area of the drowning point is found, and then the accurate drowning point is measured by the laser rangefinder. Determine whether there is a drowning person and the specific location of the drowning person, excluding specific rescue measures.

In 2019, the patent with the application number [201910382747.0] disclosed a method for online recognition of drowning based on machine vision, including image acquisition; preprocessing; image transmission: the preprocessed image will be transmitted to a cloud server through a wireless network; Offline training OpenPose: The cloud server is based on lightweight accelerated OpenPose, and offline training is suitable for models that extract key points of the human body in water; offline training classifier: After extracting the key points of the human body, a neural network-based binary classifier is trained for judgment Whether the person is drowning; server online monitoring: the cloud server runs online, and the improved lightweight acceleration OpenPose is used to take key points of the human body in the image, and judge the key points for drowning, calculate the degree of danger, and output alarm information. The utility model can be used on water small robots, water fixed cameras or underwater fixed cameras, and is used for recognizing the real-time posture of swimmers, screening and early warning of suspicious postures of drowning, and assisting lifeguards in identifying drowning in swimming pools and seasides. effect.

However, the above-mentioned patented technologies have the following defects: (1) the existing drowning prevention systems of UAVs that can implement direct rescue lack autonomy, and human control is required to achieve positioning and return and rescue throughout the whole process; (2) some UAVs only monitor (man-made remote control monitoring, online self-identification monitoring) drowning behavior, and broadcast the location and real-time environment to the relevant personnel for help, instead of directly releasing lifebuoys or life-saving ropes; (3) some patents that can directly implement rescue, The situation that the drowning person has been submerged in the water and cannot touch the surface rescue tools (such as life-saving ropes and life buoys) is not considered; (4) the rescue route of the drone is not optimally planned, which will easily delay the rescue time; (5) At present Generally, drowning is judged by water surface video, and the drowning person on the water surface is searched in the air, without considering that the drowning person may have submerged in the water.

Utility model content

The purpose of the present utility model is to overcome the above-mentioned deficiencies of the prior art and provide a UAV-based child drowning prevention monitoring system and UAV with high rescue efficiency and fast rescue speed.

The technical scheme of the present utility model is:

A child drowning prevention monitoring system based on an unmanned aerial vehicle of the present utility model comprises:

chassis;

A navigation unit including a processor and a navigation camera, surveillance camera and ultrasonic distance sensor connected to the processor;

a buoy communication unit, including a communication buoy, and the processor is connected to the server through the communication buoy;

The flight control unit includes an underwater propeller arranged at the rear of the casing and a flying-wing propeller arranged at the top of the casing. The underwater propeller is driven by a propulsion motor; the flying-wing propeller is driven by a rotor motor; the propulsion The motor and the rotor motor are connected to the output of the processor;

The charging unit includes a rechargeable battery and a battery life state detection unit, and the battery life state monitoring unit is connected to the input end of the processor;

The airbag control unit includes an inflatable airbag and an airbag valve controller; an airbag outlet is arranged on the casing, and an airbag valve controller is arranged at the airbag outlet, and the airbag valve controller is connected to the output end of the processor; the inflatable airbag is arranged on the Inside the casing, it can be released or retracted along the airbag outlet.

Further, the buoy communication unit further includes a buoy lock controller; the buoy lock controller is connected to the output end of the processor, and is used for locking or unlocking the communication buoy.

Further, the casing is provided with an autonomous charging contact and an outer guide point connected to the power supply end of the rechargeable battery, and the autonomous charging device contacts the autonomous charging contact through infrared guidance to perform autonomous charging of the rechargeable battery.

Further, the server is respectively connected to the wearable device, the drone, and the guardian terminal.

Further, a lighting unit is also included, and the lighting unit is connected to the output end of the processor.

Further, the casing adopts an IP69 waterproof design.

Further, the processor adopts Feiteng processor and is equipped with Beidou signal, electronic compass and inertial navigation unit.

