Multi-source information fused high-speed rail arrival reminding system and method

Abstract

The invention relates to the technical field of arrival reminding, in particular to a multi-source information fused high-speed rail arrival reminding system and method. A positioning module is used for acquiring a current high-speed rail position; the IMU module comprises a first IMU element and second IMU elements arranged on the two sides of the first IMU element respectively, the first IMU element is used for obtaining the angular velocity and the acceleration of a human body in a three-dimensional space, and the second IMU elements are arranged in a tumbler mode and used for obtaining the acceleration of a high-speed rail; the electrode patch module can be in contact with human skin and is used for collecting and transmitting electric signals; the voice recognition module is used for collecting high-speed railway notification voice information; the voice warning module is used for sending warning voice to remind passengers; the clock module is used for acquiring time and date and setting a clock reminding time point; the power supply module is used for supplying power to the MCU module. The acceleration in the advancing direction of the high-speed rail can be obtained through the IMU module, the running state of the high-speed rail is accurately judged to accurately predict the arrival time of the high-speed rail, and passengers are prevented from being awakened before time is reminded in advance.

Description

Technical Field

[0001] The present invention relates to the technical field of arrival reminder, and in particular to a high-speed rail arrival reminder system and method integrating multi-source information. Background Art

[0002] High-speed rail travel has become one of the important ways for modern people to travel. The development of an arrival reminder system can make passengers feel more at ease during their journey and avoid missing the time to get off the train, thereby improving the user's travel experience. Especially for those passengers who are prone to falling asleep or focusing on other matters during their journey, the arrival reminder system can effectively prevent them from missing their destination due to not noticing the station information, reducing the inconvenience and safety risks caused by missing the station. For railway operators, the arrival reminder system helps reduce train delays and station congestion caused by passengers missing the station, thereby improving overall operational efficiency.

[0003] The existing high-speed rail arrival reminder systems and implementation solutions mainly include the following: (1) Reminder system based on official railway APP or website: users can log in to their account through the official railway APP or website, find the train number for which they need to set a reminder, click the "Arrival Reminder" button, and the system will send a reminder message before the time set by the user; (2) Arrival reminder function based on smartphones / watches: add the departure and arrival time of the high-speed rail through the calendar application of smartphones / watches, and the system will automatically send a reminder message to the watch according to the time set by the user. (3) Intelligent reminder system based on single-chip microcomputer: This system is usually composed of single-chip microcomputer, LCD display circuit, Bluetooth module circuit, clock circuit and other parts. Users can update the current time and set the time point through the mobile phone Bluetooth APP. When the set time point is reached, the system will remind the user through vibration motor and other methods.

[0004] When the high-speed train is delayed due to temperature, rainfall, etc., the reminder system that only relies on the arrival time setting will undoubtedly wake up passengers long before the station arrives; when the high-speed train is delayed and stops in a tunnel or other underground project, the reminder system that relies on the positioning setting cannot accurately predict the arrival time and cannot wake up passengers before arriving at the station. In addition, the existing wake-up system mainly relies on sound, which is easily affected by the noisy environment of the high-speed train environment, and it is difficult to wake up passengers who are deeply asleep. Summary of the invention

[0005] The present invention aims to provide a high-speed rail arrival reminder system and method integrating multi-source information, aiming to solve the problem that the existing reminder system cannot accurately predict the arrival time.

[0006] The specific technical solutions are as follows:

[0007] The present invention provides a high-speed rail arrival reminder system integrating multi-source information, comprising a wearable carrier, a mobile terminal and a reminder device, wherein the wearable carrier is worn on a passenger, and the reminder device is arranged on the wearable carrier, and is characterized in that the reminder device comprises an MCU module, a positioning module, an IMU module, an electrode patch module, a voice recognition module, a voice warning module, a clock module and a power module; the positioning module, the IMU module, the voice recognition module and the clock module are all communicatively connected with the MCU module; the electrode patch module, the voice warning module and the power module are electrically connected with the MCU module;

[0008] The MCU module is communicatively connected to the mobile terminal; the positioning module is used to obtain the current position of the high-speed rail; the IMU module includes a first IMU element and a second IMU element disposed on both sides of the first IMU element, the first IMU element is used to obtain the angular velocity and acceleration of the human body in three-dimensional space, and the second IMU element is arranged in a tumbler-like manner to obtain the acceleration of the high-speed rail; the electrode patch module is arranged on a wearable carrier and can contact with human skin for collecting and transmitting electrical signals; the voice recognition module is used to collect high-speed rail notification voice information; the voice warning module is used to issue a warning voice to remind passengers; the clock module is used to obtain time and date, and set the clock reminder time point; the power module is used to power the MCU module.

[0009] Optionally, the MCU module has built-in speech recognition program, acceleration calculation program, running speed calculation program, arrival time prediction program and sleep state evaluation program, which can predict the time when the high-speed train arrives at the destination station.

[0010] Optionally, the high-speed rail arrival reminder system that integrates multi-source information also includes a Bluetooth module, and the MCU module is communicatively connected to the mobile terminal via the Bluetooth module; and / or, the high-speed rail arrival reminder system that integrates multi-source information also includes a WIFI module, and the WIFI module is communicatively connected to the MCU module, and the WIFI module is used to connect to the hotspot of the mobile terminal to update the map information built into the MCU module.

[0011] The present invention also provides a high-speed rail arrival reminder method integrating multi-source information, which uses the high-speed rail arrival reminder system integrating multi-source information as described above to obtain an accurate arrival time at a destination station, comprising the following steps:

[0012] S1: Use a mobile terminal to obtain site map information, set the destination site and advance reminder time, where the site map information includes site information along the route, arrival time, longitude and latitude corresponding to the site, and longitude and latitude of map points along the route;

[0013] S2, turn on the power module, and update the set advance reminder time and built-in map information of the MCU module according to the site map information, the destination site and the advance reminder time;

[0014] S4: Start the clock module, calculate the clock reminder time point according to the current time point and the arrival time of the destination station, and combine the set advance reminder time;

[0015] S5: Based on the clock reminder time point, the current high-speed rail coordinates, forward direction and running speed are obtained through the positioning module and the MCU module, and the current running status of the high-speed rail is determined;

[0016] S6: predicting the arrival time of the high-speed rail at the destination station according to the running status of the high-speed rail, and updating the clock reminder time point according to the predicted arrival time;

[0017] S7: When the clock reminder time point updated in S6 is reached, the voice warning module is started to emit a warning sound to remind the passenger that the station is about to be reached, and half of the electrode patches in the electrode patch module are started synchronously to simulate acupuncture to stimulate the human body;

[0018] S8: Use the first IMU element and the other half of the electrode patch in the electrode patch module to detect the current human body state of the passenger, and use the sleep state assessment program to assess whether the passenger is awake; if it is detected that the passenger is not awake, adjust the current of the electrode patch until it is detected that the passenger is awake or passively shut down, and automatically shut down the power module; wherein: the maximum current of the electrode patch does not exceed 10mA.

[0019] Optionally, the S5 includes:

[0020] Start the positioning module and two second IMU elements 10 minutes before the clock reminder time;

[0021] The positioning module receives satellite radio signals through the antenna for M consecutive times, and obtains the current high-speed rail coordinates, forward direction and running speed through the running speed calculation program of the MCU module, where: M is the set number of thresholds;

[0022] Use the two second IMU elements to obtain the linear acceleration along the X, Y, and Z axes, the rotational angular velocity along the X, Y, and Z axes, and the magnetic field strength along the X, Y, and Z axes of the current IMU module, and calculate the current high-speed rail acceleration through the acceleration solution program of the MCU module;

[0023] The current operating status of the high-speed rail is determined based on the operating speed and acceleration of the high-speed rail, specifically:

[0024] When the high-speed rail's running speed is 0 and the acceleration is -0.1m / s to 0.1m / s, the high-speed rail is in a temporary parking stage; when the high-speed rail's running speed is not 0 and the acceleration is -0.1m / s to 0.1m / s, the high-speed rail is in a uniform speed running stage; when the high-speed rail's running speed is not 0, the acceleration is a negative value and the absolute value is greater than 0.1m / s, the high-speed rail is in a deceleration running stage.

[0025] Optionally, the acceleration solution procedure in S5 includes:

[0026] ① Use the magnetometer of the second IMU element to measure the component of the Earth's magnetic field on the horizontal plane [m x ,m y ], the azimuth angle θ of the IMU module is calculated based on the component of the geomagnetic field on the horizontal plane imu , the specific calculation formula is as follows:

[0027]

[0028] ② The acceleration component [a x ,a y ,a z ] Estimate the pitch angle α and roll angle β of the second IMU element;

[0029] ③ According to the pitch angle α and roll angle β of the second IMU element, the azimuth angle of the second IMU element is corrected to obtain the corrected azimuth angle θ imu ′, the specific calculation formula is as follows:

[0030]

[0031] ④ Calculate the acceleration a in the forward direction of the high-speed rail according to the corrected azimuth. The specific calculation formula is as follows:

[0032] a=a x ·cos(Δθ)+a y sin(Δθ);

[0033] Where: Δθ is the angle between the second IMU element and the direction of the high-speed rail, Δθ=θ-θ′ imu .

