Physical sign detection early warning system based on firefighter uniform

Abstract

The invention discloses a firefighter uniform-based sign detection early warning system, which comprises a firefighter uniform main body, and a central processor, a physiological information monitoring module, an environment information monitoring module, a position estimation module and a communication module which are arranged on the firefighter uniform main body. When the early warning system measures heart rate and blood oxygen data for a long time, the measured data is stable and small in fluctuation, the Euler angle data of the human body can be obtained, the real-time movement posture of the human body can be obtained by analyzing the Euler angle data, the advancing displacement of the firefighter can be calculated through a quadratic integration method, and the real-time movement posture of the firefighter can be obtained. Therefore, the position information of the firefighter is estimated, and alarm identifiers corresponding to the data can be sent to an upper computer when the data exceed an alarm threshold value.

Description

Technical Field

[0001] The present invention relates to the technical field of fire fighting, and in particular to a vital sign detection and early warning system based on fire fighting clothing. Background Art

[0002] In recent years, various types of large buildings have been increasing in various parts of my country. At the same time, in order to meet various production activities, the internal structures of various large buildings have become increasingly complex. The huge and complex structure buildings not only bring more difficult rescue and firefighting environments to firefighters, but also pose a major threat to the personal safety of firefighters entering the fire scene, burying potential safety hazards. Due to the complex space, large site, and large number of personnel in large buildings, high requirements are placed on the rescue work of firefighters when a fire occurs. The main manifestation is that the internal passages and building structures of large buildings are complex, which will bring great rescue difficulties when a fire occurs. Firefighters need to quickly and accurately locate the source of the fire and take effective measures for rescue. At the same time, the on-site environment is extremely complex. There are high temperatures, smoke and other harmful substances at the fire scene, which cause certain damage to the body and respiratory system. Firefighters need to have outstanding physical fitness and strong adaptability. Some places also have the problem of incomplete firefighting facilities. Due to the wide range of use of large buildings, it is difficult to set up firefighting facilities in some places, which can easily lead to poor firefighting effects when a fire occurs. In addition, there are a large number of employees and residents in large buildings. When a fire occurs, effective methods need to be taken to evacuate personnel to avoid people being stranded at the scene.

[0003] In today's society, with the increase in large buildings and the diversification of production activities, firefighting and rescue work are facing unprecedented challenges. As a key equipment to ensure the safety of firefighters and improve rescue efficiency, the research status, application prospects and development trends of smart fire suits have attracted much attention. However, there are still some drawbacks and defects in the research and practice of smart fire suits, including the following:

[0004] (1) Insufficient authenticity verification. At present, the research on vital sign monitoring smart fire suits is mainly carried out in a simulation environment, lacking verification and testing in actual fire rescue scenes. Therefore, there may be insufficient authenticity verification in practical applications.

[0005] (2) The perception capability is not comprehensive enough. Although the smart fire suit is equipped with a variety of sensors and monitoring devices, its perception capability is relatively single, usually only equipped with vital sign monitoring function, ignoring the external environment and the firefighter's body posture and other issues.

[0006] (3) Data processing and algorithm optimization still need to be improved. Data processing and algorithm optimization of smart fire suits are important factors that determine their performance and effectiveness, and currently need to be further improved and enhanced.

[0007] (4) Limited high temperature and corrosion resistance. Due to the high temperature and strong corrosion at the fire scene, the materials and equipment of smart fire suits need to have sufficient high temperature and corrosion resistance, which is also a problem that needs to be further solved.

[0008] (5) Limited battery life. Smart fire suits require a lot of power, but the current battery life is limited and needs further improvement in technology.

[0009] (6) Poor experience. Since the vital sign monitoring smart fire suit needs to use some sensors and lines, the additional weight is about twice that of ordinary fire suits, which will be more uncomfortable to wear.

[0010] (7) Affected by environmental interference. Since the fire scene environment is complex and difficult to control, when firefighters wear smart fire suits for vital sign monitoring, the sensors may be greatly affected by environmental interference, thus affecting the accuracy of the monitoring data. Summary of the invention

[0011] The present invention provides a vital sign detection and early warning system based on fire-fighting clothing, aiming to solve all or part of the technical problems existing in the prior art.

[0012] The technical solution provided by the present invention is as follows:

[0013] A vital sign detection and early warning system based on a fire suit, comprising a fire suit body, and a central processor, a physiological information monitoring module, an environmental information monitoring module, a position estimation module, a display module and a communication module arranged on the fire suit body;

[0014] The physiological information monitoring module includes a heart rate and blood oxygen sensor, a posture sensor and a body temperature sensor. The heart rate and blood oxygen sensor is arranged on the inner side of the wrist of the fire suit body, and the posture sensor and the body temperature sensor are arranged on the lower side of the back of the fire suit body.

[0015] The environmental information monitoring module includes an environmental temperature sensor, which is arranged on the back outside of the fire suit body;

[0016] The position estimation module includes an inertial measurement unit, which is used to measure the posture and acceleration of the wearer and is arranged on the lower back side of the fire suit body;

[0017] The central processor is used to process the information collected by the physiological information monitoring module, the environmental information monitoring module and the position estimation module, and display the processed information on an external display screen;

[0018] The display module includes an LCD screen for displaying various data processed by the central processor;

[0019] The communication module is used to send the information collected by the physiological information monitoring module, the environmental information monitoring module and the position estimation module to the host computer, and send the alarm identifier corresponding to each data to the host computer when the data exceeds the alarm threshold.

[0020] Furthermore, the central processor selects STM32 series controller.

[0021] Furthermore, the heart rate and blood oxygen sensor is MAX30102.

[0022] Furthermore, the environmental information monitoring module uses MAX6675.

[0023] Furthermore, the position estimation module uses MPU6050, and the MPU6050 integrates a three-axis gyroscope and a three-axis accelerometer.

