Vehicle safety state prediction and emergency call system and method

Abstract

The invention discloses a vehicle safety state prediction and first-aid calling system and method, which are used as an independent system and are suitable for vehicles which are delivered out of a factory, not delivered out of the factory and have any purpose and performance, and a cloud service center is provided with a cloud service center. According to the method, the optimal model parameters of the vehicle state classification prediction model most matched with the vehicles of the corresponding types are issued based on the types of the vehicles, so that the method can be suitable for accurate detection of traffic accidents of the vehicles of different types, and the use applicability is greatly improved; meanwhile, the problem of emergency call failure caused by the fact that people in the vehicle cannot enter a call flow due to serious injury in a serious traffic accident can be avoided through automatic call, and the timeliness of rescue of the people in the vehicle is guaranteed; besides calling for help in an emergency center, on-site calling for help is carried out through the on-site calling-for-help module outside the vehicle, so that nearby or passing pedestrians and vehicles can be brought into help seeking objects, and therefore, the personnel in the vehicle can be ensured to obtain maximum rescue strength.

Description

Technical Field

[0001] The present invention belongs to the technical field of vehicle safety status monitoring and emergency call, and in particular relates to a system and method for vehicle safety status prediction and emergency call. Background Art

[0002] At present, vehicle-mounted emergency call systems have been widely used in vehicles to ensure that people in the vehicle can get timely assistance when a traffic accident occurs. Among them, the prior art 1 with application number CN201410030293 discloses a vehicle-mounted emergency call method and system. The call for help in this scheme is divided into two modes: automatic and manual. It can realize automatic and manual emergency calls when a traffic accident occurs. At the same time, the prior art 2 with application number CN117002432A discloses a vehicle collision rescue method. In this scheme, it is necessary to add a functional module to the vehicle system during vehicle production, and install sensors in multiple locations of the vehicle to determine the impact force and its position. In this way, with the help of multiple sensors on the vehicle body, an emergency call can be made when a traffic accident is detected.

[0003] However, the disadvantage of the aforementioned prior art 1 is that in some more serious traffic accidents, the people in the vehicle lose the ability to call for help independently due to serious injuries, and therefore cannot enter the call for help process, resulting in the failure of the emergency call. At the same time, the scheme does not introduce the timing and specific method of triggering automatic call for help, and in the scheme, the objects of emergency call for help are limited to the emergency center, which is not conducive to the accident vehicle and personnel obtaining maximum rescue force; and the aforementioned prior art 2 is not applicable to vehicles that have already left the factory, and its applicability is poor. In addition, since vehicles with different uses or performances can withstand different impact forces, the judgment of traffic accidents by setting thresholds in prior art 2 does not have universality; therefore, based on the aforementioned deficiencies, how to provide a vehicle safety status prediction and emergency call system with strong applicability, which can automatically trigger the emergency call process when a traffic accident occurs, and is conducive to the people in the vehicle obtaining maximum rescue force, has become a problem to be solved urgently. Summary of the invention

[0004] The purpose of the present invention is to provide a system and method for predicting the safety status of a vehicle and for emergency calls, so as to solve the problems existing in the prior art of poor applicability, the objects of help being limited to emergency centers, the inability to obtain rescue forces to the maximum extent, and the manual calls being prone to call failures due to serious injuries.

[0005] In order to achieve the above object, the present invention adopts the following technical solutions:

[0006] In a first aspect, a system for predicting vehicle safety status and emergency calling is provided, comprising:

[0007] The vehicle-mounted terminal is communicatively connected to a cloud service center, and is used to receive the optimal model parameters of the vehicle state classification prediction model that best matches the vehicle type issued by the cloud service center, and generate an optimal vehicle state classification prediction model according to the optimal model parameters;

[0008] A driving data collection module, wherein the driving data collection module is electrically connected to the vehicle terminal and is used to collect driving data of the vehicle within a first preset time period before a target time, and the target time is the time corresponding to when the vehicle changes from a moving state to a stationary state;

[0009] The vehicle terminal is used to perform feature extraction processing on the driving data to obtain driving feature data, and input the driving feature data into the optimal vehicle state classification prediction model to obtain the state information of the vehicle at the target time;

[0010] The vehicle terminal is used to initiate a voice inquiry into the vehicle when the status information is an accident status;

[0011] The vehicle-mounted terminal is also used to make an emergency call to the emergency center based on the vehicle's location information when it determines that no answer information to the voice inquiry has been received, and to control the operation of the on-site rescue module outside the vehicle to send out an on-site distress signal.

[0012] Based on the above disclosed content, the system provided by the present invention constructs an optimal vehicle state classification prediction model suitable for the vehicle by acquiring the optimal model parameters of the vehicle state classification prediction model that best matches the vehicle type; then, the driving data within the first preset time period before the vehicle stops is collected through the driving data collection module; then, the on-board terminal performs feature extraction processing on it to obtain driving feature data; then, the on-board terminal inputs the driving feature data into the aforementioned optimal vehicle state classification prediction model to obtain the vehicle state information; wherein, when the vehicle state information is an accident state, the on-board terminal initiates a voice inquiry into the vehicle, and when it is determined that no response information is received, based on the vehicle's location information, an emergency call is made to the emergency center, and an on-site distress signal is sent at the same time to ensure that the vehicle personnel obtain maximum rescue force.

[0013] Through the above design, the present invention, as an independent system, is applicable to vehicles that have been manufactured, have not been manufactured, and have any purpose and performance. The cloud service center, based on the type of vehicle, sends the optimal model parameters of the vehicle status classification prediction model that best matches the corresponding type of vehicle. In this way, it can be applied to the accurate detection of traffic accidents involving different types of vehicles, thereby greatly improving the applicability of use; at the same time, the automatic call for help of the present invention can avoid the problem of failure of emergency rescue calls due to serious injuries of occupants in more serious traffic accidents, thereby ensuring the timeliness of rescue of occupants in the vehicle; in addition, in addition to calling for help from the emergency center, the present invention also calls for help on the spot through the off-site call for help module, based on which nearby or passing pedestrians and vehicles can be included in the rescue objects, thereby ensuring that the vehicle personnel obtain maximum rescue force; therefore, the present invention is very suitable for large-scale application and promotion in the field of vehicle safety status monitoring and emergency call technology.

[0014] In a possible design, the driving data includes: vehicle information of the vehicle, and speed information, acceleration information, and angular velocity information of the vehicle within a first preset time period before a target time;

[0015] Among them, the vehicle terminal includes: a calculation and prediction module;

[0016] a calculation and prediction module, used for performing feature extraction processing on the speed information, the acceleration information and the angular velocity information, so as to obtain time series statistical features, frequency domain statistical features and segmented statistical features corresponding to the speed information, the acceleration information and the angular velocity information respectively;

[0017] A calculation and prediction module, configured to generate the driving characteristic data by using the vehicle information, and the time series statistical features, frequency domain statistical features and segment statistical features corresponding to the speed information, the acceleration information and the angular velocity information;

[0018] The calculation and prediction module is also used to input the driving characteristic data into the optimal vehicle state classification prediction model to obtain the state information of the vehicle at the target time.

[0019] In a possible design, the speed information is a discrete signal sequence;

[0020] The calculation and prediction module is used to perform second-order difference processing on the discrete signal sequence to obtain a second-order difference velocity sequence, and calculate the time series statistical characteristics corresponding to the velocity information based on the second-order difference velocity sequence;

[0021] A calculation and prediction module, used for performing Fourier transform processing on the discrete signal sequence to obtain a frequency domain speed signal sequence, and generating frequency domain statistical features corresponding to the speed information according to the frequency domain speed signal sequence;

[0022] A calculation and prediction module, used for dividing the discrete signal sequence into a plurality of molecular signal sequences according to the order of acquisition time from the beginning to the end, wherein the lengths of the plurality of molecular signal sequences are the same;

[0023] The calculation and prediction module is also used to determine the characteristic value of each sub-signal sequence, and use the characteristic data of each sub-signal sequence to form the segmented statistical features corresponding to the speed information, wherein the characteristic value of any sub-signal sequence includes the maximum value, minimum value, average value and median of any sub-signal sequence.

[0024] In a possible design, the calculation and prediction modules are used to generate the time series statistical features and frequency domain statistical features corresponding to the speed information using the following formula (1) and formula (2), respectively;

[0025]

[0026] In the above formula (1), s1 represents the time series statistical characteristics corresponding to the speed information, d 2 x[i] represents the i-th second-order differential velocity value in the second-order differential velocity sequence, wherein the i-th second-order differential velocity value is the second-order differential value corresponding to the i-th signal in the discrete signal sequence, and N represents the total number of signal points in the discrete signal sequence;

[0027] In the above formula (2), s2 represents the frequency domain statistical characteristics corresponding to the speed information, wherein X[k] represents the kth speed value in the frequency domain speed signal sequence, k c is the starting point corresponding to the high frequency component in the frequency domain velocity signal sequence, k c =f0 / 4, and f0 is the sampling frequency of the discrete signal sequence.