[0034] Optionally, in S5,

[0035] When the high-speed railway passes through an underground project, if the distance between the coordinates obtained by the positioning module before and after the underground project is at least greater than the set distance threshold, it means that the high-speed railway is running, and the positioning module is kept turned on until M consecutive satellite signals can be obtained or 1 minute before the set reminder time point or 10 minutes left before the arrival time is recognized by voice recognition. At this time, the running status of the high-speed railway is determined by combining the IMU module and the acceleration solution program;

[0036] If the coordinates obtained by the positioning module before and after crossing the underground project are not much different, the azimuth of the IMU module is used as the running direction of the high-speed rail. When the acceleration of the current forward direction of the high-speed rail obtained by the IMU module is zero, it is judged that the high-speed rail is in a stopped state, and other modules except the MCU module and the IMU module are suspended; when the acceleration of the current forward direction of the high-speed rail is not zero, the high-speed rail is running until the positioning module can obtain M consecutive satellite signals, and the calculated azimuth of the running speed solver is used as the running direction of the high-speed rail.

[0037] Optionally, the steps between S2 and S4 further include:

[0038] S3: pre-training the speech recognition program in the MCU module according to the keywords to obtain a keyword recognition model, wherein the keywords include site names and key time point words;

[0039] The S6 includes:

[0040] S6.1. Calculate the distance between the current high-speed rail coordinates and the destination station based on the vectorized data of the map information built into the MCU;

[0041] S6.2, calculate the predicted arrival time of the high-speed rail to the destination station through the arrival time prediction program of the MCU module according to the distance between the high-speed rail coordinates and the destination station, the forward direction and the running speed;

[0042] S6.3. Update the clock reminder time point according to the predicted arrival time, specifically:

[0043] When the predicted arrival time is delayed by more than 10 minutes, the clock reminder time point is updated by delaying the clock reminder time point by the time that exceeds the time limit; and the positioning module is set to obtain the high-speed rail coordinates every 10 minutes, and the process returns to S6.1;

[0044] When the clock reminder time point is within 10 minutes of the predicted arrival time, the voice recognition module is started 11 minutes before the predicted arrival at the station, and the voice from the 11th minute to the 8th minute of the predicted arrival at the station is recorded in real time. The voice is transmitted to the MCU module, and the pre-trained keyword recognition model is used to recognize the keyword, and the time point when the keyword appears is recorded; then the remaining arrival time is calculated based on the average of the predicted arrival time and the time point when the keyword appears, and the clock reminder time point is updated based on the remaining arrival time;

[0045] Within 2 minutes before the clock reminder time point, the positioning module is set to obtain the high-speed rail coordinates twice every 10 seconds on average, and return to S6.1;

[0046] When the high-speed rail is in operation, if the current coordinates of the high-speed rail cannot be obtained within 1 minute before the clock reminder time, the reminder time originally set by the clock shall prevail; if the set reminder time is within 10 minutes of arriving at the station, the clock reminder time shall be updated with the time point that appears in the keyword recognized by the voice recognition module.

[0047] 9. The high-speed rail arrival reminder method integrating multi-source information according to claim 8, characterized in that the arrival time prediction program includes:

[0048] ① Calculate the remaining distance from the current high-speed rail coordinates to the destination station, specifically:

[0049] Calculate the current high-speed rail coordinates and the coordinates of the nearest line network point in the direction of travel under the current high-speed rail coordinates;

[0050] Divide the high-speed rail line into multiple sections in a rasterized manner, mark the longitude and latitude coordinates of the section points and the actual lengths of adjacent points;

[0051] Calculate the distances of each segment between the nearest network point and the destination station, and add up the distances of each segment to get the remaining distance from the high-speed rail to the destination station;

[0052] ② Establish a remaining distance prediction model for starting deceleration and a remaining arrival time prediction model for starting deceleration, specifically:

[0053] Collect in advance at least 100 sets of high-speed rail data with different speeds when the high-speed rail arrives at the station on time, where the high-speed rail data includes the average running speed and average acceleration within 10 minutes before the train decelerates to enter the station, as well as the remaining distance from the station and the remaining arrival time when the train starts to decelerate;

[0054] The high-speed rail data is normalized using the minimum-maximum normalization method. Then, based on the BP neural network model, the average running speed and average acceleration are used as input parameters, and the remaining distance from the station when the train starts to decelerate and the remaining arrival time are used as output parameters. The above high-speed rail data is learned to obtain the remaining distance prediction model for starting deceleration and the remaining arrival time prediction model for starting deceleration.

[0055] ③ Calculate the remaining arrival time based on the current high-speed rail operation status, specifically:

[0056] If the high-speed rail is in a temporary stop phase, all modules except the MCU module and the IMU module will be suspended to temporarily delay waking up passengers;

[0057] If the high-speed rail is in a constant speed stage, the specific steps for calculating the remaining arrival time are as follows:

[0058] i. Input the average running speed and average acceleration in this stage into the remaining distance prediction model at the start of deceleration and the remaining arrival time at the start of deceleration, and output the remaining distance at the start of deceleration and the running time of the deceleration section;

[0059] ii. The remaining distance from the high-speed train to the destination station minus the remaining distance when deceleration begins is used to obtain the remaining distance of the uniform speed running section;

[0060] iii. Calculate the running time of the remaining uniform running segment according to the uniform running speed;

[0061] iv. Add the running time of the deceleration phase and the running time of the constant speed phase to obtain the remaining arrival time;

[0062] v. Calculate the high-speed rail arrival time according to the current time of the clock module, and update the clock reminder time;

[0063] vi. Recalculate i to v every 30 seconds;

[0064] If the high-speed rail is in the deceleration stage, the specific steps for calculating the remaining arrival time are as follows:

[0065] i. The running speed v measured at the i-th time point in the deceleration phase i , acceleration a i , the remaining distance from the high-speed rail to the destination d i , the theoretical remaining arrival time t can be calculated i , the specific calculation formula is as follows:

[0066]

[0067] ii. Verify whether the theoretical arrival time is consistent with the actual arrival time based on the remaining distance. If the theoretical remaining distance d i 'Greater than the actual remaining distance d i , the actual arrival time is considered to be less than the theoretical remaining arrival time, and the acceleration and arrival time are recalculated, and the recalculated arrival time is used as the actual arrival time; if the theoretical remaining distance d i 'Equal to the actual remaining distance d i , then the calculated theoretical remaining arrival time is taken as the actual arrival time;

[0068] Where: Theoretical remaining distance d i The calculation formula of ' is as follows:

[0069]

[0070] The formulas for recalculating acceleration and arrival time are as follows:

[0071]

[0072] iii. After obtaining the arrival time point, use the clock module to recalculate the clock reminder time point.

[0073] Optionally, the sleep state assessment procedure includes:

[0074] ① Establish the IMU sleep state assessment model and the electrode patch sleep state assessment model, specifically:

[0075] Establishing the IMU sleep state assessment model includes the following steps:

[0076] i. Collect at least 300 sets of signal data of the first IMU element in different states of the human head, and use professional instruments to evaluate the state of the human body, where different states of the human head include wakefulness, light sleep and deep sleep;

[0077] ii. Then the signal data of the first IMU element is subjected to low-pass filtering and denoising, and four key characteristic parameter data are extracted. The IMU data under the human body state is evaluated by professional instruments to establish three label data sets; among which: the four key features include the standard deviation of acceleration a std , the root mean square of acceleration a rms , angular velocity change rate Δω and acceleration signal periodicity P;

[0078] iii. Then, a typical and reliable classification algorithm, the support vector machine algorithm, is used to learn and analyze the data set, and the sum parameter and penalty parameter of the support vector machine are optimized using the grid search algorithm until the loss function is minimized, thereby establishing an IMU sleep state assessment model;

[0079] The steps to establish the sleep state assessment model of the electrode patch include the following:

[0080] i. Collect at least 300 sets of electromyographic signal data of the electrode patch module in different states of the human head, and use a human sleep state measuring instrument to evaluate the state of the human body;

[0081] ii. Then, the signal data of the electrode patch is amplified, and its root mean square characteristic data is extracted. The characteristic data of the electromyographic signal under the human body state is evaluated according to the human sleep state measuring instrument, and three label data sets are established;

[0082] iii. Then, the support vector machine algorithm is used to learn and analyze the data set, and the grid search algorithm is used to optimize the sum parameter and penalty parameter of the support vector machine until the loss function is minimized, thereby establishing a sleep state assessment model of the electrode patch;

[0083] ② Use the IMU sleep state assessment model and the electrode patch sleep state assessment model to assess the passenger's sleep state, specifically:

[0084] The electrode patch module is used to collect the passenger's electromyographic signal. Specifically, the electrode patch is placed close to the skin of the human neck to collect the change in skin surface potential and the weak electrical signal generated by muscle activity, i.e., electromyographic signal. The electrode patch obtains the electromyographic signal with a time window of 1 second at an interval of 2 seconds.

