Anti-drowning wearable device control method

Abstract

The invention relates to the technical field of intelligent wearable equipment, and discloses an anti-drowning wearable equipment control method, which comprises the following steps: step 1, switching on a power supply of anti-drowning wearable equipment, electrifying a mainboard circuit in the equipment, supplying power to each basic circuit module, generating a stable clock signal, and sending the clock signal to a power supply module; the reset circuit performs reset operation on a key chip in the equipment at the moment of power-on; and 2, after a central processing unit of the equipment receives the clock signal and finishes resetting, reading a firmware program stored in the equipment, and then verifying the integrity of the program in the firmware program reading process. When the equipment is started, the multi-mode sensor system, the energy management system, the communication system and the positioning system are subjected to comprehensive and fine initialization detection, so that the operation state of each subsystem of the equipment is strictly controlled, and the effects of ensuring the normal functions of each part and ensuring the operation reliability of the equipment before the equipment is put into use are achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent wearable devices, and in particular to a method for controlling an anti-drowning wearable device. Background Art

[0002] As people engage in water activities more and more frequently, drowning accidents have become a serious threat to life safety. Traditional drowning prevention measures mainly rely on manual monitoring, simple life-saving equipment and basic warning signs, but they are flawed in timeliness and accuracy.

[0003] In the field of anti-drowning wearable devices, early devices had single functions. For example, they were equipped with simple water depth sensors, which could roughly determine the depth range of the user, but could not fully understand the key information of the user's posture and physical condition in the water. As a result, in many cases, when the user was in the early stage of drowning danger, such as abnormal posture due to physical exhaustion but had not yet sunk deep, the device could not detect it in time and issue an early warning.

[0004] At the same time, previous devices lacked consideration for individual differences among users. Users of different ages, genders, and physical conditions have significant differences in their performance in the water and risk of drowning. For example, children have relatively weak physical strength and swimming skills, are at higher risk of drowning, and their physical reactions when drowning are different from those of adults. Among adults, athletes with better physical fitness and the general public have very different endurance and ability to deal with danger in the water. However, traditional equipment uses a unified standard to judge drowning risk, which cannot achieve accurate personalized monitoring.

[0005] In addition, limitations in data processing restrict the performance improvement of anti-drowning wearable devices. The data collected by sensors are easily affected by factors such as the interference of the underwater environment and human motion artifacts, and the early devices lacked effective data preprocessing mechanisms, which reduced the accuracy and reliability of the data, and further affected the accuracy of the device's judgment of drowning situations, resulting in frequent false alarms or missed alarms.

[0006] Furthermore, the energy management and communication functions of the equipment are not perfect. In terms of energy supply, a single power supply method cannot meet the needs of long-term use in complex environments, and it is easy for the power to run out and the equipment to fail. In terms of communication, it supports limited communication modes. For example, when communication is blocked in an underwater environment, drowning information cannot be transmitted in time, delaying the rescue opportunity.

[0007] Therefore, those skilled in the art provide a method for controlling an anti-drowning wearable device to solve the above-mentioned problems. Summary of the invention

[0008] In view of the deficiencies in the prior art, the present invention provides a method for controlling an anti-drowning wearable device to solve the problems raised in the above background technology.

[0009] To achieve the above objectives, the present invention is implemented through the following technical solutions: a method for controlling a wearable anti-drowning device, comprising:

[0010] Step 1: Turn on the power of the anti-drowning wearable device, power the mainboard circuit inside the device, power each basic circuit module, generate a stable clock signal, and the reset circuit resets the key chip in the device at the moment of power-on;

[0011] Step 2: After receiving the clock signal and completing the reset, the central processor of the device reads the internally stored firmware program, and then verifies the integrity of the program during the reading of the firmware program. If the verification is correct, the program is loaded into the memory. If an error is found during the verification, the error handling mechanism is triggered, and an error message is displayed on the display screen through a specific error code, and the device is stopped from starting;

[0012] Step 3: After the CPU completes its own initialization, it sends a hardware self-test command to the multimodal sensor system, energy management system, communication system, and positioning system. After receiving the command, their internal control circuits start the self-test program.

