Vehicle driving warning method and its device, equipment and medium

Abstract

The present application relates to a vehicle driving alarm method and its device, equipment, and medium, the method comprising: obtaining a longitudinal acceleration set and a vertical acceleration set corresponding to vehicle driving within a predetermined time period, when the latest longitudinal acceleration in the longitudinal acceleration set is less than a first preset threshold, obtaining the corresponding longitudinal average absolute value according to the longitudinal acceleration set, when the longitudinal average absolute value exceeds a second preset threshold, determining the vertical turbulence frequency corresponding to the bumpy driving condition according to the vertical acceleration set, when the vertical turbulence frequency is lower than a third preset threshold, sending a continuous alarm control instruction to a target alarm unit, so as to continuously control the target alarm unit to show that the vehicle is in a state of rapid deceleration. The present application realizes accurate and efficient rapid deceleration detection and can give timely alarms by comprehensively analyzing the real-time acceleration in the longitudinal and lateral directions of the vehicle during driving.

Description

Technical Field

[0001] The present application relates to vehicle safety technology, and in particular to a vehicle driving warning method and its device, equipment, and medium. Background Art

[0002] With the rapid development of autonomous driving and intelligent driving assistance systems, driving safety has become the focus of public and industry attention. In this field, the detection of sudden deceleration events is of great significance for preventing traffic accidents and improving driving safety. Traditional sudden deceleration detection systems mainly rely on external sensors such as radars and cameras to monitor vehicle status and surrounding environment. Although these sensors can provide rich information, they have some limitations, including: cost issues: the cost of sensors such as radars and cameras is relatively high, which limits their application in economical vehicles; environmental interference: the performance of these sensors is easily affected by weather and lighting conditions, and their performance will be significantly reduced in low visibility conditions such as rain and fog; misjudgment problem: due to reliance on external environmental information, these sensors may make misjudgments in complex road conditions and vibration environments, affecting accuracy and reliability. Summary of the invention

[0003] The primary purpose of the present application is to solve at least one of the above problems and provide a vehicle driving warning method and its device, equipment, and medium.

[0004] According to one aspect of the present application, a vehicle driving alarm method is provided, comprising the following steps:

[0005] Obtaining a longitudinal axis acceleration set and a vertical axis acceleration set corresponding to the vehicle's travel within a predetermined time period;

[0006] When the latest longitudinal acceleration in the longitudinal acceleration set is less than the first preset threshold, obtaining the corresponding longitudinal average absolute value according to the longitudinal acceleration set, and determining whether the longitudinal average absolute value exceeds the second preset threshold;

[0007] When the longitudinal average absolute value exceeds the second preset threshold, determining the vertical turbulence frequency corresponding to the matching bumpy driving condition according to the vertical axis acceleration set, and judging whether the vertical turbulence frequency exceeds the third preset threshold;

[0008] When the vertical turbulence frequency is lower than a third preset threshold, a continuous warning control instruction is sent to the target warning unit to continuously control the target warning unit to display that the vehicle is in a driving state of rapid deceleration.

[0009] According to another aspect of the present application, a vehicle driving warning device is provided, comprising:

[0010] A data acquisition module, used to acquire a longitudinal axis acceleration set and a vertical axis acceleration set corresponding to the vehicle's travel within a predetermined time period;

[0011] a first judgment module, configured to obtain a corresponding longitudinal average absolute value according to the longitudinal acceleration set when the latest longitudinal acceleration in the longitudinal acceleration set is less than a first preset threshold, and to judge whether the longitudinal average absolute value exceeds a second preset threshold;

[0012] a second judgment module, configured to determine, when the longitudinal average absolute value exceeds a second preset threshold, a vertical turbulence frequency corresponding to a matching bumpy driving condition according to the vertical axis acceleration set, and to judge whether the vertical turbulence frequency exceeds a third preset threshold;

[0013] The driving warning module is used to send a continuous warning control instruction to the target warning unit when the vertical turbulence frequency is lower than a third preset threshold value, so as to continuously control the target warning unit to display that the vehicle is in a driving state of rapid deceleration.

[0014] According to another aspect of the present application, a vehicle driving warning device is provided, comprising a central processing unit and a memory, wherein the central processing unit is used to call and run a computer program stored in the memory to execute the steps of the vehicle driving warning method described in the present application.

[0015] According to another aspect of the present application, a non-volatile readable storage medium is provided, which stores a computer program implemented according to the vehicle driving alarm method in the form of computer-readable instructions, and when the computer program is called and executed by a computer, the steps included in the method are executed.

[0016] The technical solution of this application has many advantages, including but not limited to the following aspects:

[0017] The present application first monitors the longitudinal axis acceleration set of the vehicle within a predetermined time period, and determines whether the vehicle may be in a deceleration state by setting a first preset threshold value, and provides a preliminary judgment basis for subsequent judgments based on the significant change in the longitudinal axis acceleration of the vehicle when it decelerates suddenly.