[0085] The signal data of the passenger's head is measured by using the first IMU element, wherein: the first IMU element is set to detect 5 times every 5 seconds, and then detect again after an interval of 5 seconds;

[0086] The passenger's electromyographic signal and head signal data are respectively transmitted to the sleep state evaluation model of the IMU of the MCU module and the sleep state evaluation model of the electrode patch to evaluate the passenger's sleep state;

[0087] If the evaluation results of the IMU's sleep state evaluation model and the electrode patch's sleep state evaluation model are both awake, the power module is automatically turned off;

[0088] If the evaluation results of the IMU sleep state evaluation model and the electrode patch sleep state evaluation model are both light sleep, the initial current of the electrode patch is adjusted to 0.5-1mA, and the stimulation frequency is a medium frequency, low amplitude pulse wave of 20-50Hz;

[0089] If the evaluation results of the IMU sleep state evaluation model and the electrode patch sleep state evaluation model are both deep sleep state, the initial current of the electrode patch is adjusted to 1-2mA, and the stimulation frequency is adjusted to a low-frequency, high-amplitude pulse wave of 50-100Hz;

[0090] After 5 to 10 seconds of continuous stimulation, if the passenger is awake, the stimulation will be stopped; if the passenger is not awake, the current will be adjusted according to the changes in the passenger's electromyographic signals or small movements. Specifically, the current amplitude is increased by 0.5 mA every 5 seconds; if there is no response after 30 to 60 seconds of stimulation, the current will be directly increased to the maximum current to wake up; the stimulation will be stopped after 5 seconds.

[0091] The present invention has the following beneficial effects:

[0092] The second IMU element in the IMU module of the present invention is set up in a tumbler-type structure. When the high-speed train is delayed and stops in a tunnel or other underground project, the acceleration in the direction of the high-speed train can be obtained through the IMU module, and the running status of the high-speed train can be accurately judged to accurately predict the arrival time of the high-speed train, avoiding waking up passengers before the advance reminder time;

[0093] The present invention uses an IMU module and an electrode patch module to monitor the state of the human body, and when the initial awakening method is not enough to wake up the passenger, it can detect that the human body is still asleep and feedback to the MCU module. The MUC module can gradually increase the strength of the electrode patch module until the passenger is awakened;

[0094] The arrival reminder system of the present invention is worn by passengers through a wearable carrier to timely remind passengers of their arrival, thereby reducing the pressure on users to remember intermediate stops and improving travel comfort;

[0095] Passengers can be woken up by voice and current stimulation from electrode patches, which can reduce the noise impact on people around them. The main reminder method is physical touch stimulation, which can help deep sleepers gradually wake up before arriving at the station and avoid abrupt external interference. It can not only effectively prevent missed stops, but also relieve travel fatigue and significantly improve user experience and practicality.

[0096] In addition to the above-described objects, features and advantages, the present invention has other objects, features and advantages. The present invention will be further described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0097] The drawings constituting a part of this application are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0098] Figure 1 is a schematic diagram of the structure of a wearable carrier and a reminder device in an embodiment of the present invention;

[0099] Figure 2 is a schematic diagram of the circumcision of the wearable carrier and the reminder device in an embodiment of the present invention;

[0100] Figure 3 is a schematic diagram of the installation of the second IMU element in an embodiment of the present invention;

[0101] Figure 4 It is a flow chart of a high-speed rail arrival reminder system method integrating multi-source information in an embodiment of the present invention.

[0102] Description of Figure Numbers:

[0103] 1. Wearable carrier, 1.1. Inner ring, 1.2. Outer ring, 1.3. Shock-absorbing cotton block, 2. Reminder device, 2.1. MCU module + WIFI module + Bluetooth module, 2.2. Positioning module, 2.3. IMU module, 2.3.1. First IMU element, 2.3.2. Second IMU element, 2.3.3. Counterweight, 2.3.4. Fixed rod, 2.4. Electrode patch module, 2.5. Voice recognition module + voice warning module, 2.6. Clock module, 2.7. Power module. DETAILED DESCRIPTION

[0104] In order to make the purpose, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings. Several embodiments of the present invention are shown in the accompanying drawings. However, the present invention can be implemented in many different forms and is not limited to the embodiments described herein.

[0105] In the present invention, unless otherwise clearly specified and limited, the terms "installed", "connected", "connected", "fixed" and the like 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 or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal connection of two elements. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to the specific circumstances. The terms "first" and "second" are used for descriptive purposes only and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features.

[0106] In the present invention, unless otherwise clearly specified and limited, the first feature being "above" or "below" the second feature may include the first and second features being in direct contact, or may include the first and second features being in contact not directly but through another feature between them. Moreover, the first feature being "above", "above" and "above" the second feature includes the first feature being directly above and obliquely above the second feature, or simply indicates that the first feature is higher in level than the second feature. The first feature being "below", "below" and "below" the second feature includes the first feature being directly below and obliquely below the second feature, or simply indicates that the first feature is lower in level than the second feature. The terms "vertical", "horizontal", "left", "right", "above", "below" and similar expressions are for illustrative purposes only, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as limiting the present invention.

[0107] The embodiments of the present invention are described in detail below with reference to the accompanying drawings, but the present invention can be implemented in many different ways as defined and covered by the claims.

[0108] Example 1

[0109] like Figure 1 and Figure 2 As shown, this embodiment provides a high-speed rail arrival reminder system integrating multi-source information, including a wearable carrier 1, a mobile terminal and a reminder device 2, the wearable carrier 1 is worn on the passenger, the reminder device 2 is arranged on the wearable carrier 1, and the reminder device 2 includes an MCU module, a positioning module 2.2, an IMU module 2.3, an electrode patch module 2.4, a voice recognition module, a voice warning module, a clock module 2.6 and a power module 2.7; the positioning module 2.2, the IMU module 2.3, the voice recognition module and the clock module 2.6 are all connected to the MCU module in communication; the electrode patch module 2.4, the voice warning module and the power module 2.7 are electrically connected to the MCU module; the MCU module is connected to the mobile terminal in communication; the positioning module 2.2 is used to obtain the current The IMU module 2.3 includes a first IMU element 2.3.1 and a second IMU element 2.3.2 disposed on both sides of the first IMU element 2.3.1. The first IMU element 2.3.1 is used to obtain the angular velocity and acceleration of the human body in three-dimensional space. The second IMU element 2.3.2 is arranged in a tumbler-like manner to obtain the acceleration of the high-speed rail. The electrode patch module 2.4 is arranged on the wearable carrier 1 and can contact the human skin to collect and transmit electrical signals. The voice recognition module is used to collect high-speed rail notification voice information. The voice warning module is used to issue a warning voice to remind passengers. The clock module 2.6 is used to obtain the time and date, and set the clock reminder time point. The power module 2.7 is used to power the MCU module.

[0110] In this embodiment, the mobile carrier is a U-shaped pillow, which includes a filler, an inner ring 1.1 and an outer ring 1.2. The inner ring 1.1 forms a filling space for the filler, and the outer ring 1.2 is wrapped outside the inner ring 1.1, and the patches of the electrode patch module 2.4 are all arranged between the inner ring 1.1 and the outer ring 1.2 of the U-shaped pillow. A card slot is formed on the U-shaped pillow, and the reminder device 2 is embedded in the U-shaped pillow through the card slot, and the surface is covered with a shock-absorbing cotton block 1.3 to avoid causing discomfort to the human body. The reminder device 2 is arranged in a three-layer structure in the direction away from the shock-absorbing cotton block 1.3. The layer close to the shock-absorbing cotton block 1.3 is the IMU module 2.3 and the clock module 2.5 and the positioning module 2.2 arranged at both ends of the IMU module; the middle layer is the MCU module + WIFI module + Bluetooth module 2.1, and the layer away from the shock-absorbing cotton block 1.3 is the power module 2.7; and both ends of the three-layer structure are provided with a voice recognition module + voice warning module 2.5.

[0111] See also Figure 3The first IMU element 2.3.1 in the IMU module 2.3 is equipped with three single-axis accelerometers and three single-axis gyroscopes, which can directly measure the angular velocity and acceleration of an object in three-dimensional space; the second IMU element 2.3.2 adopts a tumbler-style setting, and the second IMU element 2.3.2 includes three single-axis accelerometers, three single-axis gyroscopes and a magnetometer. The IMU module 2.3 is communicated with the MCU control unit, specifically: the second IMU element 2.3.2 is set on the counterweight block 2.3.3, and the counterweight block 2.3.3 is a water drop-shaped structure with a light top and a medium bottom, and is rotatably set on the wearable carrier 1 through a fixed rod 2.3.4 to ensure that the acceleration of the second IMU element 2.3.2 in the z-axis direction must be the acceleration of gravity. When the high-speed rail is delayed and stops in a tunnel or other underground project, the acceleration in the forward direction of the high-speed rail can be obtained through the IMU module 2.3, and the operation status of the high-speed rail can be accurately judged to accurately predict the arrival time of the high-speed rail, so as to avoid waking up passengers before the advance reminder time.

[0112] The IMU module 2.3 and the electrode patch module 2.4 are used to monitor the human body state. When the initial awakening method is not enough to wake up the passenger, it can be detected that the human body is still asleep and fed back to the MCU module. The MUC module can gradually increase the strength of the electrode patch module 2.4 until the passenger is awakened. The electrode patch module 2.4 is made of conductive materials, carbon materials, etc., and can collect / transmit electrical signals. It is attached to the U-shaped pillow through the module patch, so as to directly contact the human neck skin. The module is distributed by at least 3 patches for collecting electrical signals and 3 patches for transmitting electrical signals, and is electrically connected to the MCU module through the line.

[0113] The high-speed rail arrival reminder system integrating multi-source information is worn on passengers through a wearable carrier 1 to timely remind passengers of arrival, thereby alleviating the pressure of users to remember intermediate stops and improving travel comfort.

[0114] Passengers can be woken up by voice and current stimulation from electrode patches, which can reduce the noise impact on people around them. The main reminder method is physical touch stimulation, which can help deep sleepers gradually wake up before arriving at the station and avoid abrupt external interference. It can not only effectively prevent missed stops, but also relieve travel fatigue and significantly improve user experience and practicality.

