Intelligent wearable device for water personnel and water personnel positioning method

Abstract

The application provides a kind of water personnel intelligent wearable device and water personnel positioning method, water personnel intelligent wearable device includes main control unit, positioning unit, multidimensional monitoring unit, long-distance communication unit, high energy density battery unit and alarm unit;Positioning unit obtains the high-precision positioning information of water personnel in current water area;Multidimensional monitoring unit monitors the multidimensional data of water personnel, determines whether water personnel is in the state of drowning risk;Long-distance communication unit adopts satellite communication and 5G communication and user terminal communication;High energy density battery unit adopts new lithium-sulfur battery, main control unit dynamically allocates power according to the operating state of each unit;Alarm unit generates rescue alarm signal, and main control unit sends rescue alarm signal when determining that water personnel is in the state of drowning risk.The application improves the positioning accuracy, drowning risk prediction ability and communication performance of water personnel intelligent wearable device, and improves the safety guarantee of water personnel.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of communication technology, in particular to an intelligent wearable device for water personnel and a water personnel positioning method. BACKGROUND

[0002] With the increasing frequency of water activities (such as water sports, water operations, marine rescue, etc.), the safety of water personnel has become an important concern.

[0003] Existing intelligent wearable devices focus on position tracking functions, but the positioning accuracy is limited, especially in complex water environments (such as multi-island, strong water flow interference areas), which can easily cause large deviations, making it difficult for rescue personnel to quickly and accurately locate the exact position of the distressed personnel. In addition, in terms of vital sign monitoring, most devices can only monitor simple physiological indicators such as heart rate and body temperature, and have insufficient precision in predicting drowning risks, and cannot analyze multi-dimensional data such as the posture, motion changes of personnel in water, and the pressure of water on the human body in real time, making it difficult to issue warnings in a timely manner during the critical stage before a drowning accident occurs. In terms of communication function, the existing wearable devices have a short communication distance with the outside world and poor signal stability. In an open water environment, once it exceeds a certain range, it is easy to lose contact with the rescue command center or nearby ships, affecting the timeliness and effectiveness of the rescue. SUMMARY

[0004] The present application provides an intelligent wearable device for water personnel and a water personnel positioning method, aiming to solve the problems of low positioning accuracy, insufficient drowning risk prediction ability, and poor communication performance of existing intelligent wearable devices for water personnel, and to improve the safety of water personnel.

[0005] In a first aspect, the present application provides an intelligent wearable device for water personnel, comprising a main control unit, a positioning unit, a multi-dimensional monitoring unit, a long-distance communication unit, a high-energy density battery unit, and an alarm unit; the main control unit is connected with the positioning unit, the multi-dimensional monitoring unit, the long-distance communication unit, the high-energy density battery unit, and the alarm unit respectively, for receiving and processing data of each unit;

[0006] The positioning unit uses a multi-satellite positioning system to cooperatively position with an underwater positioning base station, the multi-satellite positioning system includes GPS, Beidou, and GLONASS satellite positioning systems, the underwater positioning base station is arranged on the bottom of a regular planning water area, communicates with the intelligent wearable device for water personnel through acoustic signals, and obtains high-precision positioning information of water personnel in the current water area;

[0007] The multi-dimension monitoring unit comprises a plurality of sensors for monitoring multi-dimension data of the water personnel in real time and transmitting the multi-dimension data to the master control unit for comprehensive analysis to determine whether the water personnel is in a drowning risk state; the multi-dimension data comprises heart rate information, body temperature information, blood pressure information, water posture information, limb movement information and water pressure information on human body;

[0008] The long-distance communication unit adopts satellite communication and 5G communication, if the signal strength of satellite communication in the current water area is greater than that of 5G communication, communication is carried out between the user terminal based on satellite communication; if the signal strength of satellite communication in the current water area is less than that of 5G communication, or the distance between the current water area and the shore is less than or equal to the preset distance, communication is carried out between the user terminal based on 5G communication;

[0009] The high-energy-density battery unit adopts a new type of lithium-sulfur battery, and the master control unit dynamically allocates power according to the operating state of each unit;

[0010] The alarm unit comprises a plurality of alarm signal generation modules for generating a plurality of rescue alarm signals, and the master control unit sends a rescue alarm signal when determining that the water personnel is in a drowning risk state.

[0011] In a second aspect, the present application further provides a water personnel positioning method applied to the water personnel intelligent wearable device of the first aspect, and the water personnel positioning method comprises:

[0012] Real-time acquisition of satellite positioning signals of a multi-satellite positioning system, multi-dimension data of the water personnel, and reception of acoustic signals sent by a plurality of underwater positioning base stations;

[0013] Based on the satellite positioning signals of each satellite positioning system and the acoustic signals of each underwater positioning base station, high-precision positioning information of the water personnel in the current water area is determined;

[0014] Based on the multi-dimension data, it is determined whether the water personnel is in a drowning risk state;

[0015] If the water personnel is in a drowning risk state, a rescue alarm signal is generated based on the high-precision positioning information, and the rescue alarm signal is sent to the user terminal based on the long-distance communication unit.

[0016] In a third aspect, the present application further provides an electronic device comprising a memory for storing a computer software program and a processor for reading and executing the computer software program to realize the water personnel positioning method as described above.

[0017] In a fourth aspect, the present application also provides a non-transitory computer readable storage medium, wherein the storage medium stores a computer software program, and the computer software program, when executed by a processor, implements any of the above-mentioned water personnel positioning methods.

[0018] In a fifth aspect, the present application also provides a computer program product comprising a computer program, and the computer program, when executed by a processor, implements any of the above-mentioned water personnel positioning methods.

[0019] The water personnel intelligent wearable device provided by the embodiments of the present application can provide relatively accurate plane position information through the multi-satellite positioning system and the underwater positioning base station cooperative positioning mode in the area with good satellite signals. In complex water areas, such as the multi-island shielding satellite signals or strong water flow interference leading to the decline of satellite positioning accuracy, the underwater positioning base station can play a role. Since the sound wave propagates relatively stably in water and is not easily affected by the interference factors above the water surface, the distance and direction between the wearable device and the base station can be accurately measured, so as to correct and supplement the satellite positioning information and improve the positioning accuracy. On the other hand, the multiple sensors in the multi-dimensional monitoring unit can comprehensively collect the physiological and motion state data of the water personnel, and the comprehensive analysis of the multi-dimensional data by the master control unit can predict the drowning risk in advance, so as to timely issue an alarm in the dangerous critical stage before the drowning accident occurs completely, thereby improving the drowning risk prediction ability. In addition, the combination of satellite communication and 5G communication can ensure that the device and the outside world are not limited by geographical distance, and can realize global communication coverage. When the water personnel approaches the shore or enters the 5G base station coverage area, the device is automatically switched to 5G communication, and the high speed and low delay characteristics of 5G communication can quickly transmit a large amount of data, thereby ensuring the stability of long-distance communication, improving the efficiency and quality of data transmission, and improving the communication performance. Therefore, the problems of low positioning accuracy, insufficient drowning risk prediction ability and poor communication performance of the existing water personnel intelligent wearable device are effectively solved, and the safety protection level and emergency rescue efficiency of the water personnel are significantly improved. BRIEF DESCRIPTION OF DRAWINGS

[0020] Figure 1 is a structural schematic diagram of the water personnel intelligent wearable device provided by the present application;

[0021] Figure 2 is a flow schematic diagram of the water personnel positioning method provided by the present application;

[0022] Figure 3 is an embodiment diagram of the electronic device provided by the embodiments of the present application;

[0023] Figure 4An embodiment of a computer readable storage medium provided by the embodiment of the present application is shown in the figure. DETAILED DESCRIPTION

[0024] The technical solutions in the embodiments of the present application will be described clearly and completely in combination with the accompanying drawings of the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present application.

[0025] In the description of the present application, the terms "first", "second" are used only for the purpose of description, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined with "first", "second" can explicitly or implicitly include one or more of the features. In the description of the present application, the meaning of "multiple" is two or more, unless otherwise specifically limited.

[0026] In the description of the present application, the term "for example" is used to indicate "as an example, illustration or explanation". Any embodiment described as "for example" in the present application is not necessarily interpreted as more preferred or more advantageous than other embodiments. The following description is given in order to enable any person skilled in the art to implement and use the present application. In the following description, details are listed for the purpose of explanation. It should be understood that those skilled in the art can realize the present application without using these specific details. In other examples, well-known structures and processes will not be described in detail in order to avoid unnecessary details making the description of the present application obscure. Therefore, the present application is not intended to be limited to the shown embodiments, but is consistent with the broadest scope of principles and features disclosed.