[0028] In a possible design, the driving data acquisition module includes: a speed measurement module, an acceleration measurement module and a gyroscope measurement module;

[0029] A speed measurement module, used to collect speed information of the vehicle within a first preset time period before the target time, and send the speed information to the vehicle terminal;

[0030] An acceleration measurement module, used to collect acceleration information of the vehicle within a first preset time period before the target time, and send the information to the vehicle terminal;

[0031] A gyroscope measurement module, used for collecting angular velocity information of the vehicle within a first preset time period before the target time, and sending the information to the vehicle terminal;

[0032] The speed measurement module is further used to collect a first speed of the vehicle at the target time and a second speed within a second preset time after the target time, and transmit the first speed and each second speed to the vehicle terminal;

[0033] The vehicle-mounted terminal is also used to determine that the vehicle is in a stationary state when it is determined that the first speed is equal to each second speed, and to obtain the speed information, acceleration information and angular velocity information transmitted by the speed measurement module, the acceleration measurement module and the gyroscope measurement module within a first preset time period before the target moment, so as to generate the driving data using the obtained speed information, acceleration information and angular velocity information.

[0034] In a possible design, it also includes: a network and communication module, a GPS positioning module and a human-computer interaction module, wherein the human-computer interaction module includes a vehicle information input submodule, an emergency alarm button and a voice emergency alarm submodule, and the vehicle information input submodule is used to obtain the vehicle information of the vehicle in response to the human-computer interaction and transmit it to the vehicle terminal;

[0035] GPS positioning module, used to obtain the vehicle's location information and transmit the location information to the vehicle terminal;

[0036] The emergency alarm button and voice emergency alarm submodule are used to generate emergency alarm information and voice alarm information in response to human-computer interaction and send them to the vehicle terminal;

[0037] The vehicle-mounted terminal is used to make an emergency call to the emergency center based on the vehicle's location information when receiving emergency alarm information and / or voice alarm information, and control the operation of the off-site rescue module to send out an on-site distress signal.

[0038] In a possible design, the cloud service center is used to obtain multiple driving feature data sets, wherein each driving feature data set corresponds to a vehicle type, and each driving feature data set contains sample driving feature data of multiple sample vehicles of the corresponding vehicle type;

[0039] The cloud service center is used to use each driving feature data set as an initial training data set;

[0040] The cloud service center is used to perform parameter optimization processing on the initial vehicle state classification prediction model based on each initial training data set to obtain the optimal model parameters of the vehicle state classification prediction model for each vehicle type;

[0041] The vehicle terminal is further used to send a parameter sending request to the cloud service center, and the parameter sending request includes vehicle information;

[0042] The cloud service center is also used to filter out the optimal model parameters of the vehicle type corresponding to the target vehicle from the optimal model parameters corresponding to each vehicle type based on the vehicle information in the parameter sending request, and send them to the vehicle terminal, so that the vehicle terminal can generate the optimal vehicle state classification prediction model according to the optimal model parameters sent by the cloud service center, wherein the target vehicle is the vehicle corresponding to the vehicle terminal.

[0043] In a possible design, for any initial training data set, the cloud service center is used to divide the initial training data set into a training set and a test set;

[0044] The cloud service center is used to generate a chromosome population based on the model parameters of the initial vehicle state classification prediction model, wherein each chromosome in the chromosome population corresponds to a set of initial model parameters;

[0045] The cloud service center is used to initialize the number of evolutions g and obtain the chromosome population at the g-th evolution to determine the model parameters corresponding to each chromosome in the chromosome population at the g-th evolution, wherein when g is 1, the model parameters corresponding to each chromosome at the g-th evolution are the initial model parameters corresponding to each chromosome;

[0046] The cloud service center is used to generate a vehicle state classification prediction model corresponding to each chromosome at the g-th evolution according to the model parameters corresponding to each chromosome at the g-th evolution;

[0047] The cloud service center is used to train the vehicle state classification prediction model corresponding to each chromosome at the g-th evolution using the training set, and calculate the fitness of each chromosome at the g-th evolution according to the model output data of the vehicle state classification prediction model corresponding to each chromosome at the g-th evolution, wherein the fitness of any chromosome is the loss function value of the vehicle state classification prediction model corresponding to the any chromosome;

[0048] The cloud service center is used to determine whether the evolution stop condition is met, wherein the evolution stop condition is that the minimum fitness at the g-th evolution is less than or equal to the fitness threshold, or g is greater than or equal to the maximum number of evolutions;

[0049] The cloud service center is used to perform individual mutation processing on the chromosome population at the g-th evolution based on the differential strategy to obtain the mutation population at the g-th evolution when it is determined that the evolution stop condition is not met, and to perform crossover update processing on the mutation population at the g-th evolution to obtain the crossover population at the g-th evolution;

[0050] The cloud service center is used to calculate the fitness of each crossover individual in the crossover population at the g-th evolution, and based on the fitness of each crossover individual, perform a selection operation on each crossover individual to obtain the individual population at the g-th evolution;

[0051] The cloud service center is used to perform disturbance processing on the individual population to obtain a disturbed population at the g-th evolution;

[0052] The cloud service center is used to add 1 to g, update the chromosome population at the g-th evolution to the perturbation population at the g-1-th evolution, and re-determine the model parameters corresponding to each chromosome in the chromosome population at the g-th evolution, until the evolution stop condition is met, so as to use the vehicle state classification prediction model corresponding to the chromosome with the smallest fitness in the chromosome population that meets the evolution stop condition as the pre-trained model;

[0053] The cloud service center is also used to use the test set to test the pre-trained model, and after the test passes, use the model parameters corresponding to the pre-trained model as the optimal model parameters for a specified vehicle type, wherein the specified vehicle type is the vehicle type corresponding to any of the initial training data sets.

[0054] In a possible design, for any chromosome in the chromosome population at the g-th evolution, the cloud service center is used to generate a first random positive integer, a second random positive integer, and a third random positive integer, wherein the first random positive integer, the second random positive integer, and the third random positive integer are different from each other and are all less than or equal to the total number of chromosomes in the chromosome population;

[0055] The cloud service center is used to select the r1th chromosome, the r2th chromosome and the r3th chromosome from the chromosome population at the gth evolution, and r1, r2, r3 represent the first random positive integer, the second random positive integer and the third random positive integer respectively;

[0056] The cloud service center is used to perform an individual mutation operation on any of the chromosomes according to the r1th chromosome, the r2th chromosome, and the r3th chromosome, so as to obtain a mutant individual corresponding to the any of the chromosomes;

[0057] For any mutant individual in the mutant population during the g-time evolution, the cloud service center is used to generate a first random number and a fourth random positive integer, wherein the value interval of the first random number is (0,1), and the fourth random positive integer is less than or equal to the total number of genes of any mutant individual;

[0058] The cloud service center is further used to perform a crossover update operation on each gene in any variant individual according to the first random number and the fourth random positive integer, so as to obtain a crossover individual corresponding to any variant individual after the crossover update operation.

[0059] In a second aspect, a method for predicting a vehicle safety status and making an emergency call is provided, which is executed by a vehicle-mounted terminal in the system for predicting a vehicle safety status and making an emergency call based on the first aspect or any possible design of the first aspect, and the method comprises:

[0060] Receiving the optimal model parameters of the vehicle state classification prediction model that best matches the vehicle type issued by the cloud service center, and generating an optimal vehicle state classification prediction model based on the optimal model parameters;

[0061] Receiving driving data of the vehicle within a first preset time period before a target time sent by a driving data acquisition module, wherein the target time is a time corresponding to when the vehicle changes from a moving state to a stationary state;

[0062] Performing feature extraction processing on the driving data to obtain driving feature data, and inputting the driving feature data into the optimal vehicle state classification prediction model to obtain state information of the vehicle at the target time;

[0063] If the status information is an accident status, a voice inquiry is initiated to the vehicle, and it is determined whether an answer to the voice inquiry is received;

[0064] If not, an emergency call is made to the emergency center based on the vehicle's location information, and the on-site rescue module outside the vehicle is controlled to operate to send out an on-site distress signal.

[0065] In a third aspect, an electronic device is provided, comprising a memory, a processor and a transceiver which are communicatively connected in sequence, wherein the memory is used to store a computer program, the transceiver is used to send and receive messages, and the processor is used to read the computer program to execute the method for predicting the vehicle safety status and making an emergency call as described in the second aspect.

[0066] In a fourth aspect, a storage medium is provided, on which instructions are stored. When the instructions are executed on a computer, the method for predicting the vehicle safety status and making an emergency call as described in the second aspect is executed.