[0018] Next, after confirming that the vehicle is in a decelerating driving state, the longitudinal average absolute value determined based on the longitudinal axis acceleration set is further compared with the second preset threshold to verify whether the comprehensively considered average longitudinal deceleration increment is significant. This enhances the robustness of the deceleration detection by reducing the impact of occasional, short-term acceleration changes on the judgment, ensuring that the next step of detection is triggered when the acceleration change is continuous and significant.

[0019] Then, under the premise that the average longitudinal deceleration increment is significant, the real-time vertical axis accelerations in the vertical axis acceleration set are analyzed to determine whether the vehicle is in a bumpy driving state, and the corresponding vertical turbulence frequency is quantified and compared with the third preset threshold. This can effectively distinguish between sudden deceleration caused by vehicle vibration due to road unevenness and sudden deceleration caused by the driver's braking operation when driving on a flat road section, thereby avoiding misjudgment.

[0020] Finally, the vertical turbulence frequency is lower than the third preset threshold, and it is confirmed that the vehicle is driving on a flat road and the average deceleration increment is significant, so the vehicle is judged to be in a state of rapid deceleration, and a continuous warning control instruction is sent to the target warning unit to continuously remind the driver that the vehicle is in a state of rapid deceleration. It can be seen that the timely warning mechanism implemented is of great significance for preventing possible traffic accidents and improving driving safety.

[0021] In summary, it can not only reduce the misjudgment of sudden deceleration and improve the accuracy of the warning, but also work stably under various road conditions, unaffected by weather and lighting conditions, providing drivers with a more reliable safety guarantee, helping to effectively improve driving safety and reduce the occurrence of traffic accidents. In addition, by setting multiple preset thresholds, a flexible judgment mechanism is provided, which can be appropriately adjusted according to different driving conditions, vehicle characteristics, driving habits, etc., to adapt to various complex road conditions and driving situations. Such flexibility and adaptability further improve the accuracy and reliability of the warning. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 Provide an exemplary I MU installation location side view for this application;

[0023] Figure 2 A flow chart of an embodiment of a vehicle driving warning method of the present application;

[0024] Figure 3 This is a principle block diagram of the vehicle driving warning device of the present application;

[0025] Figure 4 This is a structural diagram of a vehicle driving warning device used in this application. DETAILED DESCRIPTION

[0026] In order to facilitate understanding of the application scenarios of the present application, an exemplary operating environment of a vehicle driving warning method is provided. The operating environment is implemented based on a vehicle driving warning device. The central processor in the vehicle driving warning device can be any processor that can provide data receiving and computing capabilities, such as an MCU, a CPU, or an MPU. The vehicle driving warning device also includes an IMU (Inertial Measurement Unit) electrically connected to the central processor. Figure 1 As shown in 100, and an alarm unit. The IMU is generally called an inertial measurement unit or an inertial measurement device, which is an electronic device in which three single-axis (x, y, z) accelerometers can measure the basic linear acceleration of the vehicle in the three axis directions. The alarm unit can be any one or any combination of a light bulb, display screen, horn, buzzer, vibrator, etc. installed in the vehicle. The vehicle driving alarm device can be independent or integrated in the vehicle's on-board system.

[0027] The central processing unit receives multiple longitudinal axis accelerations and vertical axis accelerations collected by the IMU at regular intervals, filters them, and then performs sudden deceleration analysis. When it is confirmed that the vehicle is in a sudden deceleration state, the corresponding continuous alarm instruction is sent to a pre-designated alarm unit to continuously control the alarm unit to display that the vehicle is in a sudden deceleration state. Furthermore, when it is confirmed that the vehicle has left the sudden deceleration state, the corresponding end alarm control instruction is sent to the alarm unit to end controlling the alarm unit to display that the vehicle is in a sudden deceleration state.

[0028] See also Figure 2 According to a vehicle driving warning method provided by the present application, it can be implemented as a computer program product running in a vehicle driving warning device. In one embodiment, it includes the following steps:

[0029] Step S1100, obtaining a longitudinal axis acceleration set and a vertical axis acceleration set corresponding to the vehicle's travel within a predetermined time period;

[0030] The microcontroller unit is connected to the inertial measurement device through a serial peripheral interface to realize a communication connection, and can collect the three single-axis accelerometers in the inertial measurement device, and measure the basic linear accelerations of the vehicle in the three axis directions in a time sequence of first to last within a preset time length, and for the longitudinal axis and the vertical axis, respectively collect the basic linear accelerations corresponding to the axis to form the initial acceleration set of the axis. The preset time length can be set by a person skilled in the art as needed, for example, 20ms.

[0031] In one embodiment, two circular buffers of size N are initialized and created. For the initial acceleration set of the longitudinal axis, each longitudinal axis acceleration is traversed in reverse order so that they are added to a circular buffer one by one in the order of last to first in time sequence to obtain a first circular buffer; for the initial acceleration set of the vertical axis, each vertical axis acceleration is traversed in reverse order so that they are added to another circular buffer one by one in the order of last to first in time sequence to obtain a second circular buffer. It should be noted that the initial acceleration sets of the longitudinal axis and the vertical axis respectively obtained each time thereafter can be directly written into the corresponding circular buffer in the same way, because the circular buffer has the characteristics of not needing to clear the data when it is full, and can overwrite the old data storage with the newly written data, which can reduce memory allocation and thus reduce system overhead, and is conducive to the long-term stable operation of the program. The N is preset to be the total number of accelerations corresponding to the single axis collected within the preset time length.