[0024] Furthermore, the MAX30102 is used to calculate the heart rate and blood oxygen saturation, including the following steps:

[0025] S1. Signal preprocessing

[0026] Collect the optical signal reflected by the MAX30102, convert it into a PPG signal, and perform filtering and denoising on the PPG signal;

[0027] S2. Get periodic window

[0028] Finding periodic peaks and troughs in the preprocessed PPG signal, and detecting the start and end positions of each heartbeat to obtain a periodic window;

[0029] S3. Calculate heart rate

[0030] Based on the periodic window, the heart rate is calculated;

[0031] S4. Calculate blood oxygen saturation

[0032] Based on the pre-processed PPG signal, the blood oxygen saturation is calculated by analyzing the relationship between light intensity and time.

[0033] Furthermore, estimating the position by the MPU6050 includes the following steps:

[0034] Initialize the MPU6050 and obtain acceleration data;

[0035] Initialize the Kalman filter to process the raw electrical signal data of the sensor;

[0036] The processed raw electrical signal data is transformed to obtain raw acceleration data;

[0037] Preprocessing the acceleration data, and then calculating the speed of the object using a double integration method according to the acceleration data and the time interval;

[0038] Based on the calculated velocity data and time interval, the displacement of the object is calculated using the double integration method.

[0039] The displacement data is sent to the host computer to draw the displacement image.

[0040] Furthermore, the MPU6050 is also used for performing posture prediction, and performing posture prediction using the MPU6050 includes the following steps:

[0041] Initialize the three-axis gyroscope and three-axis accelerometer of the MPU6050, and then initialize the DMP function by calling the dmpInitialize() function of the MPU6050 library;

[0042] Calibrate the three-axis gyroscope and the three-axis accelerometer;

[0043] Call the setDMPEnabled(true) function to enable the DMP function;

[0044] Use the dmpGetQuaternion() function and dmpGetGravity() function provided in the MPU6050 library to obtain the values ​​of quaternion and gravity acceleration, which are used to calculate the Euler angle of the attitude and obtain the attitude result;

[0045] The calculated posture results are output through the serial port and visualized using the host computer.

[0046] Furthermore, the quaternion and the value of gravity acceleration are used to calculate the Euler angle of the posture through the following formula:

[0047] Roll=atan2(Ay,Az)*RAD_TO_DEG

[0048] Pitch=atan2(-Ax,sqrt(Ay^2+Az^2))*RAD_TO_DEG

[0049] Yaw=atan2(2*q1*q2-2*q0*q3,2*q0^2+2*q1^2-1)*RAD_TO_DEG

[0050] Among them, Ax, Ay, and Az represent the acceleration values ​​of the three axes respectively, q0, q1, q2, and q3 represent the four components of the quaternion respectively; RAD_TO_DEG is a constant used to convert radians to degrees. Multiplying this constant will get the roll angle in degrees.

[0051] Furthermore, the central processor is communicatively connected with the physiological information monitoring module, the environmental information monitoring module, and the position estimation module via UART communication protocol and IIC communication protocol.

[0052] Compared with the prior art, the present invention has the following beneficial effects:

[0053] 1. The vital sign detection and early warning system based on fire suits provided by the present invention measures heart rate and blood oxygen data for a long time, and the measured data is relatively stable, no invalid data appears, and the fluctuation of heart rate and blood oxygen data is also small;

[0054] 2. The vital sign detection and early warning system based on fire suits provided by the present invention can obtain the Euler angle data of the human body through the MPU6050 motion sensor, and the real-time motion posture of the human body can be obtained by analyzing the Euler angle data;

[0055] 3. The fire suit-based vital sign detection and early warning system provided by the present invention can calculate the displacement of the firefighters through the quadratic integration method, and then estimate the position information of the firefighters;

[0056] 4. The fire-fighting suit-based vital sign detection and warning system provided by the present invention uses a K-type thermocouple to measure the ambient temperature. The maximum temperature that the thermocouple can measure is 1100°C. At the same time, MAX6675 is used to perform temperature compensation on the thermocouple to improve its accuracy.

[0057] 5. By comparing the common wearable devices on the market with the heart rate and blood oxygen devices designed in this article, it is found that the heart rate and blood oxygen detection device designed in this invention has stable measurement data output, a heart rate error value of 0.91%, and a blood oxygen error value of 0.7%. The data is reliable and meets the design and use requirements. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] Figure 1 It is an overall structural framework diagram of a vital sign detection and early warning system based on fire-fighting clothing in an embodiment of the present invention;

[0059] Figure 2 It is a connection diagram of a vital sign detection and early warning system based on fire-fighting clothing in an embodiment of the present invention;

[0060] Figure 3 It is a PPG curve diagram in an embodiment of the present invention;

[0061] Figure 4 Schematic diagram of Euler angles of sensors in an embodiment of the present invention;

[0062] Figure 5 This is a graph of raw data collected using a heart rate and blood oxygen sensor in an embodiment of the present invention;

[0063] Figure 6This is a data display diagram output when invalid data occurs when the heart rate and blood oxygen sensor is used for measurement in an embodiment of the present invention;

[0064] Figure 7 1 is a data diagram of continuous measurement using a heart rate and blood oxygen sensor in an embodiment of the present invention, including (a) resting heart rate and (b) resting blood oxygen;

[0065] Figure 8 This is a raw data diagram of the internal temperature of the fire-fighting suit in an embodiment of the present invention;

[0066] Fig. 9 This is a diagram showing the temperature change inside the fire suit when exercising in the fire suit at room temperature according to an embodiment of the present invention;

[0067] Fig.10 The data graphs recorded by the sensor when the paper box wrapped with the sensor in the embodiment of the present invention is free-falling from the desktop, wherein (a) is the change of the roll angle during the free-fall test; (b) is the change of the pitch angle during the free-fall test; (c) is the change of the heading angle during the free-fall test;

[0068] Fig.11 This is a graph of raw acceleration data recorded by the sensor when a paper box wrapped with the sensor in an embodiment of the present invention freely falls from a desktop;

[0069] Fig.12 This is a data graph measured when a human body falls to the left rear in an embodiment of the present invention;

[0070] Fig.13 is a force vector diagram in space in an embodiment of the present invention;

[0071] Fig.14 It is a curve diagram for estimating indoor straight line position in an embodiment of the present invention;

[0072] Fig.15 This is a graph of the original data when a thermocouple is used for measurement in an embodiment of the present invention;

[0073] Fig.16 It is a temperature curve image of natural cooling after the thermocouple is heated in the embodiment of the present invention;

[0074] Fig.17 This is a diagram showing the temperature change over time in the fire suit in an embodiment of the present invention, where T1 is the temperature measured by the system; T2 is the data measured by the thermometer;

[0075] Fig.18 This is a graph showing changes in heart rate over time in an embodiment of the present invention, where HR1 is a system measurement value and HR2 is a wearable device measurement value;

[0076] Fig.19This is a graph of blood oxygen changes over time in an embodiment of the present invention, where SPO21 is the system measurement value and SPO22 is the wearable device measurement value.

