Multi - area Scenario Mode Adaptive System and Method for Rear Loading Machine Screen

Abstract

The present invention discloses a multi-zone scenario mode adaptive system and method for a rear-mounted vehicle screen, and specifically relates to the field of vehicle screen control technology, including: by acquiring multi-mode feature data, such as rear-seat gravity difference, sound frequency, driving time, weather impact index, etc., different scenario coefficients are calculated respectively to form scenario feature data and input into a scenario prediction model to predict scenario modes such as children's mode, wake-up mode, nap mode and dating mode, construct and train the scenario prediction model based on multi-dimensional feature data, and preset a safety level for the prediction result to optimize the adaptability and intelligent performance of the vehicle screen. The present invention identifies different driving scenarios through the precise collection and calculation of multi-mode feature data, and then adaptively provides appropriate scenario modes, effectively improving the in-vehicle interactive experience and driving safety, and providing innovative support for intelligent vehicle systems.

Description

Technical Field

[0001] The present invention relates to the technical field of vehicle screen control, and more specifically, to a multi-region scenario mode adaptive system and method for a rear-mounted vehicle screen. Background Art

[0002] In order to adapt to different user needs and usage scenarios, a multi-zone scenario mode adaptive system for aftermarket car screens has been proposed to achieve intelligent adjustment of the in-car screen to different usage environments. However, traditional car screens rely on fixed mode configurations, and users need to switch manually. For example, when switching from navigation mode to entertainment mode, there are cumbersome operations, and different users, weather, and driving scenarios (such as night driving, rainy days or high-speed driving, congestion) have different requirements for screen brightness, color temperature, and functional interfaces, which the existing system cannot automatically identify and adjust.

[0003] In order to solve this problem, in the existing methods, for example, the patent application with publication number CN115840599A discloses a scenario mode management method and system, which includes: if a scenario mode startup operation is detected, a prompt message is output; if the scenario mode is selected as a custom scenario mode, the custom function preference settings are obtained, the set custom scenario mode is uploaded to the server, and the vehicle scenario mode is updated to the custom scenario mode; if the scenario mode is selected as an application scenario mode, the target application scenario mode is set to the vehicle scenario mode according to the selection operation; if a scenario mode sharing request is received, a sharing instruction is sent to the server so that the server sends the scenario mode to the target user, and the sharing instruction includes sharing target user information and scenario mode information to be shared. Although this method can customize the scenario mode according to user needs, research and application of this method and the existing technology have found that there are at least the following defects:

[0004] When the driver uses a customized scenario mode, he or she does not fully consider whether the current scenario mode is consistent with the vehicle's safe driving state. For example, when the vehicle is in a high-speed driving state, if the rear-mounted screen displays too much entertainment content or complex interactive interfaces, it may distract the driver's attention and increase cognitive load, thereby interfering with driving decisions and reducing driving safety.

[0005] To this end, the present invention provides a multi-region scenario mode adaptive system and method for a rear-mounted vehicle screen. Summary of the invention

[0006] In order to overcome the above-mentioned defects of the prior art, the present invention provides a multi-zone scenario mode adaptive system and method for a rear-mounted vehicle screen to solve the problems raised in the above-mentioned background technology.

[0007] To achieve the above-mentioned purpose, the present invention provides the following technical solutions: In the first aspect, a multi-region scenario mode adaptive method for a rear-mounted vehicle screen comprises:

[0008] Step 1: Acquire first mode feature data, and calculate a first scene coefficient according to the first mode feature data; the first mode feature data includes a rear-row gravity difference, a first timestamp, and a sound frequency;

[0009] Step 2: Acquire second mode characteristic data, and calculate a second scenario coefficient according to the second mode characteristic data; the second mode characteristic data includes continuous driving time, weather impact index and fatigue assessment index;

[0010] Step 3: Acquire the third mode characteristic data, and calculate the third scenario coefficient according to the third mode characteristic data; the third mode characteristic data includes an idle speed value, a second timestamp, and a main seat inclination angle;

[0011] Step 4: Acquire fourth mode feature data, and calculate a fourth scene coefficient according to the fourth mode feature data; the fourth mode feature data includes identity recognition data, total number of rides, and music timestamp;

[0012] Step 5: Mark the first scene coefficient, the second scene coefficient, the third scene coefficient and the fourth scene coefficient as scene feature data, and input the scene feature data into a pre-built scenario prediction model to obtain corresponding prediction results; the prediction results include a child mode, a wake-up mode, a nap mode and a dating mode;

[0013] Step 6: Preset the safety level for the prediction results, and analyze the prediction results output by the scenario prediction model to obtain the current scenario mode.

[0014] Furthermore, the method for calculating and obtaining the first scene coefficient according to the first mode feature data includes:

[0015] The rear gravity difference, the first timestamp and the sound frequency are marked as , and ;

[0016] The rear gravity difference, the first timestamp and the sound frequency are formally calculated to obtain the first scene coefficient, which is calculated as follows:

[0017] , ;

[0018] In the formula, represents the first scene coefficient, is the weight factor of the rear-row gravity difference, and The value of is 1 or 0. Indicates the real-time gravity value. Indicates the standard gravity value;

[0019] The first time period is preset. If the first timestamp belongs to the first time period, The value of is 1, if the first timestamp does not belong to the preset first time period, The value of is 0; if the sound frequency belongs to the child's sound frequency, The value of is 1, if the sound frequency does not belong to the children's sound frequency, The value of is 0.