[0067] In a fifth aspect, a computer program product comprising instructions is provided, which, when executed on a computer, causes the computer to execute the method for predicting the vehicle safety status and making an emergency call as described in the second aspect.

[0068] Beneficial effects:

[0069] (1) As an independent system, the present invention is applicable to vehicles that have been manufactured, those that have not been manufactured, and vehicles of any purpose and performance. The cloud service center, based on the type of vehicle, issues the optimal model parameters of the vehicle status classification prediction model that best matches the corresponding type of vehicle. In this way, it can be applied to the accurate detection of traffic accidents involving different types of vehicles, thereby greatly improving the applicability of use. At the same time, the automatic call for help of the present invention can avoid the problem of failure of emergency rescue calls due to serious injuries of occupants in more serious traffic accidents, thereby ensuring the timeliness of the rescue of occupants. In addition, in addition to calling for help from the emergency center, the present invention also calls for help on the spot through the off-site call for help module. Based on this, nearby or passing pedestrians and vehicles can be included in the rescue objects, thereby ensuring that the vehicle occupants receive the maximum rescue force. Therefore, the present invention is very suitable for large-scale application and promotion in the field of vehicle safety status monitoring and emergency call technology.

[0070] (2) The present invention adopts an improved differential evolution algorithm to optimize model parameters. In this way, while having a higher convergence speed, the result can have the ability to jump out of the local optimal solution, taking into account the efficiency, performance and accuracy of the algorithm. BRIEF DESCRIPTION OF THE DRAWINGS

[0071] Figure 1 A schematic diagram of the architecture of a system for predicting vehicle safety status and emergency calling provided by an embodiment of the present invention;

[0072] Figure 2 A flowchart of the system for predicting vehicle safety status and emergency calling provided by an embodiment of the present invention;

[0073] Figure 3 A schematic flow chart of the steps of a method for predicting vehicle safety status and making an emergency call provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0074] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the present invention will be briefly introduced below in combination with the drawings and the description of the embodiments or the prior art. Obviously, the following description of the structure of the drawings is only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. It should be noted that the description of these embodiments is used to help understand the present invention, but does not constitute a limitation of the present invention.

[0075] 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 only used to distinguish one unit from another unit. For example, a first unit can be referred to as a second unit, and similarly, a second unit can be referred to as a first unit without departing from the scope of the exemplary embodiments of the present invention.

[0076] It should be understood that the term "and / or" that may appear in this article is merely a description of the association relationship between associated objects, indicating that three relationships may exist. For example, A and / or B can represent three situations: A exists alone, B exists alone, and A and B exist at the same time. The term " / and" that may appear in this article describes another type of association object relationship, indicating that two relationships may exist. For example, A / and B can represent two situations: A exists alone, and A and B exist alone. In addition, the character " / " that may appear in this article generally indicates that the previous and next associated objects are in an "or" relationship.

[0077] Example:

[0078] See also Figure 1 As shown, the vehicle safety status prediction and emergency call system provided by the present embodiment may include but is not limited to: a cloud service center, a vehicle-mounted terminal, a driving data collection module and an off-vehicle on-site rescue module; wherein the cloud service center is used to train vehicle status classification prediction models corresponding to different types of vehicles to obtain optimal model parameters corresponding to different types of vehicles; the driving data collection module is used to collect driving data of the vehicle before parking; and the vehicle-mounted terminal can play the role of model deployment and vehicle status prediction based on the driving data collected by the driving data collection module; at the same time, it also plays the role of automatic and manual emergency calls. In addition, the off-vehicle on-site rescue module serves the purpose of on-site rescue.

[0079] The following discloses the specific working process of each of the above modules:

[0080] In this embodiment, the vehicle-mounted terminal is communicatively connected to a cloud service center, and is used to receive the optimal model parameters of the vehicle state classification prediction model that best matches its own vehicle type and is issued by the cloud service center, and generate the optimal vehicle state classification prediction model based on the optimal model parameters; in specific implementation, a database is set up in the cloud service center, and the database stores sample driving feature data of sample vehicles of different types, and the cloud service center is used to train the initial vehicle state classification prediction model (i.e., the untrained model) based on the sample driving feature data of sample vehicles of different types, so as to obtain the optimal model parameters of different types of vehicles; then, the vehicle-mounted terminal can send a parameter issuance request to the cloud service center (which contains vehicle information, such as vehicle model, purpose, displacement and / or motor power, etc., and the vehicle information facilitates the matching of vehicle types), so as to obtain the optimal model parameters that best match its own vehicle type; wherein, for example, the aforementioned initial vehicle state classification prediction model can be, but is not limited to, a logistic regression model, and the optimization process of the optimal model parameters is described in detail below.

[0081] In this way, after the vehicle-mounted terminal constructs the optimal vehicle state classification prediction model based on the received optimal model parameters, it can use the driving data collected by the data acquisition module to predict the vehicle state; among them, the data collection process of the driving data acquisition module and the process of vehicle state prediction by the vehicle-mounted terminal are shown as follows.

[0082] In this embodiment, the driving data acquisition module is electrically connected to the vehicle terminal and is used to collect the driving data of the vehicle within a first preset time before the target time, and the target time is the time corresponding to when the vehicle changes from a moving state to a stationary state; in specific implementation, the driving data acquisition module may include, but is not limited to: a speed measurement module, an acceleration measurement module and a gyroscope measurement module; wherein the specific collection process of the driving data is:

[0083] The speed measurement module is used to collect the speed information of the vehicle within a first preset time before the target moment, and send it to the vehicle-mounted terminal; similarly, the acceleration measurement module is used to collect the acceleration information of the vehicle within a first preset time before the target moment, and send it to the vehicle-mounted terminal; and the gyroscope measurement module is used to collect the angular velocity information of the vehicle within a first preset time before the target moment, and send it to the vehicle-mounted terminal; at the same time, the speed measurement module is also used to collect the first speed of the vehicle at the target moment and the second speed within a second preset time after the target moment, and transmit the first speed and each second speed to the vehicle-mounted terminal; in this way, the vehicle-mounted terminal can determine whether the vehicle is in a stopped state according to the continuous length of time when the vehicle's speed is 0.

[0084] That is, the vehicle-mounted terminal is also used to determine that the vehicle is in a stationary state when it is determined that the first speed is equal to each second speed, and obtain the speed information, acceleration information and angular velocity information transmitted by the speed measurement module, the acceleration measurement module and the gyroscope measurement module within a first preset time period before the target moment, so as to generate the driving data using the obtained speed information, acceleration information and angular velocity information; in this embodiment, the example speed information, acceleration information and angular velocity information are all discrete signal sequences, that is, including the speed, acceleration and angular velocity at different times within the first preset time period before the target moment, and the speed, acceleration and angular velocity are all speed information in three directions, that is, including the speed, acceleration and angular velocity information in the vehicle's forward direction, the lateral direction of the vehicle's forward direction, and the vertical upward direction.

[0085] At the same time, the driving data may also include but is not limited to vehicle information of the vehicle, such as vehicle model, purpose, displacement and / or motor power, etc.; in this way, the driving data before the vehicle stops can be composed based on the vehicle information, as well as the aforementioned speed information, acceleration information and angular velocity information.

[0086] Based on the above explanation, the vehicle-mounted terminal obtains the speed of the vehicle in real time. When the speed is detected to be 0 and the duration is greater than the second preset duration (that is, after the vehicle speed is 0, the speed within the second preset duration after the corresponding moment is 0), the vehicle's moving state is determined to be stationary; at this time, it is necessary to obtain the acceleration information, speed information and angular velocity information transmitted by the above-mentioned acceleration measurement module, speed measurement module and gyroscope measurement module within the first preset time before the target moment, and combine them with the vehicle information to form the driving data before the vehicle stops.

[0087] In this way, after obtaining the driving data before the vehicle stops, the vehicle terminal can use the aforementioned optimal vehicle state classification prediction model to predict the vehicle state, that is: the vehicle terminal is used to perform feature extraction processing on the driving data to obtain driving feature data, and input the driving feature data into the optimal vehicle state classification prediction model to obtain the state information of the vehicle at the target time; in specific implementation, feature extraction is performed on the driving data, mainly on the speed information, acceleration information and angular velocity, and the specific extraction process is as follows:

[0088] In a specific implementation, for example, the vehicle-mounted terminal may include, but is not limited to: a calculation and prediction module; wherein the calculation and prediction module is used to perform feature extraction processing on the speed information, the acceleration information and the angular velocity information, so as to obtain the time series statistical features, frequency domain statistical features and segmented statistical features corresponding to the speed information, the acceleration information and the angular velocity information, respectively; then, the calculation and prediction module uses the vehicle information, and the time series statistical features, frequency domain statistical features and segmented statistical features corresponding to the speed information, the acceleration information and the angular velocity information to generate the driving feature data.