[0115] The voice recognition module is mainly composed of two microphones placed on the upper and lower sides of the instrument body. The sound collected by the microphones is transmitted to the MCU module. The voice recognition program of the MCU module recognizes the key site information and arrival time information. The voice recognition module is communicatively connected to the MCU module; the voice warning module is a miniature buzzer component, and the voice warning module is electrically connected to the MCU module.

[0116] Among them, the MCU module has built-in speech recognition program, acceleration calculation program, running speed calculation program, arrival time prediction program and sleep state evaluation program, which can predict the time when the high-speed rail arrives at the destination station. In this embodiment, the MCU module is a single-chip microcomputer, which is equipped with a USB interface and a type-C interface, which are respectively used to connect to the mobile terminal and charge. The single-chip microcomputer is embedded with a speech recognition program, an acceleration calculation program, a running speed calculation program, an arrival time prediction program and a sleep state evaluation program for predicting the arrival time of the high-speed rail.

[0117] Furthermore, the high-speed rail arrival reminder system integrating multi-source information also includes a Bluetooth module, and the MCU module is connected to the mobile terminal through the Bluetooth module. The Bluetooth module is composed of a Bluetooth wireless transceiver, a baseband controller, etc., and is used to receive the station information transmitted from the mobile terminal, transmit it to the MCU module to interpret the station name, arrival time, and longitude and latitude coordinates, and the Bluetooth module is connected to the MCU module.

[0118] Furthermore, the high-speed rail arrival reminder system integrating multi-source information also includes a WIFI module, which is connected to the MCU module in communication, and the WIFI module is used to connect to the hotspot of the mobile terminal to update the map information built into the MCU module. The WIFI module is composed of a Wifi chip, a transceiver, a wireless local area network (WLAN) transceiver, an antenna, etc., and is used to connect to the mobile phone hotspot for the MCU module to automatically update the internally stored high-speed rail route map and communicate with the MCU module.

[0119] In this embodiment, the clock module 2.6 is composed of a counter, a memory, a stable oscillator, etc., which is used to provide accurate time and date, transmit it to the MCU module, and update the time in real time with the mobile terminal; the clock module 2.6 is connected to the MCU module in communication

[0120] The power module 2.7 is composed of a micro rechargeable lithium battery of at least 10000mah, which is electrically connected to the MCU module to transmit power to each module.

[0121] The mobile terminal can be a mobile phone / notebook. The mobile terminal app has built-in corresponding high-speed rail station reading program, map analysis program and arrival reminder program, etc. The mobile terminal is connected to the reminder system body through a Bluetooth module.

[0122] Example 2

[0123] like Figure 4As shown, a high-speed rail arrival reminder method integrating multi-source information adopts the high-speed rail arrival reminder system integrating multi-source information as above to obtain the accurate arrival time at the destination station. The method in this embodiment is different from the traditional high-speed rail arrival reminder method in which the IMU module 2.3 is installed on the high-speed rail. Instead, it is based on the human body and the IMU module 2.3 is installed on a U-shaped pillow. Taking into account the human-machine state, the two second IMU elements 2.3.2 for measuring the human posture in the IMU module 2.3 are arranged in a tumbler style to ensure the horizontal measurement of the IMU and avoid the problem that the azimuth angle of the IMU module 2.3 is inconsistent with the running direction of the high-speed rail due to different sleeping postures of the human body; and the forward direction of the high-speed rail is determined by GPS, and the angle between the two directions is calculated according to the self-direction calculated by the IMU, and then the acceleration in the forward direction of the high-speed rail is accurately calculated.

[0124] On the other hand, the prediction of arrival time requires obtaining the coordinates of the high-speed rail and the station, and calculating the straight-line distance between the two points as the remaining distance from the high-speed rail to the station. However, the high-speed rail line is not a straight line, but may be composed of arcs and straight lines. The method in this embodiment obtains the map grid points of the stations along the line, and then calculates the segments of each grid point, which can obtain the remaining distance with higher accuracy;

[0125] In addition, when a train is delayed, it must run at a constant speed for a period of time and then decelerate before it can arrive; the conventional method only predicts the remaining time to arrive at the station by speed and distance, without taking into account the remaining distance required for deceleration when about to arrive at the station, while the method in this embodiment establishes a remaining distance prediction model for starting deceleration and a remaining arrival time prediction model for starting deceleration; effectively improving the accuracy of arrival time prediction.

[0126] The specific steps include:

[0127] (I) Import site data and select the arrival site.

[0128] S1: Use a mobile terminal to obtain site map information, set the destination site and advance reminder time, where the site map information includes site information along the route, arrival time, longitude and latitude corresponding to the site, and longitude and latitude of map points along the route;

[0129] The mobile terminal obtains the station information and arrival time along the designated high-speed rail vehicle number through the high-speed rail station reading program, obtains the longitude and latitude corresponding to the station and the longitude and latitude of the map points along the line through the map analysis level, and sets the destination station and advance reminder time according to the arrival reminder program

[0130] (ii) Update the reminder settings and maps of the MCU module.

[0131] S2, turn on the power module 2.7, and update the set advance reminder time and built-in map information of the MCU module according to the site map information, the destination site and the advance reminder time;

[0132] The MCU module is connected to the mobile terminal via the Bluetooth connection module, and the site information data, advance reminder time, map information and arrival site are imported into the MCU module. The WIFI module is connected to the wifi of the mobile terminal, and the map information built into the MCU module is updated according to the site map information.

[0133] (3) Establish a keyword recognition program.

[0134] S3: Pre-train the speech recognition program in the MCU module according to the keywords to obtain a keyword recognition model, wherein: the keywords include the site name and key time point words; the key time point words include "10 minutes", "5 minutes", "about to arrive", "reach", etc.

[0135] (iv) Start the clock module 2.6 and calculate the reminder time point.

[0136] S4: Start the clock module 2.6, and calculate the clock reminder time point according to the current time point and the arrival time of the destination station, combined with the set advance reminder time;

[0137] (5) Start positioning to obtain the current coordinates and running speed of the high-speed rail; (6) Start IMU module 2.3 to determine the current running status of the high-speed rail.

[0138] S5: Based on the clock reminder time point, the current high-speed rail coordinates, forward direction and running speed are obtained through the positioning module 2.2 and the MCU module, and the current running status of the high-speed rail is determined;

[0139] S5 includes:

[0140] Start the positioning module 2.2 and the two second IMU elements 2.3.2 10 minutes before the clock reminder time;

[0141] The positioning module 2.2 is used to continuously receive satellite radio signals through the antenna M times, and the current high-speed rail coordinates, forward direction and running speed are obtained through the running speed calculation program of the MCU module, where: M is a set number of thresholds, and in this embodiment, M=10;

[0142] The two second IMU elements 2.3.2 are used to obtain the linear acceleration along the X, Y, and Z axes, the rotational angular velocity along the X, Y, and Z axes, and the magnetic field strength along the X, Y, and Z axes of the current IMU module 2.3, and the acceleration of the current high-speed rail is calculated through the acceleration solution program of the MCU module;

[0143] The current operating status of the high-speed rail is determined based on the operating speed and acceleration of the high-speed rail, specifically:

[0144] When the high-speed rail's running speed is 0 and the acceleration is -0.1m / s to 0.1m / s, the high-speed rail is in a temporary parking stage; when the high-speed rail's running speed is not 0 and the acceleration is -0.1m / s to 0.1m / s, the high-speed rail is in a uniform speed running stage; when the high-speed rail's running speed is not 0, the acceleration is a negative value and the absolute value is greater than 0.1m / s, the high-speed rail is in a deceleration running stage.

[0145] The acceleration solver in S5 includes:

[0146] ① Use the magnetometer of the second IMU element 2.3.2 to measure the component of the Earth's magnetic field on the horizontal plane [m x ,m y ], calculate the azimuth angle θ of IMU module 2.3 according to the component of the geomagnetic field on the horizontal plane imu , the specific calculation formula is as follows:

[0147]

[0148] ② The acceleration component [a x ,a y ,a z ] Estimate the pitch angle α and roll angle β of the second IMU element 2.3.2;

[0149] ③ According to the pitch angle α and roll angle β of the second IMU element 2.3.2, the azimuth angle of the second IMU element 2.3.2 is corrected to obtain

[0150] Corrected azimuth angle θ imu ′, the specific calculation formula is as follows:

[0151]

[0152] ④ Calculate the acceleration a in the forward direction of the high-speed rail according to the corrected azimuth. The specific calculation formula is as follows:

[0153] a=a x ·cos(Δθ)+a y sin(Δθ);

[0154] Where: Δθ is the angle between the second IMU element 2.3.2 and the direction of the high-speed rail, Δθ=θ-θ′ imu .

[0155] In S5,

[0156] When the high-speed railway passes through an underground project, if the coordinates obtained by the positioning module 2.2 before and after the underground project are at least greater than the set distance threshold, it means that the high-speed railway is running, and the positioning module 2.2 is kept turned on until M consecutive satellite signals can be obtained or 1 minute before the reminder time is set or 10 minutes are left before the arrival time through voice recognition. At this time, the IMU module 2.3 and the acceleration solution program are combined to determine the running status of the high-speed railway; in this embodiment, the distance threshold is set to be the distance value calculated according to the speed obtained before the crossing and the time interval between the two positionings.