[0027] Optionally, referring to Figure 1 shown, Figure 1 is a structural schematic diagram of the water personnel intelligent wearable device provided by the present application, which includes a main control unit, a positioning unit, a multi-dimensional monitoring unit, a long-distance communication unit, a high-energy density battery unit and an alarm unit. Among them, the main control unit in the embodiment of the present application is connected with the positioning unit, the multi-dimensional monitoring unit, the long-distance communication unit, the high-energy density battery unit and the alarm unit respectively, for receiving and processing the data of each unit.

[0028] In an embodiment, the positioning unit in the embodiment of the present application adopts a multi-satellite positioning system in cooperation with an underwater positioning base station, wherein the multi-satellite positioning system includes a GPS (Global Navigation Satellite System), a Beidou satellite positioning system and a GLONASS (Galileo Navigation Satellite System) satellite positioning system. The underwater positioning base station is arranged on the bottom of a normal planning water area and communicates with the intelligent wearable device of the water personnel through acoustic signals. Therefore, when the positioning unit positions the intelligent wearable device of the water personnel, the positioning unit can position according to the satellite positioning signals of each satellite positioning system and the acoustic signals of each underwater positioning base station. For example, in an area with good satellite signals, the multi-satellite positioning system can provide relatively accurate plane position information. When the satellite positioning accuracy is reduced due to complex water areas such as island shielding satellite signals or strong water flow interference, the underwater positioning base station can play a role. Since the acoustic wave propagates relatively stably in water and is not easily affected by interference factors above the water surface, the distance and direction between the wearable device and the base station can be accurately measured, so as to correct and supplement the satellite positioning information, thereby obtaining high-precision positioning information of the water personnel in the current water area.

[0029] In an embodiment, the multi-dimensional monitoring unit in the embodiment of the present application includes a plurality of sensors, such as a heart rate sensor, a body temperature sensor, a blood pressure acquisition sensor, an underwater posture acquisition sensor, a limb movement acquisition sensor and a water pressure on human body acquisition sensor. Therefore, the multi-dimensional monitoring unit can obtain heart rate information, body temperature information, blood pressure information, underwater posture information, limb movement information and water pressure on human body information of the water personnel, and transmit the above data to the main control unit.

[0030] The heart rate sensor is usually based on the PPG (Photoplethysmogram) technology. The sensor emits light of a specific wavelength (such as green light) to the skin surface. The absorption and reflection of light by blood changes periodically with the heartbeat. By detecting the change in reflected light intensity, the sensor converts it into an electrical signal, and then calculates the heart rate value. For example, in each heartbeat cycle, when the blood fills the blood vessels, it absorbs more light, and the reflected light intensity decreases. When the heart relaxes, the reflected light intensity increases. The sensor samples the light intensity change at a certain sampling frequency (such as 100 Hz), and collects a series of discrete data points to form the original sequence HR(t) of heart rate data.

[0031] The body temperature sensor adopts a temperature sensor such as a thermistor or a thermocouple. The body temperature sensor is based on the thermoelectric properties of materials, and the resistance value or electromotive force thereof changes with temperature. The body temperature sensor is attached to a suitable part of the human body (e.g., the inner wrist or the chest) to perceive the body surface temperature in a contact manner. The body temperature sensor collects temperature data at a relatively low sampling frequency (e.g., 1 Hz) to obtain a body temperature data sequence BT(t).

[0032] The blood pressure collection sensor utilizes the oscillometric blood pressure measurement principle. By detecting the pressure change caused by the pulse fluctuation of the blood vessel wall during the inflation and deflation of the cuff, the blood pressure collection sensor obtains the systolic pressure SBP(t) and diastolic pressure DBP(t) values.

[0033] The underwater posture collection sensor uses an inertial measurement unit (IMU), which usually integrates an accelerometer, a gyroscope, and a magnetometer. The accelerometer is used to measure the acceleration of the human body in three coordinate axis directions. For example, when the human body has an upward or downward motion in the water, the acceleration in the vertical direction will change. When there is a translation or rotation motion, the acceleration in the horizontal direction will also change. The gyroscope mainly measures the angular velocity of the human body around the three coordinate axes, which can reflect the rotational motion of the human body, such as rolling, yawing, etc. The magnetometer can assist in determining the direction information. The IMU collects data at a high sampling frequency (e.g., 100 Hz) to obtain the original data sequences of the pitch angle θ(t), the roll angle φ(t), and the yaw angle ψ(t), which reflect the posture changes of the human body in the water.

[0034] The limb motion collection sensor is installed on the key parts of the human body limbs (e.g., wrist, elbow, knee, ankle, etc.) by an acceleration sensor. When the limbs move, the acceleration sensor will detect the corresponding acceleration change. For example, when the arm is stroking, the acceleration sensor at the wrist will collect periodic acceleration changes, and the acceleration components A x (t), A y (t), and A z (t) in different directions can reflect the direction, amplitude, and frequency of the limb motion. The sensor collects data at a high sampling frequency (e.g., 100 Hz).

[0035] ​The pressure collection sensor uses a pressure sensor array distributed at different positions on the human body surface to perceive the water pressure on each part of the human body. For example, pressure sensors are placed at positions such as the chest, back, abdomen, etc. When the human body is in different postures (such as standing, floating, sinking, etc.) in the water, the water pressure on each part will change. The pressure sensor converts the pressure into an electrical signal to collect data at a certain sampling frequency (such as 10 Hz) to obtain a water pressure data sequence P(t).

[0036] Further, the main control unit comprehensively analyzes the heart rate information, body temperature information, blood pressure information, water posture information, limb movement information, and water pressure information of the water personnel to determine whether the water personnel is in a drowning risk state. It should be noted that each sensor in the multi-dimensional monitoring unit has a waterproof and corrosion-resistant coating to adapt to the long-term underwater use environment.

[0037] In an embodiment, the long-distance communication unit adopts satellite communication and 5G communication. If the signal strength of satellite communication in the current water area is greater than that of 5G communication, the long-distance communication unit communicates with the user terminal based on satellite communication. If the signal strength of satellite communication in the current water area is less than that of 5G communication, or the distance between the current water area and the shore is less than or equal to a preset distance, the long-distance communication unit communicates with the user terminal based on 5G communication.

[0038] Therefore, the water personnel intelligent wearable device can transmit the high-precision positioning information of the water personnel in the current water area, the real-time heart rate information, body temperature information, blood pressure information, water posture information, limb movement information, and water pressure information of the water personnel, and the data of whether the water personnel is in a drowning risk state to the user terminal in real time through the long-distance communication unit. It should be noted that the long-distance communication unit also includes a signal enhancement antenna, which adopts a foldable design and can be stored in a specific part of the wearable device when not in use, and can be unfolded to enhance the receiving and sending capability of the communication signal when in use.

[0039] In an embodiment, the high-energy-density battery unit in the embodiment of the present application adopts a new type of lithium-sulfur battery, and the main control unit dynamically allocates power according to the operating state of each unit. It should be noted that the high-energy-density battery unit also has a wireless charging function, which can be wirelessly charged when the water personnel is in a specific area (such as a charging area on a docking pier or a rescue ship).

[0040] In an embodiment, the alarm unit includes a plurality of alarm signal generation modules, which can generate a plurality of rescue alarm signals, such as light alarm signals, sound alarm signals, light and sound alarm signals, etc. When the main control unit determines that the water personnel is in a drowning risk state, it sends a rescue alarm signal.