[0089] Specifically, since the feature extraction processes of velocity information, angular velocity information, and acceleration information are the same, the following takes velocity information as an example to illustrate the specific calculation process of its time series statistical features, frequency domain statistical features, and segmented statistical features:

[0090] In this embodiment, it has been explained above that the speed information is a discrete signal sequence. Therefore, the calculation and prediction module is used to perform second-order difference processing on the discrete signal sequence to obtain a second-order differential speed sequence, and based on the second-order differential speed sequence, calculate the time series statistical characteristics corresponding to the speed information.

[0091] Wherein, for any signal point in the discrete signal sequence corresponding to the speed information, the second-order difference processing formula of any signal point can be, but is not limited to, as shown in the following formula (3):

[0092] d 2 x[i]=x[i+2]-2x[i+1]+x[i] (3)

[0093] In the above formula (3), x[i] represents the i-th signal point in the discrete signal sequence corresponding to the speed information, d 2 x[i] represents the i-th second-order differential velocity value in the second-order differential velocity sequence, that is, the second-order differential value corresponding to the aforementioned i-th signal, i=1,2,...,N, and N represents the total number of signal points in the discrete signal sequence, that is, the total number of velocity values ​​in the velocity information.

[0094] After completing the second-order difference processing of the speed information, the time series statistical characteristics corresponding to the speed information can be calculated based on this; optionally, the time series statistical characteristics corresponding to the speed information can be calculated by, for example but not limited to, using the following formula (1).

[0095]

[0096] In the above formula (1), s1 represents the time series statistical characteristics corresponding to the speed information, d 2x[i] represents the i-th second-order differential velocity value in the second-order differential velocity sequence, wherein the i-th second-order differential velocity value is the second-order differential value corresponding to the i-th signal in the discrete signal sequence, and N represents the total number of signal points in the discrete signal sequence.

[0097] After calculating the time series statistical features corresponding to the speed information based on the above formulas (1) and (3), the frequency domain statistical features can be calculated, that is:

[0098] The calculation and prediction module is used to perform Fourier transform processing on the discrete signal sequence to obtain a frequency domain speed signal sequence, and generate frequency domain statistical features corresponding to the speed information based on the frequency domain speed signal sequence; in this embodiment, the following formula (4) can be used as an example but is not limited to perform Fourier transform processing on the aforementioned discrete signal sequence.

[0099]

[0100] In the above formula (4), X[k] represents the frequency domain signal corresponding to the i-th signal point in the above discrete signal sequence, j is an imaginary unit, and k=01,2,...,N-1.

[0101] Thus, after the Fourier transform of the speed information is completed by the aforementioned formula (4), the frequency domain statistical characteristics corresponding to the speed information can be calculated based on this. For example, the following formula (2) can be used, but is not limited to, to calculate the aforementioned frequency domain statistical characteristics.

[0102]

[0103] In the above formula (2), s2 represents the frequency domain statistical characteristics corresponding to the speed information, wherein X[k] represents the kth speed value in the frequency domain speed signal sequence, k c is the starting point corresponding to the high frequency component in the frequency domain velocity signal sequence, k c =f0 / 4, and f0 is the sampling frequency of the discrete signal sequence.

[0104] Thus, through the above method, the frequency domain statistical features of the speed information can be extracted; then, the segmented statistical features can be calculated, and the process is as follows:

[0105] The calculation and prediction module is used to divide the discrete signal sequence into a number of sub-signal sequences (wherein the lengths of the sub-signal sequences are the same) in the order of acquisition time from front to back; then, it is used to determine the characteristic value of each sub-signal sequence; finally, the characteristic data of each sub-signal sequence can be used to form the segmented statistical features corresponding to the speed information. In this embodiment, the characteristic value of any sub-signal sequence can include but is not limited to the maximum value, minimum value, average value and median of any sub-signal.

[0106] Thus, through the aforementioned method, the extraction of the three characteristic information of the velocity information can be completed; and then, with the same principle, the extraction of the time series statistical features, frequency domain statistical features and segmented statistical features of the angular velocity information and acceleration information can be completed.

[0107] Next, the calculation and prediction module is used to generate the driving characteristic data by utilizing the vehicle information, as well as the time series statistical characteristics, frequency domain statistical characteristics and segmented statistical characteristics corresponding to the speed information, the acceleration information and the angular velocity information; finally, the driving characteristic data is input into the optimal vehicle state classification prediction model to obtain the state information of the vehicle at the target time.

[0108] In a specific implementation, for example, the state information of the vehicle at the target time is an accident state or a normal state. Therefore, the vehicle terminal (i.e., the aforementioned calculation and prediction module) is used to determine whether to enter the emergency call process according to the state information of the vehicle, that is:

[0109] The vehicle-mounted terminal is used to initiate a voice inquiry into the vehicle when the status information is an accident status; then, when the vehicle-mounted terminal determines that no answer information to the voice inquiry is received, it makes an emergency call to the emergency center based on the vehicle's location information, and controls the operation of the off-vehicle on-site rescue module to send out an on-site distress signal; in this way, in addition to calling for help from the emergency center, this embodiment also uses the off-vehicle on-site rescue module to include nearby or passing pedestrians and vehicles as objects of assistance, so that when an accident occurs on a mountain road or a hidden road, making it difficult for vehicles and personnel to be found (such as a vehicle running off the road and falling into a woods or a gully), maximum rescue force can be obtained.

[0110] Furthermore, when the on-board terminal receives the answer information corresponding to the voice inquiry and recognizes that the answer information is confirming that the vehicle's safety status is normal, the vehicle's status information is marked as normal. At this time, no emergency call is made; at the same time, if the answer information received is confirming that the vehicle's safety status is an accident and / or requires an emergency call, it is also necessary to make an emergency call to the emergency center based on the vehicle's location information, and control the operation of the off-site rescue module to send out an on-site distress signal.

[0111] For further information, see Figure 1 As shown, the present embodiment can also perform a manual alarm call, that is, for example, the system can also include: a network and communication module, a GPS positioning module, a human-computer interaction module, a vehicle system communication module and a storage module (the aforementioned modules are arranged in the vehicle-mounted terminal, and the off-vehicle on-site emergency call module also belongs to the vehicle-mounted terminal); wherein, the human-computer interaction module includes a vehicle information entry submodule, an emergency alarm button and a voice emergency alarm submodule, and the vehicle information entry submodule is used to obtain the vehicle information of the vehicle in response to the human-computer interaction, and transmit it to the vehicle-mounted terminal; in this way, the vehicle information entry can be completed; at the same time, the vehicle system communication module is also communicatively connected to the vehicle's vehicle computer, which can read information such as the vehicle's motor power, so as to obtain complete vehicle information in combination with the vehicle information entry submodule.

[0112] Furthermore, the GPS positioning module is used to obtain the vehicle's location information and transmit the location information to the on-board terminal; the emergency alarm button and voice emergency alarm sub-module are used to generate emergency alarm information and voice alarm information in response to human-computer interaction, and send them to the on-board terminal; finally, the on-board terminal can be used to make an emergency call to the emergency center based on the vehicle's location information when receiving the emergency alarm information and / or voice alarm information, and control the operation of the on-site rescue module outside the vehicle to send out an on-site distress signal; in this way, the emergency call can be completed by the people in the vehicle themselves.

[0113] In addition, for example, the on-site distress signal can be but is not limited to an audio-visual signal, that is, the speaker plays a pre-set and recorded distress sound, and the flash emits visible light of different colors to attract the attention of passers-by and vehicles near the scene and achieve the purpose of calling for help; at the same time, for example, the aforementioned storage module is used to store the aforementioned driving data and vehicle information, so that the on-board terminal can call up the data when predicting the vehicle status.

[0114] Therefore, through the above explanation, the system provided by this embodiment is applicable to vehicles that have been manufactured, have not been manufactured, and have any purpose and performance, and the cloud service center, based on the type of vehicle, sends the optimal model parameters of the vehicle state classification prediction model that best matches the corresponding type of vehicle. In this way, it can be applied to the accurate detection of traffic accidents involving different types of vehicles, thereby greatly improving the applicability of use; at the same time, the automatic call for help of the present invention can avoid the problem of failure of emergency rescue calls due to serious injuries of occupants in more serious traffic accidents, thereby ensuring the timeliness of the rescue of occupants in the vehicle; in addition, in addition to calling for help from the emergency center, the present invention also performs on-site calls for help through the off-vehicle on-site call module, based on which nearby or passing pedestrians and vehicles can be included in the rescue objects, thereby ensuring that the vehicle personnel obtain the maximum rescue force; therefore, the present system is very suitable for large-scale application and promotion.

[0115] In a possible design, the second aspect of this embodiment, based on the first aspect of the embodiment, describes the specific process of the cloud service center training the initial vehicle state classification prediction model for different types of vehicles to obtain the optimal model parameters for different types of vehicles. The parameter optimization process is shown below.