[0013] Step 4: After each functional system completes the hardware self-test, the self-test result is fed back to the central processor through the data bus. The central processor evaluates the self-test result of each system according to the preset normal range and fault judgment standard. If the system self-tests normally, it proceeds to the next step. If there is a system or component failure, the central processor controls the warning circuit of the device to make the device inform the user through the preset indicator light flashing and sound alarm mode, so as to record the fault information in the local storage unit, which is convenient for subsequent maintenance personnel to troubleshoot and repair the fault;

[0014] Step 5: If the device completes the above hardware self-check and no abnormal conditions are found, the central processor sends an activation command to the display and interaction unit. After receiving the command, the display screen lights up and displays the user information entry interface, the operation buttons and the touch operation interface enter an operable state, and receive the personal information entered by the user;

[0015] Step 6: The user inputs height information by means of the operation button and the touch operation interface. After the height information is input, it is converted into a digital signal by the input processing circuit of the display and interaction unit, and transmitted to the storage unit of the device and marked as the height data storage area. Then the weight data is input and stored in the weight data storage area after processing. Then the age information is input and stored in the age data storage area. Finally, the gender information is input and stored in the gender data storage area.

[0016] Step 7: After successfully entering the user information, the central processor sends a signal to start data collection to each sensor in the multimodal sensor system according to the preset program instructions. At the same time, the central processor activates the sensor data transmission control circuit, which is responsible for coordinating the data transmission channels between each sensor and the central processor to ensure that the data is accurately and quickly transmitted to the central processor;

[0017] Step 8: Each sensor collects corresponding data according to the instructions, and then transmits it to the central processor after processing through the internal circuit to form a multi-modal data set, providing a comprehensive basic data source for subsequent analysis;

[0018] Step 9: After the multimodal sensor system successfully completes the data collection task, the central processing unit immediately starts the data preprocessing program. First, the central processing unit reads the filter algorithm configuration parameters for each sensor data in the storage unit to prepare for the subsequent filter calculation;

[0019] Step 10: For the raw data collected by each sensor, perform filtering processing operations based on the read filtering algorithm configuration parameters. For the data collected by the accelerometer and gyroscope, use a specific Kalman filtering algorithm for fusion processing. After filtering and calibration operations, provide a reliable data foundation for subsequent in-depth and accurate data analysis.

[0020] Preferably, in step 3, the accelerometer self-check of the multimodal sensor system includes checking the measurement range. Assume that the design measurement range of the accelerometer is A max and A min ,

[0021] Among them, A max is the maximum measured value in the positive direction, A min is the maximum measured value in the negative direction,

[0022] During the self-test, apply the standard test acceleration A test , if A min test max If the accelerometer can accurately output the corresponding signal, the measurement range check passes, otherwise it is judged as a fault.

[0023] Preferably, in step six, after receiving the user's age information, the display and interaction unit divides the age into children (0-12 years old), teenagers (13-19 years old), adults (20-59 years old) and the elderly (60 years old and above) according to preset age classification rules, and marks and stores them so as to adopt different age-related parameters in drowning risk assessment.

[0024] Preferably, in step eight, when the pressure sensor collects water pressure data and converts it into water depth data, according to the formula: ​​

[0025]

[0026] Among them, h is the water depth, P is the water pressure, ρ is the density of water, and g is the acceleration due to gravity.

[0027] Preferably, in step nine, a moving average filtering algorithm is used in the filtering processing algorithm for the heart rate sensor data.

[0028] Assume that the heart rate data sequence collected by the heart rate sensor is HR i (i=1,2,…,n), filtered data HR filtered The calculation formula is:

[0029]

[0030] Among them, k is the current data point index, m is the moving average window length,

[0031] HR filtered Represents the filtered heart data, HR i Represents the i-th data point in the original heart rate data sequence.

[0032] Preferably, in step 10, when the Kalman filter algorithm processes the accelerometer and gyroscope data, the state vector x of the system is set to be [θ, ω, α] (where θ is the angle, ω is the angular velocity, and α is the angular acceleration), and the prediction equation of the Kalman filter is:

[0033]

[0034] Among them, F k is the state transfer matrix, B k is the control input matrix, u k is the control input vector, is the prior state estimate at time step k, x k-1 is the posterior state estimate at time step k-1;

[0035]

[0036] in, is the a priori estimated covariance at time step k,

[0037] P k-1 The posterior estimated covariance at time step k-1 is,

[0038] Q k is the process noise covariance matrix, representing the kinetic noise not modeled in the model;

[0039] The update equation is:

[0040]

[0041] Among them, K k is the Kalman gain at time step k,

[0042] H k is the observation matrix, which maps the state space to the observation space,

[0043] R k is the observation noise covariance matrix, representing the noise in the observations,

[0044] x k is the posterior state estimate at time step k, z k is the observed value at time step k,

[0045] P k is the posterior estimated covariance at time step k, and I is the identity matrix.