[0020] Furthermore, the continuous driving time is obtained by collecting the driving time counter;

[0021] The method for obtaining the weather impact index includes:

[0022] Acquiring weather characteristic data from a local meteorological station according to a current location of the vehicle, the weather characteristic data including wind speed, rainfall, visibility, and light intensity;

[0023] After normalizing the weather characteristic data, a formula calculation is performed to obtain the weather impact index, which is calculated as follows: ;

[0024] In the formula, represents the weather impact index, Indicates wind speed, Indicates rainfall, Indicates visibility, Indicates the light intensity, , , and are the corresponding weight factors respectively.

[0025] Furthermore, the method for obtaining the fatigue assessment index includes:

[0026] Acquiring driver fatigue characteristic data, wherein the fatigue characteristic data includes the number of sharp turns, speed difference, and braking delay duration;

[0027] The fatigue characteristic data is calculated in a formula to obtain the fatigue assessment index, and the calculation formula is:

[0028] ;

[0029] In the formula, represents the fatigue assessment index, Indicates the number of sharp turns, Indicates the speed difference, Indicates the braking delay time. represents the weight factor for the number of sharp turns, represents the weight factor of the speed difference, A weighting factor representing the braking delay duration.

[0030] Furthermore, the method for obtaining the second scene coefficient includes:

[0031] The continuous driving time, weather impact index and fatigue assessment index are calculated comprehensively to obtain the second scenario coefficient, which is calculated as follows:

[0032] ;

[0033] In the formula, represents the second scenario coefficient, Ls represents the continuous driving time, is the weight factor of continuous driving time, is a correction constant greater than zero and less than 1.

[0034] Furthermore, the method for obtaining the third scene coefficient includes:

[0035] The idle speed value, the second timestamp and the main seat inclination angle are marked as , and ;

[0036] The idle speed value, the second timestamp and the main seat inclination angle are normalized to calculate the third scenario coefficient, and the calculation formula is:

[0037] ;

[0038] In the formula, represents the third scenario coefficient, represents the weighting factor of the idle speed value, represents the weight factor of the main seat inclination, The value of is 1 or 0;

[0039] The second time period is preset. If the second timestamp belongs to the preset second time period, The value of is 1, if the second timestamp does not belong to the preset second time period, The value of is 0.

[0040] Furthermore, the method for obtaining the fourth scene coefficient includes:

[0041] The identification data, the total number of rides, and the music timestamp are marked as , and ;

[0042] Formulate the identification data, the total number of rides and the music timestamp to obtain the fourth scenario coefficient; ;

[0043] In the formula, represents the fourth scenario coefficient, represents the weight factor of the total number of passengers, represents the weight factor of the music timestamp;

[0044] The identity recognition data is preset, and different values ​​are set according to the successful state or failed state of automatic connection between the driver's and passenger's mobile device and the vehicle screen. The identity recognition data corresponding to the successful state of automatic connection between the driver's and passenger's mobile device and the vehicle screen is set to 1, and the identity recognition data corresponding to the automatic connection recognition state between the driver's and passenger's mobile device and the vehicle screen is set to 0;

[0045] The preset driving number threshold is 2. If the total number of driving is greater than or less than the preset driving number threshold, then The value is 0. If the total number of passengers is equal to the preset threshold, then The value is 1; preset music time period, if the music timestamp belongs to the preset music time period, The value of is 1, if the music timestamp does not belong to the preset music time period, The value of is 0.

[0046] Furthermore, the specific training process of the scenario prediction model includes:

[0047] Collect a group A of scene feature data in advance, where A is an integer greater than 1, set corresponding prediction results for the scene feature data, set different digital labels for the child mode, the wake-up mode, the nap mode, and the dating mode, and set corresponding prediction results for different scene feature data in group A in sequence;

[0048] Mark the prediction results as digital labels, and convert the scene feature data and the corresponding digital labels into a corresponding set of feature vectors;

[0049] Each group of feature vectors is used as the input of a scenario prediction model. The scenario prediction model takes a group of digital labels corresponding to each group of scene feature data as output, and takes the actual digital labels corresponding to each group of scene feature data as prediction targets, where the actual digital labels are pre-set digital labels corresponding to the scene feature data; minimizing the sum of prediction errors of all scene feature data is used as a training target; the scenario prediction model is trained until the sum of prediction errors reaches convergence, and the training is stopped. The scenario prediction model is a deep neural network model.

[0050] Furthermore, the method for pre-sorting the prediction results includes:

[0051] The awakening mode in the prediction results is set to the first safety level in advance; the child mode in the prediction results is set to the second safety level; the nap mode and the dating mode in the prediction results are set to the third safety level; the first safety level> the second safety level> the third safety level.

[0052] Furthermore, the method of analyzing the prediction results output by the scenario prediction model to obtain the current scenario mode includes:

[0053] Obtaining a real-time safety factor, taking the difference between the real-time safety factor and the standard safety factor as the safety factor difference, comparing the safety factor difference with a preset threshold, and if the safety factor difference is greater than the preset threshold, taking the prediction result output by the scenario prediction model as the current scenario mode; if the safety factor difference is less than or equal to the preset threshold, taking a higher-level scenario mode corresponding to the prediction result output by the scenario prediction model as the current scenario mode;

[0054] Wherein, the acquisition of the real-time safety factor includes:

[0055] The first scenario coefficient, the second scenario coefficient, the third scenario coefficient and the fourth scenario coefficient are calculated comprehensively to obtain the real-time safety factor, and the calculation formula is:

[0056] ;

[0057] In the formula, represents the real-time safety factor, , , and is the corresponding weight factor, and 1> > > > >0.

