Motorcycle fatigue driving warning device and warning method

Abstract

The present invention discloses a motorcycle fatigue driving warning device and a warning method, wherein the device comprises a timer monitoring unit, an electrocardiogram signal fatigue monitoring unit and a reminder unit connected in sequence. The warning device is low in cost, convenient to use, stable and efficient. The warning method can effectively prevent motorcycle drivers from driving fatigued due to long-term continuous driving, monitor the driver's fatigue state in real time, and make corresponding warning measures for the driver in different fatigue states, thereby reducing the accident rate and further ensuring road traffic safety.

Description

Technical Field

[0001] The invention relates to traffic safety technology, in particular to a motorcycle fatigue driving warning device and a warning method. Background Art

[0002] In modern society, fatigue driving causes a lot of economic losses. During the Spring Festival, motorcycles have become the first choice for most migrant workers to return home due to their low price, low fuel consumption and easy operation. A few people like to travel long distances by motorcycle.

[0003] Drivers who drive continuously for a long time without adequate rest are prone to psychological and physical fatigue, which will reduce their driving skills and eventually lead to fatigue driving. Fatigue driving not only harms the health of drivers but also increases the probability of traffic accidents. Therefore, designing a stable, accurate and efficient fatigue driving warning device can not only help protect the physical and mental health of drivers, but also effectively prevent the occurrence of road traffic accidents.

[0004] A Chinese patent with publication number CN205722338U discloses a motorcycle overtime driving reminder system: it uses a timer to count the current driving time. When the driver exceeds the driving time, the LCD screen of the motorcycle dashboard displays graphics and text to remind the driver, and at the same time changes the color of the overtime driving warning light on the dashboard. However, most of the time, the driver is visually observing the road conditions ahead and the surrounding environment, and the visual reminder on the dashboard may be ignored. In addition, distracting attention to watch the LCD screen will increase the driver's visual burden and fatigue, and driving fatigue cannot be effectively prevented and stopped.

[0005] The Chinese patent publication number CN205405809U discloses a fatigue driving detection alarm system based on a smart watch: by monitoring the driver's heart rate and continuous driving time, it determines whether the driver is in a fatigue driving state, and promptly reminds the driver through voice and vibration alarms. However, the standard for judging the driver's fatigue state is unclear, and it fails to make different warning measures for the driver's different degrees of fatigue, and cannot provide effective fatigue suppression. Summary of the invention

[0006] The purpose of the present invention is to provide a motorcycle fatigue driving warning device and a warning method in view of the deficiencies of the prior art. The warning device is low in cost, easy to use, stable and efficient. The warning method can effectively prevent motorcycle drivers from driving fatigued due to long-term continuous driving, monitor the driver's fatigue state in real time, and make corresponding warning measures for the driver in different fatigue states, thereby reducing the accident rate and further ensuring road traffic safety.

[0007] The technical solution for achieving the purpose of the present invention is:

[0008] A motorcycle fatigue driving warning device comprises a timer monitoring unit, an electrocardiogram fatigue monitoring unit and a reminder unit connected in sequence, wherein the timer monitoring unit is provided with a single-chip microcomputer and a speed sensor connected to the single-chip microcomputer, the single-chip microcomputer is provided with a single-chip microcomputer control module and a timer, the electrocardiogram fatigue monitoring unit is provided with an electrocardiogram signal acquisition module, an electrocardiogram signal data processing module and a machine learning algorithm module connected in sequence, the reminder unit is provided with an instrument panel background light module, a voice prompt module, a left handlebar vibration module, a left handlebar current stimulation module and a body shaking module which are electrically connected, the single-chip microcomputer is installed below the instrument panel of the motorcycle, the speed sensor is installed on the side of the front wheel of the motorcycle, the electrocardiogram fatigue monitoring unit is placed in a smart bracelet worn by the motorcycle driver, the electrocardiogram fatigue monitoring unit is connected to the single-chip microcomputer control module via Bluetooth, and the single-chip microcomputer control module is electrically connected to the instrument panel background light module, the voice prompt module, the left handlebar vibration module, the left handlebar current stimulation module and the body shaking module via a wire.