[0041] The embodiment of the present application can provide more accurate plane position information by the multi-satellite positioning system and the underwater positioning base station cooperative positioning mode in the area with good satellite signal. When the satellite positioning accuracy is reduced due to the complex water area, such as the satellite signal blocked by multiple islands or the strong water flow interference, the underwater positioning base station can play a role. Since the sound wave propagates relatively stably in water and is not easily affected by the interference factors above the water surface, the distance and direction between the wearable device and the base station can be accurately measured, so as to correct and supplement the satellite positioning information and improve the positioning accuracy. On the other hand, the multiple sensors in the multi-dimensional monitoring unit can comprehensively collect the physiological and motion state data of the water personnel, and the main control unit can comprehensively analyze the multi-dimensional data to predict the drowning risk in advance, and timely issue an alarm in the dangerous critical stage before the drowning accident occurs, thereby improving the drowning risk prediction ability. Furthermore, the satellite communication and 5G communication combination mode is adopted. In a wide water area, satellite communication can ensure that the device and the outside world are not limited by geographical distance, and can realize global communication coverage. When the water personnel are close to the shore or enter the 5G base station coverage area, the 5G communication is automatically switched to, and the high speed and low delay characteristics of 5G communication can quickly transmit a large amount of data, which not only ensures the stability of long-distance communication, but also improves the efficiency and quality of data transmission, thereby improving the communication performance. Therefore, the problems of low positioning accuracy, insufficient drowning risk prediction ability and poor communication performance of the existing intelligent wearable device for water personnel are effectively solved, and the safety protection level and emergency rescue efficiency of the water personnel are significantly improved.

[0042] Optionally, referring to Figure 2 , Figure 2 is a flowchart of the water personnel positioning method provided by the present application. The execution subject of the water personnel positioning method in the embodiment of the present application is the intelligent wearable device for water personnel. The wearable device is used to briefly describe the intelligent wearable device for water personnel described above, and therefore, the water personnel positioning method provided by the embodiment of the present application comprises the following steps:

[0043] Step 10, real-time acquisition of satellite positioning signals of a multi-satellite positioning system, multi-dimensional data of the water personnel, and reception of sound wave signals sent by multiple underwater positioning base stations.

[0044] Specifically, the wearable device in the embodiment of the present application can receive the signals of the GPS satellite system, the Beidou satellite system and the GLONASS satellite system in real time, and therefore, the wearable device can acquire the satellite positioning signals of the multi-satellite positioning system in real time. The satellite positioning signals include ephemeris data, pseudo-range information and Doppler shift data, and therefore, it can be understood that the wearable device can acquire the ephemeris data, the pseudo-range information and the Doppler shift data of the multi-satellite positioning system in real time.

[0045] Further, the wearable device can obtain multi-dimensional data of the water personnel in real time through the internal sensor, wherein the multi-dimensional data comprises heart rate information, body temperature information, blood pressure information, water posture information, limb movement information and water pressure information on the human body, so it can be understood that the wearable device can obtain the heart rate information, body temperature information, blood pressure information, water posture information, limb movement information and water pressure information on the human body of the water personnel in real time.

[0046] Further, the wearable device can receive the sound wave signals sent by the plurality of underwater positioning base stations in real time, wherein the sound wave signals comprise base station number information and signal sending time of the underwater positioning base stations, so it can be understood that the wearable device receives the sound wave signals sent by the plurality of underwater positioning base stations in real time, and analyzes the sound wave signals sent by each underwater positioning base station to obtain the base station number information and signal sending time of each underwater positioning base station.

[0047] Step 20, based on the satellite positioning signals of each satellite positioning system and the sound wave signals of each underwater positioning base station, the high-precision positioning information of the water personnel in the current water area is determined.

[0048] Further, the wearable device performs positioning analysis according to the ephemeris data, pseudo-range information and Doppler shift data of each satellite positioning system, and the signal sending time of each underwater positioning base station, to obtain the high-precision positioning information of the water personnel in the current water area, as described in steps 201 to 204.

[0049] Step 30, based on the multi-dimensional data, it is determined whether the water personnel is in a drowning risk state.

[0050] Further, the wearable device analyzes the state of the water personnel according to the heart rate information, body temperature information, blood pressure information, water posture information, limb movement information and water pressure information on the human body of the water personnel obtained in real time, and determines in real time whether the water personnel is in a drowning risk state, as described in steps 301 to 304.

[0051] Step 40, if the water personnel is in a drowning risk state, a rescue alarm signal is generated based on the high-precision positioning information, and the rescue alarm signal is sent to the user terminal based on the long-distance communication unit.

[0052] Further, if it is determined that the water-borne person is in a drowning risk state, the wearable device generates a rescue alarm signal according to the high-precision positioning information. If the signal strength of satellite communication in the current water area is greater than the signal strength of 5G communication, the wearable device sends the rescue alarm signal to the user terminal through satellite communication in the long-distance communication unit. If the signal strength of satellite communication in the current water area is less than the signal strength of 5G communication, or the distance between the current water area and the shore is less than or equal to the preset distance, the wearable device sends the rescue alarm signal to the user terminal through 5G communication in the long-distance communication unit.

[0053] The embodiment of the present application can provide more accurate plane position information by the multi-satellite positioning system and the underwater positioning base station cooperative positioning method. In complex water areas, such as multi-island shielding satellite signals or strong water flow interference leading to satellite positioning accuracy decline, the underwater positioning base station can play a role. Since sound waves propagate relatively stably in water and are not easily affected by interference factors above the water surface, the distance and direction between the wearable device and the base station can be accurately determined, thereby correcting and supplementing the satellite positioning information and improving the positioning accuracy. On the other hand, the multiple sensors in the multi-dimensional monitoring unit can comprehensively collect the physiological and motion state data of the water-borne person, and the main control unit can comprehensively analyze the multi-dimensional data to predict the drowning risk in advance, thereby issuing an alarm in time at the dangerous critical stage before the drowning accident occurs, thereby improving the drowning risk prediction ability. Furthermore, by using the combination of satellite communication and 5G communication, in a wide water area, satellite communication can ensure that the device is connected with the outside world without geographical distance limitation, and can realize global communication coverage. When the water-borne person approaches the shore or enters the 5G base station coverage area, the device automatically switches to 5G communication, and uses the high speed and low delay characteristics of 5G communication to quickly transmit a large amount of data, thereby ensuring the stability of long-distance communication, improving the efficiency and quality of data transmission, and improving the communication performance. Therefore, the positioning accuracy of the existing intelligent wearable device for water-borne persons is low, the drowning risk prediction ability is insufficient, and the communication performance is poor. The safety protection level and emergency rescue efficiency of the water-borne person are significantly improved.

[0054] In an embodiment, steps 201 to 204 are described as follows:

[0055] Step 201, based on the ephemeris data and Doppler shift data of each satellite positioning system, the satellite clock bias and relative speed variation of each satellite positioning system are calculated respectively.

[0056] Optionally, the ephemeris data in the embodiments of the present application includes satellite orbit parameters and related information of satellite clock. The satellite clock bias is mainly caused by the difference between the physical characteristics of the satellite clock (such as the frequency stability of the oscillator) and the ideal clock. Therefore, the wearable device calculates the satellite clock bias of each satellite positioning system according to the ephemeris data of each satellite positioning system. The specific process is as follows: extracting the basic parameters of the satellite clock in the ephemeris data, wherein the ephemeris data provides the polynomial coefficients of the satellite clock, and the polynomial expression of the satellite clock bias is Δt s =a0+a1(t-t0)+a2(t-t0) 2 +....+a n (t-t0) n , wherein t0 represents a reference time, a0, a1, a2,..., a n are polynomial coefficients obtained from the ephemeris data.

[0057] According to the special theory of relativity, due to the high-speed movement of the satellite, the satellite clock will be slower than the ground clock, and the frequency change amount is Δf s / f=-v 2 / 2c 2 , wherein v represents the speed of the satellite, c represents the speed of light, and f represents the nominal frequency of the satellite clock. According to the general theory of relativity, due to the weak gravitational field of the satellite, the satellite clock will be faster than the ground clock, and the frequency change amount is Δf g / f=ΔU / 2c 2 , wherein ΔU is the gravitational potential difference between the satellite and the ground. Therefore, the total change amount of the satellite clock frequency is Δf s / f+Δf g / f.

[0058] In an embodiment, the frequency bias Δf(t1) of the satellite clock at a certain time t1 is known. Then, the change amount of the satellite clock bias in a time interval Δt is By continuously updating this bias value, the satellite clock bias corrected based on the ephemeris data and the relativistic effect is obtained, and the satellite clock bias of each satellite positioning system is obtained n represents the number of satellite positioning systems.