[0058] In a second aspect, the present invention provides a multi-region scenario mode adaptive system for a rear-mounted vehicle screen; a multi-region scenario mode adaptive method for implementing the above-mentioned rear-mounted vehicle screen comprises:

[0059] A first mode analysis module, used to obtain first mode feature data, and calculate a first scene coefficient according to the first mode feature data; the first mode feature data includes a rear-row gravity difference, a first timestamp, and a sound frequency;

[0060] A second mode analysis module, used to obtain second mode characteristic data, and calculate a second scenario coefficient according to the second mode characteristic data; the second mode characteristic data includes continuous driving time, weather impact index and fatigue assessment index;

[0061] A third mode analysis module, used to obtain third mode characteristic data, and calculate a third scenario coefficient according to the third mode characteristic data; the third mode characteristic data includes an idle speed value, a second timestamp, and a main seat inclination angle;

[0062] A fourth mode analysis module, used to obtain fourth mode feature data, and calculate a fourth scene coefficient according to the fourth mode feature data; the fourth mode feature data includes identity recognition data, total number of rides and music timestamp;

[0063] A scenario prediction module, marking the first scenario coefficient, the second scenario coefficient, the third scenario coefficient and the fourth scenario coefficient as scenario feature data, and inputting the scenario feature data into a pre-built scenario prediction model to obtain corresponding prediction results; the prediction results include a child mode, a wake-up mode, a nap mode and a dating mode;

[0064] The safety analysis module is used to pre-set the safety level for the prediction results and analyze the prediction results output by the scenario prediction model to obtain the current scenario mode.

[0065] Technical effects and advantages of the present invention:

[0066] The present invention identifies different driving scenarios through the precise collection and calculation of multi-mode feature data, and then adaptively provides appropriate scenario modes (children mode, wake-up mode, nap mode and dating mode). By acquiring and analyzing data such as rear-seat gravity difference, driving time, idling state, ambient weather and other data in real time, it can flexibly adapt to different driving needs. The scenario mode obtained through scenario prediction model analysis and combined with safety level preset can provide drivers with a more reasonable and safe interface experience in different scenarios, which not only enhances the personalized driving experience, but also optimizes driving safety and comfort, making the system more intelligent and humane. BRIEF DESCRIPTION OF THE DRAWINGS

[0067] Figure 1 This is a flow chart of the multi-region scenario mode adaptive method of the after-installed vehicle screen of Example 1;

[0068] Figure 2 This is a flow chart of a method for calculating and obtaining a first scene coefficient according to first mode feature data in Embodiment 1;

[0069] Figure 3 The training flow chart of the scenario prediction model of Example 1;

[0070] Figure 4 This is a schematic structural diagram of the multi-zone scenario mode adaptive system of the after-market vehicle screen of Example 2. DETAILED DESCRIPTION

[0071] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0072] In addition, the accompanying drawings are only schematic illustrations of the present invention and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. The functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor methods and / or microcontroller methods.

[0073] It should be understood that, although the terms "first", "second", etc. may be used herein to describe various units, these units should not be limited by these terms. These terms are used only to distinguish one unit from another unit. For example, without departing from the scope of the exemplary embodiments, a first unit may be referred to as a second unit, and a similar second unit may be referred to as a first unit. The term "and / or" used herein includes any and all combinations of one or more of the listed associated items.

[0074] Example 1

[0075] See also Figure 1 As shown, this embodiment discloses a multi-region scenario mode adaptive method for a rear-mounted vehicle screen, the method comprising:

[0076] Step 1: Acquire first mode feature data, and calculate a first scene coefficient according to the first mode feature data;

[0077] It should be noted that: the first mode characteristic data includes the rear-seat gravity difference, the first timestamp and the sound frequency; wherein the rear-seat gravity difference is obtained by measuring the pressure sensor in the rear of the car and calculating the difference between the real-time gravity value and the standard gravity value; the standard gravity value is the preset maximum weight value of the child, and the reference value is 45 kg. The specific value is set by those skilled in the art according to the actual situation and is not specifically limited here. The first timestamp provides time information through the electronic clock of the car and generates the first timestamp. The sound frequency is the sound frequency received by the microphone in the car.

[0078] See also Figure 2As shown, in implementation, the method for calculating and obtaining the first scene coefficient according to the first mode feature data includes:

[0079] The rear gravity difference, the first timestamp and the sound frequency are marked as , and ;

[0080] The rear gravity difference, the first timestamp and the sound frequency are formally calculated to obtain the first scene coefficient, which is calculated as follows:

[0081] , ;

[0082] In the formula, represents the first scene coefficient, is the weight factor of the rear-row gravity difference, and The value of is 1 or 0. Indicates the real-time gravity value. Indicates standard gravity value.

[0083] It should be noted that: the first time period is preset, if the first timestamp belongs to the preset first time period, The value of is 1, if the first timestamp does not belong to the preset first time period, The value of is 0; if the sound frequency belongs to the child's sound frequency, The value of is 1, if the sound frequency does not belong to the children's sound frequency, The value of is 0.

[0084] An exemplary description is that the preset first time period includes:

[0085] School hours: 3:00pm to 6:00pm, Monday to Friday;

[0086] Family time: Saturday and Sunday from 9:00am to 8:00pm.