[0009] A motorcycle fatigue driving warning method, comprising the above-mentioned motorcycle fatigue driving warning device, the method comprising the following steps:

[0010] 1) Timer-based fatigue driving warning:

[0011] 1-1) After the motorcycle is started, the speed sensor installed on the motorcycle detects the speed of the vehicle. If the speed is higher than 3.5km / h, the timer of the timer monitoring unit starts counting and records the driving time;

[0012] 1-2) When the timer reaches 2 hours, the MCU control module controls the reminder unit to remind the driver of overtime driving, triggers the dashboard background light module to change color and the voice prompt module. If the speed sensor detects that the vehicle speed is continuously lower than 3.5 km / h for 10 minutes or more during the timer, the timer is reset and returns to step 1-1);

[0013] 1-3) When the timer reaches 3 hours, the MCU control module controls the reminder unit to remind the driver of overtime driving, trigger the left handlebar vibration module and voice prompt module. If the speed sensor detects that the vehicle speed is continuously lower than 3.5 km / h for 15 minutes or more during the timer, the timer is reset and the process returns to step 1-1);

[0014] 1-4) When the timer reaches 4 hours, the single-chip control module controls the reminder unit to remind the driver of overtime driving, trigger the left handlebar current stimulation module and the voice prompt module. If the speed sensor detects that the vehicle speed is continuously lower than 3.5 km / h for 20 minutes or more during the timer, the timer is reset and the process returns to step 1-1);

[0015] 1-5) When the timer reaches 5 hours, it is determined that the motorcycle driver is driving for an excessively long time. The single-chip control module controls the reminder unit to remind the driver of the overtime driving, trigger the dashboard background light module to change color, the voice prompt module, the left handlebar vibration module, the left handlebar current stimulation module and the body shaking module. When the speed sensor detects that the vehicle speed is lower than 3.5 km / h, the combined reminder stops working temporarily. If the driver continues to drive the vehicle within 30 minutes, the combined reminder continues to work. If the driver stops to rest for 30 minutes, the combined reminder is completely turned off, and the timer is reset, and the process returns to step 1-1);

[0016] 2) Fatigue driving warning based on ECG signals:

[0017] 2-1) The motorcycle driver uses a smart bracelet to collect the driver's ECG signal in real time, and uses the wavelet threshold denoising method to remove baseline drift, myoelectric interference and power frequency interference from the collected ECG signal;

[0018] 2-2) Use the Pan&Tompkins algorithm to detect the R wave of the ECG signal after noise removal. The detection algorithm steps of Pan&Tompkins are as follows:

[0019] 2-2-1) Initialize the signal threshold THR_SIGNAL and the noise threshold THR_NOISE. The initialization formulas of the signal threshold THR_SIGNAL and the noise threshold THR_NOISE are shown in formula (1) and formula (2):

[0020] (1),

[0021] (2),

[0022] Among them, THR_SIGNAL is the initial signal threshold, THR_NOISE is the initial noise threshold, pre_ecg is the ECG signal value after denoising, and fs is the sampling frequency of the signal;

[0023] 2-2-2) Compare the current ECG signal maximum value MAX_PEAK with the signal threshold THR_SIGNAL. , then MAX_PEAK is determined as the peak value of the R wave, the position of the R wave at this time is recorded, and the signal level value is updated at the same time. The calculation formula for updating the signal level value is shown in formula (3):

[0024] (3),

[0025] Among them, SIGNAL_LEVEL is the signal level value, and MAX_PEAK is the maximum value of the current ECG signal;

[0026] like , then MAX_PEAK is determined as noise, and the noise level value is updated at the same time. The calculation formula for updating the noise level value is shown in formula (4):

[0027] (4),

[0028] Among them, NOISE_LEVEL is the noise level value, and MAX_PEAK is the maximum value of the current ECG signal;