[0032] Since the ring buffer also has the first-in-first-out feature, we start traversing the first ring buffer directly from its head pointer and read N 1 Each longitudinal acceleration of N constitutes a longitudinal acceleration set. 1 Less than or equal to the size of the first ring buffer, the specific value can be set by those skilled in the art as needed. And, from the second ring buffer, directly start traversing from its head pointer, read N 2 Each vertical axis acceleration of N constitutes a vertical axis acceleration set. 2 Smaller than the size of the second ring buffer, the specific value can be set by those skilled in the art as needed.

[0033] Step S1200: when the latest longitudinal acceleration in the longitudinal acceleration set is less than the first preset threshold, obtaining the corresponding longitudinal average absolute value according to the longitudinal acceleration set, and determining whether the longitudinal average absolute value exceeds the second preset threshold;

[0034] The first preset threshold is zero. It is determined whether the first longitudinal acceleration in the longitudinal acceleration set, i.e., the latest longitudinal acceleration, is less than zero. If it is less than zero, it means that the vehicle is in a decelerating driving state at this time. All longitudinal accelerations in the longitudinal acceleration set are summed and then divided by the total number of these longitudinal accelerations to obtain the longitudinal average absolute value. Then, it is determined whether it exceeds the second preset threshold. The second preset threshold is used to measure whether the longitudinal average value is large enough. If it is large enough, it means that the vehicle has significantly reduced its driving speed in a short period of time because the driver brakes quickly enough. The specific value can be set by those skilled in the art as needed according to the disclosure herein.

[0035] Step S1300: when the longitudinal average absolute value exceeds the second preset threshold, determining the vertical turbulence frequency corresponding to the bumpy driving condition according to the vertical axis acceleration set, and judging whether the vertical turbulence frequency exceeds the third preset threshold;

[0036] At this time, it is determined whether each vertical axis acceleration in the vertical axis acceleration set belongs to the normal fluctuation value range for driving on a flat road section, and the total number of vertical axis accelerations that do not belong to the normal fluctuation value range is determined as the vertical turbulence frequency, and it is determined whether it exceeds the third preset threshold. The third preset threshold is used to measure whether the vertical turbulence frequency is large enough. If it is large enough, it means that the vehicle is turbulent in the vertical axis direction, that is, it is not driving on a flat road section. The specific value can be set as needed by technical personnel in this field based on the disclosure herein.

[0037] Step S1400: When the vertical turbulence frequency is lower than a third preset threshold, a continuous alarm control instruction is sent to a target alarm unit to continuously control the target alarm unit to display that the vehicle is in a state of rapid deceleration.

[0038] The alarm unit controlled by the continuous alarm control instruction can be any one or more of a speaker, a buzzer, a display screen, a vibrator, etc., and the corresponding display can be any one or more of a loud audio, a striking light, a strong seat vibration, etc. A person skilled in the art can pre-set the continuous alarm control instruction as needed according to the disclosure herein for ready call. The controlled alarm unit is the target alarm unit.

[0039] At this time, it means that the vehicle is traveling on a flat road section and the driving speed is greatly reduced in a short period of time. Therefore, the preset continuous alarm control instruction is called and sent to the target alarm unit to continuously control the target alarm unit to show that the vehicle is in a state of rapid deceleration, so that the driver of the notified vehicle will currently perform his driving behavior as sudden braking, thereby stimulating the driver to observe the road conditions more vigilantly and drive carefully.

[0040] It can be understood that each time the predetermined time has passed, the above steps S1100 to S1400 will be triggered and executed based on the corresponding predetermined time, so as to form a continuous monitoring loop in an iterative manner, thereby ensuring real-time and continuous monitoring of the vehicle's sudden deceleration state, and promptly issuing an alarm to the driver when it is confirmed that the vehicle is in a sudden deceleration state, thereby improving driving safety.

[0041] It is not difficult to understand from the above embodiments that the technical solution of the present application has many advantages, including but not limited to the following aspects:

[0042] The present application first monitors the longitudinal axis acceleration set of the vehicle within a predetermined time period, and determines whether the vehicle may be in a deceleration state by setting a first preset threshold value, and provides a preliminary judgment basis for subsequent judgments based on the significant change in the longitudinal axis acceleration of the vehicle when it decelerates suddenly.

[0043] Next, after confirming that the vehicle is in a decelerating driving state, the longitudinal average absolute value determined based on the longitudinal axis acceleration set is further compared with the second preset threshold to verify whether the comprehensively considered average longitudinal deceleration increment is significant. This enhances the robustness of the deceleration detection by reducing the impact of occasional, short-term acceleration changes on the judgment, ensuring that the next step of detection is triggered when the acceleration change is continuous and significant.