[0087] If the first timestamp is 3:30 pm on Monday, which is within the first preset time period, then The value of is 1. If the first timestamp is 9:30 p.m. on Monday, which is not within the preset first time period, then The value of is 0.

[0088] It is also noted that the method of determining whether the sound frequency is that of a child includes: collecting sound through a microphone and using a spectrum feature analysis device to detect whether the sound frequency conforms to the spectrum of a child's sound. The child's sound frequency is between 300 and 4000 Hz, and the adult's sound frequency is between 85 and 255 Hz. When the sound frequency is 400 Hz, The value of is 1, when the sound audio is 215Hz, The value of is 0.

[0089] In this step, the system can determine whether there are passengers in the back row through the gravity difference in the back row, especially recognize the presence of children, thereby triggering specific scenario modes such as child mode. The first timestamp provides current time information. Combined with the preset first time period (such as school time or family activity time), it can identify whether it is a high-frequency time period for children, which helps to automatically switch to child mode and improve the intelligence of the system. By collecting the sound frequency in the car, it can identify the unique sound characteristics of children, further assist in identifying the possibility of children riding, and improve the recognition accuracy.

[0090] Step 2: Acquire second mode characteristic data, and calculate a second scenario coefficient according to the second mode characteristic data; the second mode characteristic data includes continuous driving time, weather impact index and fatigue assessment index;

[0091] It should be noted that: the continuous driving time is obtained by collecting the driving time counter;

[0092] Wherein, the method for obtaining the weather impact index includes:

[0093] Acquiring weather characteristic data from a local meteorological station according to a current location of the vehicle, the weather characteristic data including wind speed, rainfall, visibility, and light intensity;

[0094] It should be noted that: according to the current geographical location of the car, relevant real-time weather information is obtained by accessing the local meteorological station or weather data source. Specifically, the wind speed refers to the current wind speed information. The greater the wind speed, the greater the driving risk to the car. The rainfall refers to the precipitation in the area, which affects the driving line of sight and the slipperiness of the road. The greater the rainfall, the greater the driving risk to the car. The visibility refers to the current line of sight range. The lower the visibility, the greater the driving risk to the car. The light intensity refers to the detection of the current light intensity. The weaker the light intensity, the greater the driving risk to the car.

[0095] After normalizing the weather characteristic data, a formula calculation is performed to obtain the weather impact index, which is calculated as follows:

[0096] ;

[0097] In the formula, represents the weather impact index, Indicates wind speed, Indicates rainfall, Indicates visibility, Indicates the light intensity, , , and are the corresponding weight factors, and , , and is not zero, Represents the logarithmic function with base e.

[0098] Wherein, the method for obtaining the fatigue assessment index includes:

[0099] Acquiring driver fatigue characteristic data, wherein the fatigue characteristic data includes the number of sharp turns, speed difference, and braking delay duration;

[0100] It should be noted that the number of sharp turns refers to the number of times the steering wheel rotates rapidly to a larger angle in a short period of time. The steering wheel angle sensor quickly rotates to a preset angle in a short period of time and is recorded as a sharp turn. The speed difference refers to the difference in the rapid change of the speed of the car in a short period of time. The reference value for the short time is 2 seconds. The specific value is set by those skilled in the art according to actual conditions and is not specifically limited here. The brake delay time refers to the reaction time between the driver facing a situation where braking is required (such as an obstacle ahead, a sharp turn, etc.) and actually stepping on the brakes. The forward-looking radar and camera detect the distance to the obstacle or vehicle ahead. When the obstacle distance is lower than the safety threshold, the driver's reaction time begins to be recorded. It is worth noting that any method for obtaining the brake delay time in the prior art is acceptable and is not specifically described here.

[0101] The fatigue characteristic data is calculated in a formula to obtain the fatigue assessment index, and the calculation formula is:

[0102] ;

[0103] In the formula, represents the fatigue assessment index, Indicates the number of sharp turns, Indicates the speed difference, Indicates the braking delay time. represents the weight factor for the number of sharp turns, represents the weight factor of the speed difference, A weighting factor representing the braking delay duration.

[0104] In implementation, the method for obtaining the second scene coefficient includes:

[0105] The continuous driving time, weather impact index and fatigue assessment index are calculated comprehensively to obtain the second scenario coefficient, which is calculated as follows:

[0106] ;

[0107] In the formula, represents the second scenario coefficient, Ls represents the continuous driving time, is the weight factor of continuous driving time, is a correction constant greater than zero and less than 1;

[0108] This step collects continuous driving time data, obtains the weather impact index and fatigue assessment index, and comprehensively assesses the degree of driving risk. The larger the value of the second scenario coefficient, the greater the probability of activating the awakening mode, thereby reducing the potential impact of the environment and fatigue on the driver, providing the driver with a safer driving experience.

[0109] Step 3: Acquire the third mode characteristic data, and calculate the third scenario coefficient according to the third mode characteristic data; the third mode characteristic data includes an idle speed value, a second timestamp, and a main seat inclination angle;

[0110] It should be noted that the idle speed value refers to the speed of the engine in neutral or parking state, which is obtained by collecting data from the speed sensor and converting it. The second timestamp is provided and generated by the electronic clock of the car. The main seat inclination angle refers to the inclination of the backrest of the main driver's seat relative to the horizontal plane; the inclination angle of the seat backrest is detected by the seat angle sensor installed on the adjustment device of the main driver's seat.