[0029] 2-2-3) After the five R wave peaks are determined using the signal threshold THR_SIGNAL and the noise threshold THR_NOISE, the signal threshold THR_SIGNAL and the noise threshold THR_NOISE are updated. The calculation formulas for updating the signal threshold THR_SIGNAL and the noise threshold THR_NOISE are shown in formula (5) and formula (6):

[0030] (5),

[0031] (6),

[0032] Among them, THR_SIGNAL is the signal threshold, THR_NOISE is the noise threshold, NOISE_LEVEL is the noise level value, and SIGNAL_LEVEL is the signal level value;

[0033] 2-2-4) Repeat steps 2-2-2) and 2-2-3) to detect all R wave peaks in the ECG signal;

[0034] 2-3) Calculate each RR interval based on the time difference between adjacent R waves. The calculation formula of the RR interval is shown in formula (7):

[0035] (7),

[0036] Among them, RR_period is the RR interval of the ECG signal, RR_time is the interval between two adjacent R waves, and FS is the sampling frequency of the signal;

[0037] 2-4) Set 100 memory cells in the smart bracelet, and each memory cell is used to store the most recent 100 RR interval values;

[0038] 2-5) Calculate the ECG signal characteristic value SDNN based on the values ​​of N consecutive RR intervals. The calculation formula of SDNN value is shown in formula (8) and formula (9):

[0039] (8),

[0040] (9),

[0041] in, is the value of the i-th RR interval, is the mean of N consecutive RR intervals, SDNN represents the standard deviation of the RR interval;

[0042] 2-6) The single-chip control module transmits the driver’s driving time to the smart bracelet via Bluetooth;

[0043] 2-7) The smart bracelet collects the ECG signal feature value SDNN, driving time and fatigue status of the driver while driving. The machine learning algorithm module uses the ECG signal feature value SDNN, driving time and fatigue status as input variables, and calls the machine learning algorithm to build a fatigue driving habit model. The fatigue driving habit model predicts the driver's ECG signal feature value SDNN, and accurately calculates and corrects the predicted value through the actual measurement value, so that the predicted value is infinitely close to the true value;

[0044] 2-8) Compare the calculated and predicted ECG signal feature value SDNN with the three preset thresholds representing different fatigue levels when α is 150ms, β is 200ms, and γ is 250ms: When the device determines that the driver's fatigue state is mild fatigue, the ECG signal fatigue monitoring unit transmits the fatigue state information to the single-chip control module via Bluetooth, and the single-chip control module controls the reminder unit to remind fatigue driving, trigger the instrument panel background light module to change color and the voice prompt module; when When the device determines that the driver's fatigue state is moderate fatigue, the ECG signal fatigue monitoring unit transmits the fatigue state information to the single-chip control module via Bluetooth, and the single-chip control module controls the reminder unit to remind the driver of fatigue driving, triggering the left handlebar vibration module and voice prompt module; when When the device determines that the driver's fatigue state is severe fatigue, the ECG signal fatigue monitoring unit transmits the fatigue state information to the single-chip control module via Bluetooth. The single-chip control module controls the reminder unit to provide fatigue driving reminders, triggering the left handlebar current stimulation module and voice prompt module.

[0045] This technical solution divides the driver's fatigue status into three categories: if the driver is in a state of mild fatigue, the dashboard background light changes color and voice prompts are triggered, and the speaker on the motorcycle plays: You are mildly fatigued, please stop and rest for 10 minutes or more; if the driver is in a state of moderate fatigue, the left handlebar vibrates and voice prompts are triggered, and the speaker on the motorcycle plays: You are moderately fatigued, please stop and rest for 15 minutes or more; if the driver is in a state of severe fatigue, the left handlebar current stimulation and voice prompts are triggered, and the speaker on the motorcycle plays: You are severely fatigued, please stop and rest for 20 minutes or more. The driver is made aware of his own fatigue status, and the corresponding early warning measures are used to relieve the driver's fatigue, reduce the probability of traffic accidents, and further ensure road traffic safety.