[0077] The reference numerals are as follows:

[0078] 1- Fire suit body, 2- Central processor, 3- Physiological information monitoring module, 4- Environmental information monitoring module, 5- Position estimation module, 6- Communication module, 7- Display screen, 8- Host computer. DETAILED DESCRIPTION

[0079] In order to make the purpose, technical scheme and advantages of the embodiments of the present application clearer, the technical scheme in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the embodiments described below are part of the embodiments of the present application, rather than all of the embodiments. The components of the embodiments of the present application described and shown in the drawings here can be arranged and designed in various different configurations.

[0080] Therefore, the detailed description of the embodiments of the present application provided below in conjunction with the accompanying drawings is intended to represent only selected embodiments of the present application, and is not intended to limit the scope of protection claimed in the present application. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present application.

[0081] In the description of the embodiments of the present invention, it should be noted that, unless otherwise clearly specified and limited, the terms "set", "install", "connect", and "connect" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection, an electrical connection, or can communicate with each other; it can be directly connected, or indirectly connected through an intermediate medium, and it can be the internal connection of two elements or the interaction relationship between two elements. For ordinary technicians in this field, the specific meanings of the above terms in the embodiments of the present invention can be understood according to specific circumstances.

[0082] See also Figure 1 The present invention provides a vital sign detection and early warning system based on fire suits, including a fire suit main body 1, and a central processor 2, a physiological information monitoring module 3, an environmental information monitoring module 4, a position estimation module 5, a communication module 6 and a display screen 7 arranged on the fire suit main body 1.

[0083] The physiological information monitoring module 3 includes a heart rate and blood oxygen sensor, a posture sensor and a body temperature sensor. The heart rate and blood oxygen sensor is arranged on the inner side of the wrist of the fire suit body, and the posture sensor and the body temperature sensor are arranged on the lower side of the back of the fire suit body.

[0084] The environmental information monitoring module 4 includes an environmental temperature sensor, which is arranged on the back outside of the fire-fighting suit body.

[0085] The position estimation module 5 includes an inertial measurement unit, which is used to measure the posture and acceleration of the wearer.

[0086] The central processor 2 is used to process the information collected by the physiological information monitoring module 3 , the environmental information monitoring module 4 and the position estimation module 5 , and display the processed information on the external display screen 7 .

[0087] The communication module 6 is used to send the information collected by the physiological information monitoring module 3 , the environmental information monitoring module 4 and the position estimation module 5 to the host computer 8 .

[0088] The display screen is used to display the information collected by the physiological information monitoring module 3 , the environmental information monitoring module 4 and the position estimation module 5 .

[0089] Optionally, the central processor uses an STM32 series controller.

[0090] Optionally, the heart rate and blood oxygen sensor is MAX30102.

[0091] Optionally, the environmental information monitoring module 4 uses MAX6675.

[0092] Optionally, the position estimation module 5 uses MPU6050, and the MPU6050 integrates a three-axis gyroscope and a three-axis accelerometer.

[0093] Optionally, the MAX30102 is used to calculate the heart rate and blood oxygen saturation, including the following steps:

[0094] S1. Signal preprocessing

[0095] Collect the optical signal reflected by the MAX30102, convert it into a PPG signal, and perform filtering and denoising on the PPG signal;

[0096] S2. Get periodic window

[0097] Finding periodic peaks and troughs in the preprocessed PPG signal, and detecting the start and end positions of each heartbeat to obtain a periodic window;

[0098] S3. Calculate heart rate

[0099] Based on the periodic window, the heart rate is calculated;

[0100] S4. Calculate blood oxygen saturation

[0101] Based on the pre-processed PPG signal, the blood oxygen saturation is calculated by analyzing the relationship between light intensity and time.

[0102] Specifically, the basic principle of using MAX30102 to measure human heart rate and blood oxygen is to utilize the different reflectivity of light when blood vessels contract and relax, collect the reflected light signal, and convert it into a PPG curve. Then, the algorithm is used for analysis and calculation to obtain the human heart rate and blood oxygen data. Figure 3 It is a typical PPG curve, that is, the original image received by the sensor. It can be divided into two parts: direct current signal (DC) and alternating current signal (AC). The AC signal is relatively simple and is obtained by reflecting the beating arterial blood. The peak corresponds to the contraction of the heart and the trough corresponds to the diastole of the heart. The DC signal can be divided into the following three parts from top to bottom, namely, non-pulsating arterial blood, venous blood and other tissues. For different people of different ages, genders and physical conditions, the corresponding DC values ​​will also be different. Therefore, relative values ​​are usually used when calculating blood oxygen.

[0103] The algorithm for calculating heart rate in this application uses the time domain method, and the time domain method can be used to calculate heart rate and blood oxygen saturation using classic waveform analysis. At present, commercial smart bracelets and smart watches often use photoelectric sensors to obtain heart rate and blood oxygen data, that is, use PPG signals to obtain.

[0104] The PPG signal reflects the dynamic changes in the blood flow in the human body's peripheral blood vessels. The basic principle is to illuminate the skin surface with a light source. After the light is reflected and diffused by the skin layer, it passes through the thin-walled capillaries in the skin layer, and is reflected again by the skin surface before being received by the photoelectric sensor. When the blood in the capillaries flows, it absorbs the light passing through, and its absorption amount changes with the vasodilation, causing the PPG waveform to fluctuate. By analyzing these changes, the heart rate and blood oxygen saturation can be calculated. The basic process of the algorithm is shown in the figure below.