[0111] The idle speed value, the second timestamp and the main seat inclination angle are marked as , and ;

[0112] The idle speed value, the second timestamp and the main seat inclination angle are normalized to calculate the third scenario coefficient, and the calculation formula is:

[0113] ;

[0114] In the formula, represents the third scenario coefficient, represents the weighting factor of the idle speed value, represents the weight factor of the main seat inclination, The value of is 1 or 0.

[0115] It should be noted that: the second time period is preset, if the second timestamp belongs to the preset second time period, The value of is 1, if the second timestamp does not belong to the preset second time period, The value of is 0;

[0116] Exemplarily, the preset second time period includes:

[0117] Rest time: 7:00 to 8:00, 12:00 to 14:00, 18:00 to 19:00 every day;

[0118] If the second timestamp is 7:30, which belongs to the preset second time period, then The value of is 1. If the second timestamp is 8:30, which does not belong to the preset second time period, then The value of is 0.

[0119] In this step, when the idle speed is high and the main seat inclination angle is large, the car screen system automatically turns on the nap mode, automatically adjusts the car screen brightness, display content or audio selection, so that the content is more in line with the driver's status, meets the driver's needs for short rest and enhances the comfort and safety of the in-car experience.

[0120] Step 4: Acquire fourth mode feature data, and calculate a fourth scene coefficient according to the fourth mode feature data; the fourth mode feature data includes identity recognition data, total number of rides, and music timestamp;

[0121] It should be noted that the total number of passengers refers to the number of passengers including the driver, and the number of passengers is counted through seat sensor detection; the music timestamp refers to the time period when the car audio system plays music of a specific category (such as romantic, soft), which is used to determine whether it meets the dating scene. The music playback record is used to count the playback time period of romantic music.

[0122] The identity recognition data is pre-set, and different values ​​are set according to the successful state or failed state of automatic connection between the driver's and passenger's mobile device and the vehicle screen. The identity recognition data corresponding to the successful state of automatic connection between the driver's and passenger's mobile device and the vehicle screen is set to 1, and the identity recognition data corresponding to the automatic connection recognition state between the driver's and passenger's mobile device and the vehicle screen is set to 0.

[0123] The identification data, the total number of rides, and the music timestamp are marked as , and ;

[0124] Formulate the identification data, the total number of rides and the music timestamp to obtain the fourth scenario coefficient;

[0125] ;

[0126] In the formula, represents the fourth scenario coefficient, represents the weight factor of the total number of passengers, represents the weight factor of the music timestamp;

[0127] The preset driving number threshold is 2. If the total number of driving is greater than or less than the preset driving number threshold, then The value is 0. If the total number of passengers is equal to the preset threshold, then The value is 1.

[0128] It should be noted that: preset music time period, if the music timestamp belongs to the preset music time period, The value of is 1, if the music timestamp does not belong to the preset music time period, The value of is 0; the preset music time period is customized by technicians in this field or drivers and passengers, and is not specifically limited.

[0129] Step 5: Mark the first scenario coefficient, the second scenario coefficient, the third scenario coefficient and the fourth scenario coefficient as scenario feature data, and input the scenario feature data into a pre-built scenario prediction model to obtain corresponding prediction results.

[0130] It should be noted that the prediction results include children's mode, awakening mode, nap mode and dating mode.

[0131] The child mode is used for operations such as children's entertainment, safety settings and parental control, and is used when a family has children riding, thereby improving the children's riding experience and reducing interference to the driver;

[0132] The awakening mode includes operations such as fatigue prompting and interactive music playback, which is used for long-distance driving and fatigue driving to keep the driver awake and reduce risks;

[0133] The nap mode includes operations such as lighting adjustment, external traffic noise reduction and temperature adjustment, and is used for short breaks when parking or idling, thereby providing a comfortable rest environment;

[0134] The dating mode includes operations such as ambient lighting and romantic music, and is used for couples traveling or private communication, thereby improving the communication atmosphere and comfort.

[0135] See also Figure 3 As shown, in implementation, the specific training process of the scenario prediction model includes:

[0136] A group of scene feature data is collected in advance, where A is an integer greater than 1, and corresponding prediction results are set for the scene feature data. Different digital labels are set for the child mode, the wake-up mode, the nap mode, and the dating mode. For example, the digital label is set to 1 for the child mode, the digital label is set to 2 for the wake-up mode, the digital label is set to 3 for the nap mode, and the digital label is set to 4 for the dating mode; the prediction results corresponding to the scene feature data are collected by technicians in this field during the diagnosis of historical scene feature data. The technicians in this field judge in turn according to actual experience that under the conditions of different scene feature data in group A, the different scene feature data in group A are set to corresponding prediction results in turn;

[0137] Mark the prediction results as digital labels, and convert the scene feature data and the corresponding digital labels into a corresponding set of feature vectors;

[0138] Each set of feature vectors is used as the input of the scenario prediction model. The scenario prediction model uses a set of digital labels corresponding to each set of scenario feature data as output, and uses the actual digital labels corresponding to each set of scenario feature data as the prediction target. The actual digital labels are pre-set digital labels corresponding to the scenario feature data. The training target is to minimize the sum of the prediction errors of all scenario feature data. The calculation formula of the prediction error is: ,in, is the prediction error, is the group number of the feature vector corresponding to the scene feature data, For the The numerical labels corresponding to the scene feature data of the group, For the The actual numerical labels corresponding to the scene feature data of the group are obtained; the scenario prediction model is trained until the sum of the prediction errors reaches convergence and the training is stopped.