[0044] Then, under the premise that the average longitudinal deceleration increment is significant, the real-time vertical axis accelerations in the vertical axis acceleration set are analyzed to determine whether the vehicle is in a bumpy driving state, and the corresponding vertical turbulence frequency is quantified and compared with the third preset threshold. This can effectively distinguish between sudden deceleration caused by vehicle vibration due to road unevenness and sudden deceleration caused by the driver's braking operation when driving on a flat road section, thereby avoiding misjudgment.

[0045] Finally, the vertical turbulence frequency is lower than the third preset threshold, and it is confirmed that the vehicle is driving on a flat road and the average deceleration increment is significant, so the vehicle is judged to be in a state of rapid deceleration, and a continuous warning control instruction is sent to the target warning unit to continuously remind the driver that the vehicle is in a state of rapid deceleration. It can be seen that the timely warning mechanism implemented is of great significance for preventing possible traffic accidents and improving driving safety.

[0046] In summary, it can not only reduce the misjudgment of sudden deceleration and improve the accuracy of the warning, but also work stably under various road conditions, unaffected by weather and lighting conditions, providing drivers with a more reliable safety guarantee, helping to effectively improve driving safety and reduce the occurrence of traffic accidents. In addition, by setting multiple preset thresholds, a flexible judgment mechanism is provided, which can be appropriately adjusted according to different driving conditions, vehicle characteristics, driving habits, etc., to adapt to various complex road conditions and driving situations. Such flexibility and adaptability further improve the accuracy and reliability of the warning.

[0047] In a further embodiment, step S1400, when the vertical turbulence frequency is lower than the third preset threshold, after sending a continuous alarm control instruction to the target alarm unit, comprises the following steps:

[0048] Step S1510: when the predetermined time period has passed again, obtaining the corresponding longitudinal average absolute value according to the longitudinal axis acceleration set within the predetermined time period, and determining whether the longitudinal average absolute value exceeds the second preset threshold;

[0049] At this time, a set of longitudinal accelerations within the predetermined time period is obtained, all longitudinal accelerations therein are summed, and then divided by the total number of these longitudinal accelerations to obtain a longitudinal average absolute value.

[0050] Step S1520: When the longitudinal average absolute value is lower than or equal to a second preset threshold, sending an end alarm control instruction to the target alarm unit to end controlling the target alarm unit to display that the vehicle is in a driving state of rapid deceleration.

[0051] At this time, it means that the vehicle has left the driving state of rapid deceleration, so there is no need to continue to control the target alarm unit to display that the vehicle is in a state of rapid deceleration, so the micro control unit constructs an end alarm control instruction and sends it to the device, so that the device ends the display after receiving the instruction. It is not difficult to understand that the driver of the vehicle can know that the current vehicle has left the driving state of rapid deceleration through the end of the display while driving the vehicle.

[0052] In this embodiment, by determining whether the longitudinal average value corresponding to the longitudinal axis acceleration set exceeds the second preset threshold value, when it is lower than or equal to the second preset threshold value, it is determined that the vehicle has left the rapid deceleration driving state, and the display of the vehicle in the rapid deceleration driving state ends, and the latest driving state of the vehicle is updated in time, and the driver is informed of the latest driving state without interfering with the driving of the driver of the vehicle.

[0053] In a further embodiment, obtaining a longitudinal axis acceleration set and a vertical axis acceleration set corresponding to vehicle travel within a predetermined time period includes the following steps:

[0054] Step S1110, from a preset first bidirectional linked list, a first preset number of longitudinal accelerations within a preset time length is obtained according to a last-to-first time sequence to construct a longitudinal acceleration set;

[0055] The first bidirectional linked list stores multiple longitudinal accelerations of the vehicle in the longitudinal direction during the driving process within a preset time period. The head node in the bidirectional linked list stores the longitudinal acceleration that is first in time sequence, and each node thereafter stores each longitudinal acceleration corresponding to the sequence that comes first and last.

[0056] Starting from the tail node in the first doubly linked list, according to the predecessor pointers of the tail node and the nodes before it, the required number of nodes are traversed one by one, thereby obtaining the longitudinal accelerations of the corresponding nodes in the last-to-first order to form a longitudinal acceleration set. The required number is the first preset number, which is less than or equal to the number of nodes in the first doubly linked list, and the specific number can be set by those skilled in the art as needed.

[0057] Step S1120: Acquire a second preset number of vertical axis accelerations within a preset time length from a preset second bidirectional linked list according to a last-to-first time sequence to construct a vertical axis acceleration set.

[0058] The second bidirectional linked list stores multiple vertical axis accelerations in the vertical axis direction during the vehicle's driving process within a preset time period. The head node in the bidirectional linked list stores the vertical axis acceleration that is first in time sequence, and each node thereafter stores each vertical axis acceleration corresponding to the first-to-last time sequence.