[0157] If the coordinates obtained by the positioning module 2.2 before and after crossing the underground project are not much different, the azimuth of the IMU module 2.3 is used as the running direction of the high-speed rail. When the acceleration of the current forward direction of the high-speed rail obtained by the IMU module 2.3 is zero, it is judged that the high-speed rail is in a stopped state. Except for the MCU module and the IMU module 2.3, other modules are suspended and passengers are temporarily delayed to save power. When the acceleration of the current forward direction of the high-speed rail is not zero, the high-speed rail is running. Until the positioning module 2.2 can obtain M consecutive satellite signals, the calculated azimuth of the running speed solver is used as the running direction of the high-speed rail.

[0158] (VII) Predict the arrival time of high-speed trains and update reminder time points.

[0159] S6: predicting the arrival time of the high-speed rail at the destination station according to the running status of the high-speed rail, and updating the clock reminder time point according to the predicted arrival time;

[0160] S6 includes:

[0161] S6.1. Calculate the distance between the current high-speed rail coordinates and the destination station based on the vectorized data of the map information built into the MCU;

[0162] S6.2, calculate the predicted arrival time of the high-speed rail to the destination station through the arrival time prediction program of the MCU module according to the distance between the high-speed rail coordinates and the destination station, the forward direction and the running speed;

[0163] S6.3. Update the clock reminder time point according to the predicted arrival time, specifically:

[0164] When the predicted arrival time is delayed by more than 10 minutes, the clock reminder time point is updated by delaying the clock reminder time point by the time that exceeds the time limit; and the positioning module 2.2 is set to obtain the high-speed rail coordinates once every 10 minutes, and the process returns to S6.1;

[0165] When the clock reminder time point is within 10 minutes of the predicted arrival time, the voice recognition module is started 11 minutes before the predicted arrival at the station, and the voice from the 11th minute to the 8th minute of the predicted arrival at the station is recorded in real time. The voice is transmitted to the MCU module, and the pre-trained keyword recognition model is used to recognize the keyword, and the time point when the keyword appears is recorded; then the remaining arrival time is calculated based on the average of the predicted arrival time and the time point when the keyword appears, and the clock reminder time point is updated based on the remaining arrival time;

[0166] Within 2 minutes before the clock reminder time point, the positioning module 2.2 is set to obtain the high-speed rail coordinates twice every 10 seconds on average, and return to S6.1;

[0167] When the high-speed rail is in operation, if the current coordinates of the high-speed rail cannot be obtained within 1 minute before the clock reminder time, the reminder time originally set by the clock shall prevail; if the set reminder time is within 10 minutes of arriving at the station, the clock reminder time shall be updated with the time point that appears in the keyword recognized by the voice recognition module.

[0168] The arrival time prediction procedure includes:

[0169] ① Calculate the remaining distance from the current high-speed rail coordinates to the destination station, specifically:

[0170] Calculate the current high-speed rail coordinates and the coordinates of the nearest line network point in the direction of travel under the current high-speed rail coordinates;

[0171] Divide the high-speed rail line into multiple sections in a rasterized manner, mark the longitude and latitude coordinates of the section points and the actual lengths of adjacent points;

[0172] Calculate the distances of each segment between the nearest network point and the destination station, and add up the distances of each segment to get the remaining distance from the high-speed rail to the destination station;

[0173] ② Establish a remaining distance prediction model for starting deceleration and a remaining arrival time prediction model for starting deceleration, specifically:

[0174] Collect in advance at least 100 sets of high-speed rail data with different speeds when the high-speed rail arrives at the station on time, where the high-speed rail data includes the average running speed and average acceleration within 10 minutes before the train decelerates to enter the station, as well as the remaining distance from the station and the remaining arrival time when the train starts to decelerate;

[0175] The high-speed rail data is normalized using the minimum-maximum normalization method. Then, based on the BP neural network model, the average running speed and average acceleration are used as input parameters, and the remaining distance from the station when the train starts to decelerate and the remaining arrival time are used as output parameters. The above high-speed rail data is learned to obtain the remaining distance prediction model for starting deceleration and the remaining arrival time prediction model for starting deceleration.

[0176] ③ Calculate the remaining arrival time based on the current high-speed rail operation status, specifically:

[0177] If the high-speed rail is in a temporary stop phase, all modules except the MCU module and IMU module 2.3 will be suspended to temporarily wake up passengers;

[0178] If the high-speed rail is in a constant speed stage, the specific steps for calculating the remaining arrival time are as follows:

[0179] i. Input the average running speed and average acceleration in this stage into the remaining distance prediction model at the start of deceleration and the remaining arrival time at the start of deceleration, and output the remaining distance at the start of deceleration and the running time of the deceleration section;

[0180] ii. The remaining distance from the high-speed train to the destination station minus the remaining distance when deceleration begins is used to obtain the remaining distance of the uniform speed running section;

[0181] iii. Calculate the running time of the remaining uniform running segment according to the uniform running speed;

[0182] iv. Add the running time of the deceleration phase and the running time of the constant speed phase to obtain the remaining arrival time;

[0183] v. Calculate the high-speed train arrival time according to the current time of the clock module 2.6, and update the clock reminder time;

[0184] vi. Recalculate i to v every 30 seconds;

[0185] If the high-speed rail is in the deceleration stage, the specific steps for calculating the remaining arrival time are as follows:

[0186] i. The running speed v measured at the i-th time point in the deceleration phase i , acceleration a i , the remaining distance from the high-speed rail to the destination d i , the theoretical remaining arrival time t can be calculated i , the specific calculation formula is as follows:

[0187]

[0188] ii. Verify whether the theoretical arrival time is consistent with the actual arrival time based on the remaining distance. If the theoretical remaining distance d i 'Greater than the actual remaining distance d i , the actual arrival time is considered to be less than the theoretical remaining arrival time, and the acceleration and arrival time are recalculated, and the recalculated arrival time is used as the actual arrival time; if the theoretical remaining distance d i 'Equal to the actual remaining distance d i, then the calculated theoretical remaining arrival time is taken as the actual arrival time;

[0189] Where: Theoretical remaining distance d i The calculation formula of ' is as follows:

[0190]

[0191] The formulas for recalculating acceleration and arrival time are as follows:

[0192]

[0193] iii. After obtaining the arrival time point, the clock module 2.6 is used to recalculate the clock reminder time point.

[0194] (VIII) Start reminder device 2.

[0195] S7: When the updated clock reminder time point in S6 is reached, the voice warning module is started to emit a warning sound to remind passengers that they are about to reach the station, and half of the electrode patches in the electrode patch module 2.4 are simultaneously started to simulate acupuncture stimulation of the human body; the warning sound is required to be less than 60 decibels.

[0196] (IX) Determine the sleeping state of the human body.

[0197] S8: Use the first IMU element 2.3.1 and the other half of the electrode patch in the electrode patch module 2.4 to detect the current human body state of the passenger, and use the sleep state assessment program to assess whether the passenger is awake; if it is detected that the passenger is not awake, adjust the current of the electrode patch until it is detected that the passenger is awake or passively shut down, and automatically shut down the power module 2.7; wherein: the maximum current of the electrode patch does not exceed 10mA.

[0198] The sleep status assessment process includes:

[0199] ① Establish the IMU sleep state assessment model and the electrode patch sleep state assessment model, specifically:

[0200] Establishing the IMU sleep state assessment model includes the following steps:

[0201] i. Collect at least 300 sets of signal data of the first IMU element 2.3.1 in different states of the human head, and use professional instruments to evaluate the state of the human body, where different states of the human head include wakefulness, light sleep and deep sleep;

[0202] Sleep state assessment mainly relies on the IMU element and electrode patch in the middle of the reminder system body.

[0203] IMU components mainly detect human movement through accelerometers and gyroscopes. Accelerometers are used to detect changes in acceleration of the human body in all directions, mainly reflecting the displacement or static state of the human body. When the human body is in a sleeping state, especially in the deep sleep stage, the body is usually in a relatively static state, with almost no large acceleration changes. The gyroscope detects the rotation or tilt changes of the head, which can describe the changes in the posture of the human body in a plane or space. When awake, the head will have a large rotation or movement, while in the sleeping state, especially in the deep sleep stage, the head movement is less. Set the IMU to detect 5 times every 5s, and then detect again after an interval of 5s.

[0204] ii. Then the signal data of the first IMU element 2.3.1 is subjected to low-pass filtering and denoising, and four key characteristic parameter data are extracted. The IMU data under the human body state is evaluated by professional instruments to establish three label data sets; among which: the four key features include the standard deviation of acceleration a std , the root mean square of acceleration a rms , angular velocity change rate Δω and acceleration signal periodicity P;

[0205] The three-axis data (X, Y, and Z axes) of the accelerometer and gyroscope will record the acceleration and rotation changes of the head in real time. The data acquisition frequency can be set to 50-100Hz to ensure sufficient accuracy. At the same time, the raw data is filtered during the acquisition process to remove high-frequency noise and external vibration. Since the IMU element is mainly attached to the side close to the human neck and is greatly affected by changes in human posture, when the human posture changes, the IMU element will receive high-frequency amplitude models. Therefore, a low-pass filter or a band-pass filter is used to remove frequency components greater than 5Hz and filter out most of the vibrations caused by high-speed rail operation.