[0046] According to the technical solution, when the driver's continuous driving time reaches 2 hours, 3 hours, and 4 hours, the single-chip control module controls the instrument panel background light to change color and voice prompt, the left handlebar to vibrate and voice prompt, and the left handlebar to stimulate current and voice prompt to remind the driver of overtime driving. When the timer reaches 5 hours in total, the single-chip control module simultaneously starts the instrument panel background light to change color, voice prompt, left handlebar vibration, left handlebar current stimulation, and body shaking to remind the driver that he is driving for an excessively long time. This can effectively prevent the driver from driving fatigued due to excessive driving time and causing traffic accidents, thereby improving driving safety.

[0047] Compared with the prior art, the advantages of this technical solution are:

[0048] 1. This technical solution uses a timer to measure the driving time. At the same time, the smart bracelet calculates and predicts the driver's ECG signal characteristic value in real time and compares it with the preset threshold value, which can realize overtime driving warning and real-time monitoring and warning of the driver's fatigue state, so that the driver can realize fatigue in time and take the initiative to stop and rest, improve road traffic safety, and protect the driver's life and property safety;

[0049] 2. The smart bracelet uses machine learning algorithms to build a fatigue driving habit model to predict the driver's ECG signal feature value SDNN, improve monitoring accuracy, and make fatigue status judgment more accurate. In addition, the smart bracelet is low-cost, comfortable, convenient and fast to use, and is easily accepted by users.

[0050] This warning device is low-cost, easy to use, stable and efficient. This warning method can effectively prevent motorcycle drivers from driving fatigued due to long-term continuous driving, monitor the driver's fatigue state in real time, and make corresponding warning measures for the driver in different fatigue states, thereby reducing the accident rate and further ensuring road traffic safety. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 It is a structural schematic diagram of an embodiment;

[0052] Figure 2 It is a flow chart of a timer-based fatigue driving warning method in an embodiment;

[0053] Figure 3 Schematic diagram of the process of the fatigue driving warning method based on electrocardiogram signal in the embodiment. DETAILED DESCRIPTION

[0054] The content of the present invention is further described below in conjunction with the drawings and embodiments, but the present invention is not limited thereto.

[0055] Example:

[0056] Reference Figure 1 A motorcycle fatigue driving warning device comprises a timer monitoring unit, an electrocardiogram fatigue monitoring unit and a reminder unit connected in sequence, the timer monitoring unit is provided with a single-chip microcomputer and a speed sensor connected to the single-chip microcomputer, the single-chip microcomputer is provided with a single-chip microcomputer control module and a timer, the electrocardiogram fatigue monitoring unit is provided with an electrocardiogram signal acquisition module, an electrocardiogram signal data processing module and a machine learning algorithm module connected in sequence, the reminder unit is provided with an electrically connected instrument panel background light module, a voice prompt module, a left handlebar vibration module, a left handlebar current stimulation module and a body shaking module, the single-chip microcomputer is installed under the instrument panel of the motorcycle, the speed sensor is installed on the side of the front wheel of the motorcycle, the electrocardiogram fatigue monitoring unit is placed in a smart bracelet worn by the motorcycle driver, the electrocardiogram fatigue monitoring unit is connected to the single-chip microcomputer control module via Bluetooth, and the single-chip microcomputer control module is electrically connected to the instrument panel background light module, the voice prompt module, the left handlebar vibration module, the left handlebar current stimulation module and the body shaking module via wires.

[0057] The single chip microcomputer model is AT89S52, the speed sensor model is EN150, and the voice prompt module uses a speaker.