[0059] Further, the wearable device calculates the relative speed change amount of each satellite positioning system according to the Doppler shift data of each satellite positioning system. The specific analysis is as follows: the Doppler shift is caused by the relative movement between the satellite and the wearable device. When the satellite approaches the wearable device, the received signal frequency will increase; when the satellite moves away from the wearable device, the received signal frequency will decrease. Therefore, the frequency f0 of the satellite transmitted signal is obtained, and the frequency f received by the wearable device ist According to the Doppler frequency shift formula f t =f0(1+v r / c), where v r This is the radial relative velocity between the satellite and the wearable device (i.e., the relative velocity along the line connecting them), where c is the speed of light. Further, the radial relative velocity v at a given moment is calculated. r =c(f t -f0) / f0. Further, the received frequency is continuously measured. In one embodiment, the frequency received at time t1 is f1, and the frequency received at time t2 is f2. Then, the change in radial relative velocity during the time interval t1-t2 is Δv. r = c(f2-f1) / f0, thus obtaining the relative velocity change Δv for each satellite positioning system. i .

[0060] Step 202: Based on the coordinates, pseudorange information, satellite clock deviation, and relative velocity change of each satellite positioning system, construct a first objective function to solve for the positioning information of personnel on the water and the clock deviation of the intelligent wearable devices of personnel on the water.

[0061] Furthermore, the wearable device acquires the coordinates (x, y) of each satellite positioning system. i ,y i ,z i Based on the coordinates of each satellite positioning system, the satellite clock offset, and the relative velocity change, as well as the coordinates of the personnel on the water and the clock offset of the intelligent wearable device for the personnel on the water, a preliminary positioning equation is constructed. This preliminary positioning equation can be expressed as:

[0062]

[0063] Where, d i The preliminary positioning equation represents the actual distance between the i-th satellite positioning system and the wearable device, (x i ,y i ,z i Let (x, y, z) represent the coordinates of the i-th satellite positioning system, and let Δt represent the coordinates of the personnel on the water to be solved. u Let f represent the clock skew of the smart wearable device for waterborne personnel to be solved. i Let t represent the satellite signal frequency of the i-th satellite positioning system. i This represents the signal propagation time of the i-th satellite positioning system.

[0064] Furthermore, based on the preliminary positioning equation and the pseudorange information of each satellite positioning system, the wearable device constructs an error equation for each satellite positioning system, whereby the error equation can be expressed as:

[0065]

[0066] wherein e i represents the error equation of the i-th satellite positioning system, ρ i represents the pseudo-range information of the i-th satellite positioning system.

[0067] Further, the above nonlinear equation set is linearized by Taylor series expansion to obtain a linearized matrix equation, and the linearized matrix equation is the first objective function for solving the positioning information of the water personnel and the clock bias of the intelligent wearable device of the water personnel, therefore, the first objective function can be expressed as:

[0068] E=A·X+B.

[0069] wherein the first objective function E represents the error equation e i of all satellite positioning systems n ] T ; A represents a coefficient matrix constructed by the coordinates of the water personnel to be solved and the clock bias of the intelligent wearable device of the water personnel to be solved, that is, A=[x,y,z,Δt u ] T ; B represents a vector containing satellite coordinates and pseudo-range information.

[0070] Step 203, solving the first objective function to obtain the initial positioning information of the water personnel and the clock bias correction number of the intelligent wearable device of the water personnel;

[0071] Further, the wearable device solves the first objective function by least square method, that is, X=(A T A) -1 A T E, to obtain the initial positioning information (x0, y0, z0) of the water personnel and the clock bias correction number Δt u0 of the intelligent wearable device of the water personnel.

[0072] Step 204, based on the clock bias correction number and the sound wave signal of each underwater positioning base station, the initial positioning information is corrected to obtain the high-precision positioning information of the water personnel in the current water area.

[0073] Further, the wearable device corrects the initial positioning information according to the clock bias correction number and the sound wave signal of each underwater positioning base station to obtain the high-precision positioning information of the water personnel in the current water area, which is specifically described in steps 2041 to 2045.

[0074] The embodiment of the present application can provide more accurate plane position information by the multi-satellite positioning system and the underwater positioning base station cooperative positioning mode in the area with good satellite signals. When the satellite positioning accuracy is reduced due to the satellite signal blocked by multiple islands or the strong water flow interference in the complex water area, the underwater positioning base station can play a role. Since the sound wave propagation in water is relatively stable and is not easily affected by the interference factors above the water surface, the distance and direction between the wearable device and the base station can be accurately measured, so as to correct and supplement the satellite positioning information, improve the positioning accuracy, and thus improve the safety guarantee level and emergency rescue efficiency of the water personnel.

[0075] In an embodiment, steps 2041 to 2045 are described as follows:

[0076] Step 2041, based on the signal sending time in the sound wave signal of each underwater positioning base station, the signal receiving time when the intelligent wearable device of the water personnel receives the sound wave signal of each underwater positioning base station, and the water temperature of the current water area at each signal receiving time, the distance between the intelligent wearable device of the water personnel and each underwater positioning base station is determined.

[0077] Optionally, the wearable device can determine the signal sending time in the sound wave signal of each underwater positioning base station according to the information carried in the sound wave signal of each underwater positioning base station.

[0078] Optionally, the wearable device can receive the sound wave signal from m underwater positioning base stations, so as to determine the signal receiving time when the wearable device receives the sound wave signal of each underwater positioning base station, and the signal receiving time of the m underwater positioning base stations is recorded as t r1 ,t r2 ,...,t rm .

[0079] Further, the wearable device obtains the water temperature of the current water area at each signal receiving time, and determines the propagation speed of the sound wave signal at each signal receiving time according to the water temperature at each signal receiving time, wherein the calculation formula of the propagation speed of the sound wave signal at each signal receiving time is as follows:

[0080]

[0081] wherein v wj represents the propagation speed of the sound wave signal at the jth signal receiving time t rj , t rj represents the signal receiving time when the wearable device receives the sound wave signal of the jth underwater positioning base station; T j represents the water temperature of the current water area at the jth signal receiving time t rj , unit: degree Celsius.

[0082] Further, the wearable device determines the distance between the intelligent wearable device of the water personnel and each underwater positioning base station according to the signal sending time in the acoustic wave signal of each underwater positioning base station, the signal receiving time when the intelligent wearable device of the water personnel receives the acoustic wave signal of each underwater positioning base station, and the water temperature of the current water area at each signal receiving time, and the specific calculation formula is as follows:

[0083] d j =v wj (t rj -t sendj ), wherein d j represents the distance between the intelligent wearable device of the water personnel and the jth underwater positioning base station, t sendj represents the signal sending time of the jth underwater positioning base station.

[0084] Step 2042, based on the coordinates of each underwater positioning base station and the distance between the intelligent wearable device of the water personnel and each underwater positioning base station, a second objective function for solving the positioning information of the water personnel is constructed.

[0085] Step 2043, the second objective function is solved to obtain the positioning correction amount.

[0086] Further, the wearable device constructs a second objective function for solving the positioning information of the water personnel according to the coordinates of each underwater positioning base station, the positioning information of the water personnel to be solved, and the distance between the intelligent wearable device of the water personnel and each underwater positioning base station, and the second objective function can be represented as:

[0087]

[0088] wherein (x j ,y j ,z j ) represents the coordinates of the jth underwater positioning base station, (x,y,z) represents the coordinates of the water personnel to be solved, η j represents the acoustic wave propagation measurement error of the jth underwater positioning base station.

[0089] Further, for m, m≥3 base stations, a nonlinear equation set can be obtained, and after linearization processing by Taylor series expansion, the positioning correction amount based on the underwater base station is solved based on the least square method, wherein the positioning correction amount can be represented as Δx b ,Δy b ,Δz b .

[0090] Step 2044, the initial positioning information is corrected based on the positioning correction amount to obtain the final positioning information of the water personnel.

[0091] Further, the wearable device adds the positioning correction amount to the initial positioning information to correct the initial positioning information, and obtains the final positioning information of the water personnel, and thus the final positioning information (x f ,y f ,z f ) of the water personnel can be expressed as:

[0092] (x f ,y f ,z f )=(x0+Δx b ,y0+Δy b ,z0+Δz b ).

[0093] In step 2045, the initial positioning information and the final positioning information are fused based on the clock bias correction number to obtain high-precision positioning information.

[0094] Further, the wearable device fuses the initial positioning information and the final positioning information according to the clock bias correction number to obtain high-precision positioning information, which is specifically described in steps 20451 to 20452.