Further, the navigation camera is a VSLAM camera; the surveillance camera is a 270° high-definition camera.

An unmanned aerial vehicle of the present invention includes the drone-based child drowning prevention monitoring system described in any of the foregoing.

Further, the navigation camera is arranged on the lower side of the casing, the monitoring camera is arranged on the front end of the casing; the communication buoy and the buoy lock controller are arranged on the upper side of the rear of the casing; the airbag outlet is arranged on the navigation camera the back end.

The beneficial effects of the present utility model:

(1) By setting the inflatable airbag and the airbag valve controller, the human body will take the initiative to enter the water for rescue when there is a risk of drowning, which greatly reduces the risk;

(2) By setting the communication buoy, on the one hand, the drone can still provide lower control delay and video images when it enters the water for rescue; on the other hand, when the drone is launched, the communication buoy is locked in the buoy lock in advance, It will not hinder the drone from entering the water and get quick rescue;

(3) The autonomous drone is equipped with an underwater propeller, and the underwater propeller is turned on after entering the water, which has better rescue time than only relying on the rotor drive;

(4) By establishing communication with the server, the drone can fly and monitor behavior outside the rescue stage without human intervention. In the water rescue stage, the server can remotely control and release rescue equipment in real time, with high flexibility, strong stability, and operation. low cost;

(5) Set up autonomous charging points for drones near waters with high drowning risk, and the drones can be quickly and autonomously charged after landing on the autonomous charging points;

(6) Actively intervene in the human body's launching behavior through amphibious autonomous drones, and take the initiative to enter the water for rescue when there is a risk of drowning, which greatly reduces the risk.

Description of drawings

1 is a block diagram of a circuit structure of a rescue system according to an embodiment of the present invention;

2 is a schematic structural diagram of an unmanned aerial vehicle according to an embodiment of the present invention;

3 is a schematic structural diagram of an unmanned aerial vehicle in another direction according to an embodiment of the present invention;

4 is a schematic diagram of a monitoring state of an unmanned aerial vehicle according to an embodiment of the present invention.

Description of the accompanying drawings: 1. Chassis; 2. Surveillance camera; 3. Autonomous charging contact; 4. Underwater propeller; 5. Flying wing propeller; 6. Propulsion motor; 7. Buoy lock.

Detailed ways

The present utility model will be further described in detail below with reference to the accompanying drawings and specific embodiments.

As shown in Figures 1 to 3: a drone-based child drowning prevention monitoring system, including a casing 1, on which is provided a flight control unit, a navigation unit, a buoy communication unit, a charging unit, and an airbag Control unit and lighting unit.

The casing 1 adopts IP69 waterproof design, which can realize search and rescue by diving to ten meters underwater.

The navigation unit includes a processor, a navigation camera, a monitoring camera 2, an ultrasonic distance sensor and a camera pan/tilt connected to the processor. Among them, the processor is preferably based on the self-controllable Feiteng processor to replace foreign processors based on Intel and nVidia as autonomous positioning and navigation, and obtain its own position by carrying Beidou signals, electronic compass, and inertial navigation. Differential GNSS positioning technology receives satellite signals (including Beidou satellites or GPS). The navigation camera is located on the lower side of the casing, and preferably a VSLAM camera is used. The camera is in the shape of a round hole. The navigation camera cooperates with the processor to design a VSLAM algorithm for visual fusion, so that the drone can calculate its own position and build it at the same time. The environment map solves the problem of positioning and map construction when the UAV is moving in an unknown environment. The monitoring camera is arranged at the front end of the casing, and adopts a 270° high-definition camera, which is used for shooting the external environment and the scene in the water, and the dead angle can be avoided by 270° rotation. The angle of the surveillance camera is adjusted through the camera pan/tilt. The ultrasonic distance sensor is used to detect obstacles to avoid obstacles.