[0105] The following steps are used to calculate heart rate and blood oxygen based on PPG signals using the time domain method:

[0106] (1) Signal preprocessing

[0107] After receiving the PPG signal, some preprocessing operations are required. Generally, the signal needs to be filtered and denoised to make it smooth and eliminate noise interference.

[0108] (2) Obtaining periodic windows

[0109] In the PPG signal, the waveform generated by each heartbeat is periodic. Therefore, by setting a fixed-size sliding window, we can find the periodic peaks and troughs in the signal and detect the start and end positions of each heartbeat.

[0110] (3) Calculate heart rate

[0111] Based on the periodic window, the time interval between each heartbeat can be calculated, and then the heart rate can be calculated. For example, if three heartbeats are found through the sliding window, and the time intervals between them are calculated to be T1 and T2, then the heart rate HR can be calculated as:

[0112] HR=60 / ((T1+T2) / 2) (1)

[0113] (4) Calculate blood oxygen saturation

[0114] Based on the PPG signal, the blood oxygen saturation can be calculated by analyzing the relationship between light intensity and time. Generally speaking, blood oxygen saturation is inversely proportional to the Pleth Variability Index (PVI). PVI represents the amplitude variation range of the PPG waveform, which indicates the filling degree of capillaries. When the PVI is higher, the blood oxygen saturation is lower.

[0115] The basic principle of blood oxygen saturation calculation is as follows:

[0116] First, based on experience, the ratio of the larger light intensity to the smaller light intensity is used as the characteristic value. This characteristic value (AC) is compared with the subsequent signal average value (DC), and the calculation result is amplified to form an indicator called a ratio (R). Blood oxygen saturation is usually calculated using the following formula:

[0117] SpO2=(AB) / (A+B) (2)

[0118] (2) In the formula, A and B are known parameters, corresponding to the ratio values ​​when the blood oxygen saturation is 100% and 0%.

[0119] When calculating blood oxygen, in order to ensure the accuracy and reliability of the measurement results, this application uses RED and IR light to measure blood oxygen, which is called dual-wavelength blood oxygen saturation measurement method. According to a large number of studies, blood oxygen saturation and R are linearly related, see formula (3) and formula (4):

[0120] SpO2=-45.06*R 2 +30.354*R+94.845 (3)

[0121]

[0122] The following are some of the main procedures used when calculating heart rate and blood oxygen:

[0123] #include <stdio.h>

[0124] #include <stdlib.h>

[0125] #include <math.h>

[0126] #define SIGNAL_LENGTH 1000

[0127] #define HRV_LENGTH 5

[0128] / * Function declarations * /

[0129] void get_ppg_signal(double* ppg_signal, int signal_length, double sampling_rate);

[0130] void preprocess_ppg(double* ppg_signal, int signal_length);

[0131] void detect_r_peak(double* ppg_signal, int signal_length, double* r_peak, int hrv_length);

[0132] double compute_hr(double* r_peak, int hrv_length, double sampling_rate);

[0133] double compute_spo2(double* ppg_signal, int signal_length);

[0134] int main() {

[0135] / * Define PPG signal data * /

[0136] double ppg_signal[SIGNAL_LENGTH]; double r_peak[HRV_LENGTH];

[0137] / * Collect PPG signal * /

[0138] double sampling_rate = 50; / / Sampling frequency is 50HZ

[0139] get_ppg_signal(ppg_signal,SIGNAL_LENGTH,sampling_rate);

[0140] / *Preprocess PPG signal* /

[0141] preprocess_ppg(ppg_signal,SIGNAL_LENGTH);

[0142] / *Detect R peak* /

[0143] detect_r_peak(ppg_signal,SIGNAL_LENGTH,r_peak,HRV_LENGTH);

[0144] / *Calculate heart rate* /

[0145] double hr=compute_hr(r_peak,HRV_LENGTH,sampling_rate);

[0146] printf("Calculated heart rate: %.2f bpm\n",hr);

[0147] / *Calculate blood oxygen saturation* /

[0148] double spo2=compute_spo2(ppg_signal,SIGNAL_LENGTH);

[0149] printf("The calculated blood oxygen saturation is: %.2f%%\n",spo2*100); return0;}

[0150] / *Generate simulated PPG signal* /

[0151] Optionally, estimating the position by the MPU6050 includes the following steps:

[0152] Step 201, initialize the MPU6050 and obtain acceleration data.

[0153] Step 202: pre-process the acceleration data, and then calculate the velocity of the object using a double integration method based on the acceleration data and the time interval.

[0154] Step 203: Calculate the displacement of the object using a double integration method according to the calculated velocity data and time interval.

[0155] Specifically, the MPU6050 integrates a three-axis gyroscope and a three-axis accelerometer to measure the attitude and acceleration of an object. This application uses the MPU6050 to perform a double integration of the collected data to estimate the position of the object. The main algorithm idea is as follows:

[0156] First, initialize MPU6050, obtain acceleration data, and then pre-process the acceleration data. Since MPU6050 may have noise, it is necessary to use Kalman filtering to smooth the acceleration data. Then, according to the acceleration data and time interval, the speed of the object is calculated using the double integration method. According to the calculated speed data and time interval, the displacement of the object is calculated using the double integration method. By looping through the above steps, the position information of the object is updated in real time.

[0157] The double integration method is a method for calculating displacement based on acceleration data. It obtains displacement data by integrating acceleration data twice, and can effectively reduce the accumulation of integration errors better than the single integration method.

[0158] Assuming that the acceleration data is a(t), t is time, and the starting point of integration is t0, then according to Newton's second law and the definition of acceleration, the differential equations of velocity v(t) and displacement x(t) can be obtained:

[0159] v(t)= ∫a(t)dt (5)

[0160] x(t) = ∫v(t)dt (6)

[0161] At this time, the relationship between displacement x(t) and acceleration a(t) is:

[0162] x(t) = ∫∫a(t)dt 2 (7)

[0163] In simple terms, the double integration method is to first set x(t0) to 0, and then perform two discrete integrations on the acceleration data to obtain the velocity and displacement data at each moment.