[0095] The embodiment of the present application can provide relatively accurate plane position information by the multi-satellite positioning system and the underwater positioning base station cooperative positioning mode in the area with good satellite signals. When the satellite positioning accuracy is reduced due to complex water area, the underwater positioning base station can play a role. Since the sound wave propagates relatively stably in water and is not easily affected by interference factors above the water surface, the distance and direction between the wearable device and the base station can be accurately measured, so as to correct and supplement the satellite positioning information, improve the positioning accuracy, and thus improve the safety guarantee level and emergency rescue efficiency of the water personnel.

[0096] In an embodiment, steps 20451 to 20452 are described as follows:

[0097] In step 20451, a first weight coefficient of the initial positioning information and a second weight coefficient of the final positioning information are determined based on the clock bias correction number.

[0098] Specifically, the wearable device calculates the first weight coefficient of the initial positioning information according to the clock bias correction number Δt u0 , and the specific calculation process of the first weight coefficient is as follows: the wearable device calculates the satellite signal strength S1 and the satellite signal signal-to-noise ratio SNR1 of the multi-satellite positioning system according to the clock bias correction number Δt u0 , and the calculation formula of the satellite signal strength S1 is as follows:

[0099] S1=S 10*exp(-aDeltat u0 ).

[0100] wherein S 10 represents the signal strength in an ideal state without clock error, exp() represents an exponential function, and a represents a preset attenuation coefficient.

[0101] The calculation formula of the satellite signal signal-to-noise ratio SNR1 is as follows:

[0102] SNR1=S1 / [N0*(1+beta*abs(Deltat u0 )).

[0103] wherein N0 represents a basic noise level, and beta represents a noise influence coefficient.

[0104] Further, the wearable device acquires the transmission power of the acoustic wave signal of each underwater positioning base station, the gain of the intelligent wearable device of the water personnel, and the propagation attenuation of the acoustic wave in water.

[0105] Further, the wearable device calculates the base station signal strength and the base station signal signal-to-noise ratio of each underwater positioning base station according to the transmission power of the acoustic wave signal of each underwater positioning base station, the gain of the intelligent wearable device of the water personnel, and the propagation attenuation of the acoustic wave in water, wherein the calculation formula of the base station signal strength of each underwater positioning base station is as follows:

[0106] A j =A j *(mu*f j +lambda).

[0107] wherein S 2j represents the base station signal strength of the jth underwater positioning base station, P tj represents the transmission power of the acoustic wave signal of the jth underwater positioning base station, G r represents the gain of the wearable device, d j represents the distance between the wearable device and the jth underwater positioning base station, A j represents the propagation attenuation of the acoustic wave signal of the jth underwater positioning base station in water, f j represents the acoustic wave frequency of the acoustic wave signal of the jth underwater positioning base station, mu represents the absorption coefficient of water in the current area, and lambda represents the scattering coefficient of water in the current area.

[0108] The calculation formula of the base station signal signal-to-noise ratio of each underwater positioning base station is as follows:

[0109]

[0110] wherein SNR 2j represents the base station signal signal-to-noise ratio of the jth underwater positioning base station, Ar represents the effective area of the wearable device receiving the sound wave; P 1j represents the ambient background noise of the current area at the jth signal receiving moment t rj based on the type of water environment; P 2j represents the water flow noise of the current area at the jth signal receiving moment t rj , ρ j represents the density of water in the current area at the jth signal receiving moment t rj , v j represents the flow velocity of the current area at the jth signal receiving moment t rj , L represents the size of the wearable device.

[0111] Further, the wearable device obtains the base station signal strength and base station signal signal-to-noise ratio of the underwater positioning base station according to the mean value of the base station signal strength and the base station signal signal-to-noise ratio of each underwater positioning base station, wherein the calculation formula of the base station signal strength S2 of the underwater positioning base station is as follows

[0112]

[0113] The calculation formula of the base station signal signal-to-noise ratio of the underwater positioning base station is as follows:

[0114]

[0115] Further, the wearable device determines the first weight coefficient ω1 according to the base station signal strength, the base station signal signal-to-noise ratio, the satellite signal strength and the satellite signal signal-to-noise ratio, and the specific calculation formula is as follows:

[0116]

[0117] Further, the wearable device determines the second weight coefficient ω2 of the final positioning information according to the first weight coefficient, that is, the second weight coefficient ω2 = 1-ω1.

[0118] Step 20452, based on the initial positioning information and its corresponding first weight coefficient and the final positioning information and its corresponding second weight coefficient, determine the high-precision positioning information.

[0119] Further, the wearable device performs weighted calculation according to the initial positioning information and its corresponding first weight coefficient and the final positioning information and its corresponding second weight coefficient, to obtain the high-precision positioning information, therefore, the high-precision positioning information (x final , y final , z final ) can be represented as:

[0120] (x final , y final , zfinal )=(ω1x0+ω2x f ,ω1y0+ω2y f ,ω1z0+ω2z f ).

[0121] This invention utilizes a multi-satellite positioning system in conjunction with an underwater positioning base station for location services. In areas with good satellite signal coverage, the multi-satellite positioning system can provide relatively accurate planar position information. When complex water conditions cause a decrease in satellite positioning accuracy, the underwater positioning base station plays a crucial role. Because sound waves propagate relatively stably in water and are less susceptible to interference from factors above the water surface, it can accurately determine the distance and orientation between the wearable device and the base station, thereby correcting and supplementing the satellite positioning information, improving positioning accuracy, and ultimately enhancing the safety and emergency rescue efficiency for personnel on the water.

[0122] In one embodiment, steps 301 to 304 are described as follows:

[0123] Step 301: Filter the multi-dimensional data to obtain filtered multi-dimensional data, and extract risk assessment data features related to drowning risk from the filtered multi-dimensional data.

[0124] Optionally, in the multi-dimensional data of this embodiment of the invention, the following data are included: heart rate data HR(t) (unit: beats / minute), body temperature data BT(t) (unit: degrees Celsius), blood pressure data (systolic blood pressure SBP(t) and diastolic blood pressure DBP(t), unit: mmHg), and water attitude data (pitch angle θ(t), roll angle). And yaw angle ψ(t), unit: degrees), limb movement data (A x (t),A y (t),A z (t), unit: m / s 2 The data includes the pressure of water on the human body, P(t) (unit: Pascal), where t represents the time of data collection.

[0125] Furthermore, the wearable device filters the collected multi-dimensional data to remove noise interference and obtain filtered multi-dimensional data.

[0126] For heart rate data, since PPG signals are susceptible to noise interference such as motion artifacts, this embodiment of the invention employs an adaptive filtering algorithm, such as an adaptive filter based on the least mean square (LMS) algorithm, to dynamically adjust the filter coefficients according to the error between the input signal and the desired signal (noise-free heart rate signal). In one embodiment, the input raw heart rate data sequence is HR. raw(t), the reference signal (a signal related to heart rate collected by other relatively stable sensors or a model signal constructed according to historical heart rate data) is HR ref (t), the coefficient vector of the filter is ω(t), and the output of the filter is HR filtered (t) = ω T (t)HR raw (t), the error e(t) = HR ref (t) - HR filtered (t), by constantly updating the coefficient ω(t+1) = ω(t) + μe(t)HR raw (t), so that the filter gradually adapts to the signal characteristics, removes noise interference, and obtains more accurate heart rate data HR(t), wherein μ is a step factor for controlling the convergence speed.

[0127] For body temperature data, since its change is relatively slow and the collection environment is relatively stable, the embodiment of the present application adopts simple mean filtering. In an embodiment, the collected original body temperature data sequence is BT raw (t), the data in a certain time window (such as the past 10 sampling points) is averaged to obtain the filtered body temperature data Wherein, N is the window size.

[0128] For blood pressure data, since the measurement process is relatively stable and the noise influence is small, the embodiment of the present application adopts median filtering. The original blood pressure data sequence SBP raw (t) and DBP raw (t) is replaced by the median value of the data in the time window (such as the past 5 sampling points), to obtain the filtered systolic pressure SBP(t) and diastolic pressure DBP(t).

[0129] For posture data, Kalman filtering algorithm is used to fuse and filter the posture data. In an embodiment, the state vector of the posture angle measured by the IMU is Wherein, are the angular velocities of the pitch angle, the roll angle and the yaw angle, respectively, and the measurement vector is Wherein, θ m , ψ m is the posture angle directly measured by the IMU.