[0059] Starting from the tail node in the second doubly linked list, according to the predecessor pointers of the tail node and the nodes before it, the required number of nodes are traversed one by one, thereby obtaining the vertical axis accelerations of the corresponding nodes in the order of last to first to form a vertical axis acceleration set. The required number is the second preset number, which is less than or equal to the number of nodes in the second doubly linked list, and the specific number can be set by those skilled in the art as needed.

[0060] In this embodiment, by traversing and reading the corresponding number of longitudinal axis accelerations and vertical axis accelerations in the first bidirectional linked list and the second bidirectional linked list one by one from the end, a longitudinal axis acceleration set and a vertical axis acceleration set are correspondingly constructed, without the need to look up the timestamps of each acceleration, and the execution is very efficient and accurate.

[0061] In a further embodiment, step S1110, obtaining a first preset number of longitudinal accelerations within a preset time length from a preset first bidirectional linked list according to a last-to-first time sequence, and before constructing a longitudinal acceleration set, includes the following steps:

[0062] Step S1101, obtaining an original three-axis acceleration set measured by an inertial measurement device within a preset time period;

[0063] The microcontroller unit is connected to the inertial measurement device based on the communication connection realized by the serial peripheral interface, and collects the three single-axis accelerometers in the inertial measurement device. Within a preset time length, each basic linear acceleration of the vehicle in the three axis directions is measured according to the time sequence arranged in a first-to-last order. For each axis, the basic linear accelerations corresponding to the axis are aggregated to form the acceleration set of the axis, and then the acceleration sets of the three axes are aggregated to form the original three-axis acceleration set. The predetermined time length can be set by a technician in this field as needed, for example, 20ms. The serial peripheral interface (SPI) is a synchronous peripheral interface that enables the microcontroller unit to communicate with the inertial measurement device in a serial manner to exchange information.

[0064] Step S1102, performing low-pass filtering on the original three-axis acceleration set to obtain a corresponding target three-axis acceleration set;

[0065] In one embodiment, a first-order low-pass filter is used to filter the acceleration set of each axis in the original three-axis acceleration set, and high-frequency noise introduced by measurement errors, environmental noise and other factors is removed from the processed acceleration set, while retaining the corresponding low-frequency motion information. Thus, the acceleration sets after the filtering are collected to form the target three-axis acceleration set. The exemplary formula of the first-order low-pass filter is as follows:

[0066] y[n]=α·x[n]|+(1-α)·y[n-1]

[0067] Wherein: x[n] is the input acceleration at the current moment, y[n] is the output acceleration at the current moment, y[n-1] is the output acceleration at the previous moment, α is the filter coefficient, and its value range is 0 to 1. The smaller the value of α, the smaller the weight of x[n], the smoother the output waveform, but the response sensitivity is reduced. Personnel in this field can set α as needed according to the disclosure herein. It is recommended to set the value range of α to 0.1 to 0.3 to balance the degree of smoothness and response sensitivity.

[0068] In other embodiments, the low-pass filtering process may also be implemented by a Butterworth filter implemented based on a digital filter design method, or a discrete Fourier transform (DFT). For example, the acceleration set of each axis in the original acceleration axis is transformed from the time domain to the frequency domain using a discrete Fourier transform, filtering is implemented in the frequency domain, and after removing the high-frequency components, it is converted back to the time domain.

[0069] Step S1103: storing each longitudinal axis acceleration in the target three-axis acceleration set in a first double-linked list, and storing each vertical axis acceleration in the target three-axis acceleration set in a second double-linked list.

[0070] Initialize and create two bidirectional linked lists for the longitudinal axis and the vertical axis in the target three-axis acceleration set respectively, arrange the longitudinal axis accelerations of the longitudinal axis in the target three-axis acceleration set in a first-to-last order, and add them one by one in sequence to the bidirectional linked list of the longitudinal axis to obtain a first bidirectional linked list, wherein each node includes a corresponding longitudinal axis acceleration, a predecessor pointer pointing to a previous node, and a successor pointer pointing to a next node, and the head node stores the longitudinal axis acceleration with the earliest timing sequence, and add the vertical axis accelerations of the vertical axis in the target three-axis acceleration set arranged in a first-to-last order to the bidirectional linked list of the vertical axis one by one to obtain a second bidirectional linked list, wherein each node includes a corresponding vertical axis acceleration, a predecessor pointer pointing to a previous node, and a successor pointer pointing to a next node, and the head node stores the vertical axis acceleration with the earliest timing sequence.

[0071] Further in this step, if the first bidirectional linked list has stored multiple corresponding longitudinal accelerations obtained after the last preset duration, after the first bidirectional linked list is cleared, multiple corresponding longitudinal accelerations obtained after the current preset duration are stored. If the second bidirectional linked list has stored multiple corresponding vertical accelerations obtained after the last preset duration, after the second bidirectional linked list is cleared, multiple corresponding vertical accelerations obtained after the current preset duration are stored. Thus, it is ensured that the first bidirectional linked list and the second bidirectional linked list store the latest acceleration data corresponding to the real-time acquisition and filtering processing, and avoid the waste of storage resources caused by storing relatively old acceleration data.