[0058] A motorcycle fatigue driving warning method, comprising the above-mentioned motorcycle fatigue driving warning device, the method comprising the following steps:

[0059] like Figure 2 As shown, 1) Timer-based fatigue driving warning:

[0060] 1-1) After the motorcycle is started, the speed sensor installed on the motorcycle detects the speed of the vehicle. If the speed is higher than 3.5km / h, the timer of the timer monitoring unit starts counting and records the driving time;

[0061] 1-2) When the timer reaches 2 hours, the MCU control module controls the reminder unit to remind the driver of overtime driving, triggers the dashboard background light module to change color and the voice prompt module. If the speed sensor detects that the vehicle speed is continuously lower than 3.5 km / h for 10 minutes or more during the timer, the timer is reset and returns to step 1-1);

[0062] 1-3) When the timer reaches 3 hours, the MCU control module controls the reminder unit to remind the driver of overtime driving, trigger the left handlebar vibration module and voice prompt module. If the speed sensor detects that the vehicle speed is continuously lower than 3.5 km / h for 15 minutes or more during the timer, the timer is reset and the process returns to step 1-1);

[0063] 1-4) When the timer reaches 4 hours, the single-chip control module controls the reminder unit to remind the driver of overtime driving, trigger the left handlebar current stimulation module and the voice prompt module. If the speed sensor detects that the vehicle speed is continuously lower than 3.5 km / h for 20 minutes or more during the timer, the timer is reset and the process returns to step 1-1);

[0064] 1-5) When the timer reaches 5 hours, it is determined that the motorcycle driver is driving for an excessively long time. The single-chip control module controls the reminder unit to remind the driver of the overtime driving, trigger the dashboard background light module to change color, the voice prompt module, the left handlebar vibration module, the left handlebar current stimulation module and the body shaking module. When the speed sensor detects that the vehicle speed is lower than 3.5 km / h, the combined reminder stops working temporarily. If the driver continues to drive the vehicle within 30 minutes, the combined reminder continues to work. If the driver stops to rest for 30 minutes, the combined reminder is completely turned off, and the timer is reset, and the process returns to step 1-1);

[0065] like Figure 3 As shown, 2) Fatigue driving warning based on ECG signal:

[0066] 2-1) The motorcycle driver uses a smart bracelet to collect the driver's ECG signal in real time, and uses the wavelet threshold denoising method to remove baseline drift, myoelectric interference and power frequency interference from the collected ECG signal;

[0067] 2-2) Use the Pan&Tompkins algorithm to detect the R wave of the ECG signal after noise removal. The detection algorithm steps of Pan&Tompkins are as follows:

[0068] 2-2-1) Initialize the signal threshold THR_SIGNAL and the noise threshold THR_NOISE. The initialization formulas of the signal threshold THR_SIGNAL and the noise threshold THR_NOISE are shown in formula (1) and formula (2):

[0069] (1),

[0070] (2),

[0071] Among them, THR_SIGNAL is the initial signal threshold, THR_NOISE is the initial noise threshold, pre_ecg is the ECG signal value after denoising, and fs is the sampling frequency of the signal;

[0072] 2-2-2) Compare the current ECG signal maximum value MAX_PEAK with the signal threshold THR_SIGNAL. ， Then MAX_PEAK is determined as the peak value of the R wave, the position of the R wave at this time is recorded, and the signal level value is updated at the same time. The calculation formula for updating the signal level value is shown in formula (3):

[0073] (3),

[0074] Among them, SIGNAL_LEVEL is the signal level value, and MAX_PEAK is the maximum value of the current ECG signal;

[0075] if , then MAX_PEAK is determined as noise, and the noise level value is updated at the same time. The calculation formula for updating the noise level value is shown in formula (4):

[0076] (4),

[0077] Among them, NOISE_LEVEL is the noise level value, and MAX_PEAK is the maximum value of the current ECG signal;

[0078] 2-2-3) After the five R wave peaks are determined using the signal threshold THR_SIGNAL and the noise threshold THR_NOISE, the signal threshold THR_SIGNAL and the noise threshold THR_NOISE are updated. The calculation formulas for updating the signal threshold THR_SIGNAL and the noise threshold THR_NOISE are shown in formula (5) and formula (6):