[0046] Preferably, in the step 10, in the baseline correction operation of the skin galvanic response sensor data, the initially collected n skin conductivity data is set as G i (i=1, 2, ..., n), calculate the baseline value:

[0047]

[0048] The corrected data is:

[0049] G corrected =G i -G baseline ,

[0050] Among them, G corrected represents the corrected skin conductivity data, G i represents the original skin conductivity data point, G baseline Represents the baseline skin conductivity value.

[0051] Preferably, in step 3, when the lithium-ion battery of the energy management system self-checks, the current power Q battery , if Q battery Low preset minimum power threshold Q min , then the battery power failure is determined, where Q min Set according to the minimum power required for normal operation of the device.

[0052] Preferably, in step eight, the image data collected by the optical sensor is compressed before being transmitted to the central processor, using the JPEG compression algorithm, and the compression ratio is automatically adjusted according to the current communication bandwidth and storage capacity of the device. Assume that the amount of original image data is D original , the amount of compressed image data is D compressed , compression ratio:

[0053]

[0054] Among them, D original is the original image data, D compressed is the amount of compressed image data, and CR is the compression ratio.

[0055] Preferably, in step 4, if there is a fault, in addition to recording the fault information, the central processing unit attempts to restart the faulty component. If the self-test fails after the restart, it is determined to be a permanent hardware fault.

[0056] The present invention provides a control method for a wearable device for preventing drowning, which has the following beneficial effects:

[0057] 1. The present invention performs comprehensive and detailed initialization detection on the multimodal sensor system, energy management system, communication system and positioning system when the device is started, thereby strictly controlling the operating status of each subsystem of the device, thereby ensuring that each part of the device functions normally before being put into use and ensuring the reliability of the device operation.

[0058] 2. The present invention allows users to enter personal information such as height, weight, age and gender, and applies the information to a drowning risk assessment algorithm, thereby achieving a personalized assessment of drowning risks for different individual characteristics, thereby achieving an effect that is more in line with the user's actual situation and more accurately prevents drowning accidents.

[0059] 3. The present invention collects multi-dimensional data by utilizing a multimodal sensor system, including an accelerometer, a gyroscope, a pressure sensor, a heart rate sensor, a galvanic skin response sensor, and an optical sensor, and fuses and comprehensively analyzes the collected multi-dimensional data, thereby realizing a comprehensive multi-angle judgment of the user's true state in the water, and achieving a more timely and accurate effect of detecting signs of drowning.

[0060] 4. The present invention performs professional preprocessing on the collected raw data, adopts the Kalman filter algorithm to process the accelerometer and gyroscope data, and calibrates and normalizes other sensor data to remove data interference factors, ensure the reliability of subsequent data analysis, and improve the performance and accuracy of the equipment's anti-drowning monitoring. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] Figure 1 It is a flow chart of the present invention. DETAILED DESCRIPTION

[0062] In order to make the technical personnel in the technical field understand the scheme of the present invention, the technical scheme in the embodiment of the present invention will be clearly and completely described below in combination with the drawings in the embodiment of the present invention. Obviously, the described embodiment is a partial embodiment of the present invention, not a complete embodiment. Based on the embodiment of the present invention, other embodiments obtained by ordinary technicians in the field without creative work should fall within the scope of protection of the present invention.

[0063] The present invention is described in detail below in conjunction with the accompanying drawings:

[0064] Example:

[0065] Please refer to the attached Figure 1 , an embodiment of the present invention provides a method for controlling a wearable anti-drowning device, comprising:

[0066] Step 1: Turn on the power of the anti-drowning wearable device, power the mainboard circuit inside the device, power each basic circuit module, generate a stable clock signal, and the reset circuit resets the key chip in the device at the moment of power-on;

[0067] Step 2: After receiving the clock signal and completing the reset, the central processor of the device reads the internally stored firmware program, and then verifies the integrity of the program during the reading of the firmware program. If the verification is correct, the program is loaded into the memory. If an error is found during the verification, the error handling mechanism is triggered, and an error message is displayed on the display screen through a specific error code, and the device is stopped from starting;

[0068] Step 3: After the CPU completes its own initialization, it sends a hardware self-test command to the multimodal sensor system, energy management system, communication system, and positioning system. After receiving the command, their internal control circuits start the self-test program.