[0072] In this embodiment, by using a bidirectional linked list structure to store the acceleration data of the longitudinal axis and the vertical axis of the vehicle respectively, the time sequence of the acceleration data can be ensured, so that the acceleration set within a specific time length can be obtained conveniently and accurately in a last-to-first time sequence, especially the latest acceleration can be retrieved in real time in dynamic monitoring later. At the same time, by low-pass filtering the original three-axis acceleration data, high-frequency noise is effectively removed, the main low-frequency motion characteristics are retained, the data is ensured to be smooth and reliable, and the dynamic acceleration changes of the vehicle can be accurately captured.

[0073] In a further embodiment, step S1400, sending a continuous alarm control instruction to a target alarm unit to continuously control the target alarm unit to display that the vehicle is in a state of rapid deceleration, comprises the following steps:

[0074] Step S2400: acquiring real-time traffic data, and using a preset rapid deceleration evaluation model to determine driving rationality based on the real-time traffic data, wherein the real-time traffic data includes any one or more of road type information, traffic congestion information, road slope, etc.;

[0075] The real-time traffic data includes all data describing the real-time driving section and its traffic conditions.

[0076] The road type information in the real-time traffic data is used to describe the corresponding road type and can be obtained through the Internet, such as: expressways, main roads, secondary roads, branch roads, rural roads, dedicated roads, pedestrian streets, bicycle lanes, emergency lanes, etc.

[0077] The traffic congestion information in the real-time traffic data is used to describe the real-time traffic congestion status, such as: smooth, slow, congested, severe congestion, etc.

[0078] The road slope in the real-time traffic data is used to quantify the slope of the real-time driving section. For example, it can be acquired through a slope sensor and a mass sensor, or it can be estimated by measuring the basic linear acceleration of the vehicle in the three axes based on an inertial measurement device through an extended Kalman filter and a least squares method to perform mass estimation and slope estimation.

[0079] The sudden deceleration assessment model is pre-trained to a convergence state, and learns the ability to determine driving rationality based on real-time traffic data. The specific training process can be flexibly implemented by those skilled in the art according to the reasoning process of the model disclosed below. The model structure of the sudden deceleration assessment model can be a text feature network followed by a classifier. The text feature network is suitable for extracting text features of the input text. The recommended model is the BERT model, and any other model such as Transfomer Encoder, RoBERTa, XLM-RoBERTa, MPNet, BiLSTM, GPT, etc. can also be used. Of course, those skilled in the art can also build a text feature network by themselves, as long as it can extract text features of the input text. The classifier can be implemented using a fully connected layer or MLP.

[0080] The real-time traffic data is input into the sudden deceleration evaluation model, and the text feature network in the sudden deceleration evaluation model extracts the deep semantic feature information corresponding to the real-time traffic data, and vectorizes the deep semantic feature information to obtain a text feature vector, wherein the deep semantic feature information represents the characteristics of the real-time traffic. Then, the classifier connected to the feature extraction network classifies and maps the text feature vector, obtains the classification probabilities mapped to the preset binary categories, and obtains the classification probability of the positive category in the binary category as the driving rationality. The binary category includes a positive category and a negative category, and the positive category represents that the driver drives the vehicle to a state of sudden deceleration under the real-time traffic condition, and such driving behavior is reasonable; the negative category represents that the driver drives the vehicle to a state of sudden deceleration under the real-time traffic condition, and such driving behavior is unreasonable.

[0081] Step S2410: Determine an alarm level that matches the driving rationality according to the driving rationality, call a continuous alarm control instruction corresponding to the alarm level and send it to a target alarm unit.

[0082] Multiple warning levels can be pre-set, for example, from low to high, they can be prompts, low-level warnings, serious warnings, and emergency warnings, and each warning level corresponds to a different degree of driving rationality. For example, the driving rationality is at the lowest level, indicating that sudden deceleration is extremely unreasonable under the current road conditions, and the driving rationality belongs to the value range (0, 0.1); the driving rationality is at a relatively close to the lowest level, indicating that the sudden deceleration behavior under the current road conditions is not very reasonable, and the driving rationality belongs to the value range [0, 1, 0.3); the driving rationality is at a medium level, indicating that the sudden deceleration behavior under the current road conditions has a certain rationality, but there is also an unreasonable risk, and the driving rationality belongs to [0.3, 0.7]; the driving rationality is close to the highest level, indicating that sudden deceleration is reasonable under the current road conditions.

[0083] Furthermore, a continuous alarm control instruction corresponding to each alarm level is constructed for ready call. For example: the higher the alarm level, the stronger the alarm display intensity, and the display can be any one or any combination of controlling the horn or buzzer installed in the vehicle to emit a specified alarm audio, controlling the light strip installed in the vehicle to emit a specified alarm light effect, controlling the vibrator in the seat installed in the vehicle to vibrate, etc., and the display intensity can be any one or any combination of the audio loudness, the light effect color and / or brightness and / or style, the vibration frequency and / or amplitude, etc. It is not difficult to understand that each continuous alarm control instruction maps and associates the alarm unit controlled by it.