[0139] The scenario prediction model is specifically a deep neural network model, which includes an input layer, a hidden layer and an output layer; each hidden layer includes multiple neurons, each neuron is connected to the neurons in the next layer, and the connection contains weights, which determine the importance and influence of data transmission in the neural network; an activation function is applied to each neuron between the hidden layer and the output layer, and the activation function introduces nonlinearity, allowing the network to learn more complex patterns and features.

[0140] Step 6: Preset the safety level for the prediction results, and analyze the prediction results output by the scenario prediction model to obtain the current scenario mode.

[0141] In implementation, the method of pre-ranking the prediction results includes:

[0142] The awakening mode in the prediction results is set to the first safety level in advance; the child mode in the prediction results is set to the second safety level; the nap mode and the dating mode in the prediction results are set to the third safety level; the first safety level> the second safety level> the third safety level.

[0143] In implementation, the method of analyzing the prediction results output by the scenario prediction model to obtain the current scenario mode includes:

[0144] Obtain the real-time safety factor, take the difference between the real-time safety factor and the standard safety factor as the safety factor difference, compare the safety factor difference with the preset threshold, if the safety factor difference is greater than the preset threshold, then take the prediction result output by the scenario prediction model as the current scenario mode; if the safety factor difference is less than or equal to the preset threshold, then take the higher-level scenario mode corresponding to the prediction result output by the scenario prediction model as the current scenario mode.

[0145] In implementation, the acquisition of the real-time safety factor includes:

[0146] The first scenario coefficient, the second scenario coefficient, the third scenario coefficient and the fourth scenario coefficient are calculated comprehensively to obtain the real-time safety factor, and the calculation formula is:

[0147] ;

[0148] In the formula, represents the real-time safety factor, , , and is the corresponding weight factor, and 1> > > > >0.

[0149] It should be noted that the preset threshold is determined by those skilled in the art based on actual needs and a large amount of experimental data.

[0150] An example is as follows: when the safety factor difference exceeds a preset threshold, and the prediction result output by the scenario prediction model is child mode, the child mode is used as the current scenario mode; when the safety factor difference is less than or equal to the preset threshold, and the prediction result output by the scenario prediction model is child mode, but due to the current low safety factor, for driving safety reasons, the child mode is upgraded to a refreshing mode with a higher safety factor to ensure higher driving safety.

[0151] This embodiment identifies different driving scenarios through the precise collection and calculation of multi-mode feature data, and then adaptively provides appropriate scenario modes (children mode, wake-up mode, nap mode, and dating mode). By acquiring and analyzing data such as rear-seat gravity difference, driving time, idling state, ambient weather, etc. in real time, it can flexibly adapt to different driving needs. The scenario mode obtained through scenario prediction model analysis and combined with safety level preset can provide drivers with a more reasonable and safe interface experience in different scenarios, which not only enhances the personalized driving experience, but also optimizes driving safety and comfort, making the system more intelligent and humane.

[0152] Example 2

[0153] See also Figure 4 As shown, this embodiment provides a multi-region scenario mode adaptive system for a rear-mounted vehicle screen, including: a first mode analysis module, a second mode analysis module, a third mode analysis module, a fourth mode analysis module, a scenario prediction module and a safety analysis module, and the above modules are connected by wired and / or wireless means to achieve data transmission between each other;

[0154] A first mode analysis module, used to obtain first mode feature data, and calculate a first scene coefficient according to the first mode feature data; the first mode feature data includes a rear-row gravity difference, a first timestamp, and a sound frequency;

[0155] A second mode analysis module, used to obtain second mode characteristic data, and calculate a second scenario coefficient according to the second mode characteristic data; the second mode characteristic data includes continuous driving time, weather impact index and fatigue assessment index;

[0156] A third mode analysis module, used to obtain third mode characteristic data, and calculate a third scenario coefficient according to the third mode characteristic data; the third mode characteristic data includes an idle speed value, a second timestamp, and a main seat inclination angle;

[0157] A fourth mode analysis module, used to obtain fourth mode feature data, and calculate a fourth scene coefficient according to the fourth mode feature data; the fourth mode feature data includes identity recognition data, total number of rides and music timestamp;

[0158] A scenario prediction module, marking the first scenario coefficient, the second scenario coefficient, the third scenario coefficient and the fourth scenario coefficient as scenario feature data, and inputting the scenario feature data into a pre-built scenario prediction model to obtain corresponding prediction results; the prediction results include a child mode, a wake-up mode, a nap mode and a dating mode;

[0159] The safety analysis module is used to pre-set the safety level for the prediction results and analyze the prediction results output by the scenario prediction model to obtain the current scenario mode.

[0160] Example 3

[0161] This embodiment provides an electronic device. The electronic device may include one or more processors and one or more memories. The memories store computer-readable codes, and when the computer-readable codes are executed by the one or more processors, the multi-region scenario mode adaptive method of the rear-mounted vehicle screen as described above may be executed.