The airbag control unit includes an inflatable airbag and an airbag valve controller; an airbag outlet is arranged in the middle of the lower side of the casing for releasing the inflatable airbag, and the airbag valve controller is located on the outlet side of the airbag, and controls the inflation by controlling the opening and closing of the valve. The air bag is released and retracted; the air bag valve controller is connected to the output end of the processor. The airbag outlet is preferably arranged at the rear end of the navigation camera. When the inflatable air bag needs to be released, the processor will send a control command to the air bag valve controller to control the valve to open.

The charging unit includes a rechargeable battery and a battery life state detection unit, and the battery life state monitoring unit is connected to the input end of the processor; when the battery life state monitoring unit detects that the power of the rechargeable battery is lower than a preset reference power, it will send the The processor sends a charging request signal. There are two autonomous charging contacts 3 connected to the power end of the rechargeable battery and an infrared guidance point on the side of the casing. When the drone lands at the designated position, the autonomous charging device contacts the autonomous charging contact 3 through infrared guidance. Self-charging.

The flight control unit includes an underwater propeller 4 arranged at the rear of the casing and a flying-wing propeller 5 arranged at the top of the casing. The underwater propeller is used for underwater travel and is driven by a propulsion motor 6; The propeller is mainly used as steering power and is driven by the rotor motor, and the number of rotor motors is four. The propulsion motor and the rotor motor are connected to the output of the processor.

The lighting unit is an LED light, used for underwater or nighttime lighting, and the LED light is connected to the output end of the processor. LED lights can be placed anywhere on the casing.

The buoy communication unit includes a communication buoy and a buoy lock controller arranged on the upper side of the rear of the casing, and the buoy lock controller is connected to the processor for locking or unlocking the communication buoy. When the drone is in normal flight, it is locked to the casing. When the drone receives the water entry command, the buoy lock controller is controlled to open the buoy lock 7, and the communication buoy is released. The communication buoy is a hollow transmitting antenna, which is preferably a 5-meter long communication cable in this embodiment. After the drone enters the water, the buoyancy will make the buoy float on the water surface, providing a high-bandwidth communication relay for the drone entering the water, preferably using a 4G or 5G mobile network for communication.

The processor communicates with the server through a communication buoy, and the server is a supercomputer cloud platform. And the server communicates with the wearable device, the drone, and the guardian terminal respectively; the server is used to judge the risk level of human drowning according to the drowning risk function, perform early warning classification, and send early warning instructions to the wearable device and the guardian terminal; if the guardian terminal chooses When the drone travels, the server establishes communication with the processor of the drone to obtain the battery life status information and location information of the drone. The information dispatches the UAV array, and at the same time informs the human customer service of the rescue center to prepare for manual intervention in rescue activities at any time; the server is also used to obtain the image information captured by the UAV at the target location, and acts according to the image information to determine whether the human body is in a drowning state. If it is judged to be drowning, the server remotely commands the drone to carry out rescue in real time.

As shown in Figure 4: If the drone of this embodiment does not receive the request for support instruction sent by the server, the drone is charged in place or in a waiting state, and if it receives a rescue instruction from the server, it will select the closest to the water. The drone is dispatched to the target location, and the other drones are in a waiting state. If the power of the drone to go is weakened, the server will dispatch the standby drone to go, and the drone that needs to be charged will be found near the water. The charging device performs autonomous charging,

In dealing with the drowning problem of outdoor human bodies (especially children), the anti-drowning system and method designed by the present invention use autonomous drones in two stages respectively.