[0069] Step 4: After each functional system completes the hardware self-test, the self-test result is fed back to the central processor through the data bus. The central processor evaluates the self-test result of each system according to the preset normal range and fault judgment standard. If the system self-tests normally, it proceeds to the next step. If there is a system or component failure, the central processor controls the warning circuit of the device to make the device inform the user through the preset indicator light flashing and sound alarm mode, so as to record the fault information in the local storage unit, which is convenient for subsequent maintenance personnel to troubleshoot and repair the fault;

[0070] Step 5: If the device completes the above hardware self-check and no abnormal conditions are found, the central processor sends an activation command to the display and interaction unit. After receiving the command, the display screen lights up and displays the user information entry interface, the operation buttons and the touch operation interface enter an operable state, and receive the personal information entered by the user;

[0071] Step 6: The user inputs height information by means of the operation button and the touch operation interface. After the height information is input, it is converted into a digital signal by the input processing circuit of the display and interaction unit, and transmitted to the storage unit of the device and marked as the height data storage area. Then the weight data is input and stored in the weight data storage area after processing. Then the age information is input and stored in the age data storage area. Finally, the gender information is input and stored in the gender data storage area.

[0072] Step 7: After successfully entering the user information, the central processor sends a signal to start data collection to each sensor in the multimodal sensor system according to the preset program instructions. At the same time, the central processor activates the sensor data transmission control circuit, which is responsible for coordinating the data transmission channels between each sensor and the central processor to ensure that the data is accurately and quickly transmitted to the central processor;

[0073] Step 8: Each sensor collects corresponding data according to the instructions, and then transmits it to the central processor after processing through the internal circuit to form a multi-modal data set, providing a comprehensive basic data source for subsequent analysis;

[0074] Step 9: After the multimodal sensor system successfully completes the data collection task, the central processing unit immediately starts the data preprocessing program. First, the central processing unit reads the filter algorithm configuration parameters for each sensor data in the storage unit to prepare for the subsequent filter calculation;

[0075] Step 10: For the raw data collected by each sensor, perform filtering processing operations based on the read filtering algorithm configuration parameters. For the data collected by the accelerometer and gyroscope, use a specific Kalman filtering algorithm for fusion processing. After filtering and calibration operations, provide a reliable data foundation for subsequent in-depth and accurate data analysis.

[0076] Step 1 Benefits: Generate a stable clock signal to provide an accurate timing reference for the operation of each component of the device, ensure that each operation is performed synchronously, reset the key chip by the reset circuit, start the initial state, avoid system errors caused by abnormal state before the chip, ensure the stability and predictability of the device startup, and lay the foundation for subsequent normal operation;

[0077] Benefits of step 2: Reading and verifying the firmware program can ensure that the program is complete and undamaged, and prevent device operation failures caused by firmware errors. If the program is loaded into the memory correctly, it can ensure that the device operates as expected. If an error is found, it will be displayed and stopped in time. The user knows the cause of the device failure and avoids hardware damage or data errors that may be caused by erroneous programs.

[0078] Step 3 Benefits: Send self-check instructions to the multimodal sensor system, energy management system, communication system and positioning system to comprehensively check the key functional modules of the equipment. The self-check of the internal control circuit of each system can detect potential hardware problems in advance, prevent inaccurate monitoring data, communication interruption or positioning failure due to hardware failure in actual use, and improve the overall reliability of the equipment;

[0079] Step 4 Benefits: The CPU evaluates the self-test results according to the standards. If normal, the process continues to ensure that the equipment can enter the next normal working state. If there is a fault, the user is informed through the warning circuit and the information is recorded, which is convenient for maintenance personnel to quickly locate the problem, which helps to find equipment problems in time and repair them, reduce safety hazards caused by equipment failures, improve equipment maintenance efficiency, and reduce equipment use costs;