[0162] The system according to the implementation mode of the present application can also be implemented with the help of the architecture of the electronic device disclosed in the present application. The electronic device may include a bus, one or more CPUs, a read-only memory (ROM), a random access memory (RAM), a communication port connected to a network, an input / output, a hard disk, etc. A storage device in the electronic device, such as a ROM or a hard disk, can store the multi-region scenario mode adaptation method for the rear-mounted vehicle screen provided in the present application. Furthermore, the electronic device may also include a user interface. Of course, the architecture disclosed in the present application is only exemplary. When implementing different devices, one or more components in the electronic device disclosed in the present application may be omitted according to actual needs.

[0163] Example 4

[0164] The present embodiment provides a computer-readable storage medium storing instructions, which, when executed on a computer, enable the computer to execute the multi-zone scenario mode adaptation method for a rear-mounted vehicle screen of Embodiment 1.

[0165] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0166] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A multi-region scenario mode adaptive method for a rear-mounted vehicle screen, characterized in that: include: Step 1: Acquire first mode feature data, and calculate a first scene coefficient according to the first mode feature data; The first mode characteristic data includes a rear row gravity difference, a first timestamp, and a sound frequency; The rear-seat gravity difference, the first timestamp and the sound frequency are calculated by formulating a formula to obtain a first scene coefficient, wherein the rear-seat gravity difference is used to indicate whether the weight of the rear-seat passenger meets the child standard, and the first timestamp is used to indicate whether it belongs to a preset child riding time period; Voice frequency is used to identify children’s voice characteristics; Step 2: Acquire second mode characteristic data, and calculate a second scenario coefficient according to the second mode characteristic data; the second mode characteristic data includes continuous driving time, weather impact index and fatigue assessment index; The continuous driving time, weather impact index and fatigue assessment index are calculated comprehensively to obtain the second scenario coefficient, wherein the continuous driving time is used to assess the fatigue risk, the weather impact index is used to assess the driving environment risk and the fatigue assessment index is used to judge the driver's state; Step 3: Acquire the third mode characteristic data, and calculate the third scenario coefficient according to the third mode characteristic data; the third mode characteristic data includes an idle speed value, a second timestamp, and a main seat inclination angle; Normalizing the idle speed value, the second timestamp, and the main seat inclination angle to calculate a third scenario coefficient, wherein the idle speed value is used to evaluate the parking state, the main seat inclination angle is used to evaluate the seat tilt, and the second timestamp is used to evaluate the rest time period; Step 4: Acquire fourth mode feature data, and calculate a fourth scene coefficient according to the fourth mode feature data; the fourth mode feature data includes identity recognition data, total number of rides, and music timestamp; The identity recognition data, the total number of passengers and the music timestamp are calculated in a formula to obtain the fourth scenario coefficient; wherein the identity recognition data is preset, and different values ​​are set according to the automatic connection success state or connection failure state of the driver's mobile device and the vehicle screen, the total number of passengers is used to obtain whether there are two people in the car, and the music timestamp is used to obtain the romantic music playing time period; Step 5: Mark the first scene coefficient, the second scene coefficient, the third scene coefficient and the fourth scene coefficient as scene feature data, and input the scene feature data into a pre-built scenario prediction model to obtain corresponding prediction results; the prediction results include a child mode, a wake-up mode, a nap mode and a dating mode; Step 6: Preset the safety level for the prediction results, and analyze the prediction results output by the scenario prediction model to obtain the current scenario mode.

2. The multi-region scenario mode adaptive method for the aftermarket vehicle screen according to claim 1 is characterized in that: The method for calculating and obtaining the first scene coefficient according to the first mode feature data includes: The rear gravity difference, the first timestamp and the sound frequency are marked as c. and ; The calculation formula for the first scene coefficient is: ， ； In the formula, represents the first scene coefficient, is the weight factor of the rear-row gravity difference, and The value of is 1 or 0. Indicates the real-time gravity value. Indicates the standard gravity value; The first time period is preset. If the first timestamp belongs to the first time period, The value of is 1, if the first timestamp does not belong to the preset first time period, The value of is 0; if the sound frequency belongs to the child's sound frequency, The value of is 1, if the sound frequency does not belong to the children's sound frequency, The value of is 0.

3. The multi-region scenario mode adaptive method of the aftermarket vehicle screen according to claim 2 is characterized in that: The continuous driving time is obtained by collecting the driving time counter; The method for obtaining the weather impact index includes: Acquiring weather characteristic data from a local meteorological station according to a current location of the vehicle, the weather characteristic data including wind speed, rainfall, visibility, and light intensity; After normalizing the weather characteristic data, a formula calculation is performed to obtain the weather impact index, which is calculated as follows: ； In the formula, represents the weather impact index, Indicates wind speed, Indicates rainfall, Indicates visibility, Indicates the light intensity, , , and are the corresponding weight factors respectively.

4. The multi-region scenario mode adaptive method for the aftermarket vehicle screen according to claim 3 is characterized in that: The method for obtaining the fatigue assessment index includes: Acquiring driver fatigue characteristic data, wherein the fatigue characteristic data includes the number of sharp turns, speed difference, and braking delay duration; The fatigue characteristic data is calculated in a formula to obtain the fatigue assessment index, and the calculation formula is: ； In the formula, represents the fatigue assessment index, Indicates the number of sharp turns, Indicates the speed difference, Indicates the braking delay time. represents the weight factor for the number of sharp turns, represents the weight factor of the speed difference, A weighting factor representing the braking delay duration.

5. The multi-region scenario mode adaptive method for the aftermarket vehicle screen according to claim 4 is characterized in that: The calculation formula of the second scenario coefficient is: ； In the formula, represents the second scenario coefficient, Ls represents the continuous driving time, is the weight factor of continuous driving time, is a correction constant greater than zero and less than 1.