The first stage is the early warning stage, which includes:

通过穿戴式设备上传的位置数据p得到人体位置的精确评估p`，计算线段p<sub>c</sub>-p`分别与p₁-p₂,......p_n-1-p_n线段是否相交，如果相交则分别计算p`与这些线段的垂直距离；如果相交数为偶数，则风险因子D等于所有垂直距离中最小值。如果没有相交，或者相交数为奇数，则风险因子D等于-1；风险因子D即为距离d。根据距离d计算潜在的溺水风险，其中，溺水风险函数定义为R＝f(d，s，w，h，t)，式中，s为所处季节，w为天气状态，h为湿度，t为温度。根据溺水风险函数判断人体溺水的风险等级，进行预警分级。Children wear wearable devices, which can be in the form of watches, bracelets, or school badges; the wearable device will send the child's relevant location information to the server, and the server calculates the child's distance from the dangerous water area d according to the child's location. is: marking the boundary of dangerous waters, represented by an ordered set of points P={p ₁ , p ₂ , p ₃ , p ₄ , p ₅ ,...,p _n }; calculating the average value of the boundary of dangerous waters

Accurate evaluation p` of the human body position is obtained through the position data p uploaded by the wearable device, and calculate whether the line segments p <sub>c</sub> -p` intersect with the line segments p ₁ -p ₂ ,......p _n-1 -p _n respectively, If they intersect, the vertical distances between p` and these line segments are calculated respectively; if the number of intersections is even, the risk factor D is equal to the minimum value among all vertical distances. If there is no intersection, or the number of intersections is odd, the risk factor D is equal to -1; the risk factor D is the distance d. Calculate the potential drowning risk according to the distance d, where the drowning risk function is defined as R=f(d, s, w, h, t), where s is the season, w is the weather state, h is the humidity, t for the temperature. According to the drowning risk function, the risk level of human drowning is judged, and the early warning classification is carried out.

The labeling method of hazardous waters is managed through the combination of UGC (user-generated content) and PGC (professional-generated content), that is, provided by the combination of users (such as students' parents, teachers, etc.) and experts (such as officials, administrators, etc.) Label data. In the system initialization stage, the system administrator will import the basic data of key water areas, which comes from the government water conservancy facilities and other surveying and mapping departments, and the server provides interface access data.

Among them, the information of water width and depth changes more drastically during the flood season. Especially in the rainy season, the water volume changes greatly, and the water level and channel width in the morning and noon will change greatly. Therefore, it is necessary to be able to dynamically map the real-time width and depth of the water area. The system performs dynamic mapping and tracking of important dangerous waters, and combines positioning technology with UAV technology. The drone collects the ordered point set P' for each flight, and replaces the ordered point set P of the water area on the server.

The warning classification includes: automatically triggering three-level warnings near water, near water, and entering water according to the position information of the human body; when the human body approaches the dangerous waters, the system issues a level III warning to the human body; The relevant contacts of the human body; when it is detected that the human body may enter the water, the system will issue a level 1 warning to the human body, and at the same time notify the relevant contacts of the human body again.

When the server sends a signal to the guardian terminal to notify the child of the dangerous situation, the guardian can choose whether to send out the drone. If the guardian needs to dispatch the drone, the guardian terminal will send a request command to the server to request the drone support. After the server receives the command, it will notify the drone according to the battery life status of the nearby drone and the distance from the monitoring site. arrive at the monitoring site. The server sends the child's location to the drone. After the drone obtains the target location, it will plan a shortest path to the monitoring location based on the map information and current wind direction information. Among them, the server will uniformly schedule the drone array to cover the near-water human body in a full cycle according to the flying drone, the ready-to-fly drone, the nearby drone and the wind direction information. After the drone takes off, the server plans the standby drone to prepare according to the battery life status of the current drone array, and at the same time informs the rescue center's human customer service to prepare for manual intervention at any time. The drone will be remotely controlled to the vicinity of the child's location for monitoring and observation, and the real-time video under monitoring will be uploaded to the server, which will then be sent to the guardian's client (the drone monitoring screen can be simultaneously transmitted to the rescue through the operator's network). The center, parents, and nearby enthusiastic people form a joint rescue). The server will automatically schedule the next drone to take over the monitoring of the current drone before the battery life runs out. If the child is successfully persuaded to leave or the child is not in danger of drowning, the drone will return to the starting point and recharge itself. Among them, the autonomous charging point of the drone is located near the waters with a high risk of drowning, so that the drone can quickly and autonomously charge after landing on the autonomous charging point.