[0116] In specific applications, the cloud service center is used to obtain multiple driving feature data sets and use each driving feature data set as an initial training data set; wherein each driving feature data set corresponds to a vehicle type, and each driving feature data set contains sample driving feature data of multiple sample vehicles of the corresponding vehicle type.

[0117] In this embodiment, the vehicles are first classified and managed according to their purpose, displacement and / or motor power, and then the cloud service center receives sample driving feature data uploaded by the vehicle terminals of multiple sample vehicles of different vehicle types; thus, the initial vehicle state classification prediction model can be trained according to the vehicle type, thereby obtaining a vehicle state classification prediction model adapted to each vehicle type.

[0118] Optionally, each driving feature data set is obtained by extracting features from driving data sets of different vehicle types; wherein each driving data set also corresponds to a vehicle type, and each driving data set contains sample driving data of multiple sample vehicles of the corresponding vehicle type; in this embodiment, the content contained in the sample driving data is compatible with the content contained in the driving data in the first aspect of the embodiment, and the feature extraction process can be referred to the first aspect of the aforementioned embodiment, and will not be repeated here.

[0119] At the same time, the on-board terminal of the sample vehicle will also perform data labeling processing when uploading the sample feature data, that is, obtain the label of each sample feature data (that is, the label of the initial training data). In this embodiment, the label of each sample feature data is mainly divided into normal state and accident state, which can be represented by different numbers, such as 1 or 0.

[0120] Specifically, the label is determined as follows: if the on-board terminal determines that within the second preset time range after the target moment, the vehicle's traveling state changes from stationary to moving (that is, the first speed is not equal to each second speed) or receives a command to stop the start-stop switch in the vehicle's locomotive system (that is, through the vehicle system communication module, it is connected to the vehicle's vehicle system to read the vehicle's start-stop switch status), then the label of the marked sample feature data is a normal state (that is, the corresponding vehicle safety state is normal); conversely, if the on-board terminal determines that within the second preset time range after the target moment, the vehicle's traveling state has been in a stationary state; or receives information sent by the emergency alarm button and / or the voice emergency alarm submodule in the human-computer interaction module, then the label of the sample feature data is marked as an accident state (that is, the corresponding vehicle safety state is an accident).

[0121] Furthermore, when the aforementioned sample characteristic data is still not marked as a normal state within a third preset time period starting from the target moment, then the data is deleted from the storage module.

[0122] In this way, after data labeling, several initial training sets can be obtained. Then, the cloud service center optimizes the model parameters of the vehicle state classification prediction model corresponding to different vehicle types based on different driving feature data sets (i.e., different training data sets). The process is shown below.

[0123] The cloud service center is used to perform parameter optimization processing on the initial vehicle state classification prediction model based on each initial training data set to obtain the optimal model parameters of the vehicle state classification prediction model for each vehicle type; in this embodiment, an improved differential evolution algorithm is used to optimize the optimal model parameters of different vehicle types, and the process is shown below.

[0124] In this embodiment, since the optimization process for the optimal model parameters corresponding to different vehicle types is the same, the following takes any initial training data set as an example to perform the parameter optimization process, that is: for any initial training data set, the cloud service center is used to divide the any initial training data set into a training set and a test set; of course, the division ratio can be specifically set according to actual use, and is not specifically limited here, and before performing parameter optimization, the any initial training data set will be normalized to unify the scale of data of different dimensions, thereby facilitating subsequent model processing of the data.

[0125] Then, the cloud service center is used to generate a chromosome population based on the model parameters of the initial vehicle state classification prediction model, wherein each chromosome in the chromosome population corresponds to a set of initial model parameters; in specific implementation, it is equivalent to initializing the individual, wherein, it has been explained above that the initial vehicle state classification prediction model is a logistic regression model, therefore, the aforementioned model parameters may include but are not limited to: penalty regularization term parameters, the maximum order of the nonlinear mapping function and the corresponding parameter vector, and the parameter vector contains the coefficients of the nonlinear mapping function.

[0126] Optionally, one of the methods for initializing genes in a chromosome is given below; wherein, for any gene in each chromosome (i.e., any model parameter, such as a penalty regularization term parameter, the maximum order of a nonlinear mapping function, etc.), the initial value of any gene (i.e., any model parameter) can be obtained by, for example but not limited to, using the following formula (5).

[0127] H L,D =rand(L,D)×(H up -H low )+H low (5)

[0128] In the above formula (5), (L, D) represents the search space of parameters, L and D represent the population size (i.e., the total number of chromosomes) and the total number of genes of any chromosome (i.e., the total number of model parameters), rand (L, D) represents the random number in the search space, and H up ,H low Then they represent the gene search upper limit and gene search lower limit of any gene respectively.

[0129] Thus, through the aforementioned formula (5), the initialization of each model parameter can be completed, thereby obtaining different initial model parameters, based on which the initial chromosome population can be generated.

[0130] Next, the initial chromosome population can be iteratively evolved to find the optimal model parameters for the vehicle type corresponding to any initial training data set during the iterative evolution process; wherein the iterative evolution process is as follows.

[0131] The cloud service center is used to initialize the number of evolutions g and obtain the chromosome population at the g-th evolution to determine the model parameters corresponding to each chromosome in the chromosome population at the g-th evolution, wherein when g is 1, the model parameters corresponding to each chromosome at the g-th evolution are the initial model parameters corresponding to each chromosome.

[0132] After determining the chromosome population at the g-th evolution, different vehicle state classification prediction models can be constructed for the model parameters corresponding to different chromosomes in each evolution, and the aforementioned training set can be used for training, so as to calculate the fitness of each chromosome in this evolution based on the output of each model; wherein the aforementioned process can be but is not limited to as shown below.

[0133] The cloud service center is used to generate a vehicle state classification prediction model corresponding to each chromosome at the g-th evolution according to the model parameters corresponding to each chromosome at the g-th evolution; then, the training set is used to train the vehicle state classification prediction model corresponding to each chromosome at the g-th evolution, and the fitness of each chromosome at the g-th evolution is calculated according to the model output data of the vehicle state classification prediction model corresponding to each chromosome at the g-th evolution; in this embodiment, the model training is carried out with each training data in the training set as input and the vehicle state of the sample vehicle corresponding to each training data as output (that is, the probability of normal state and accident state, and the maximum value is taken as the model output).

[0134] At the same time, the fitness of any chromosome (representing any chromosome in the chromosome population at the g-th evolution) is the loss function value of the vehicle state classification prediction model corresponding to the chromosome, and the fitness calculation formula is:

[0135]

[0136] In the above formula (6), Fit represents the fitness value, It represents the probability that the vehicle status information corresponding to the b-th training data output by the target model is an accident status when the b-th training data in the training set is used as input, m is the total number of training data in the training set, λ is the penalty regularization term parameter in the model parameter corresponding to any chromosome, θ j represents the jth coefficient of the nonlinear mapping function, J represents the total number of coefficients of the nonlinear mapping function, y (b) represents the label data of the b-th training data, wherein the target model is the vehicle state classification prediction model corresponding to any chromosome, and In the formula, x (b) represents the bth training data, f θ () represents a nonlinear mapping function.

[0137] Thus, the fitness corresponding to each chromosome at the g-th evolution can be calculated by the above formula (6). It can be seen from the above formula (6) that the smaller the loss function value, the smaller the fitness and the better the model performance. Therefore, in this embodiment, the evolution is terminated when the minimum fitness in the chromosome population is less than or equal to the fitness threshold, thereby obtaining the optimal model parameters corresponding to the vehicle type of any initial training data set.

[0138] The specific iterative process is as follows:

[0139] The cloud service center is used to determine whether the evolution stop condition is met, wherein the evolution stop condition is that the minimum fitness at the g-th evolution is less than or equal to the fitness threshold, or g is greater than or equal to the maximum number of evolutions; in this embodiment, the fitness threshold can be specifically set according to actual use and is not specifically limited here; and when it is determined that the evolution stop condition is not met, individual mutation, crossover and other operations need to be performed, and the process is:

[0140] The cloud service center is used to perform individual mutation processing on the chromosome population at the g-th evolution based on the differential strategy to obtain the mutant population at the g-th evolution when it is judged that the evolution stop condition is not met, and to perform crossover update processing on the mutant population at the g-th evolution to obtain the crossover population at the g-th evolution.

[0141] In specific applications, this embodiment adopts a differential strategy to achieve individual variation, wherein any chromosome in the chromosome population at the g-th evolution is taken as an example to illustrate the individual variation operation:

[0142] In a specific implementation, for any chromosome in the chromosome population during the g-th evolution, the cloud service center is used to generate a first random positive integer, a second random positive integer, and a third random positive integer, wherein the first random positive integer, the second random positive integer, and the third random positive integer are different from each other and are all less than or equal to the total number of chromosomes in the chromosome population (referring to the total number of chromosomes in the initial chromosome population).