[0084] Of course, those skilled in the art can also set these warning levels and their corresponding driving rationality levels as needed according to the above disclosure, as well as the continuous warning control instructions corresponding to each warning level.

[0085] The driving rationality is matched with the degree of driving rationality corresponding to each warning level. When the driving rationality belongs to the quantitative value range that characterizes the degree of driving rationality corresponding to the warning level, it is confirmed that the driving rationality matches the warning level, and the continuous alarm control instruction corresponding to the alarm level is called and sent to the alarm unit controlled by it, that is, the target alarm unit.

[0086] In this embodiment, the rationality of the driving behavior of the driver driving the vehicle to a state of sudden deceleration under the corresponding real-time road conditions is quantitatively analyzed based on the real-time road condition data, so as to realize the rationality evaluation of the sudden deceleration behavior, and trigger the corresponding alarm mechanism accordingly. In this process, the sudden deceleration evaluation model based on deep learning is used to process and analyze the real-time road condition data, so as to extract the deep semantic feature information, and classify and map it to the quantitative index of the corresponding driving rationality, so as to ensure the accuracy and efficiency of the evaluation. In addition, it is possible to identify whether the sudden deceleration behavior is reasonable, and dynamically adjust the alarm level according to the different degrees of driving rationality, and issue an alarm to the driver in at least one way (such as audio, light, seat vibration, etc.) through the alarm unit, so as to respond to the real-time road conditions more accurately, and timely remind and / or prompt the driver to pay attention to the potential driving risks, thereby helping to improve driving safety and road traffic efficiency, and helping the driver to make more reasonable driving decisions under various complex road conditions, reduce traffic accidents caused by sudden deceleration, and improve the overall traffic fluency.

[0087] In a further embodiment, step S1400, sending a continuous alarm control instruction to a target alarm unit to continuously control the target alarm unit to display that the vehicle is in a state of rapid deceleration, comprises the following steps:

[0088] Step S3400: using a preset habit recognition model to determine the confidence level corresponding to the driving data belonging to the driving habit according to the longitudinal axis acceleration set and the vertical axis acceleration set;

[0089] The habit recognition model is pre-trained to a convergence state, and acquires the ability to determine the confidence corresponding to the driving data belonging to the driving habit based on multiple longitudinal accelerations and vertical accelerations. The specific training process can be flexibly implemented by those skilled in the art according to the reasoning process of the model disclosed below. The model structure of the sudden deceleration evaluation model can be a text feature network followed by a classifier. The text feature network is suitable for extracting text features of the input text. The recommended model is the BERT model, and any other model such as Transfomer Encoder, RoBERTa, XLM-RoBERTa, MPNet, BiLSTM, GPT, etc. can also be used. Of course, those skilled in the art can also build a text feature network by themselves, as long as it can extract the text features of the input text. The classifier can be implemented using a fully connected layer or MLP.

[0090] Each longitudinal acceleration in the longitudinal acceleration set and each vertical acceleration in the vertical acceleration set are input into the habit recognition model, and the text feature network in the habit recognition model extracts the deep semantic feature information corresponding to these accelerations, and vectorizes the deep semantic feature information to obtain a text feature vector, wherein the deep semantic feature information represents the characteristics of the vehicle's sudden deceleration. Then, the classifier connected to the feature extraction network classifies and maps the text feature vector, obtains the classification probabilities mapped to the preset binary categories, and obtains the classification probability of the positive category in the binary category as the confidence corresponding to the driving data of these accelerations belonging to the driving habit. The binary category includes a positive category and a negative category, and the positive category represents that the vehicle's sudden deceleration belongs to the driving habit; the negative category represents that the vehicle's sudden deceleration does not belong to the driving habit.

[0091] Step S3410: Determine an alarm level that matches the confidence level according to the confidence level, call a continuous alarm control instruction corresponding to the alarm level and send it to a target alarm unit.

[0092] Multiple warning levels are preset, from low to high, they can be prompts, low-level warnings, serious warnings, and emergency warnings, and each warning level corresponds to a different confidence interval. For example, if the confidence is very high, it means that the current driving behavior is very consistent with the driver's habits, and only a low-level prompt may be needed to inform the driver that the vehicle currently being driven is in a state of rapid deceleration; if the confidence is low, it means that the driving behavior deviates from the habit, and a higher level of warning may be required. Specifically, the corresponding confidence value range can be set for each warning level, for example: when the confidence is lower than 0.1, it is an emergency warning, when the confidence is between 0.1 and 0.3, it is a serious warning, when the confidence is between 0.3 and 0.7, it is a low-level warning, and when the confidence is higher than 0.7, it is a prompt. These value ranges can be adjusted according to actual conditions and safety requirements.