In this embodiment, the shortest path planned by the UAV refers to the shortest path of the UAV from the starting point to the target monitoring location, and the planning method includes the following steps: S1: The UAV first converts the three-dimensional map information grid S2: Assign the obstacles (mainly mountains and tall buildings) in the 3D map to the highest cost weight; S3: Use the artificial potential field method to calculate the cost weight of each grid of the 3D grid according to the starting point and end point information of the UAV ; S4: According to the wind direction and wind size, gradually increase the cost weight of each grid along the wind direction; S5: Use the D* algorithm to search for the lowest cost path from the starting point to the end point in the three-dimensional grid.

The second stage is the rescue stage, which includes:

After the drone uploads the video content of the child into the water in real time to the server, the server conducts behavior analysis on the real-time video to determine whether the person entering the water is in a state of drowning. When launching, the flying-wing propeller is mainly used as steering power, and the UAV is equipped with an underwater linear thruster to improve the traveling speed after entering the water. The operator controls the drone to reach the drowning person, and remotely controls the release of the inflatable airbag, which provides reliable buoyancy for the drowning person to float on the water, and buys more time for the rescue of lifeguards or nearby people. In addition, the communication buoy can still provide low control delay and video images to the server during water rescue, so as to monitor the underwater dynamics of the drowning person.

In the drowning rescue problem, the utility model designs a drone-server-drone method, that is, the drone transmits the captured video to the server in real time, judges whether the drowning behavior occurs through behavior recognition, and then operates the background operation without The man-machine directly searches and releases the airbag for rescue. The drone not only performs the work of judging whether the drowning behavior occurs, but also performs the work of direct rescue. The continuous identification of drowning behavior and drowning rescue can improve the rescue efficiency to a certain extent. In view of the situation that the drowning person may have been submerged in the water and cannot reach the floating objects rescued on the water surface, the amphibious design of the utility model for the drone can realize the function of releasing the inflatable airbag underwater, and the drone can not only search underwater In the case of a drowning person, the inflatable airbag can also be released directly at the side of the drowning person.

In addition, the drone designed by the utility model has the functions of autonomous flight, autonomous positioning, autonomous return and autonomous charging, which improves the autonomy of the drone to a certain extent, shortens the rescue time, and improves the rescue efficiency.

Among them, autonomous flight is realized through two levels of path planning algorithms: global path planning and local path planning. Before the algorithm is started, the system will first rasterize the 3D map, and then calculate the current position of the UAV through the D* global path planning algorithm. The complete flight path to the monitoring site is added, and the dynamic characteristics of the current UAV are added to use the DWA algorithm to approximate the global path, and the approximate operation result controls the flight control unit to fly to the accident location.

The aforementioned shortest path algorithm is the D* algorithm (global path planning hierarchy), and the DWA algorithm (DynamicWindow Approach, translated as dynamic window method) described later is the local path planning hierarchy. By using the global navigation algorithm of D*, an optimal global path can be planned for the robot. This algorithm does not consider the dynamic problem of the actual flight of the machine, but provides the theoretically feasible shortest path. The above "dynamic problem" may be: In actual flight, the path change problem caused by the changing flight speed, assuming that the machine needs to turn in flight, if the flight speed is too fast at this time, then the radian of the turning path will be relatively large, otherwise the radian of the path will be relatively small. DWA is a local path planning algorithm based on the dynamic window method, adding the dynamic characteristics of the speed of the machine when it is actually moving, to deal with the above-mentioned dynamic problems when the machine is actually moving. The combination of the two embodies two levels of global path planning and local path planning.

Autonomous positioning is achieved in the following ways: with the use of the autonomously controllable Feiteng processor, Beidou signal, electronic compass, inertial navigation and VSLAM algorithm, the UAV builds an environmental map while calculating its own position, and solves the problem of the UAV in the The problem of positioning and map construction when moving in an unknown environment, to achieve autonomous positioning.