[0143] After generating three random positive integers, the cloud service center is used to select the r1th chromosome, the r2th chromosome and the r3th chromosome from the chromosome population at the gth evolution, wherein r1, r2, r3 represent the first random positive integer, the second random positive integer and the third random positive integer respectively; for example, assuming that the three random positive integers are 3, 4 and 5, then the 3rd, 4th and 5th chromosomes are selected from the chromosome population at the gth evolution; of course, the above example is only for illustration, and when the generated random positive integers are different, the selected chromosomes are also different.

[0144] After obtaining the r1th chromosome, the r2th chromosome and the r3th chromosome, the cloud service center is used to perform an individual mutation operation on any of the chromosomes according to the r1th chromosome, the r2th chromosome and the r3th chromosome to obtain a mutant individual corresponding to any of the chromosomes; in specific implementation, the following formula (7) may be used, for example but not limited to, to perform an individual mutation operation on any of the chromosomes.

[0145] U a (g) = H r1 (g)+F(H r2 (g)-H r3 (g)) (7)

[0146] In the above formula (7), U a (g) represents the variant individual corresponding to any of the chromosomes, H r1 (g),H r2 (g),H r3 (g) represents the r1th chromosome, the r2th chromosome and the r3th chromosome respectively, and F represents the scaling factor, wherein the first random positive integer, the second random positive integer and the third random positive integer are not equal to a, and a represents the index number of any chromosome in the chromosome population at the gth evolution.

[0147] In this way, through the above formula (7), the mutation operation of any chromosome can be completed, and then, according to the same principle, the mutation of the remaining chromosomes can be completed; and after obtaining the mutant population at the g-th evolution, the mutant population can be cross-updated to obtain the crossover population at the g-th evolution.

[0148] Optionally, the following still takes any mutant individual in the mutant population as an example to illustrate the crossover update operation of the individual, that is: for any mutant individual in the mutant population during g evolutions, the cloud service center is used to generate a first random number and a fourth random positive integer (wherein the value interval of the first random number is (0,1), and the fourth random positive integer is less than or equal to the total number of genes of the any mutant individual); then, the cloud service center is used to perform a crossover update operation on each gene in the any mutant individual according to the first random number and the fourth random positive integer, so as to obtain the crossover individual corresponding to the any mutant individual after the crossover update operation.

[0149] In a specific application, for example, a cloud service center may, but is not limited to, use the following formula (8) to perform a crossover update operation on the genes in any variant individual to obtain a crossover individual corresponding to any variant individual.

[0150]

[0151] In the above formula (8), v a,d represents the dth gene in the crossover individual corresponding to any variant individual, u a,d represents the dth gene in any variant individual, R1 represents the first random number, d rand represents the fourth random positive integer, CR represents the crossover probability, h a,d represents the d-th gene in the chromosome of any variant individual corresponding to the g-th evolution, where d=1, 2, 3, ..., D, and D represents the total number of genes in any variant individual.

[0152] Thus, through the aforementioned formula (8), the crossover update operation of each mutant individual can be completed; at the same time, based on the aforementioned formula (8), it can be known that the crossover update operation of this embodiment is essentially to perform gene crossover according to the crossover probability, and to ensure that at least one dimension of the gene in the gene crossover operation comes from the mutated gene, thereby ensuring that the population has the ability to evolve, that is, if the first condition in formula (8) is met, the gene is retained, otherwise, the gene needs to be discarded and the original gene is retained; based on this, gene selection can be achieved.

[0153] After obtaining the mutant population at the g-th evolution, individual selection can be performed according to the fitness of each mutant individual in the mutant population, that is: the cloud service center is used to calculate the fitness of each crossover individual in the crossover population at the g-th evolution, and based on the fitness of each crossover individual, select each crossover individual to obtain the individual population at the g-th evolution; in this embodiment, the calculation method of the fitness of each crossover individual can refer to the fitness calculation process of each chromosome at the g-th evolution mentioned above, that is, first construct a vehicle state classification prediction model corresponding to each crossover individual according to the model parameters corresponding to each crossover individual; then, use the training set for training; finally, calculate the loss function value based on the model output, and then obtain the fitness of each crossover individual.

[0154] Furthermore, this embodiment takes any crossover individual in the crossover population during the g-th evolution as an example to illustrate the individual selection process, namely: judging whether the fitness of any crossover individual is less than the fitness of the target individual, if so, retaining the any crossover individual; otherwise, updating the any crossover individual to the target individual; wherein the target individual is the chromosome corresponding to the any crossover individual in the chromosome population during the g-th evolution.

[0155] Specifically, the following formula (9) can be used to represent the aforementioned individual selection operation:

[0156]

[0157] In the above formula (9), V a (g) represents any crossover individual, H a (g+1) The selected individual corresponding to any crossover individual, Fit(V a (g)) represents the fitness of any crossover individual, H a (g) represents the target individual, Fit(H a (g)) represents the fitness of the target individual.

[0158] Thus, through the above formula (9), the selection of each mutant individual can be completed, that is, the mutant individual with the best fitness compared to the original chromosome is retained, and the mutant individual with a fitness less than the fitness of the original chromosome is replaced with the original chromosome (that is, the chromosome before mutation); then, in order to avoid falling into the local optimum, this embodiment also performs random perturbation, that is:

[0159] The cloud service center is used to perform perturbation processing on the individual population to obtain the perturbed population at the g-th evolution. In the specific implementation, any selected individual in the individual population is taken as an example for explanation, and the following formula (10) can be used, but not limited to, to perform perturbation processing on it.

[0160]

[0161] In the above formula (10), H a ′(g+1) represents the disturbance individual corresponding to any selected individual, γ represents the disturbance factor, γ∈[0,0.5], rand() represents taking a random number, S r Represents a set of random natural numbers, where the length of the random natural number is β, and the value of any element in the set is less than or equal to the total number of chromosomes in the initial chromosome population, and β∈[0,0.5].

[0162] In this way, through the above formula (10), the disturbance processing of each selected individual in the individual population can be completed, and then the fitness can be recalculated, and according to the fitness, it is determined whether it is necessary to continue evolution until the evolution stops when the evolution condition is met; wherein, the iterative evolution process is:

[0163] The cloud service center is used to add 1 to g, update the chromosome population at the g-th evolution to the perturbation population at the g-1-th evolution, and re-determine the model parameters corresponding to each chromosome in the chromosome population at the g-th evolution, until the evolution stop condition is met, so as to use the vehicle state classification prediction model corresponding to the chromosome with the smallest fitness in the chromosome population that meets the evolution stop condition as the pre-trained model.

[0164] In this way, through the aforementioned improved differential evolution algorithm, the optimal vehicle state classification prediction model for the vehicle type corresponding to any of the aforementioned initial training data sets can be obtained; at this time, the model test can be performed, that is, the vehicle state classification prediction model corresponding to the chromosome with the smallest fitness in the chromosome population when the aforementioned evolution stop condition is met is used as the trained model; then, the cloud service center is also used to use the test set to test the pre-trained model, and after the test passes, the model parameters corresponding to the pre-trained model are used as the optimal model parameters for the specified vehicle type, wherein the specified vehicle type is the vehicle type corresponding to any of the initial training data sets.

[0165] , verify the model classification accuracy and compare it with the preset accuracy threshold. If ≥, the model parameters of the pre-trained model are used as the optimal model parameters for the vehicle type corresponding to any of the initial training data sets; otherwise, it is necessary to recheck the sample data or regenerate the initial chromosome population, that is, reinitialize the model parameters, and then repeat the aforementioned iterative evolution process until the test passes.

[0166] In this way, the optimal model parameters corresponding to each vehicle type can be obtained through the improved differential evolution algorithm mentioned above, and then, the optimal model parameters can be sent to the vehicle terminals corresponding to different vehicle types according to the parameter sending request of the vehicle terminal; specifically, the parameter sending process is:

[0167] The vehicle-mounted terminal is also used to send a parameter sending request to the cloud service center, and the parameter sending request includes vehicle information; in this way, the cloud service center is used to screen out the optimal model parameters of the vehicle type corresponding to the target vehicle from the optimal model parameters corresponding to each vehicle type based on the vehicle information in the parameter sending request, and send them to the vehicle-mounted terminal, so that the vehicle-mounted terminal generates the optimal vehicle state classification prediction model according to the optimal model parameters sent by the cloud service center, wherein the target vehicle is the vehicle corresponding to the vehicle-mounted terminal.

[0168] Based on this, the deployment of the most suitable vehicle status classification prediction model for different types of vehicles can be completed for different types of vehicles; of course, this embodiment can regularly retrain and update the model parameters, and re-send them to the on-board terminal of the corresponding vehicle type after the update to ensure the accuracy of the model's classification of vehicle status.