[0080] Step 5 Benefits: After the hardware is normal, the display and interaction unit is activated to prepare for user information entry, making the device operation process coherent, allowing users to smoothly enter personal information, providing a data basis for subsequent personalized drowning risk assessment, and improving the accuracy of the device's drowning risk judgment for different users;

[0081] Step 6 Benefits: User information is processed and stored in categories. Height, weight, age and gender information are used for a variety of personalized functions. For example, height affects the calibration of sensor measurement range, and age and gender affect drowning risk assessment algorithm parameters, so that the device can accurately monitor and evaluate according to individual user characteristics, better adapting to the drowning prevention needs of different user groups;

[0082] Step 7 Benefits: The central processor starts sensor data collection and activates the transmission control circuit to ensure the timeliness of data collection and the accuracy and efficiency of data transmission. Each sensor can respond quickly to start collecting data. The transmission control circuit ensures that multi-sensor data is transmitted to the central processor in an orderly and fast manner to avoid data loss or transmission errors, providing a complete and accurate data basis for subsequent analysis;

[0083] Step 8 Benefits: Each sensor collects corresponding data and processes and transmits it to form a multimodal data set. The multimodal data can reflect the user's state in the water in multiple dimensions. The comprehensive data can comprehensively and accurately analyze whether the user is in danger of drowning, thereby improving the accuracy and reliability of drowning monitoring;

[0084] Step 9 Benefits: The CPU reads the filter algorithm configuration parameters to prepare for filter calculations. Preset parameters based on device performance and sensor characteristics to ensure that the filter algorithm is optimized for different sensor data, improve the pertinence and effectiveness of data preprocessing, and make subsequent data processing more in line with the actual operating environment and data characteristics of the device;

[0085] Step 10 Benefits: Filtering the raw data according to the configuration parameters, especially using the Kalman filter algorithm to fuse the accelerometer and gyroscope, can effectively remove noise interference and improve the accuracy of attitude data. The data after filtering and calibration is more reliable, providing a guarantee for in-depth and accurate data analysis, and improving the accuracy of the equipment's drowning risk assessment and warning, effectively reducing the risk of false alarms and missed alarms.

[0086] Summary: The steps of the control method are closely linked, from the circuit basis and chip status check of the equipment startup initialization, to the functional system self-test and fault handling, to the user information entry and multimodal sensor data collection, transmission, and preprocessing, forming a complete and rigorous process. Each step has a specific role and benefit, which together ensure the reliability of equipment operation, the accuracy of data, the accuracy of drowning monitoring and the personalized adaptability to different users, providing solid technical support and process guarantee for anti-drowning wearable devices to effectively play their safety protection functions.

[0087] In step 3, the accelerometer self-check of the multimodal sensor system includes checking the measurement range. Assume that the design measurement range of the accelerometer is A max and A min ,

[0088] Among them, A max is the maximum measured value in the positive direction, A min is the maximum measured value in the negative direction,

[0089] During the self-test, apply the standard test acceleration A test , if A min test max If the accelerometer can accurately output the corresponding signal, the measurement range check passes, otherwise it is judged as a fault;

[0090] In step 3, the lithium-ion battery of the energy management system self-checks and detects the current power Q battery , if Q battery Low preset minimum power threshold Q min , then the battery power failure is determined, where Q min Set according to the minimum power required for normal operation of the device.

[0091] The accelerometer is an important part of the multimodal sensor system, working together with the gyroscope sensor to reflect the complete movement posture of the human body in water. If there is a problem with the measurement range, the posture information obtained by the multimodal sensor system will be biased, which will affect the device's ability to judge whether the user is in danger of drowning.

[0092] ​​Lithium-ion batteries are an important energy supply for equipment. Whether the battery is fully charged or not is directly related to whether the equipment can operate normally. During the self-test, the current battery level is detected and compared with the preset minimum battery level threshold. If Q battery <Q min If the battery power is insufficient to support the normal operation of the device, the system can know in advance whether the battery power is insufficient to support the normal operation of the device, avoid sudden power failure of the device due to power exhaustion during use, and enable functions such as sensor data collection, drowning risk monitoring, and alarm and communication functions in dangerous situations to continue to operate stably and continuously protect the safety of users.