[0079] (5),

[0080] (6),

[0081] Among them, THR_SIGNAL is the signal threshold, THR_NOISE is the noise threshold, NOISE_LEVEL is the noise level value, and SIGNAL_LEVEL is the signal level value;

[0082] 2-2-4) Repeat steps 2-2-2) and 2-2-3) to detect all R wave peaks in the ECG signal;

[0083] 2-3) Calculate each RR interval based on the time difference between adjacent R waves. The calculation formula of the RR interval is shown in formula (7):

[0084] (7),

[0085] Among them, RR_period is the RR interval of the ECG signal, RR_time is the interval between two adjacent R waves, and FS is the sampling frequency of the signal;

[0086] 2-4) Set 100 memory cells in the smart bracelet, and each memory cell is used to store the most recent 100 RR interval values;

[0087] 2-5) Calculate the ECG signal characteristic value SDNN based on the values ​​of N consecutive RR intervals. The calculation formula of SDNN value is shown in formula (8) and formula (9):

[0088] (8),

[0089] (9),

[0090] in, is the value of the i-th RR interval, is the mean of N consecutive RR intervals, SDNN represents the standard deviation of the RR interval;

[0091] 2-6) The single-chip control module transmits the driver’s driving time to the smart bracelet via Bluetooth;

[0092] 2-7) The smart bracelet collects the ECG signal feature value SDNN, driving time and fatigue status of the driver while driving. The machine learning algorithm module uses the ECG signal feature value SDNN, driving time and fatigue status as input variables, and calls the machine learning algorithm to build a fatigue driving habit model. The fatigue driving habit model predicts the driver's ECG signal feature value SDNN, and accurately calculates and corrects the predicted value through the actual measurement value, so that the predicted value is infinitely close to the true value;

[0093] 2-8) Compare the calculated and predicted ECG signal feature value SDNN with the three preset thresholds representing different fatigue levels when α is 150ms, β is 200ms, and γ is 250ms: When the device determines that the driver's fatigue state is mild fatigue, the ECG signal fatigue monitoring unit transmits the fatigue state information to the single-chip control module via Bluetooth, and the single-chip control module controls the reminder unit to remind fatigue driving, trigger the dashboard background light module to change color, and the voice prompt module to prompt. In this example, the speaker on the motorcycle plays: You are mildly fatigued, please stop and rest for 10 minutes or more; when When the device determines that the driver's fatigue state is moderate fatigue, the ECG signal fatigue monitoring unit transmits the fatigue state information to the single-chip control module via Bluetooth, and the single-chip control module controls the reminder unit to give a fatigue driving reminder, triggering the left handlebar vibration module and the voice prompt module to prompt. In this example, the speaker on the motorcycle plays: You are moderately fatigued, please stop and rest for 15 minutes or more; when When the device determines that the driver's fatigue state is severe fatigue, the ECG signal fatigue monitoring unit transmits the fatigue state information to the single-chip control module via Bluetooth. The single-chip control module controls the reminder unit to give a fatigue driving reminder, triggering the left handlebar current stimulation module and the voice prompt module to prompt. In this example, the speaker on the motorcycle plays: You are severely fatigued, please stop and rest for 20 minutes or more.

[0094] In this example, after the motorcycle is started, when the timer reaches 4 hours in total, the single-chip control module in the fatigue driving warning device controls the reminder unit to give an overtime driving reminder, triggers the left handlebar current stimulation and plays a voice reminder, and the speaker on the motorcycle plays: You have been driving for a total of 4 hours, please stop and rest for 20 minutes or more; when the timer reaches 4 and a half hours in total, if the ECG signal fatigue monitoring unit detects that the driver is in a state of moderate fatigue, the ECG signal fatigue monitoring unit transmits the fatigue status information to the single-chip control module via Bluetooth, and the single-chip control module controls the reminder unit to give a fatigue driving reminder, triggers the left handlebar to vibrate and plays a voice reminder, and the speaker on the motorcycle plays: You are moderately fatigued, please stop and rest for 15 minutes or more.