[0164] Assume that the time interval is Δt, that is, Δt = t(i+1)-t(i), and the acceleration is discretized as follows:

[0165] a(i) = a(t0 + iΔt) (8)

[0166] Then, the velocity v(i) and displacement x(i) are:

[0167] v(i)=v(i-1)+(a(i)+a(i-1)) / 2 *Δt (9)

[0168] x(i)=x(i-1)+(v(i)+v(i-1)) / 2*Δt (10)

[0169] Here, acceleration a(i) is the discrete acceleration data, while v(i) and x(i) are the estimated values ​​of velocity and displacement at time point i.

[0170] Here are some of the main procedures used when estimating position:

[0171] int main(void){

[0172] float a[SAMPLE_BUF_SIZE]; / / buffer for storing acceleration values

[0173] float v[SAMPLE_BUF_SIZE]; / / buffer for storing speed values

[0174] float x[SAMPLE_BUF_SIZE]; / / buffer for storing displacement values

[0175] float time=0; / / initial time

[0176] float last_v=0; / / The speed at the last moment

[0177] float last_x=0; / / displacement at the last moment

[0178] / / Clear acceleration, velocity, and displacement data

[0179] memset(a,0,sizeof(float)*SAMPLE_BUF_SIZE);

[0180] memset(v,0,sizeof(float)*SAMPLE_BUF_SIZE);

[0181] memset(x,0,sizeof(float)*SAMPLE_BUF_SIZE);

[0182] / / Read acceleration data

[0183] / / TODO: Get data from the sensor and store it in buffer a

[0184] / / Calculate the displacement by double integration method

[0185] for(int i=0; i <SAMPLE_BUF_SIZE;i++){

[0186] / / Calculate the current speed

[0187] v[i]=last_v+(a[i]+a[i-1]) / 2*SAMPLE_INTERVAL;

[0188] / / Calculate the displacement at the current moment

[0189] x[i]=last_x+(v[i]+last_v) / 2*SAMPLE_INTERVAL;

[0190] / / Update the speed and displacement of the previous moment

[0191] last_v = v[i];

[0192] last_x = x[i];

[0193] / / Output the estimated displacement value

[0194] printf("t=%.2f,x=%.2f\n",time,x[i]);

[0195] / / Update the time of the current sampling point

[0196] time += SAMPLE_INTERVAL;

[0197] }

[0198] return 0;

[0199] }

[0200] In the program, the arrays storing acceleration, velocity, and displacement values ​​are first defined, and then they are all initialized to zero. Then the acceleration data is read in a loop, the velocity and displacement at each moment are calculated using the double integration method, and the estimated displacement value is output to the terminal.

[0201] When calculating the velocity value, we use the definition of displacement change rate: v(t) = Δx / Δt

[0202] Among them, Δt is the sampling time interval SAMPLE_INTERVAL. Since we know the displacement value last_x at the previous moment and the displacement value x[i] at the current moment, we can get Δx by subtracting them and calculate the velocity value v[i].

[0203] When calculating displacement, we use the definition of displacement:

[0204] x(t)=∫v(t)dt (11)

[0205] The velocity v(t) has been obtained, so the double integration method can be used for integration to obtain the estimated displacement value.

[0206] Optionally, the MPU6050 is also used for posture prediction, and the posture prediction using the MPU6050 includes the following steps:

[0207] Step 301, initialize the three-axis gyroscope and three-axis accelerometer of the MPU6050, and then initialize the DMP function by calling the dmpInitialize() function of the MPU6050 library;

[0208] Step 302: calibrate the three-axis gyroscope and the three-axis accelerometer;

[0209] Step 303: Call the setDMPEnabled(true) function to enable the DMP function.

[0210] Step 304: Use the dmpGetQuaternion() function and the dmpGetGravity() function provided in the MPU6050 library to obtain the values ​​of the quaternion and the gravitational acceleration, which are used to calculate the Euler angle of the posture and obtain the posture result;

[0211] Step 305: Output the calculated posture result through the serial port and use the host computer for visualization processing.

[0212] Specifically, the present application uses the (DMP) digital motion processor function provided by MPU6050 to help predict human posture without filtering. DMP can perform posture calculation inside MPU6050. By calling the function of MPU6050 function library to enable DMP function, the quaternion and acceleration values ​​calculated by raw data are obtained to calculate the Euler angle of posture (see Figure 4 ).

[0213] Using the quaternion and the value of gravity acceleration, the Euler angle of the posture can be calculated by the following formula:

[0214] Roll=atan2(Ay,Az)*RAD_TO_DEG

[0215] Pitch=atan2(-Ax,sqrt(Ay^2+Az^2))*RAD_TO_DEG

[0216] Yaw=atan2(2*q1*q2-2*q0*q3,2*q0^2+2*q1^2-1)*RAD_TO_DEG

[0217] Among them, Ax, Ay, Az represent the acceleration values ​​of the three axes respectively, and q0, q1, q2, q3 represent the four components of the quaternion respectively. atan2 is an extended form of the inverse tangent function, which accepts two parameters, Ay and Az, and calculates the attitude angle by the ratio of these two parameters. RAD_TO_DEG is a constant used to convert radians to degrees. Multiplying this constant will get the roll angle in degrees.

[0218] The following is part of the main program for posture prediction using DMP:

[0219] #include "I2Cdev.h"

[0220] #include "MPU6050.h"

[0221] #include "helper_3dmath.h"

[0222] MPU6050 MPU;

[0223] int16_t ax,ay,az; int16_t gx,gy,gz; Quaternion q;

[0224] void setup(){

[0225] Serial.begin(115200);

[0226] Wire.begin();

[0227] mpu.initialize();

[0228] mpu.dmpInitialize();

[0229] mpu.setDMPEnabled(true);

[0230] delay(1000);

[0231] }

[0232] Void loop(){

[0233] mpu.resetFIFO();

[0234] while(!mpu.dmpReady()){}mpu.getIntStatus();

[0235] mpu.getQuaternion(&q);

[0236] mpu.getAcceleration(&ax,&ay,&az);

[0237] mpu.getRotation(&gx,&gy,&gz);

[0238] float Roll=atan2(2.0*qx*qy-2.0*qw*qz,2.0*qw*q.w+2.0*qx*qx-1);

[0239] float Pitch=-asin(2.0*qx*q.z+2.0*qw*qy);

[0240] float Yaw=atan2(2.0*qy*qz-2.0*qw*qx,2.0*qw*q.w+2.0*qz*qz-1);

[0241] Roll = Roll * 180.0 / PI;

[0242] Pitch = Pitch * 180.0 / PI;

[0243] Yaw = Yaw * 180.0 / PI;

[0244] Serial.print("Roll:"); Serial.print(Roll);

[0245] Serial.print("Pitch:"); Serial.print(Pitch);

[0246] Serial.print("Yaw:"); Serial.println(Yaw);

[0247] delay(10);

[0248] }

[0249] Optionally, the central processor 2 is communicatively connected with the physiological information monitoring module 3 , the environmental information monitoring module 4 , and the position estimation module 5 via a UART communication protocol and / or an IIC communication protocol.