[0093] Once a battery failure is determined, the user or relevant maintenance personnel will receive a prompt in time and take corresponding solutions, such as timely charging or replacing the battery. This will help improve the efficiency and reliability of the equipment and reduce the occurrence of equipment failure to work properly and drowning monitoring interruptions due to insufficient battery power. It ensures that the equipment is always in normal use and provides a stable energy basis for the realization of anti-drowning functions.

[0094] In summary, the self-inspection of the accelerometer measurement range and lithium-ion battery power in step three, the key aspects of sensor function assurance and energy supply stability lay the foundation for the reliable operation of anti-drowning wearable devices.

[0095] In step six, after receiving the user's age information, the display and interaction unit divides the age into children (0-12 years old), teenagers (13-19 years old), adults (20-59 years old) and the elderly (60 years old and above) according to the preset age classification rules, and marks and stores them to adopt different age-related parameters in drowning risk assessment.

[0096] In step eight, when the pressure sensor collects water pressure data and converts it into water depth data, according to the formula:

[0097]

[0098] Among them, h is the water depth, P is the water pressure, ρ is the density of water, and g is the acceleration due to gravity.

[0099] The storage by age classification in step six and the use of formulas to convert water pressure into water depth data in step eight bring significant advantages to the drowning risk assessment function of anti-drowning wearable devices from the perspective of user individual characteristics and environmental data utilization. Age classification storage makes risk assessment more in line with the characteristics of people of different age groups and realizes personalized and accurate monitoring. The conversion of water pressure into water depth data provides intuitive and practical key environmental information for risk assessment, and works with other data to help more accurately and comprehensively judge drowning risks. The two work together to further improve the effectiveness and reliability of the equipment in drowning prevention, allowing the equipment to better serve the safety and security needs of different users in various water environments.

[0100] In step nine, a moving average filtering algorithm is used in the filtering processing algorithm for the heart rate sensor data.

[0101] Assume that the heart rate data sequence collected by the heart rate sensor is HR i (i=1,2,…,n), filtered data HR filtered The calculation formula is:

[0102]

[0103] Among them, k is the current data point index, m is the moving average window length,

[0104] HR filtered Represents the filtered heart data, HR i Represents the i-th data point in the original heart rate data sequence.

[0105] In step 10, when the Kalman filter algorithm processes the accelerometer and gyroscope data, the state vector x of the system is set to [θ, ω, α] (where θ is the angle, ω is the angular velocity, and α is the angular acceleration). The prediction equation of the Kalman filter is:

[0106]

[0107] Among them, F k is the state transfer matrix, B k is the control input matrix, u k is the control input vector, is the prior state estimate at time step k, x k-1 is the posterior state estimate at time step k-1;

[0108]

[0109] in, is the a priori estimated covariance at time step k,

[0110] P k-1 The posterior estimated covariance at time step k-1 is,

[0111] Q k is the process noise covariance matrix, representing the kinetic noise not modeled in the model;

[0112] The update equation is:

[0113]

[0114] Among them, K k is the Kalman gain at time step k,

[0115] Hk is the observation matrix, which maps the state space to the observation space,

[0116] R k is the observation noise covariance matrix, representing the noise in the observations,

[0117] x k is the posterior state estimate at time step k, z k is the observed value at time step k,

[0118] P k is the posterior estimated covariance at time step k, and I is the identity matrix;

[0119] In step 10, in the baseline correction operation of the skin galvanic response sensor data, the initially collected skin conductivity data is set as G i (i=1, 2, ..., n), calculate the baseline value:

[0120]

[0121] The corrected data is:

[0122] G corrected =G i -G baseline ,

[0123] Among them, G corrected represents the corrected skin conductivity data, G i represents the original skin conductivity data point, G baseline Represents the baseline skin conductivity value.

[0124] The moving average filter algorithm in step nine processes the heart rate data, and the Kalman filter algorithm in step ten processes the baseline correction operation of the accelerometer and gyroscope data and the galvanic skin response sensor data. The angles of processing different sensor data provide key support for anti-drowning wearable devices.

[0125] Moving average filtering ensures that heart rate data is stable and reliable for analysis, Kalman filtering integrates multi-sensor data to accurately obtain human posture, and baseline correction of skin galvanic response sensor data eliminates interference and highlights drowning signals. The three work together to significantly improve the quality and accuracy of data collected by the device, enhance the device's ability to assess drowning risks based on multi-modal sensor data, and effectively improve the overall performance and reliability of anti-drowning wearable devices, enabling more accurate and timely monitoring and prevention of drowning accidents.