[0130] The state transition matrix F of the system is constructed based on a human kinematics model and characteristics of the sensor, and describes a transition rule of the state vector from one time to the next time without external measurement input, i.e., a change relationship of the attitude angle and the angular velocity in adjacent time intervals based on human motion and physical characteristics of the sensor itself, such as considering natural rotation, translation and other motion modes of the human body in water and response characteristics of components of the IMU, and the like. The measurement matrix H is determined according to a correlation between the attitude angle measured by the IMU and each element in the state vector, and reflects a relationship between the measurement vector and the state vector. Uncertain changes of the state vector caused by various unpredictable factors (such as irregular flow disturbance of the human body in water, slight performance fluctuations of the IMU itself, and the like) in the state transition process are analyzed, and the process noise covariance matrix Q is constructed. The measurement noise covariance matrix R is constructed based on statistical characteristics of noise contained in each measurement value in the measurement vector Z, and reflects a size and correlation of measurement errors caused by factors such as sensor accuracy limitation, environmental interference and the like in the measurement process.

[0131] Therefore, the prediction step of the Kalman filter is:

[0132]

[0133] wherein. represents a state estimation value at the last time based on the state vector prior estimation value at the current time obtained by prediction, i.e., a prediction of the attitude angle and the angular velocity at the current time according to the state transition rule before combining the measurement data at the current time; represents a corresponding prior estimation error covariance matrix.

[0134] The update step is:

[0135]

[0136]

[0137] wherein, K k represents a Kalman gain; represents a state vector estimation value at the current time obtained by updating after combining the measurement data Z k at the current time, which is a more accurate estimation result of the real attitude state after fusion and optimization of the prior estimation value and the measurement data; P k represents a corresponding updated estimation error covariance matrix, and I represents a unit matrix.

[0138] The data of the accelerometer, the gyroscope and the magnetometer are fused by Kalman filtering to improve the accuracy and stability of the attitude data, and filtered attitude data θ(t), and ψ(t) are obtained.

[0139] For the limb motion data, since the limb motion data has more high-frequency noise, the embodiment of the present application adopts a Butterworth low-pass filter. According to the frequency characteristics of the limb motion, in an embodiment, the cutoff frequency is, for example, 10 Hz, and the noise components higher than this frequency are filtered out. The original limb motion acceleration data sequence is A xraw (t), A yraw (t), A zraw (t), and the filtered data is A x (t), A y (t), A z (t).

[0140] For the water pressure data, the embodiment of the present application adopts a weighted moving average filter. Since the importance of the pressure sensors at different positions to the overall determination of drowning may be different, different weights ω i are assigned to the pressure sensors at different positions according to experience or experiment. In an embodiment, the original water pressure data sequence is P iraw (t), and i represents the sensors at different positions, and the filtered water pressure data is:

[0141]

[0142] wherein G is the number of pressure sensors.

[0143] Further, the wearable device extracts risk assessment data features related to the drowning risk from the filtered multi-dimensional data, wherein the risk assessment data features include a heart rate change rate, a body temperature change rate, a blood pressure change rate, an attitude change feature, a limb motion activity and a water pressure change feature.

[0144] wherein the heart rate change rate is ΔHR(t)=[HR(t)-HR(t-Δt)] / Δt, the body temperature change rate is ΔBT(t)=[BT(t)-BT(t-Δt)] / Δt, the blood pressure change rate is ΔSBP(t)=[SBP(t)-SBP(t-Δt)] / Δt and ΔDBP(t)=[DBP(t)-DBP(t-Δt)] / Δt, the attitude change feature is Δθ(t)=θ(t)-θ(t-Δt), Δψ(t)=ψ(t)-ψ(t-Δt).

[0145] The limb motion activity is the limb motion energy ​Calculate the limb motion energy change rate ΔE(t) = [E(t) - E(t - Δt)] / Δt.

[0146] Calculate the water pressure change feature: water pressure change rate ΔP(t) = [P(t) - P(t - Δt)] / Δt; water pressure distribution feature, divide the human body surface into multiple regions (head, chest, abdomen, limbs, etc.), measure the pressure value of each region by multiple pressure sensors, and the pressure value of the i-th region is P i (t), calculate the proportion of the pressure of each region to the total pressure Where n is the number of regions.

[0147] Step 302, input the risk assessment data features into the pre-constructed risk assessment model to obtain the risk assessment probability value output by the risk assessment model.

[0148] Optionally, the wearable device of the embodiment of the present application is built-in with a pre-trained risk assessment model, and the risk assessment model is trained based on sample data features and corresponding risk assessment label results, and the specific suspension process is described in steps 50 to 70. Therefore, the wearable device inputs the risk assessment data features into the pre-constructed risk assessment model to obtain the risk assessment result output by the risk assessment model, and the risk assessment result in the embodiment of the present application is a risk assessment probability value, wherein the value range of the risk assessment probability value is [0, 1], 0 represents no drowning risk, and 1 represents high drowning risk.

[0149] Step 303, if the risk assessment probability value is greater than or equal to the preset probability threshold, it is determined that the water personnel is in a drowning risk state; step 304, if the risk assessment probability value is less than the preset probability threshold, it is determined that the water personnel is in no drowning risk.

[0150] Further, the wearable device compares the risk assessment probability value with the preset probability threshold in value size to obtain a comparison result, wherein the preset probability threshold is set according to actual conditions, such as 0.55, 0.58, etc. If it is determined that the comparison result is that the risk assessment probability value is greater than or equal to the preset probability threshold, the wearable device determines that the water personnel is in a drowning risk state. If it is determined that the comparison result is that the risk assessment probability value is less than the preset probability threshold, the wearable device determines that the water personnel is in no drowning risk.

[0151] According to the embodiment of the present application, the comprehensive physiological and motion state data of the water personnel combined with the risk assessment model can quickly and accurately predict the risk assessment probability value, so as to quickly and accurately determine whether the water personnel is in a drowning risk state according to the risk assessment probability value, thereby significantly improving the safety guarantee level and emergency rescue efficiency of the water personnel.

[0152] In an embodiment, the training constraint condition is that the loss function value of each sample data feature is less than or equal to a preset loss threshold value, and the loss difference value of adjacent two loss function values is less than a preset difference threshold value, and the loss difference value presents a decreasing trend, and steps 50 to 70 are described as follows:

[0153] In step 50, each sample data feature is input into the initial evaluation model that is initially trained to obtain a risk evaluation prediction result of each sample data feature output by the initial evaluation model.

[0154] Optionally, the number M of input layer nodes of the initial evaluation model in the embodiment of the present application is equal to the feature dimension of the sample data feature. model The sample data feature in the embodiment of the present application includes heart rate, body temperature, blood pressure, posture, limb action and water pressure, and therefore, the number M of input layer nodes of the initial evaluation model is equal to 6. model The number of hidden layer nodes is equal to 64. Wherein, l model is the number of output layer nodes, represents a rounding up function, l model = 1 in the embodiment of the present application, a model is an adjustment constant between 1 and 10, for example, a model = 5, the number of output layer nodes is 1, and the risk evaluation probability value is taken as an example, which is in a range of 0 to 1, 0 represents no drowning risk, and 1 represents high drowning risk.

[0155] Therefore, the wearable device inputs each sample data feature into the initial evaluation model that is initially trained to obtain a risk evaluation prediction result of each sample data feature output by the initial evaluation model.

[0156] In step 60, the loss function value of each sample data feature is calculated based on the loss function of the initial evaluation model and the risk evaluation label result and the risk evaluation prediction result of each sample data feature.

[0157] Further, the wearable device calculates the loss function value of each sample data feature according to the loss function of the initial evaluation model and the risk evaluation label result and the risk evaluation prediction result of each sample data feature, and the loss function can be expressed as:

[0158] Wherein, Loss i represents the loss function value of the i th sample data feature, represents the risk evaluation label result of the i th sample data feature, represents the risk evaluation prediction result of the i th sample data feature, and δ x represents a preset weight coefficient.

[0159] Step 70, if it is determined that the loss function value of each sample data feature does not satisfy the training constraint condition, adjusting the model parameters of the initial evaluation model based on the risk assessment label result and the risk assessment prediction result of each sample data feature until the loss function value of each sample data feature output by the adjusted evaluation model satisfies the training constraint condition, and obtaining the risk assessment model.