Claims

1. A motorcycle fatigue driving warning device, characterized in that: It includes a timer monitoring unit, an electrocardiogram fatigue monitoring unit and a reminder unit connected in sequence, the timer monitoring unit is provided with a single-chip microcomputer and a speed sensor connected to the single-chip microcomputer, the single-chip microcomputer is provided with a single-chip microcomputer control module and a timer, the electrocardiogram fatigue monitoring unit is provided with an electrocardiogram signal acquisition module, an electrocardiogram signal data processing module and a machine learning algorithm module connected in sequence, the reminder unit is provided with an instrument panel background light module, a voice prompt module, a left handlebar vibration module, a left handlebar current stimulation module and a body shaking module which are electrically connected, the single-chip microcomputer is installed under the instrument panel of the motorcycle, the speed sensor is installed on the side of the front wheel of the motorcycle, the electrocardiogram fatigue monitoring unit is placed in a smart bracelet worn by the motorcycle driver, the electrocardiogram fatigue monitoring unit is connected to the single-chip microcomputer control module via Bluetooth, and the single-chip microcomputer control module is electrically connected to the instrument panel background light module, the voice prompt module, the left handlebar vibration module, the left handlebar current stimulation module and the body shaking module via wires.

2. A motorcycle fatigue driving warning method, comprising the motorcycle fatigue driving warning device according to claim 1, characterized in that: The method comprises the following steps: 1) Timer-based fatigue driving warning: 1-1) After the motorcycle is started, the speed sensor installed on the motorcycle detects the speed of the vehicle. If the speed is higher than 3.5km / h, the timer of the timer monitoring unit starts counting and records the driving time; 1-2) When the timer reaches 2 hours, the MCU control module controls the reminder unit to remind the driver of overtime driving, triggers the dashboard background light module to change color and the voice prompt module. If the speed sensor detects that the vehicle speed is continuously lower than 3.5 km / h for 10 minutes or more during the timer, the timer is reset and returns to step 1-1); 1-3) When the timer reaches 3 hours, the MCU control module controls the reminder unit to remind the driver of overtime driving, trigger the left handlebar vibration module and voice prompt module. If the speed sensor detects that the vehicle speed is continuously lower than 3.5 km / h for 15 minutes or more during the timer, the timer is reset and the process returns to step 1-1); 1-4) When the timer reaches 4 hours, the single-chip control module controls the reminder unit to remind the driver of overtime driving, trigger the left handlebar current stimulation module and the voice prompt module. If the speed sensor detects that the vehicle speed is continuously lower than 3.5 km / h for 20 minutes or more during the timer, the timer is reset and the process returns to step 1-1); 1-5) When the timer reaches 5 hours, it is determined that the motorcycle driver is driving for an excessively long time. The single-chip control module controls the reminder unit to remind the driver of the overtime driving, trigger the dashboard background light module to change color, the voice prompt module, the left handlebar vibration module, the left handlebar current stimulation module and the body shaking module. When the speed sensor detects that the vehicle speed is lower than 3.5 km / h, the combined reminder stops working temporarily. If the driver continues to drive the vehicle within 30 minutes, the combined reminder continues to work. If the driver stops to rest for 30 minutes, the combined reminder is completely turned off, and the timer is reset, and the process returns to step 1-1); 2) Fatigue driving warning based on ECG signals: 2-1) The motorcycle driver uses a smart bracelet to collect the driver's ECG signal in real time, and uses the wavelet threshold denoising method to remove baseline drift, myoelectric interference and power frequency interference from the collected ECG signal; 2-2) Use the Pan&Tompkins algorithm to detect the R wave of the ECG signal after noise removal. The detection algorithm steps of Pan&Tompkins are as follows: 2-2-1) Initialize the signal threshold THR_SIGNAL and the noise threshold