[0206] The four key features are:

[0207] The standard deviation of acceleration is a std : Used to measure the fluctuation degree of acceleration signal within a period of time. The calculation formula is as follows:

[0208]

[0209] Wherein, n is the number of any time point after reaching the clock reminder time point updated in S6; a n is the acceleration value of the nth sampling point in the time period, N is the number of sampling points in the time window, N = 5; is the mean value of acceleration in the time window,

[0210] The root mean square of acceleration a rms: Reflects the intensity of the acceleration signal, used to distinguish between intense motion and static state. The calculation formula is as follows:

[0211]

[0212] Angular velocity change rate Δω: used to describe the degree of change in the head rotation rate.

[0213]

[0214] In the formula, ω n is the angular velocity value of the nth sampling point, |ω n+1 -ω n | is the angular velocity change value of adjacent sampling points.

[0215] Acceleration signal periodicity P: reflects the periodic characteristics of the motion signal and is used to detect turning over or periodic small movements. The calculation formula is as follows:

[0216]

[0217] Where argmax|FFT(a(t))| is the main frequency of the signal; a(t) is the acceleration change curve value within the time period.

[0218] iii. Then, a typical and reliable classification algorithm, the support vector machine algorithm, is used to learn and analyze the data set, and the sum parameter and penalty parameter of the support vector machine are optimized using the grid search algorithm until the loss function is minimized, thereby establishing an IMU sleep state assessment model;

[0219] The steps to establish the sleep state assessment model of the electrode patch include the following:

[0220] i. Collect at least 300 sets of electromyographic signal data of the electrode patch module 2.4 in different states of the human head, and use a human sleep state measuring instrument to evaluate the state of the human body;

[0221] ii. Then, the signal data of the electrode patch is amplified, and its root mean square characteristic data is extracted. The characteristic data of the electromyographic signal under the human body state is evaluated according to the human sleep state measuring instrument, and three label data sets are established;

[0222] iii. Then, the support vector machine algorithm is used to learn and analyze the data set, and the grid search algorithm is used to optimize the sum parameter and penalty parameter of the support vector machine until the loss function is minimized, thereby establishing a sleep state assessment model of the electrode patch;

[0223] ② Use the IMU sleep state assessment model and the electrode patch sleep state assessment model to assess the passenger’s sleep state.

[0224] The two sleep state assessment models trained above are burned into the MCU module, where the IMU module 2.3 collects data once every 5 seconds, and the electrode patch obtains the electromyographic signal with a time window of 1 second once every 2 seconds. When both items are assessed as awake, it can be considered that the passenger has woken up.

[0225] Acupuncture function is realized through electrode patches, and the current is transmitted to the surface and deep nerve endings of the skin through the electrode patches. This function has been widely used in massagers. The effect of acupuncture stimulation is simulated by applying current to the skin, such as using a low-frequency pulse generator circuit to generate a stimulation signal (frequency range 10-100Hz, pulse width about 50-200μs), and the output voltage is controlled within a safe range (0-20V).

[0226] Specifically:

[0227] The electrode patch module 2.4 is used to collect the passenger's electromyographic signal. Specifically, the electrode patch is placed close to the skin of the human neck to collect the changes in skin surface potential and the weak electrical signals generated by muscle activity, namely electromyographic signals. Among them, the electrode patch obtains the electromyographic signal with a time window of 1 s every 2 s. The electromyographic activity is significantly reduced during light sleep and deep sleep, and is higher when awake.

[0228] The signal data of the passenger's head is measured by the first IMU element 2.3.1, wherein: the first IMU element 2.3.1 is set to detect 5 times every 5 seconds, and then detect again after an interval of 5 seconds;

[0229] The passenger's electromyographic signal and head signal data are respectively transmitted to the sleep state evaluation model of the IMU of the MCU module and the sleep state evaluation model of the electrode patch to evaluate the passenger's sleep state;

[0230] If the evaluation results of the IMU sleep state evaluation model and the electrode patch sleep state evaluation model are both awake, the power module 2.7 is automatically turned off;

[0231] If the evaluation results of the IMU sleep state evaluation model and the electrode patch sleep state evaluation model are both light sleep, the initial current of the electrode patch is adjusted to 0.5-1mA, and the stimulation frequency is a medium frequency, low amplitude pulse wave of 20-50Hz;

[0232] If the evaluation results of the IMU sleep state evaluation model and the electrode patch sleep state evaluation model are both deep sleep state, the initial current of the electrode patch is adjusted to 1-2mA, and the stimulation frequency is adjusted to a low-frequency, high-amplitude pulse wave of 50-100Hz;

[0233] After 5 to 10 seconds of continuous stimulation, if the passenger is awake, the stimulation will be stopped; if the passenger is not awake, the current will be adjusted according to the changes in the passenger's electromyographic signals or small movements. Specifically, the current amplitude is increased by 0.5 mA every 5 seconds; if there is no response after 30 to 60 seconds of stimulation, the current will be directly increased to the maximum current to wake up; the stimulation will be stopped after 5 seconds.

[0234] In this embodiment, the language recognition program in the MCU module includes:

[0235] 1. First, select Baidu's open source speech recognition program PaddleSpeech, which is suitable for Chinese broadcast word recognition, as the basic program;

[0236] 2. Then collect audio data containing high-speed rail broadcast keywords, provide corresponding text labels for these audio data, and ensure that each audio file contains clear keyword pronunciation; pre-process the audio data, including noise reduction, volume adjustment, format conversion, etc., to ensure that the quality and format of the audio data meet the training requirements of PaddleSpeech;

[0237] 3. Deploy the relevant programs and training sample data to the cloud computing service platform, then select a suitable training script in the recipes directory of PaddleSpeech, and then configure the appropriate parameters according to the "station name" keywords along the national high-speed rail lines and time keywords such as "10 minutes", "5 minutes", "about to arrive", "arrive", etc., start training the speech recognition model, and use the test data set to evaluate the performance of the model. By adjusting the training parameters and sample data until its test performance is excellent, convert the trained PaddleSpeech model into a format suitable for offline mobile deployment. After that, connect to the cloud computing platform through the WIFI module, and deploy the tested program to the MCU module, so that it can still be used when the network is disconnected.

[0238] 4. Secondly, the RNNoise deep learning algorithm, which is recognized to have a good noise reduction effect, is selected as the basic algorithm for audio noise suppression. On this basis, the audio data of radio station broadcasts in a noise-free state and audio data with noisy noise are prepared. The program and sample data are deployed to the cloud computing platform, and then the architecture and core parameters of RNNoise are arranged, and the noise reduction model is trained. The performance of the model is evaluated using the test data set to enable it to have good noise reduction performance. The trained model is deployed to the MCU module for easy use when the network is disconnected.

[0239] 5. Finally, select the corresponding site name so that the speech recognition model only recognizes the site name and the corresponding time-related keyword detection. After the microphone obtains the surrounding sound, the noise in the sound is removed through the noise reduction model, and then the keywords in the broadcast sound are identified, and the time point when the keywords appear is recorded.

[0240] In this embodiment, the specific process of interpreting the site data information is:

[0241] The station data information interpretation is completed on the mobile terminal app, which has built-in 12306 website address and map latitude and longitude coordinate information and corresponding crawling technology. This technology logs in to your account through the https: / / kyfw.12306.cn website, and then enters the high-speed rail vehicle number to crawl and retrieve information through the python crawler technology in developer mode. The station information structure is: {"arrive_day_str":"**","arrive_time":"**","station_train_code":"**","station_name":"**","arrive_day_diff":"**","OT":[],"start_time":"**","wz_num":"**","station_no":"**","running_time":"**"}. The data of each station and the arrival time in the information are obtained by analyzing and saving them in the keyword dictionary.

[0242] After that, the longitude and latitude coordinates of the train stations are retrieved from the Chinese High-speed Rail and Airline Database (CRAD). Then, an electronic map such as the China Railway Online Map http: / / cnrail.geogv.org / zhcn / about is called using the API Key, and the high-speed rail stations with the corresponding longitude and latitude coordinates are identified using the geocoding function. The high-speed rail lines between the stations are crawled and divided into multiple segments in a rasterized manner, and the longitude and latitude coordinates of the segment points and the actual lengths of the adjacent points are marked.

[0243] After completing the above data interpretation, the data information points are transmitted to the MCU module, and the map information along the route of the MCU module is updated.

[0244] In this embodiment, the running speed calculation program refers to the running speed calculation method based on the GPS / Beidou chip, and the calculation steps are as follows:

[0245] GPS / Beidou chips generally use GNSS data settlement method, and calculate the position by measuring the time difference of the signal reaching the receiver. The output data generally includes key information such as latitude and longitude, elevation, speed, positioning time, etc., and inputs the key information into the MCU module; then, the running speed of the train can be obtained by calculating the distance between two positioning points, and the real distance d between two adjacent positioning points can be calculated by the Haversine formula:

[0246]

[0247] Where r is the radius of the earth, about 6371000m; Δφ = φ 2 -φ 1 is the latitude difference between the two positioning points, φ 2 ,φ 1 are the dimensions of front and rear positioning respectively; Δλ=λ 2 -λ 1 is the longitude difference between the two positioning points, λ 2 , 1 They are the longitudes of the front and rear positioning respectively.

[0248] The time interval Δt can be calculated based on the positioning time of the two positioning points, from which the speed v of the high-speed rail at this time can be calculated as:

[0249]

[0250] Then, based on the latitude and longitude coordinates of the two positionings, the azimuth θ of the high-speed rail's forward direction (relative to the north direction) can be calculated:

[0251]

[0252] This embodiment also includes a readable storage medium on which computer program instructions are stored. When the computer program instructions are executed by a processor, the high-speed rail arrival reminder method integrating multi-source information as described above is implemented.