[0160] Further, it is determined whether the loss function value of each sample data feature satisfies the training constraint condition. If it is determined that the loss function value of each sample data feature does not satisfy the training constraint condition, that is, at least one loss function value of the sample data feature is greater than the preset loss threshold, or / and, at least one loss difference value between the loss function value of the adjacent ith sample data feature and the loss function value of the (i+1)th sample data feature is greater than or equal to the preset difference threshold, or / and, at least one loss difference value does not show a decreasing trend, that is, the loss difference value between the loss function value of the (i+1)th sample data feature and the loss function value of the (i+2)th sample data feature is greater than or equal to the loss difference value between the loss function value of the ith sample data feature and the loss function value of the (i+1)th sample data feature, then an adjustment coefficient is calculated according to the risk assessment label result and the risk assessment prediction result of each sample data feature, wherein the calculation formula of the adjustment coefficient is as follows:

[0161]

[0162] wherein A model represents the adjustment coefficient, N samlpe represents the number of sample data features, σ(y ture ) represents the standard deviation of the risk assessment label result, σ(y pre ) represents the standard deviation of the risk assessment prediction result, represents the mean value of the risk assessment label result, represents the mean value of the risk assessment prediction result, min[·] represents the minimum value function, and max[·] represents the maximum value function.

[0163] Further, the model parameters of the initial evaluation model are adjusted by the adjustment coefficient, and the present embodiment can adjust the weight matrix and the bias vector of the activation function in the hidden layer in the initial evaluation model. For example, each weight coefficient in the weight matrix of the kth hidden layer is adjusted to each bias term in the bias vector is adjusted to

[0164] Further, each sample data feature is input into the adjusted evaluation model to obtain a loss function value of each sample data feature output by the adjusted evaluation model until the loss function value of each sample data feature meets the training constraint condition, that is, the loss function values of all sample data features are less than or equal to a preset loss threshold, and the loss difference between the loss function value of each adjacent ith sample data feature and the loss function value of an (i+1)th sample data feature is less than a preset difference threshold, and all loss differences show a decreasing trend, that is, the loss difference between the loss function value of the (i+1)th sample data feature and the loss function value of an (i+2)th sample data feature is less than the loss difference between the loss function value of the ith sample data feature and the loss function value of the (i+1)th sample data feature, to obtain the risk evaluation model.

[0165] The embodiment of the present application trains the risk evaluation model, so that subsequent risk evaluation probability values can be quickly and accurately predicted according to comprehensive physiological and motion state data of the water personnel combined with the risk evaluation model, so that whether the water personnel is in a drowning risk state can be quickly and accurately determined according to the risk evaluation probability values, and the safety guarantee level and emergency rescue efficiency of the water personnel are significantly improved.

[0166] Referring to Figure 3 , Figure 3 An embodiment of an electronic device provided by the embodiment of the present application is shown in the figure. As shown in the figure, Figure 3 the embodiment of the present application provides an electronic device 300, which includes a memory 310, a processor 320, and a computer program 311 stored in the memory 310 and executable on the processor 320, and the processor 320 implements the following steps when executing the computer program 311:

[0167] real-time acquisition of satellite positioning signals of a multi-satellite positioning system, multi-dimensional data of the water personnel, and reception of acoustic signals sent by a plurality of underwater positioning base stations;

[0168] determination of high-precision positioning information of the water personnel in the current water area based on the satellite positioning signals of each satellite positioning system and the acoustic signals of each underwater positioning base station;

[0169] determination of whether the water personnel is in a drowning risk state based on the multi-dimensional data;

[0170] if the water personnel is in a drowning risk state, generating a rescue alarm signal based on the high-precision positioning information, and sending the rescue alarm signal to a user terminal based on a long-distance communication unit.

[0171] Referring to Figure 4 , Figure 4 An embodiment of a computer-readable storage medium provided by the embodiment of the present application is shown in the figure. As shown in the figure, Figure 4As shown, the embodiment provides a computer readable storage medium 400, which stores a computer program 311, and the computer program 311 is executed by a processor to implement the following steps:

[0172] Real-time acquisition of satellite positioning signals of a multi-satellite positioning system, multi-dimensional data of the water personnel, and reception of acoustic wave signals sent by a plurality of underwater positioning base stations;

[0173] Based on the satellite positioning signals of each satellite positioning system and the acoustic wave signals of each underwater positioning base station, high-precision positioning information of the water personnel in the current water area is determined;

[0174] Based on the multi-dimensional data, it is determined whether the water personnel is in a drowning risk state;

[0175] If the water personnel is in a drowning risk state, a rescue alarm signal is generated based on the high-precision positioning information, and the rescue alarm signal is sent to a user terminal based on a long-distance communication unit.

[0176] On the other hand, the present application also provides a computer program product, which comprises a computer program, the computer program can be stored on a non-transitory computer readable storage medium, and the computer program is executed by a processor, and the computer can execute the water personnel positioning method provided by the above-mentioned methods, and the water personnel positioning method comprises:

[0177] Real-time acquisition of satellite positioning signals of a multi-satellite positioning system, multi-dimensional data of the water personnel, and reception of acoustic wave signals sent by a plurality of underwater positioning base stations;

[0178] Based on the satellite positioning signals of each satellite positioning system and the acoustic wave signals of each underwater positioning base station, high-precision positioning information of the water personnel in the current water area is determined;

[0179] Based on the multi-dimensional data, it is determined whether the water personnel is in a drowning risk state;

[0180] If the water personnel is in a drowning risk state, a rescue alarm signal is generated based on the high-precision positioning information, and the rescue alarm signal is sent to a user terminal based on a long-distance communication unit.

[0181] The device embodiments described above are only schematic, wherein the units shown as separate components can or can not be physically separate, and the components shown as units can or can not be physical units, i.e., they can be located in one place, or distributed on multiple network units. Part or all of the units can be selected according to actual needs to achieve the purpose of the embodiment scheme. Those skilled in the art can understand and implement without creative labor.

[0182] Those skilled in the art can clearly understand the technical solutions of the various embodiments from the above description of the embodiments, and the various embodiments can be implemented by means of software with the necessary general hardware platforms, and of course, can also be implemented by hardware. Based on such understanding, the above technical solutions, essentially or in other words, the part of the prior art that makes a contribution, can be embodied in the form of a software product, which can be stored in a computer readable storage medium, such as a ROM / RAM, a magnetic disk, an optical disk, and the like, and includes a number of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0183] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, rather than limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: it can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement for some technical features therein; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. An intelligent wearable device for a person on water, characterized by, The device comprises a master control unit, a positioning unit, a multi-dimensional monitoring unit, a long-distance communication unit, a high-energy-density battery unit and an alarm unit; the master control unit is connected with the positioning unit, the multi-dimensional monitoring unit, the long-distance communication unit, the high-energy-density battery unit and the alarm unit respectively, and is used for receiving and processing data of each unit; The positioning unit adopts a multi-satellite positioning system and cooperates with an underwater positioning base station to position, the multi-satellite positioning system comprises a GPS, a Beidou and a GLONASS satellite positioning system, and the underwater positioning base station is arranged on the bottom of a normal planning water area and communicates with the intelligent wearable device of the water personnel through a sound wave signal to obtain high-precision positioning information of the water personnel in the current water area; The multi-dimensional monitoring unit comprises a plurality of sensors and is used for monitoring multi-dimensional data of the water personnel in real time and transmitting the multi-dimensional data to the master control unit for comprehensive analysis to determine whether the water personnel is in a drowning risk state; The multi-dimensional data comprises heart rate information, body temperature information, blood pressure information, water posture information, limb movement information and water pressure information on the human body; The long-distance communication unit adopts satellite communication and 5G communication, if the signal strength of the satellite communication in the current water area is greater than that of the 5G communication, the satellite communication is used to communicate with the user terminal; if the signal strength of the satellite communication in the current water area is less than that of the 5G communication or the distance between the current water area and the shore is less than or equal to a preset distance, the 5G communication is used to communicate with the user terminal; The high-energy-density battery unit adopts a new type of lithium-sulfur battery, and the master control unit dynamically allocates power according to the operating state of each unit; The alarm unit comprises a plurality of alarm signal generation modules and is used for generating a plurality of rescue alarm signals, and the master control unit sends a rescue alarm signal when determining that the water personnel is in a drowning risk state; The method comprises the following steps: Filtering the multi-dimensional data to obtain filtered multi-dimensional data, and extracting risk assessment data features related to the drowning risk from the filtered multi-dimensional data; Inputting the risk assessment data features into a pre-constructed risk assessment model to obtain a risk assessment probability value output by the risk assessment model; the risk assessment model is trained based on sample data features and corresponding risk assessment label results; If the risk assessment probability value is greater than or equal to a preset probability threshold, it is determined that the water personnel is in a drowning risk state; If the risk assessment probability value is less than the preset probability threshold, it is determined that the water personnel is not in a drowning risk; The training constraint condition is that the loss function value of each sample data feature is less than or equal to a preset loss threshold, the loss difference value of adjacent two loss function values is less than a preset difference threshold, and the loss difference value presents a decreasing trend; the training process of the risk assessment model is as follows: Inputting each sample data feature into an initially trained initial evaluation model to obtain a risk assessment prediction result of each sample data feature output by the initial evaluation model; The loss function value of each sample data feature is calculated based on the loss function of the initial evaluation model in combination with the risk assessment label result and the risk assessment prediction result of each sample data feature; If it is determined that the training constraint condition is not met based on the loss function value of each sample data feature, the model parameters of the initial evaluation model are adjusted based on the risk assessment label result and the risk assessment prediction result of each sample data feature until the loss function value of each sample data feature output by the adjusted evaluation model meets the training constraint condition, and the risk assessment model is obtained; The loss function is represented as: wherein, represents the loss function value of the i-th sample data feature, represents the risk assessment label result of the i-th sample data feature, represents the risk assessment prediction result of the i-th sample data feature, represents a preset weight coefficient;​​​ The model parameters of the initial evaluation model are adjusted by an adjustment coefficient, and the calculation formula of the adjustment coefficient is as follows: ； wherein, denotes an adjustment coefficient, denotes a number of sample data features, denotes a standard deviation of risk assessment label results, denotes a standard deviation of risk assessment prediction results, denotes a mean of risk assessment label results, denotes a mean of risk assessment prediction results, denotes a minimum function, denotes a maximum function.