THR_NOISE. The initialization formulas of the signal threshold THR_SIGNAL and the noise threshold THR_NOISE are shown in formula (1) and formula (2): （1）， （2）， Among them, THR_SIGNAL is the initial signal threshold, THR_NOISE is the initial noise threshold, pre_ecg is the ECG signal value after denoising, and fs is the sampling frequency of the signal; 2-2-2) Compare the current ECG signal maximum value MAX_PEAK with the signal threshold THR_SIGNAL. , then MAX_PEAK is determined as the peak value of the R wave, the position of the R wave at this time is recorded, and the signal level value is updated at the same time. The calculation formula for updating the signal level value is shown in formula (3): （3）， Among them, SIGNAL_LEVEL is the signal level value, and MAX_PEAK is the maximum value of the current ECG signal; if , then MAX_PEAK is determined as noise, and the noise level value is updated at the same time. The calculation formula for updating the noise level value is shown in formula (4): （4）， Among them, NOISE_LEVEL is the noise level value, and MAX_PEAK is the maximum value of the current ECG signal; 2-2-3) After the five R wave peaks are determined using the signal threshold THR_SIGNAL and the noise threshold THR_NOISE, the signal threshold THR_SIGNAL and the noise threshold THR_NOISE are updated. The calculation formulas for updating the signal threshold THR_SIGNAL and the noise threshold THR_NOISE are shown in formula (5) and formula (6): （5）， （6）， Among them, THR_SIGNAL is the signal threshold, THR_NOISE is the noise threshold, NOISE_LEVEL is the noise level value, and SIGNAL_LEVEL is the signal level value; 2-2-4) Repeat steps 2-2-2) and 2-2-3) to detect all R wave peaks in the ECG signal; 2-3) Calculate each RR interval based on the time difference between adjacent R waves. The calculation formula of the RR interval is shown in formula (7): （7）， Among them, RR_period is the RR interval of the ECG signal, RR_time is the interval between two adjacent R waves, and FS is the sampling frequency of the signal; 2-4) Set 100 memory cells in the smart bracelet, and each memory cell is used to store the most recent 100 RR interval values; 2-5) Calculate the ECG signal characteristic value SDNN based on the values ​​of N consecutive RR intervals. The calculation formula of SDNN value is shown in formula (8) and formula (9): （8）， （9）， in, is the value of the i-th RR interval, is the mean of N consecutive RR intervals, SDNN represents the standard deviation of the RR interval; 2-6) The single-chip control module transmits the driver’s driving time to the smart bracelet via Bluetooth; 2-7) The smart bracelet collects the ECG signal feature value SDNN, driving time and fatigue status of the driver while driving. The machine learning algorithm module uses the ECG signal feature value SDNN, driving time and fatigue status as input variables, and calls the machine learning algorithm to build a fatigue driving habit model. The fatigue driving habit model predicts the driver's ECG signal feature value SDNN, and accurately calculates and corrects the predicted value through the actual measurement value, so that the predicted value is infinitely close to the true value; 2-8) Compare the calculated and predicted ECG signal feature value SDNN with the three preset thresholds representing different fatigue levels when α is 150ms, β is 200ms, and γ is 250ms: When the device determines that the driver's fatigue state is mild fatigue, the ECG signal fatigue monitoring unit transmits the fatigue state information to the single-chip control module via Bluetooth, and the single-chip control module controls the reminder unit to remind fatigue driving, trigger the instrument panel background light module to change color and the voice prompt module; when When the device determines that the driver's fatigue state is moderate fatigue, the ECG signal fatigue monitoring unit transmits the fatigue state information to the single-chip control module via Bluetooth, and the single-chip control module controls the reminder unit to remind the driver of fatigue driving, triggering the left handlebar vibration module and voice prompt module; when When the device determines that the driver's fatigue state is severe fatigue, the ECG signal fatigue monitoring unit transmits the fatigue state information to the single-chip control module via Bluetooth. The single-chip control module controls the reminder unit to provide fatigue driving reminders, triggering the left handlebar current stimulation module and voice prompt module.

Images