[0126] In step 8, the image data collected by the optical sensor is compressed before being transmitted to the central processor. The JPEG compression algorithm is used, and the compression ratio is automatically adjusted according to the current communication bandwidth and storage capacity of the device. Suppose the amount of original image data is Doriginal , the amount of compressed image data is D compressed , compression ratio:

[0127]

[0128] Among them, D original is the original image data, D compressed is the amount of compressed image data, and CR is the compression ratio.

[0129] In step 4, if there is a fault, in addition to recording the fault information, the central processing unit attempts to restart the faulty component. If the self-test fails after restarting, it is determined to be a permanent hardware fault.

[0130] The compression processing of image data in step 8 and the restart and determination operation of faulty components in step 4 bring many benefits to anti-drowning wearable devices through important aspects of data processing and equipment failure response. Image data compression optimizes communication transmission and saves storage resources, ensuring the efficiency of equipment information flow and the rationality of storage;

[0131] The relevant processing mechanism of faulty components improves the self-repair capability of the equipment and accurately locates permanent faults, which is conducive to improving the stability of the equipment and the efficiency of subsequent maintenance. The two work together to further improve the function and operation guarantee system of the equipment, so that the anti-drowning wearable device can play its role in anti-drowning monitoring and ensuring user safety more stably and efficiently.

[0132] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for controlling a wearable device to prevent drowning, characterized in that: include: Step 1: Turn on the power of the anti-drowning wearable device, power the mainboard circuit inside the device, power each basic circuit module, generate a stable clock signal, and the reset circuit resets the key chip in the device at the moment of power-on; Step 2: After receiving the clock signal and completing the reset, the central processor of the device reads the internally stored firmware program, and then verifies the integrity of the program during the reading of the firmware program. If the verification is correct, the program is loaded into the memory. If an error is found during the verification, the error handling mechanism is triggered, and an error message is displayed on the display screen through a specific error code, and the device is stopped from starting; Step 3: After the CPU completes its own initialization, it sends a hardware self-test command to the multimodal sensor system, energy management system, communication system, and positioning system. After receiving the command, their internal control circuits start the self-test program. Step 4: After each functional system completes the hardware self-test, the self-test result is fed back to the central processor through the data bus. The central processor evaluates the self-test result of each system according to the preset normal range and fault judgment standard. If the system self-tests normally, it proceeds to the next step. If there is a system or component failure, the central processor controls the warning circuit of the device to make the device inform the user through the preset indicator light flashing and sound alarm mode, so as to record the fault information in the local storage unit, which is convenient for subsequent maintenance personnel to troubleshoot and repair the fault; Step 5: If the device completes the above hardware self-check and no abnormal conditions are found, the central processor sends an activation command to the display and interaction unit. After receiving the command, the display screen lights up and displays the user information entry interface, the operation buttons and the touch operation interface enter an operable state, and receive the personal information entered by the user; Step 6: The user inputs height information by means of the operation button and the touch operation interface. After the height information is input, it is converted into a digital signal by the input processing circuit of the display and interaction unit, and transmitted to the storage unit of the device and marked as the height data storage area. Then the weight data is input and stored in the weight data storage area after processing. Then the age information is input and stored in the age data storage area. Finally, the gender information is input and stored in the gender data storage area. Step 7: After successfully entering the user information, the central processor sends a signal to start data collection to each sensor in the multimodal sensor system according to the preset program instructions. At the same time, the central processor activates the sensor data transmission control circuit, which is responsible for coordinating the data transmission channels between each sensor and the central processor to ensure that the data is accurately and quickly transmitted to the central processor; Step 8: Each sensor collects corresponding data according to the instructions, and then transmits it to the central processor after processing through the internal circuit to form a multi-modal data set, providing a comprehensive basic data source for subsequent analysis; Step 9: After the multimodal sensor system successfully completes the data collection task, the central processing unit immediately starts the data preprocessing program. First, the central processing unit reads the filter algorithm configuration parameters for each sensor data in the storage unit to prepare for the subsequent filter calculation; Step 10: For the raw data collected by each sensor, perform filtering processing operations based on the read filtering algorithm configuration parameters. For the data collected by the accelerometer and gyroscope, use a specific Kalman filtering algorithm for fusion processing. After filtering and calibration operations, provide a reliable data foundation for subsequent in-depth and accurate data analysis.