2. A method for locating a person on water, implemented based on the intelligent wearable device for a person on water according to claim 1, characterized in that, The water personnel positioning method comprises: Real-time acquisition of satellite positioning signals of a plurality of satellite positioning systems, multi-dimensional data of the water personnel, and sound wave signals transmitted by a plurality of underwater positioning base stations; Based on the satellite positioning signals of each satellite positioning system and the sound wave signals of each underwater positioning base station, high-precision positioning information of the water personnel in the current water area is determined; Based on the multi-dimensional data, it is determined whether the water personnel is in a drowning risk state; If the water personnel is in a drowning risk state, a rescue alarm signal is generated based on the high-precision positioning information, and the rescue alarm signal is transmitted to a user terminal based on a long-distance communication unit; The multi-dimensional data includes heart rate information, body temperature information, blood pressure information, water posture information, limb movement information, and water pressure information on the human body; Based on the multi-dimensional data, it is determined whether the water personnel is in a drowning risk state, comprising: The multi-dimensional data is filtered to obtain filtered multi-dimensional data, and risk assessment data features related to the drowning risk are extracted from the filtered multi-dimensional data; The risk assessment data features are input into a pre-constructed risk assessment model to obtain a risk assessment probability value output by the risk assessment model; the risk assessment model is trained based on sample data features and their corresponding risk assessment label results; If the risk assessment probability value is greater than or equal to a preset probability threshold, it is determined that the water personnel is in a drowning risk state; If the risk assessment probability value is less than the preset probability threshold, it is determined that the water personnel is not in a drowning risk; The training constraint condition is that the loss function value of each sample data feature is less than or equal to a preset loss threshold, the loss difference value of adjacent two loss function values is less than a preset difference threshold, and the loss difference value shows a decreasing trend; the training process of the risk assessment model is as follows: Each sample data feature is input into an initial evaluation model that has been initially trained to obtain a risk assessment prediction result of each sample data feature output by the initial evaluation model; The loss function value of each sample data feature is calculated based on the loss function of the initial evaluation model in combination with the risk assessment label result and the risk assessment prediction result of each sample data feature; If the loss function value of each sample data feature does not satisfy the training constraint condition, the model parameters of the initial evaluation model are adjusted based on the risk assessment label result and the risk assessment prediction result of each sample data feature until the loss function value of each sample data feature output by the adjusted evaluation model satisfies the training constraint condition, and the risk assessment model is obtained. The loss function is represented as: wherein, a loss function value of a first sample data feature, a loss function value of an i-th sample data feature, a risk assessment label result of an i-th sample data feature, a risk assessment prediction result of an i-th sample data feature, a risk assessment prediction result of an i-th sample data feature, a risk assessment prediction result of an i-th sample data feature, a preset weight coefficient; The model parameters of the initial evaluation model are adjusted by adjusting the adjustment coefficient, and the calculation formula of the adjustment coefficient is as follows: ； wherein, denotes an adjustment coefficient, denotes a number of sample data features, denotes a standard deviation of risk assessment label results, denotes a standard deviation of risk assessment prediction results, denotes a mean of risk assessment label results, denotes a mean of risk assessment prediction results, denotes a minimum function, denotes a maximum function.

3. The method of claim 2, wherein, The satellite positioning signal includes ephemeris data, pseudo-range information and Doppler shift data; The high-precision positioning information of the water personnel in the current water area is determined based on the satellite positioning signal of each satellite positioning system and the acoustic signal of each underwater positioning base station, including: Based on the ephemeris data and the Doppler shift data of each satellite positioning system, the satellite clock bias and the relative speed variation of each satellite positioning system are calculated respectively; Based on the coordinates, pseudo-range information, satellite clock bias and relative speed variation of each satellite positioning system, a first objective function for solving the positioning information of the water personnel and the clock bias of the intelligent wearable device of the water personnel is constructed; The first objective function is solved to obtain the initial positioning information of the water personnel and the clock bias correction number of the intelligent wearable device of the water personnel; Based on the clock bias correction number and the acoustic signal of each underwater positioning base station, the initial positioning information is corrected to obtain the high-precision positioning information of the water personnel in the current water area.

4. The method of claim 3, wherein, The high-precision positioning information of the water personnel in the current water area is obtained by correcting the initial positioning information based on the clock bias correction number and the acoustic signal of each underwater positioning base station, including: Based on the signal sending time in the acoustic signal of each underwater positioning base station, the signal receiving time when the intelligent wearable device of the water personnel receives the acoustic signal of each underwater positioning base station, and the water temperature of the current water area at each signal receiving time, the distance between the intelligent wearable device of the water personnel and each underwater positioning base station is determined; Based on the coordinates of each underwater positioning base station and the distance between the intelligent wearable device of the water personnel and each underwater positioning base station, a second objective function for solving the positioning information of the water personnel is constructed; The second objective function is solved to obtain the positioning correction amount; Based on the positioning correction amount, the initial positioning information is corrected to obtain the final positioning information of the water personnel; The initial positioning information and the final positioning information are fused based on the clock bias correction number to obtain the high-precision positioning information.

5. The method of claim 4, wherein, The initial positioning information and the final positioning information are fused based on the clock bias correction number to obtain the high-precision positioning information, including: Based on the clock bias correction number, the first weight coefficient of the initial positioning information and the second weight coefficient of the final positioning information are determined; Based on the initial positioning information, the corresponding first weight coefficient and the final positioning information, and the corresponding second weight coefficient, the high-precision positioning information is determined.

6. The method of claim 5, wherein, The first weight coefficient of the initial positioning information is determined based on the clock bias correction number, comprising: determining the base station signal strength and base station signal signal-to-noise ratio based on the transmission power of the acoustic wave signal of each underwater positioning base station, the gain of the intelligent wearable device of the water personnel, and the propagation attenuation of the acoustic wave in water; determining the satellite signal strength and satellite signal signal-to-noise ratio based on the clock bias correction number; determining the first weight coefficient based on the base station signal strength, the base station signal signal-to-noise ratio, the satellite signal strength and the satellite signal signal-to-noise ratio.

7. An electronic device comprising: The memory and the processor, characterized in that the memory has stored thereon a computer software program, and the processor reads and executes the computer software program to realize the water personnel positioning method according to any one of claims 2 to 6.

8. A non-transitory computer-readable storage medium, comprising: The storage medium has stored thereon a computer software program, and the computer software program is executed by the processor to realize the water personnel positioning method according to any one of claims 2 to 6.

Images