Autonomous return is the reverse process of autonomously flying to the monitoring point. When the drone finishes monitoring or the standby drone arrives at the scene, the drone will get the return instruction given by the server. Since the take-off point of the drone may have been used by other candidate drones, the server will dispatch its nearest available home point to instruct it to return, and the path search algorithm during the return period is the same as the autonomous flight path algorithm.

Autonomous charging is that when the drone landed at the designated position, the autonomous charging device will automatically charge by contacting with the autonomous charging contact through infrared guidance.

To sum up, the utility model can prevent the similar drowning danger and directly rescue the drowning person when the drowning danger occurs. When the human body is close to the dangerous area where drowning may occur, the autonomous drone will be dispatched and provide surveillance video of the area where drowning may occur to the guardian terminal APP. When the danger of drowning occurs, the server uses the customer service agent to remotely command the drone to enter the water to release rescue equipment for rescue. The rescue process is timely and autonomous, with less intervention and fast rescue.

Claims (10)

1. a child anti-drowning monitoring system based on unmanned aerial vehicle, is characterized in that, comprises: chassis; A navigation unit including a processor and a navigation camera, surveillance camera and ultrasonic distance sensor connected to the processor; a buoy communication unit, including a communication buoy, and the processor is connected to the server through the communication buoy; The flight control unit includes an underwater propeller arranged at the rear of the casing and a flying-wing propeller arranged at the top of the casing. The underwater propeller is driven by a propulsion motor; the flying-wing propeller is driven by a rotor motor; the propulsion The motor and the rotor motor are connected to the output of the processor; The charging unit includes a rechargeable battery and a battery life state detection unit, and the battery life state monitoring unit is connected to the input end of the processor; The airbag control unit includes an inflatable airbag and an airbag valve controller; an airbag outlet is arranged on the casing, and an airbag valve controller is arranged at the airbag outlet, and the airbag valve controller is connected to the output end of the processor; the inflatable airbag is arranged on the Inside the casing, it can be released or retracted along the airbag outlet. 2. The drone-based child drowning prevention monitoring system according to claim 1, wherein the buoy communication unit further comprises a buoy lock controller; the buoy lock controller is connected to the output end of the processor for Lock or unlock the communication buoy. 3. The drone-based anti-drowning monitoring system for children according to claim 1, wherein the casing is provided with an autonomous charging contact and an outer guide point connected to the power supply end of the rechargeable battery, and the autonomous charging device passes through the The infrared guide is in contact with the autonomous charging contact for autonomous charging of the rechargeable battery. 4. The drone-based child drowning prevention monitoring system according to claim 1, 2 or 3, wherein the server is respectively connected to wearable devices, drones, and guardian terminals in communication. 5. The drone-based anti-drowning monitoring system for children according to claim 1, 2 or 3, further comprising a lighting unit, the lighting unit being connected to the output end of the processor. 6. The drone-based anti-drowning monitoring system for children according to claim 1, 2 or 3, wherein the casing adopts an IP69 waterproof design. 7. The drone-based child drowning prevention monitoring system according to claim 1, 2 or 3, wherein the processor adopts a Feiteng processor, and is equipped with a Beidou signal, an electronic compass, and an inertial navigation unit. 8. The drone-based anti-drowning monitoring system for children according to claim 1, 2 or 3, wherein the navigation camera is a VSLAM camera; the monitoring camera is a 270° high-definition camera. 9 . An unmanned aerial vehicle, characterized in that it comprises the drone-based child drowning prevention monitoring system according to any one of claims 1 to 8 . 10 . 10. The drone according to claim 9, wherein the navigation camera is arranged on the lower side of the casing, and the monitoring camera is arranged at the front end of the casing; the communication buoy and the buoy lock controller are arranged on the upper side of the rear of the casing.

Images