2. A method for controlling a wearable anti-drowning device according to claim 1, characterized in that: In step 3, the accelerometer self-check of the multimodal sensor system includes checking the measurement range. Assume that the design measurement range of the accelerometer is A max and A min , Among them, A max is the maximum measured value in the positive direction, A min is the maximum measured value in the negative direction, During the self-test, apply the standard test acceleration A test , if A min test max If the accelerometer can accurately output the corresponding signal, the measurement range check passes, otherwise it is judged as a fault.​​ 3. A method for controlling a wearable anti-drowning device according to claim 1, characterized in that: In step six, after receiving the user's age information, the display and interaction unit divides the age into children (0-12 years old), teenagers (13-19 years old), adults (20-59 years old) and the elderly (60 years old and above) according to the preset age classification rules, and marks and stores them to adopt different age-related parameters in drowning risk assessment.

4. The method for controlling a wearable device for preventing drowning according to claim 1, characterized in that: In step eight, when the pressure sensor collects water pressure data and converts it into water depth data, according to the formula: Among them, h is the water depth, P is the water pressure, ρ is the density of water, and g is the acceleration due to gravity.

5. The method for controlling a wearable anti-drowning device according to claim 1, characterized in that: In step nine, a moving average filtering algorithm is used in the filtering processing algorithm for the heart rate sensor data. Assume that the heart rate data sequence collected by the heart rate sensor is HR i (i=1,2,…,n), filtered data HR filtered The calculation formula is: Among them, k is the current data point index, m is the moving average window length, HR filtered Represents the filtered heart data, HR i Represents the i-th data point in the original heart rate data sequence.

6. The method for controlling a wearable device for preventing drowning according to claim 1, characterized in that: In step 10, when the Kalman filter algorithm processes the accelerometer and gyroscope data, the state vector x of the system is set to be [θ, ω, α] (where θ is the angle, ω is the angular velocity, and α is the angular acceleration), and the prediction equation of the Kalman filter is: Among them, F k is the state transfer matrix, B k is the control input matrix, u k is the control input vector, is the prior state estimate at time step k, x k-1 is the posterior state estimate at time step k-1; in, is the a priori estimated covariance at time step k, P k-1 The posterior estimated covariance at time step k-1 is, Q k is the process noise covariance matrix, representing the kinetic noise not modeled in the model; The update equation is: Among them, K k is the Kalman gain at time step k, H k is the observation matrix, which maps the state space to the observation space, R k is the observation noise covariance matrix, representing the noise in the observations, x k is the posterior state estimate at time step k, z k is the observed value at time step k, P k is the posterior estimated covariance at time step k, and I is the identity matrix.

7. The method for controlling a wearable anti-drowning device according to claim 1, characterized in that: In the step 10, in the baseline correction operation of the skin galvanic response sensor data, the initially collected n skin conductivity data is set as G i (i=1, 2, ..., n), calculate the baseline value: The corrected data is: G corrected =G i -G baseline , Among them, G corrected represents the corrected skin conductivity data, G i represents the original skin conductivity data point, G baseline Represents the baseline skin conductivity value.

8. The method for controlling a wearable anti-drowning device according to claim 1, characterized in that: In step 3, the lithium-ion battery of the energy management system self-checks and detects the current power Q battery , if Q battery Low preset minimum power threshold Q min , then the battery power failure is determined, where Q min Set according to the minimum power required for normal operation of the device.

9. The method for controlling a wearable device for preventing drowning according to claim 1, characterized in that: In step 8, the image data collected by the optical sensor is compressed before being transmitted to the central processor. The JPEG compression algorithm is used, and the compression ratio is automatically adjusted according to the current communication bandwidth and storage capacity of the device. Assume that the amount of original image data is D original , the amount of compressed image data is D compressed , compression ratio: Among them, D original is the original image data, D compressed is the amount of compressed image data, and CR is the compression ratio.

10. The method for controlling a wearable anti-drowning device according to claim 1, characterized in that: In the step 4, if there is a fault, in addition to recording the fault information, the central processing unit attempts to restart the faulty component. If the self-test fails after the restart, it is determined to be a permanent hardware fault.

Images