[0253] It should be noted that the device embodiments described above are merely schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. In addition, in the accompanying drawings of the device embodiments provided by the present invention, the connection relationship between the modules indicates that there is a communication connection between them, which may be specifically implemented as one or more communication buses or signal lines. A person of ordinary skill in the art may understand and implement it without paying any creative effort.

[0254] This embodiment also includes an electronic device, including: at least one processor, at least one memory, and computer program instructions stored in the memory. When the computer program instructions are executed by the processor, the high-speed rail arrival reminder method that integrates multi-source information is implemented as described above.

[0255] Exemplarily, the computer program may be divided into one or more modules / units, one or more modules / units are stored in a memory and executed by a processor to implement the present invention. One or more modules / units may be a series of computer program instruction segments capable of implementing specific functions, and the instruction segments are used to describe the execution process of the computer program in an electronic device.

[0256] Electronic devices can be computing devices such as mobile phones, desktop computers, notebooks, PDAs, and cloud servers. Electronic devices can include, but are not limited to, processors and memories. For example, electronic devices can also include input and output devices, network access devices, buses, etc.

[0257] The processor may be a central processing unit (CPU), other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc. The processor is the control center of an electronic device, and uses various interfaces and lines to connect various parts of the entire electronic device.

[0258] The memory can be used to store computer programs and / or modules. The processor implements the computer program by running or executing the computer programs and / or modules stored in the memory, and calling the data stored in the memory. The memory can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, an application required for at least one function (such as a sound playback function, an image playback function, etc.), etc.; the data storage area can store data created according to the use of the mobile phone (such as audio data, a phone book, etc.), etc. In addition, the memory can include a high-speed random access memory, and can also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0259] Among them, if the module / unit integrated in the electronic device is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device capable of carrying computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium. It should be noted that the content contained in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, the computer-readable medium does not include electric carrier signals and telecommunication signals.

[0260] The above are only preferred embodiments of the present invention and are not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A high-speed rail arrival reminder system integrating multi-source information, comprising a wearable carrier (1), a mobile terminal and a reminder device (2), wherein the wearable carrier (1) is worn by a passenger, and the reminder device (2) is arranged on the wearable carrier (1), characterized in that: The reminder device (2) comprises an MCU module, a positioning module (2.2), an IMU module (2.3), an electrode patch module (2.4), a voice recognition module, a voice warning module, a clock module (2.6) and a power module (2.7); the positioning module (2.2), the IMU module (2.3), the voice recognition module and the clock module (2.6) are all communicatively connected to the MCU module; the electrode patch module (2.4), the voice warning module and the power module (2.7) are electrically connected to the MCU module; The MCU module is communicatively connected to the mobile terminal; the positioning module (2.2) is used to obtain the current position of the high-speed rail; the IMU module (2.3) comprises a first IMU element (2.3.1) and a second IMU element (2.3.2) arranged on both sides of the first IMU element (2.3.1), the first IMU element (2.3.1) being used to obtain the angular velocity and acceleration of the human body in three-dimensional space, and the second IMU element (2.3.2) being arranged in a tumbler-like manner and being used to obtain the acceleration of the high-speed rail; the electrode patch module (2.4) is arranged on the wearable carrier (1) and is capable of contacting with human skin and being used to collect and transmit electrical signals; the voice recognition module is used to collect high-speed rail notification voice information; the voice warning module is used to issue a warning voice to remind passengers; the clock module (2.6) is used to obtain time and date and set a clock reminder time point; and the power supply module (2.7) is used to supply power to the MCU module.

2. The high-speed rail arrival reminder system integrating multi-source information according to claim 1 is characterized in that: The MCU module has built-in speech recognition program, acceleration calculation program, running speed calculation program, arrival time prediction program and sleep state evaluation program, which can predict the time when the high-speed rail arrives at the destination station.

3. The high-speed rail arrival reminder system integrating multi-source information according to claim 2 is characterized in that: The high-speed rail arrival reminder system that integrates multi-source information also includes a Bluetooth module, and the MCU module is communicated with the mobile terminal through the Bluetooth module; and / or, the high-speed rail arrival reminder system that integrates multi-source information also includes a WIFI module, and the WIFI module is communicated with the MCU module, and the WIFI module is used to connect to the hotspot of the mobile terminal to update the map information built into the MCU module.

4. A high-speed rail arrival reminder method integrating multi-source information, using the high-speed rail arrival reminder system integrating multi-source information as claimed in claim 3 to obtain accurate arrival time at the destination station, characterized in that: The steps include: S1: Use a mobile terminal to obtain site map information, set the destination site and advance reminder time, where the site map information includes site information along the route, arrival time, longitude and latitude corresponding to the site, and longitude and latitude of map points along the route; S2, turn on the power module (2.7), and update the set advance reminder time and built-in map information of the MCU module according to the site map information, the destination site and the advance reminder time; S4: Start the clock module (2.6), calculate the clock reminder time point according to the current time point and the arrival time of the destination station, and combine the set advance reminder time; S5: Based on the clock reminder time point, the current high-speed rail coordinates, forward direction and running speed are obtained through the positioning module (2.2) and the MCU module, and the current running status of the high-speed rail is determined; S6: predicting the arrival time of the high-speed rail at the destination station according to the running status of the high-speed rail, and updating the clock reminder time point according to the predicted arrival time; S7: When the updated clock reminder time point in S6 is reached, the voice warning module is activated to emit a warning sound to remind the passenger that the station is about to be reached, and half of the electrode patches in the electrode patch module (2.4) are synchronously activated to simulate acupuncture stimulation of the human body; S8: Using the first IMU element (2.3.1) and the other half of the electrode patches in the electrode patch module (2.4) to detect the current human state of the passenger, and using the sleep state assessment program to assess whether the passenger is awake; if it is detected that the passenger is not awake, the current of the electrode patch is adjusted until the passenger is awake or passively shut down, and the power module (2.7) is automatically turned off; wherein: the maximum current of the electrode patch does not exceed 10mA.

5. The high-speed rail arrival reminder method integrating multi-source information according to claim 4 is characterized in that: The S5 includes: Activate the positioning module (2.2) and the two second IMU elements (2.3.2) 10 minutes before the clock reminder time point; The positioning module (2.2) is used to continuously receive satellite radio signals through the antenna M times, and the current high-speed rail coordinates, forward direction and running speed are obtained through the running speed calculation program of the MCU module, where: M is the set number of thresholds; The two second IMU elements (2.3.2) are used to obtain the linear acceleration along the X, Y, and Z axes, the rotational angular velocity along the X, Y, and Z axes, and the magnetic field strength along the X, Y, and Z axes of the current IMU module (2.3), and the acceleration of the current high-speed rail is calculated through the acceleration solution program of the MCU module; The current operating status of the high-speed rail is determined based on the operating speed and acceleration of the high-speed rail, specifically: When the high-speed rail's running speed is 0 and the acceleration is -0.1m / s to 0.1m / s, the high-speed rail is in a temporary parking stage; when the high-speed rail's running speed is not 0 and the acceleration is -0.1m / s to 0.1m / s, the high-speed rail is in a uniform speed running stage; when the high-speed rail's running speed is not 0, the acceleration is a negative value and the absolute value is greater than 0.1m / s, the high-speed rail is in a deceleration running stage.

6. The high-speed rail arrival reminder method integrating multi-source information according to claim 5 is characterized in that: The acceleration calculation procedure in S5 includes: ① Use the magnetometer of the second IMU element (2.3.2) to measure the component of the Earth's magnetic field on the horizontal plane [m x ,m y ], calculate the azimuth angle θ of the IMU module (2.3) based on the component of the geomagnetic field on the horizontal plane imu , the specific calculation formula is as follows: ② The acceleration component [a x ,a y ,a z ] Estimate the pitch angle α and roll angle β of the second IMU element (2.3.2); ③ According to the pitch angle α and roll angle β of the second IMU element (2.3.2), correct the azimuth angle of the second IMU element (2.3.2) to obtain the corrected azimuth angle θ imu ′, the specific calculation formula is as follows: ④ Calculate the acceleration a in the forward direction of the high-speed rail according to the corrected azimuth. The specific calculation formula is as follows: a=a x ·cos(Δθ)+a y ·sin(Δθ); Where: Δθ is the angle between the second IMU element (2.3.2) and the direction of the high-speed rail, Δθ = θ - θ i ' mu .

7. The high-speed rail arrival reminder method integrating multi-source information according to claim 6 is characterized in that: In S5, When the high-speed railway passes through an underground project, if the distance between the coordinates obtained by the positioning module (2.2) before and after the underground project is at least greater than the set distance threshold, it means that the high-speed railway is running, and the positioning module (2.2) is kept turned on until M consecutive satellite signals can be obtained or 1 minute before the set reminder time point or the voice recognition indicates that there are 10 minutes left before the arrival time. At this time, the IMU module (2.3) and the acceleration solution program are combined to determine the running status of the high-speed railway; If the coordinates obtained by the positioning module (2.2) before and after crossing the underground project are not much different, the azimuth of the IMU module (2.3) is used as the running direction of the high-speed rail. When the acceleration of the current forward direction of the high-speed rail obtained by the IMU module (2.3) is zero, it is judged that the high-speed rail is in a stopped state, and other modules except the MCU module and the IMU module (2.3) are suspended; when the acceleration of the current forward direction of the high-speed rail is not zero, the high-speed rail is running, until the positioning module (2.2) can obtain M consecutive satellite signals, the calculated azimuth of the running speed solution program is used as the running direction of the high-speed rail.