[0250] like Figure 2As shown, in this embodiment, the central processor 2, i.e., the main control part in the figure, adopts STM32, the physiological information monitoring module 3 selects MAX30102, the environmental information monitoring module 4 selects MAX6675, and the position estimation module 5 selects MPU6050. STM32 is connected to MAX6675 and MPU6050 via UART communication protocol, and STM32 is connected to MAX30102 via IIC communication protocol. STM32 is also connected to the display screen (i.e., the display module in the figure) and the host computer via UART communication protocol. The display screen can select an LCD screen, and the host computer can directly select a PC end.

[0251] The processing of heart rate and blood oxygen data is an important component of the fire suit system of the present invention. This application verifies the accuracy of the heart rate and blood oxygen module by testing the human heart rate and blood oxygen data under different conditions and comparing them with the data measured by common wearable devices on the market. When the heart rate and blood oxygen data are continuously output, it can also determine in real time whether the output data is reliable and output a judgment symbol. Figure 5 It is the raw data collected using the heart rate and blood oxygen sensor.

[0252] By comparing the common wearable devices on the market with the heart rate and blood oxygen device designed in this paper, it is found that the heart rate and blood oxygen detection device designed in this application has small errors and reliable data, which basically meets the design and use requirements. Figure 6 shown.

[0253] When the sensor determines that the detected data is invalid according to the algorithm, it will change the output heart rate and blood oxygen value to -999 and output the discrimination symbol 0. At this time, all the detected data are invalid and it is necessary to check whether the sensor is worn correctly. In order to verify the stability of the heart rate and blood oxygen sensor during operation, this application collects the heart rate and blood oxygen data when the sensor is working continuously. Figure 7 This is a data graph when using a heart rate and blood oxygen sensor for continuous measurement.

[0254] According to the measurement data, when measuring for a long time, the measured data is relatively stable, no invalid data appears, the fluctuations in heart rate and blood oxygen data are also small, and the error is within an acceptable range.

[0255] The temperature fluctuations inside the fire suit have a great impact on the life safety of firefighters. Since the heat insulation of fire suits is limited and the heat resistance has a certain threshold, when firefighters approach the fire source, the temperature inside the fire suit will gradually rise. In addition, damage to the fire suit may also cause a sudden rise in temperature. High temperatures can cause symptoms such as heat cramps and heat exhaustion in firefighters, which can threaten the safety of firefighters. Figure 8 and Fig. 9They are the original data of the internal temperature of the fire suit and the temperature change diagram inside the fire suit when exercising in the fire suit at room temperature.

[0256] The Euler angle data of the human body can be obtained through MPU6050. By analyzing the Euler angle data, we can get the real-time movement posture of the human body.

[0257] This application determines the posture of the human body by collecting and analyzing the changes in Euler angles. To ensure the reliability of the analysis results, the sensor will reset the triangle to zero during initialization and start recording data with the initialization state as the initial state. The directions represented by the positive and negative values ​​of each angle are as follows: Figure 4 The unit of Euler angle calculated by the algorithm is 0.1°, which represents the angle of rotation of each axis. Fig.10 , Fig.11 The data recorded by the sensor when the cardboard box wrapped in the sensor freely falls from the desktop. From the three-axis Euler angle, we can know that the cardboard box rotates 600° clockwise to the right front when falling and then lands on the ground. During the second free fall, it rotates 1500° to the left rear before landing. During the two free falls, the cardboard box is rotated 700°.

[0258] In order to measure human posture more accurately, this application has been tested many times and found that when the roll angle of the human body changes by more than ±30° or the pitch angle changes by more than 35°, the human body will be at risk of falling. Fig.12 The data is measured when the human body falls to the left rear.

[0259] The position estimation method used in this application is to first collect the original acceleration data. The original data collected by the accelerometer is a 16-bit binary complement value. The full-scale range of the acceleration output selected in this application is ±2g, so the sensitivity in each direction is 16384LSB / g. If the original data value obtained on the z-axis is 16217, then the corresponding acceleration data is 16217 / 16384=0.99g. The gravity acceleration is defined as 1g equals 9.8m / s 2 After obtaining the acceleration data of each axis, the displacement of the firefighter can be calculated by the quadratic integration method. As long as the angle of the force vector is obtained each time the data is recorded, the position information of the firefighter can be estimated.

[0260] When calculating the angle, the angle between the X, Y, Z axis and the force vector R is defined as Axr, Ayr, Azr, such as Fig.13 As shown, we can get:

[0261] cos(Axr)=Rx / R (12)

[0262] cos(Ayr)=Ry / R (13)

[0263] cos(Azr)=Rz / R (14)

[0264]

[0265] So we can use the arccos() function to get the angle between the X, Y, and Z axes and the force vector:

[0266] Axr=arccos(Rx / R) (16)

[0267] Ayr=arccos(Ry / R) (17)

[0268] Azr=arccos(Rz / R) (18)

[0269] Here, Rx, Ry, and Rz are the calculated acceleration data of each axis.

[0270] Assume that when the sensor is placed horizontally, only the Z axis feels the gravity vector, and it outputs the raw data of 16217. At this time, the acceleration on the Z axis is 0.99g, and the acceleration on the X and Y axes is 0.