[0169] Then, the vehicle's on-board terminal can collect the driving data before the vehicle stops in real time, and then extract features to obtain driving feature data; finally, input it into the deployed optimal vehicle state classification prediction model to obtain the vehicle's state information; finally, an emergency call can be made based on the vehicle's state information. The process can be referred to the first aspect of the embodiment and will not be repeated here.

[0170] Therefore, through the above explanation, see Figure 2 As shown, the working process of the system for predicting vehicle safety status and emergency calling provided by this embodiment is as follows:

[0171] (1) Establish a cloud service center machine database for vehicle status detection and active emergency calls to receive data reported by vehicle terminals on sample vehicles of different vehicle types.

[0172] (2) Classify and manage vehicles according to their purpose, displacement and / or motor power attributes.

[0173] (3) The sample vehicles are equipped with on-board terminals to collect vehicle information and driving data, and obtain pre-parking data, which are then reported to the cloud service center.

[0174] (4) Construct a vehicle status classification prediction model and train the model according to vehicle type, obtain the trained model and its parameters, and send the model parameters to the on-board terminals of various types of vehicles.

[0175] (5) The vehicle terminal collects driving data in real time and inputs the received model parameters into the model to construct an optimal vehicle state classification prediction model. Then, the feature data of the driving data is extracted and input into the optimal vehicle state classification prediction model for vehicle state prediction. The prediction result is normal or accident state.

[0176] (6) The vehicle terminal initiates a voice inquiry based on the vehicle status prediction result, determines the actual vehicle status based on the expected result of the voice inquiry, and initiates an active emergency call based on the actual vehicle status and acts as an on-site emergency call module outside the vehicle.

[0177] In one possible design, see Figure 3 As shown, the third aspect of this embodiment provides a method for vehicle safety status prediction and emergency call, wherein the method is executed based on the vehicle-mounted terminal in the vehicle safety status prediction and emergency call system described in the first aspect of the embodiment and the second aspect of the embodiment, and the operation process of the method can be but is not limited to the following steps S1 to S5.

[0178] S1. Receive the optimal model parameters of the vehicle state classification prediction model that best matches the vehicle type issued by the cloud service center, and generate an optimal vehicle state classification prediction model based on the optimal model parameters.

[0179] S2. Receive the driving data of the vehicle within a first preset time period before the target time sent by the driving data acquisition module, wherein the target time is the time corresponding to when the vehicle changes from a moving state to a stationary state.

[0180] S3. Perform feature extraction processing on the driving data to obtain driving feature data, and input the driving feature data into the optimal vehicle state classification prediction model to obtain the state information of the vehicle at the target time.

[0181] S4. If the status information is an accident status, a voice inquiry is initiated to the vehicle, and it is determined whether a response information to the voice inquiry is received.

[0182] S5. If not, an emergency call is made to the emergency center based on the vehicle's location information, and the on-site emergency call module outside the vehicle is controlled to operate to send an on-site distress signal.

[0183] The working process, working details and technical effects of this embodiment can be found in the first and second aspects of the embodiment, and will not be described in detail here.

[0184] The fourth aspect of this embodiment provides an electronic device as an example, comprising: a memory, a processor and a transceiver that are communicatively connected in sequence, wherein the memory is used to store computer programs, the transceiver is used to send and receive messages, and the processor is used to read the computer program to execute the method for vehicle safety status prediction and emergency call as described in the third aspect of the embodiment.

[0185] For example, the memory may include, but is not limited to, random access memory (RAM), read only memory (ROM), flash memory, first input first output (FIFO) and / or first in last out (FILO), etc. Specifically, the processor may include one or more processing cores, such as a 4-core processor, an 8-core processor, etc. The processor may be implemented in at least one hardware form of DSP (Digital Signal Processing), FPGA (Field-Programmable Gate Array), and PLA (Programmable Logic Array). At the same time, the processor may also include a main processor and a coprocessor. The main processor is a processor for processing data in the awake state, also known as a CPU (Central Processing Unit); the coprocessor is a low-power processor for processing data in the standby state.

[0186] In some embodiments, the processor may be integrated with a GPU (Graphics Processing Unit), which is responsible for rendering and drawing the content to be displayed on the display screen. For example, the processor may not be limited to a microprocessor of the STM32F105 series, a reduced instruction set computer (RISC) microprocessor, an X86 architecture processor, or a processor with an integrated embedded neural network processing unit (NPU); the transceiver may be, but is not limited to, a wireless fidelity (WIFI) wireless transceiver, a Bluetooth wireless transceiver, a general packet radio service technology (General Packet Radio Service, GPRS) wireless transceiver, a ZigBee protocol (a low-power local area network protocol based on the IEEE802.15.4 standard, ZigBee) wireless transceiver, a 3G transceiver, a 4G transceiver and / or a 5G transceiver, etc. In addition, the device may also include, but is not limited to, a power module, a display screen, and other necessary components.

[0187] The working process, working details and technical effects of the electronic device provided in this embodiment can be found in the first and second aspects of the embodiments, and will not be described in detail here.

[0188] The fifth aspect of this embodiment provides a storage medium storing instructions including the method for predicting the vehicle safety status and making an emergency call as described in the third aspect of the embodiment, that is, the storage medium stores instructions, and when the instructions are executed on a computer, the method for predicting the vehicle safety status and making an emergency call as described in the third aspect of the embodiment is executed.

[0189] The storage medium refers to a carrier for storing data, which may include but is not limited to a floppy disk, a CD, a hard disk, a flash memory, a USB flash drive and / or a memory stick, and the computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices.

[0190] The working process, working details and technical effects of the storage medium provided in this embodiment can be found in the first and second aspects of the embodiments, and will not be described in detail here.

[0191] The sixth aspect of this embodiment provides a computer program product comprising instructions, which, when executed on a computer, causes the computer to execute the method for predicting the vehicle safety status and making an emergency call as described in the third aspect of the embodiment, wherein the computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device.

[0192] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the protection scope of the present invention. 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 system for predicting vehicle safety status and emergency calling, characterized in that: include: The vehicle-mounted terminal is communicatively connected to a cloud service center, and is used to receive the optimal model parameters of the vehicle state classification prediction model that best matches the vehicle type issued by the cloud service center, and generate an optimal vehicle state classification prediction model according to the optimal model parameters; A driving data collection module, wherein the driving data collection module is electrically connected to the vehicle terminal and is used to collect driving data of the vehicle within a first preset time period before a target time, and the target time is the time corresponding to when the vehicle changes from a moving state to a stationary state; The vehicle terminal is used to perform feature extraction processing on the driving data to obtain driving feature data, and input the driving feature data into the optimal vehicle state classification prediction model to obtain the state information of the vehicle at the target time; The vehicle terminal is used to initiate a voice inquiry into the vehicle when the status information is an accident status; The vehicle-mounted terminal is also used to make an emergency call to the emergency center based on the vehicle's location information when it determines that no answer information to the voice inquiry has been received, and to control the operation of the on-site rescue module outside the vehicle to send out an on-site distress signal.

2. A vehicle safety status prediction and emergency call system according to claim 1, characterized in that: The driving data includes: vehicle information of the vehicle, and speed information, acceleration information and angular velocity information of the vehicle within a first preset time period before the target time; Among them, the vehicle terminal includes: a calculation and prediction module; a calculation and prediction module, used for performing feature extraction processing on the speed information, the acceleration information and the angular velocity information, so as to obtain time series statistical features, frequency domain statistical features and segmented statistical features corresponding to the speed information, the acceleration information and the angular velocity information respectively; A calculation and prediction module, configured to generate the driving characteristic data by using the vehicle information, and the time series statistical features, frequency domain statistical features and segment statistical features corresponding to the speed information, the acceleration information and the angular velocity information; The calculation and prediction module is also used to input the driving characteristic data into the optimal vehicle state classification prediction model to obtain the state information of the vehicle at the target time.

3. The system for predicting vehicle safety status and emergency calling according to claim 2, characterized in that: The speed information is a discrete signal sequence; The calculation and prediction module is used to perform second-order difference processing on the discrete signal sequence to obtain a second-order difference velocity sequence, and calculate the time series statistical characteristics corresponding to the velocity information based on the second-order difference velocity sequence; A calculation and prediction module, used for performing Fourier transform processing on the discrete signal sequence to obtain a frequency domain speed signal sequence, and generating frequency domain statistical features corresponding to the speed information according to the frequency domain speed signal sequence; A calculation and prediction module, used for dividing the discrete signal sequence into a plurality of molecular signal sequences according to the order of acquisition time from the beginning to the end, wherein the lengths of the plurality of molecular signal sequences are the same; The calculation and prediction module is also used to determine the characteristic value of each sub-signal sequence, and use the characteristic data of each sub-signal sequence to form the segmented statistical features corresponding to the speed information, wherein the characteristic value of any sub-signal sequence includes the maximum value, minimum value, average value and median of any sub-signal sequence.