6. The multi-region scenario mode adaptive method for the aftermarket vehicle screen according to claim 5, characterized in that: The method for obtaining the third scene coefficient includes: The idle speed value, the second timestamp and the main seat inclination angle are marked as , and ; The calculation formula of the third scenario coefficient is: ； In the formula, represents the third scenario coefficient, represents the weighting factor of the idle speed value, represents the weight factor of the main seat inclination, The value of is 1 or 0; The second time period is preset. If the second timestamp belongs to the preset second time period, The value of is 1, if the second timestamp does not belong to the preset second time period, The value of is 0.

7. The multi-region scenario mode adaptive method for a rear-mounted vehicle screen according to claim 6, characterized in that: The method for obtaining the fourth scenario coefficient includes: The identification data, the total number of rides, and the music timestamp are marked as , and ; ； In the formula, represents the fourth scenario coefficient, represents the weight factor of the total number of passengers, represents the weight factor of the music timestamp; The identity recognition data corresponding to the successful automatic connection state between the driver's mobile device and the vehicle screen is set to 1, and the identity recognition data corresponding to the automatic connection recognition state between the driver's mobile device and the vehicle screen is set to 0; The preset driving number threshold is 2. If the total number of driving is greater than or less than the preset driving number threshold, then The value is 0. If the total number of passengers is equal to the preset threshold, then The value is 1; preset music time period, if the music timestamp belongs to the preset music time period, The value of is 1, if the music timestamp does not belong to the preset music time period, The value of is 0.

8. The multi-region scenario mode adaptive method for a rear-mounted vehicle screen according to claim 7, characterized in that: The specific training process of the scenario prediction model includes: Collect A group of scene feature data in advance, where A is an integer greater than 1, set corresponding prediction results for the scene feature data, set different digital labels for the child mode, the wake-up mode, the nap mode, and the dating mode, and set corresponding prediction results for different scene feature data in group A in turn; Mark the prediction results as digital labels, and convert the scene feature data and the corresponding digital labels into a corresponding set of feature vectors; Each group of feature vectors is used as the input of a scenario prediction model. The scenario prediction model takes a group of digital labels corresponding to each group of scene feature data as output, and takes the actual digital labels corresponding to each group of scene feature data as prediction targets, where the actual digital labels are pre-set digital labels corresponding to the scene feature data; minimizing the sum of prediction errors of all scene feature data is used as a training target; the scenario prediction model is trained until the sum of prediction errors reaches convergence, and the training is stopped. The scenario prediction model is a deep neural network model.

9. The multi-region scenario mode adaptive method for a rear-mounted vehicle screen according to claim 8, characterized in that: Methods for pre-ranking prediction results include: The awakening mode in the prediction results is set to the first safety level in advance; the child mode in the prediction results is set to the second safety level; the nap mode and the dating mode in the prediction results are set to the third safety level; the first safety level> the second safety level> the third safety level.

10. The multi-region scenario mode adaptive method for a rear-mounted vehicle screen according to claim 9, characterized in that: Methods for analyzing the prediction results output by the scenario prediction model to obtain the current scenario mode include: Obtaining a real-time safety factor, taking the difference between the real-time safety factor and the standard safety factor as the safety factor difference, comparing the safety factor difference with a preset threshold, and if the safety factor difference is greater than the preset threshold, taking the prediction result output by the scenario prediction model as the current scenario mode; if the safety factor difference is less than or equal to the preset threshold, taking a higher-level scenario mode corresponding to the prediction result output by the scenario prediction model as the current scenario mode; Wherein, the acquisition of the real-time safety factor includes: The first scenario coefficient, the second scenario coefficient, the third scenario coefficient and the fourth scenario coefficient are calculated comprehensively to obtain the real-time safety factor, and the calculation formula is: ； In the formula, represents the real-time safety factor, , , and is the corresponding weight factor, and 1> > > > >

0.

11. A multi-region scenario mode adaptive system for a rear-mounted vehicle screen, used to implement a multi-region scenario mode adaptive method for a rear-mounted vehicle screen according to any one of claims 1 to 10, characterized in that: include: A first mode analysis module, used to obtain first mode feature data, and calculate a first scene coefficient according to the first mode feature data; The first mode characteristic data includes a rear row gravity difference, a first timestamp, and a sound frequency; A second mode analysis module, used to obtain second mode feature data, and calculate a second scene coefficient according to the second mode feature data; The second mode characteristic data includes continuous driving time, weather impact index and fatigue assessment index; A third mode analysis module, used to obtain third mode characteristic data, and calculate a third scenario coefficient according to the third mode characteristic data; the third mode characteristic data includes an idle speed value, a second timestamp, and a main seat inclination angle; A fourth mode analysis module, used to obtain fourth mode feature data, and calculate a fourth scene coefficient according to the fourth mode feature data; the fourth mode feature data includes identity recognition data, total number of rides and music timestamp; A scenario prediction module, marking the first scenario coefficient, the second scenario coefficient, the third scenario coefficient and the fourth scenario coefficient as scenario feature data, and inputting the scenario feature data into a pre-built scenario prediction model to obtain corresponding prediction results; the prediction results include a child mode, a wake-up mode, a nap mode and a dating mode; The safety analysis module is used to pre-set the safety level for the prediction results and analyze the prediction results output by the scenario prediction model to obtain the current scenario mode.

Images