[0271] Rx=0,Ry=0,Rz=-1g

[0272] Satisfy R 2 =Rx 2 +Ry 2 +Rz 2 =1, get the angle between the gravity vector and each axis

[0273] Axr = arccos(Rx / R) = 90°

[0274] Ayr=arccos(Ry / R)=90°

[0275] Azr=arccos(Rz / R)=0°

[0276] Based on the calculated three-axis acceleration values ​​and force vector angles, the quadratic integral method can be used to make a simple estimate of the position. This application selected a straight corridor with a length of about 18m for testing. The estimated results are as follows: Fig.14 The distance error in the test data is 4m, while the horizontal distance error is larger. The main reason is that the algorithm does not perform dynamic calibration on the sensor, and since there is no large-scale horizontal movement during movement, the smaller initial error shows a larger error after the integrated error accumulation.

[0277] In this application, a K-type thermocouple is used to measure the ambient temperature. The maximum temperature that the thermocouple can measure is 1100°C. At the same time, MAX6675 is used to perform temperature compensation on the thermocouple to improve its accuracy. Fig.15 , Fig.16They are respectively the original data when using thermocouple measurement and the temperature curve image of natural cooling after thermocouple heating.

[0278] The temperature inside the fire suit changes with time. Fig.17 When firefighters go from indoors to outdoor hot environments for strenuous exercise, the temperature inside the fire suit rises rapidly in a short period of time, reaches a peak and then remains stable.

[0279] The changes of heart rate and blood oxygen value over time are as follows: Fig.18 , as shown in 19. By comparing the common wearable devices on the market with the heart rate and blood oxygen devices designed in this paper, it is found that the heart rate and blood oxygen detection device designed in this application has stable measurement data output, a heart rate error value of 0.91%, and a blood oxygen error value of 0.7%. The reliable data basically meets the design and use requirements.

[0280] Heart rate and blood oxygen error analysis

[0281] Error analysis is an important step in evaluating the accuracy of heart rate and blood oxygen calculation using the time domain method. Generally speaking, error analysis can be considered from the following aspects:

[0282] (1) Data collection / processing errors

[0283] There will be certain errors in the data acquisition and preprocessing process, such as the accuracy of the signal acquisition equipment and the noise level. These errors will have a certain impact on the calculation results. The noise can be removed and the errors can be reduced by filtering or waveform transformation.

[0284] (2) Method error

[0285] Time domain methods also have errors. For example, the settings of signal filters and window sizes may affect the calculation results. In addition, some time domain methods may not be able to cope with different types of signals and interferences, and they also need to be adjusted appropriately.

[0286] (3) Benchmark error

[0287] Human physiological characteristics vary in age, gender, physical condition, etc., which may lead to certain errors in the accuracy of the calculation results. Therefore, these factors should be considered and corrected when calculating.

[0288] (4) Statistical Error

[0289] When the sample size is small, the heart rate and blood oxygen results calculated based on the time domain method may have certain statistical errors. Therefore, a larger sample size is needed to reduce the statistical error.

[0290] In practical applications, it is usually necessary to conduct several sets of experiments and repeatedly calculate the same data to evaluate the stability and accuracy of the calculation results. When the calculation results are abnormal, further data inspection and error analysis are required to obtain more accurate calculation results.

[0291] Position Estimation Error Analysis

[0292] When using the double integration method for displacement estimation, large errors often occur. The sources of the errors and their analysis are as follows:

[0293] (1) Accumulated error: Since the error is double-integrated, it accumulates over time. When the integration time is long, the error becomes more and more serious.

[0294] (2) Data sampling error: Due to the limited sampling frequency of the accelerometer, the data may be jittery or missing, resulting in errors in the estimated displacement value.

[0295] (3) Accelerometer error: The accuracy of the accelerometer will also affect the accuracy of displacement estimation. For example, rounding errors and precision truncation errors often occur in numerical calculations.

[0296] For these error sources, you can try the following methods to control the error:

[0297] (1) Strengthen data processing: In the process of sensor data processing, average filtering, median filtering and other technologies should be used to reduce data errors and jitter.

[0298] (2) Increase the sampling rate of the sensor: Increase the number of data points by increasing the sampling frequency of the sensor to reduce sampling errors.

[0299] (3) Use higher-precision accelerometers: Improve sensor design and manufacturing to increase sensor measurement accuracy and reduce errors.

[0300] (4) Calibration error: According to the error characteristics of numerical calculations, calibration methods can be used to compensate for the errors.

[0301] In summary, we can reduce the error in displacement estimation by strengthening data processing, improving sensor accuracy, optimizing algorithms and calibrating errors.

[0302] Posture Detection Error Analysis

[0303] When using DMP to predict the accuracy of posture, there are mainly two parts of error: sensor error and algorithm error.

[0304] (1) Sensor error

[0305] When calculating the attitude, the data of the accelerometer and gyroscope may have errors, which may affect the accuracy of the attitude calculation. Therefore, the accelerometer and gyroscope need to be calibrated.

[27] . Errors such as offset and nonlinearity can be eliminated through calibration.

[0306] (2) Algorithm Error

[0307] The prediction algorithm of DMP may have errors, which may introduce errors in posture calculation. The errors in the algorithm may come from the following factors:

[0308] (a) Static attitude error: The prediction error occurs because the minimum static attitude solution of the DMP may deviate from the actual attitude.

[0309] (b) Dynamic pose error: The minimum dynamic pose solution of DMP may deviate from the actual pose, resulting in prediction errors.

[0310] (c) Sensor signal processing error: When processing sensor data, there may be noise and interference, resulting in inaccurate prediction results.

[0311] The above errors may affect the prediction accuracy of the attitude, so these errors need to be compensated and processed to improve the prediction accuracy. It should be noted that the accuracy error is inevitable, but it can be controlled within a certain range. In practical applications, various means such as hardware and software can be combined to optimize the attitude prediction accuracy. For example, common sensor data processing algorithms such as filtering algorithms and Kalman filtering can be used to improve the attitude prediction accuracy.

[0312] Temperature Detection Error Analysis

[0313] When using the MAX6675 to measure thermocouple temperatures, there are several sources of error to consider:

[0314] (1) Accuracy of the temperature sensor: The temperature range of the temperature sensor built into the MAX6675 is usually between -20°C and 85°C, and its accuracy is usually low. Therefore, it is necessary to correct the measurement results of the temperature sensor through other devices or methods to improve the accuracy of temperature measurement.