4. The system for predicting vehicle safety status and emergency calling according to claim 3, characterized in that: The calculation and prediction module is used to generate the time series statistical features and frequency domain statistical features corresponding to the speed information by using the following formula (1) and formula (2) respectively; In the above formula (1), s1 represents the time series statistical characteristics corresponding to the speed information, d 2 x[i] represents the i-th second-order differential velocity value in the second-order differential velocity sequence, wherein the i-th second-order differential velocity value is the second-order differential value corresponding to the i-th signal in the discrete signal sequence, and N represents the total number of signal points in the discrete signal sequence; In the above formula (2), s2 represents the frequency domain statistical characteristics corresponding to the speed information, wherein X[k] represents the kth speed value in the frequency domain speed signal sequence, k c is the starting point corresponding to the high frequency component in the frequency domain velocity signal sequence, k c =f0 / 4, and f0 is the sampling frequency of the discrete signal sequence.

5. The system for predicting vehicle safety status and emergency calling according to claim 1, characterized in that: The driving data acquisition module includes: a speed measurement module, an acceleration measurement module and a gyroscope measurement module; A speed measurement module, used to collect speed information of the vehicle within a first preset time period before the target time, and send the speed information to the vehicle terminal; An acceleration measurement module, used to collect acceleration information of the vehicle within a first preset time period before the target time, and send the information to the vehicle terminal; A gyroscope measurement module, used for collecting angular velocity information of the vehicle within a first preset time period before the target time, and sending the information to the vehicle terminal; The speed measurement module is further used to collect a first speed of the vehicle at the target time and a second speed within a second preset time after the target time, and transmit the first speed and each second speed to the vehicle terminal; The vehicle-mounted terminal is also used to determine that the vehicle is in a stationary state when it is determined that the first speed is equal to each second speed, and to obtain the speed information, acceleration information and angular velocity information transmitted by the speed measurement module, the acceleration measurement module and the gyroscope measurement module within a first preset time period before the target moment, so as to generate the driving data using the obtained speed information, acceleration information and angular velocity information.

6. The system for predicting vehicle safety status and emergency calling according to claim 1, characterized in that: Also includes: A network and communication module, a GPS positioning module and a human-computer interaction module, wherein the human-computer interaction module includes a vehicle information input submodule, an emergency alarm button and a voice emergency alarm submodule, and the vehicle information input submodule is used to obtain the vehicle information of the vehicle in response to the human-computer interaction and transmit it to the vehicle terminal; GPS positioning module, used to obtain the vehicle's location information and transmit the location information to the vehicle terminal; The emergency alarm button and voice emergency alarm submodule are used to generate emergency alarm information and voice alarm information in response to human-computer interaction and send them to the vehicle terminal; The vehicle-mounted terminal is used to make an emergency call to the emergency center based on the vehicle's location information when receiving emergency alarm information and / or voice alarm information, and control the operation of the off-site rescue module to send out an on-site distress signal.

7. The system for predicting vehicle safety status and emergency calling according to claim 1, characterized in that: A cloud service center is used to obtain multiple driving feature data sets, wherein each driving feature data set corresponds to a vehicle type, and each driving feature data set contains sample driving feature data of multiple sample vehicles of the corresponding vehicle type; The cloud service center is used to use each driving feature data set as an initial training data set; The cloud service center is used to perform parameter optimization processing on the initial vehicle state classification prediction model based on each initial training data set to obtain the optimal model parameters of the vehicle state classification prediction model for each vehicle type; The vehicle terminal is further used to send a parameter sending request to the cloud service center, and the parameter sending request includes vehicle information; The cloud service center is also used to filter out the optimal model parameters of the vehicle type corresponding to the target vehicle from the optimal model parameters corresponding to each vehicle type based on the vehicle information in the parameter sending request, and send them to the vehicle terminal, so that the vehicle terminal can generate the optimal vehicle state classification prediction model according to the optimal model parameters sent by the cloud service center, wherein the target vehicle is the vehicle corresponding to the vehicle terminal.

8. The system for predicting vehicle safety status and emergency calling according to claim 7, characterized in that: For any initial training data set, the cloud service center is used to divide the initial training data set into a training set and a test set; The cloud service center is used to generate a chromosome population based on the model parameters of the initial vehicle state classification prediction model, wherein each chromosome in the chromosome population corresponds to a set of initial model parameters; The cloud service center is used to initialize the number of evolutions g and obtain the chromosome population at the g-th evolution to determine the model parameters corresponding to each chromosome in the chromosome population at the g-th evolution, wherein when g is 1, the model parameters corresponding to each chromosome at the g-th evolution are the initial model parameters corresponding to each chromosome; The cloud service center is used to generate a vehicle state classification prediction model corresponding to each chromosome at the g-th evolution according to the model parameters corresponding to each chromosome at the g-th evolution; The cloud service center is used to train the vehicle state classification prediction model corresponding to each chromosome at the g-th evolution using the training set, and calculate the fitness of each chromosome at the g-th evolution according to the model output data of the vehicle state classification prediction model corresponding to each chromosome at the g-th evolution, wherein the fitness of any chromosome is the loss function value of the vehicle state classification prediction model corresponding to the any chromosome; The cloud service center is used to determine whether the evolution stop condition is met, wherein the evolution stop condition is that the minimum fitness at the g-th evolution is less than or equal to the fitness threshold, or g is greater than or equal to the maximum number of evolutions; The cloud service center is used to perform individual mutation processing on the chromosome population at the g-th evolution based on the differential strategy to obtain the mutation population at the g-th evolution when it is determined that the evolution stop condition is not met, and to perform crossover update processing on the mutation population at the g-th evolution to obtain the crossover population at the g-th evolution; The cloud service center is used to calculate the fitness of each crossover individual in the crossover population at the g-th evolution, and based on the fitness of each crossover individual, perform a selection operation on each crossover individual to obtain the individual population at the g-th evolution; The cloud service center is used to perform disturbance processing on the individual population to obtain a disturbed population at the g-th evolution; The cloud service center is used to add 1 to g, update the chromosome population at the g-th evolution to the perturbation population at the g-1-th evolution, and re-determine the model parameters corresponding to each chromosome in the chromosome population at the g-th evolution, until the evolution stop condition is met, so as to use the vehicle state classification prediction model corresponding to the chromosome with the smallest fitness in the chromosome population that meets the evolution stop condition as the pre-trained model; The cloud service center is also used to use the test set to test the pre-trained model, and after the test passes, use the model parameters corresponding to the pre-trained model as the optimal model parameters for a specified vehicle type, wherein the specified vehicle type is the vehicle type corresponding to any of the initial training data sets.

9. The system for predicting vehicle safety status and emergency calling according to claim 8, characterized in that: For any chromosome in the chromosome population during the g-th evolution, the cloud service center is used to generate a first random positive integer, a second random positive integer, and a third random positive integer, wherein the first random positive integer, the second random positive integer, and the third random positive integer are different from each other and are all less than or equal to the total number of chromosomes in the chromosome population; The cloud service center is used to select the r1th chromosome, the r2th chromosome and the r3th chromosome from the chromosome population at the gth evolution, and r1, r2, r3 represent the first random positive integer, the second random positive integer and the third random positive integer respectively; The cloud service center is used to perform an individual mutation operation on any of the chromosomes according to the r1th chromosome, the r2th chromosome, and the r3th chromosome, so as to obtain a mutant individual corresponding to the any of the chromosomes; For any mutant individual in the mutant population during the g-time evolution, the cloud service center is used to generate a first random number and a fourth random positive integer, wherein the value interval of the first random number is (0,1), and the fourth random positive integer is less than or equal to the total number of genes of any mutant individual; The cloud service center is further used to perform a crossover update operation on each gene in any variant individual according to the first random number and the fourth random positive integer, so as to obtain a crossover individual corresponding to any variant individual after the crossover update operation.

10. A method for predicting vehicle safety status and emergency call, characterized in that: The method is executed by a vehicle terminal in the vehicle safety status prediction and emergency call system according to any one of claims 1 to 9, and the method comprises: Receiving the optimal model parameters of the vehicle state classification prediction model that best matches the vehicle type issued by the cloud service center, and generating an optimal vehicle state classification prediction model based on the optimal model parameters; Receiving driving data of the vehicle within a first preset time period before a target time sent by a driving data acquisition module, wherein the target time is a time corresponding to when the vehicle changes from a moving state to a stationary state; Performing feature extraction processing on the driving data to obtain driving feature data, and inputting the driving feature data into the optimal vehicle state classification prediction model to obtain state information of the vehicle at the target time; If the status information is an accident status, a voice inquiry is initiated to the vehicle, and it is determined whether an answer to the voice inquiry is received; If not, an emergency call is made to the emergency center based on the vehicle's location information, and the on-site rescue module outside the vehicle is controlled to operate to send out an on-site distress signal.

Images