8. The high-speed rail arrival reminder method integrating multi-source information according to claim 7 is characterized in that: The S2 and S4 also include: S3: pre-training the speech recognition program in the MCU module according to the keywords to obtain a keyword recognition model, wherein the keywords include site names and key time point words; The S6 includes: S6.

1. Calculate the distance between the current high-speed rail coordinates and the destination station based on the vectorized data of the map information built into the MCU; S6.2, calculate the predicted arrival time of the high-speed rail to the destination station through the arrival time prediction program of the MCU module according to the distance between the high-speed rail coordinates and the destination station, the forward direction and the running speed; S6.

3. Update the clock reminder time point according to the predicted arrival time, specifically: When the predicted arrival time is delayed by more than 10 minutes, the clock reminder time point is updated by delaying the clock reminder time point by the time that exceeds the time limit; and the positioning module (2.2) is set to obtain the high-speed rail coordinates once every 10 minutes, and return to S6.1; When the clock reminder time point is within 10 minutes of the predicted arrival time, the voice recognition module is started 11 minutes before the predicted arrival at the station, and the voice from the 11th minute to the 8th minute of the predicted arrival at the station is recorded in real time. The voice is transmitted to the MCU module, and the pre-trained keyword recognition model is used to recognize the keyword, and the time point when the keyword appears is recorded; then the remaining arrival time is calculated based on the average of the predicted arrival time and the time point when the keyword appears, and the clock reminder time point is updated based on the remaining arrival time; Within 2 minutes before the clock reminder time point, the positioning module (2.2) is set to obtain the high-speed rail coordinates twice every 10 seconds on average, and return to S6.1; When the high-speed rail is in operation, if the current coordinates of the high-speed rail cannot be obtained within 1 minute before the clock reminder time, the reminder time originally set by the clock shall prevail; if the set reminder time is within 10 minutes of arriving at the station, the clock reminder time shall be updated with the time point that appears in the keyword recognized by the voice recognition module.

9. The high-speed rail arrival reminder method integrating multi-source information according to claim 8 is characterized in that: The arrival time prediction procedure includes: ① Calculate the remaining distance from the current high-speed rail coordinates to the destination station, specifically: Calculate the current high-speed rail coordinates and the coordinates of the nearest line network point in the direction of travel under the current high-speed rail coordinates; Divide the high-speed rail line into multiple sections in a rasterized manner, mark the longitude and latitude coordinates of the section points and the actual lengths of adjacent points; Calculate the distances of each segment between the nearest network point and the destination station, and add up the distances of each segment to get the remaining distance from the high-speed rail to the destination station; ② Establish a remaining distance prediction model for starting deceleration and a remaining arrival time prediction model for starting deceleration, specifically: Collect in advance at least 100 sets of high-speed rail data with different speeds when the high-speed rail arrives at the station on time, where the high-speed rail data includes the average running speed and average acceleration within 10 minutes before the train decelerates to enter the station, as well as the remaining distance from the station and the remaining arrival time when the train starts to decelerate; The high-speed rail data is normalized using the minimum-maximum normalization method. Then, based on the BP neural network model, the average running speed and average acceleration are used as input parameters, and the remaining distance from the station when the train starts to decelerate and the remaining arrival time are used as output parameters. The above high-speed rail data is learned to obtain the remaining distance prediction model for starting deceleration and the remaining arrival time prediction model for starting deceleration. ③ Calculate the remaining arrival time based on the current high-speed rail operation status, specifically: If the high-speed rail is in a temporary stop phase, all modules except the MCU module and the IMU module (2.3) will be suspended to temporarily wake up passengers; If the high-speed rail is in a constant speed stage, the specific steps for calculating the remaining arrival time are as follows: i. Input the average running speed and average acceleration in this stage into the remaining distance prediction model at the start of deceleration and the remaining arrival time at the start of deceleration, and output the remaining distance at the start of deceleration and the running time of the deceleration section; ii. The remaining distance from the high-speed train to the destination station minus the remaining distance when deceleration begins is used to obtain the remaining distance of the uniform speed running section; iii. Calculate the running time of the remaining uniform running segment according to the uniform running speed; iv. Add the running time of the deceleration phase and the running time of the constant speed phase to obtain the remaining arrival time; v. Calculate the high-speed train arrival time according to the current time of the clock module (2.6) and update the clock reminder time; vi. Recalculate i to v every 30 seconds; If the high-speed rail is in the deceleration stage, the specific steps for calculating the remaining arrival time are as follows: i. The running speed v measured at the i-th time point in the deceleration phase i , acceleration a i , the remaining distance from the high-speed rail to the destination d i , the theoretical remaining arrival time t can be calculated i , the specific calculation formula is as follows: ii. Verify whether the theoretical arrival time is consistent with the actual arrival time based on the remaining distance. If the theoretical remaining distance d i 'Greater than the actual remaining distance d i , the actual arrival time is considered to be less than the theoretical remaining arrival time, and the acceleration and arrival time are recalculated, and the recalculated arrival time is used as the actual arrival time; if the theoretical remaining distance d i 'Equal to the actual remaining distance d i , then the calculated theoretical remaining arrival time is taken as the actual arrival time; Where: Theoretical remaining distance d i The calculation formula of ' is as follows: The formulas for recalculating acceleration and arrival time are as follows: iii. After obtaining the arrival time point, use the clock module (2.6) to recalculate the clock reminder time point.

10. The high-speed rail arrival reminder method integrating multi-source information according to claim 4, characterized in that: The sleep state assessment procedure includes: ① Establish the IMU sleep state assessment model and the electrode patch sleep state assessment model, specifically: Establishing the IMU sleep state assessment model includes the following steps: i. Collect at least 300 sets of signal data of the first IMU element (2.3.1) in different states of the human head, and use professional instruments to evaluate the state of the human body, where different states of the human head include wakefulness, light sleep and deep sleep; ii. Then the signal data of the first IMU element (2.3.1) is subjected to low-pass filtering and denoising, and four key characteristic parameter data are extracted. The IMU data under the human body state is evaluated by professional instruments, and three label data sets are established; among which: the four key features include the standard deviation of acceleration a std , the root mean square of acceleration a rms , angular velocity change rate Δω and acceleration signal periodicity P; iii. Then, a typical and reliable classification algorithm, the support vector machine algorithm, is used to learn and analyze the data set, and the sum parameter and penalty parameter of the support vector machine are optimized using the grid search algorithm until the loss function is minimized, thereby establishing an IMU sleep state assessment model; The steps to establish the sleep state assessment model of the electrode patch include the following: i. Collect at least 300 sets of electromyographic signal data of the electrode patch module (2.4) when the human head is in different states, and use a human sleep state measuring instrument to evaluate the state of the human body; ii. Then, the signal data of the electrode patch is amplified, and its root mean square characteristic data is extracted. The characteristic data of the electromyographic signal under the human body state is evaluated according to the human sleep state measuring instrument, and three label data sets are established; iii. Then, the support vector machine algorithm is used to learn and analyze the data set, and the grid search algorithm is used to optimize the sum parameter and penalty parameter of the support vector machine until the loss function is minimized, thereby establishing a sleep state assessment model of the electrode patch; ② Use the IMU sleep state assessment model and the electrode patch sleep state assessment model to assess the passenger's sleep state, specifically: The electrode patch module (2.4) is used to collect the passenger's electromyographic signal. Specifically, the electrode patch is placed close to the skin of the human neck to collect the change in skin surface potential and the weak electrical signal generated by muscle activity, i.e., the electromyographic signal; wherein: the electrode patch obtains the electromyographic signal with a time window of 1 second at an interval of 2 seconds; The signal data of the passenger's head is measured by using the first IMU element (2.3.1), wherein: the first IMU element (2.3.1) is set to detect 5 times every 5 seconds, and then detect again after an interval of 5 seconds; The passenger's electromyographic signal and head signal data are respectively transmitted to the sleep state evaluation model of the IMU of the MCU module and the sleep state evaluation model of the electrode patch to evaluate the passenger's sleep state; If the evaluation results of the IMU sleep state evaluation model and the electrode patch sleep state evaluation model are both awake, the power module (2.7) is automatically turned off; If the evaluation results of the IMU sleep state evaluation model and the electrode patch sleep state evaluation model are both light sleep, the initial current of the electrode patch is adjusted to 0.5-1mA, and the stimulation frequency is a medium frequency, low amplitude pulse wave of 20-50Hz; If the evaluation results of the IMU sleep state evaluation model and the electrode patch sleep state evaluation model are both deep sleep state, the initial current of the electrode patch is adjusted to 1-2mA, and the stimulation frequency is adjusted to a low-frequency, high-amplitude pulse wave of 50-100Hz; After 5 to 10 seconds of continuous stimulation, if the passenger is awake, the stimulation will be stopped; if the passenger is not awake, the current will be adjusted according to the changes in the passenger's electromyographic signals or small movements. Specifically, the current amplitude is increased by 0.5 mA every 5 seconds; if there is no response after 30 to 60 seconds of stimulation, the current will be directly increased to the maximum current to wake up; the stimulation will be stopped after 5 seconds.

Images