[0315] (2) Thermocouple measurement accuracy: The accuracy of thermocouple temperature measurement is affected by many factors, such as the characteristics of the thermocouple material, the position of the joint, the line impedance, etc. Therefore, methods such as cold end compensation are needed to improve the measurement accuracy.

[0316] (3) Interference from ambient temperature: Changes in ambient temperature may affect the accuracy of thermocouple measurements, so these interference factors need to be fully considered during the thermocouple measurement process.

[0317] (4) Power supply noise and other factors: The power supply noise and other factors in the circuit will also affect the normal operation of MAX6675 and affect the accuracy of measurement.

[0318] (5) External interference and other factors: If the environment in which the MAX6675 and the thermocouple are located has electromagnetic interference, vibration and other factors, it will also affect the thermocouple measurement results. Therefore, special attention should be paid to the existence of these interference factors.

[0319] In summary, when measuring the temperature of MAX6675 and thermocouples, it is necessary to carefully design and analyze the measurement system. Under the premise of ensuring consistent test conditions, appropriate error compensation and interference reduction methods should be adopted to improve the accuracy and reliability of the measurement results.

[0320] The above is only the best specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions within the technical scope disclosed in the present application shall be included in the protection scope of the present application. Therefore, the protection scope of the present application shall be based on the protection scope of the claims.

Claims

1. A vital sign detection and early warning system based on fire suits, characterized by: It includes a fire suit body, and a central processor, a physiological information monitoring module, an environmental information monitoring module, a position estimation module, a display module and a communication module arranged on the fire suit body; The physiological information monitoring module includes a heart rate and blood oxygen sensor, a posture sensor and a body temperature sensor. The heart rate and blood oxygen sensor is arranged on the inner side of the wrist of the fire suit body, the posture sensor and the body temperature sensor are arranged on the lower back of the fire suit body, and the display module is arranged on the outer side of the wrist of the fire suit; The environmental information monitoring module includes an environmental temperature sensor, which is arranged on the back outside of the fire suit body; The position estimation module includes an inertial measurement unit, which is used to measure the posture and acceleration of the wearer; The central processor is used to process the information collected by the physiological information monitoring module, the environmental information monitoring module and the position estimation module, and display the processed information on an external display screen; The display module includes an LCD screen for displaying various data processed by the central processor. The communication module is used to send the information collected by the physiological information monitoring module, the environmental information monitoring module and the position estimation module to the host computer, and send the alarm identifier corresponding to each data to the host computer when the data exceeds the alarm threshold.

2. The vital sign detection and early warning system based on firefighting clothing according to claim 1 is characterized in that: The central processor uses an STM32 series controller.

3. The vital sign detection and early warning system based on firefighting clothing according to claim 1 is characterized in that: The heart rate blood oxygen sensor is MAX30102.

4. The vital sign detection and early warning system based on firefighting clothing according to claim 1 is characterized in that: The environmental information monitoring module uses MAX6675.

5. The vital sign detection and early warning system based on firefighting suit according to claim 1 is characterized in that: The position estimation module uses MPU6050, and the MPU6050 integrates a three-axis gyroscope and a three-axis accelerometer.

6. The vital sign detection and early warning system based on firefighting suit according to claim 3 is characterized in that: The MAX30102 is used to calculate the heart rate and blood oxygen saturation, including the following steps: S1. Signal preprocessing Collect the optical signal reflected by the MAX30102, convert it into a PPG signal, and perform filtering and denoising on the PPG signal; S2. Get periodic window Finding periodic peaks and troughs in the preprocessed PPG signal, and detecting the start and end positions of each heartbeat to obtain a periodic window; S3. Calculate heart rate Based on the periodic window, the heart rate is calculated; S4. Calculate blood oxygen saturation Based on the pre-processed PPG signal, the blood oxygen saturation is calculated by analyzing the relationship between light intensity and time.

7. The vital sign detection and early warning system based on firefighting clothing according to claim 5 is characterized in that: Estimating the position by the MPU6050 includes the following steps: Initialize the MPU6050 and obtain acceleration data; The acceleration data is filtered to obtain the original data of the sensor, and then converted into the original acceleration data according to the voltage comparison table in the chip manual, and finally the speed of the object is calculated using the double integration method according to the acceleration data and the time interval; According to the calculated speed data and time interval, the double integration method is used again to calculate the displacement of the object. The displacement data is sent to the host computer to draw the path diagram.

8. The vital sign detection and early warning system based on firefighting clothing according to claim 5 or 7 is characterized in that: The MPU6050 is also used for posture prediction. The posture prediction using the MPU6050 includes the following steps: Initialize the three-axis gyroscope and three-axis accelerometer of the MPU6050, and then initialize the DMP function by calling the dmpInitialize() function of the MPU6050 library; Calibrate the three-axis gyroscope and the three-axis accelerometer; Call the setDMPEnabled(true) function to enable the DMP function; Use the dmpGetQuaternion() function and dmpGetGravity() function provided in the MPU6050 library to obtain the values ​​of quaternion and gravity acceleration, which are used to calculate the Euler angle of the attitude and obtain the attitude result; The calculated posture results are output through the serial port and visualized using the host computer.

9. The vital sign detection and early warning system based on firefighting clothing according to claim 8 is characterized in that: Using the quaternion and the value of gravity acceleration, the Euler angle of the posture is calculated by the following formula: Roll=atan2(Ay,Az)*RAD_TO_DEG Pitch=atan2(-Ax,sqrt(Ay^2+Az^2))*RAD_TO_DEG Yaw=atan2(2*q1*q2-2*q0*q3,2*q0^2+2*q1^2-1)*RAD_TO_DEG Among them, Ax, Ay, and Az represent the acceleration values ​​of the three axes respectively, q0, q1, q2, and q3 represent the four components of the quaternion respectively; RAD_TO_DEG is a constant used to convert radians to degrees.

10. The vital sign detection and early warning system based on firefighting suit according to claim 1 is characterized in that: The central processor is connected to the physiological information monitoring module, the environmental information monitoring module, and the position estimation module through a UART communication protocol and / or an IIC communication protocol.

Images