[0093] In addition, the confidence value range corresponding to the alarm level to which it belongs is determined based on the confidence level, and the preset continuous alarm control instruction corresponding to the alarm level corresponding to the value range is called. The alarm unit controlled by the continuous alarm control instruction may be any one or more of a speaker, a buzzer, a vibrator, a display screen, etc., and its intensity and form of expression may vary according to the alarm level. For example, an emergency alarm may be accompanied by loud audio and / or eye-catching lights and / or strong seat vibrations, while a prompt may be only a slight seat vibration and / or a gentle audio prompt. The called alarm control instruction is sent to the target alarm unit, i.e., the controlled alarm unit, to achieve real-time warning to the driver.

[0094] In this embodiment, the habit recognition model is used to analyze the confidence level corresponding to the driving habit of the vehicle's sudden deceleration based on the longitudinal axis acceleration set and the vertical axis acceleration set, so that the alarm level is dynamically adjusted according to the different degrees of confidence, and the alarm unit promptly issues an appropriate warning to the driver in at least one of the following ways: audio, light, seat vibration, etc., to avoid excessive warnings.

[0095] In a further embodiment, step S2410 or step S3410, calling a continuous alarm control instruction corresponding to the alarm level and sending it to a target alarm unit, comprises the following steps:

[0096] Step S4411, obtaining the frequent cumulative number of times corresponding to sending the continuous alarm control instruction;

[0097] The interval between each sending of the continuous alarm control instruction and the last sending is obtained. When the interval is less than or equal to the preset time, the frequency division cumulative value of the counter is triggered to increase by one. In this way, the current frequent cumulative number of times can be read. The preset time can be set as needed by those skilled in the art according to the disclosure herein.

[0098] Step S4412: When the accumulated number of frequent occurrences meets a preset condition, the determined alarm level is increased, and a continuous alarm control instruction corresponding to the increased alarm level is called and sent to a target alarm unit.

[0099] When the cumulative number of frequent occurrences exceeds the preset threshold, it means that too many continuous warning control instructions have been frequently issued in a short period of time. Accordingly, the determined warning level needs to be increased by one level, and then the preset continuous warning control instruction for the increased warning level is called and sent to the target warning unit to warn the driver with a relatively increased intensity. The preset threshold can be set as needed by those skilled in the art according to the disclosure herein.

[0100] In this implementation, by determining whether the cumulative number of times the continuous warning control instruction is sent is too high, the determined warning level is increased and the corresponding continuous warning control instruction is called to send to the target warning unit when the number is too high. This solves the problem of driver fatigue or ignoring warning information due to frequent warnings, ensures that the driver remains alert to the warning information, and thus improves the effectiveness of the warning. In addition, the driver is also informed that his frequent sudden deceleration driving behavior will increase the risk and he needs to drive with caution.

[0101] See also Figure 3 According to one aspect of the present application, a vehicle driving warning device is provided, comprising a data acquisition module, a first judgment module, a second judgment module and a driving warning module, wherein the data acquisition module is used to acquire a longitudinal acceleration set and a vertical acceleration set corresponding to vehicle driving within a predetermined time period; the first judgment module is used to obtain the corresponding longitudinal average absolute value according to the longitudinal acceleration set when the latest longitudinal acceleration in the longitudinal acceleration set is less than a first preset threshold value, and judge whether the longitudinal average absolute value exceeds a second preset threshold value; the second judgment module is used to determine the vertical turbulence frequency corresponding to the bumpy driving condition according to the vertical acceleration set when the longitudinal average absolute value exceeds the second preset threshold value, and judge whether the vertical turbulence frequency exceeds a third preset threshold value; the driving warning module is used to send a continuous warning control instruction to a target warning unit when the vertical turbulence frequency is lower than the third preset threshold value, so as to continuously control the target warning unit to display that the vehicle is in a state of rapid deceleration.

[0102] In a further embodiment, the driving warning module includes: a first judgment submodule, which is used to obtain the corresponding longitudinal average absolute value based on the longitudinal axis acceleration set within the predetermined time period when the predetermined time period passes again, and judge whether the longitudinal average absolute value exceeds the second preset threshold; a driving warning submodule, which is used to send an end warning control instruction to the target warning unit when the longitudinal average absolute value is lower than or equal to the second preset threshold, so as to end controlling the target warning unit to display that the vehicle is in a state of rapid deceleration.

[0103] In a further embodiment, the data acquisition module includes: a first collection submodule, used to obtain a first preset number of longitudinal axis accelerations within a preset time length from a preset first bidirectional linked list according to a last-to-first time sequence, and construct a longitudinal axis acceleration set; a second collection submodule, used to obtain a second preset number of vertical axis accelerations within a preset time length from a preset second bidirectional linked list according to a last-to-first time sequence, and construct a vertical axis acceleration set.

[0104] In a further embodiment, before the first set submodule, it includes: a data acquisition submodule, used to obtain the original three-axis acceleration set measured by the inertial measurement device within a preset time length; a filtering processing submodule, used to perform low-pass filtering on the original three-axis acceleration set to obtain the corresponding target three-axis acceleration set; a data storage submodule, used to store each longitudinal axis acceleration in the target three-axis acceleration set in a first bidirectional linked list, and store each vertical axis acceleration in the target three-axis acceleration set in a second bidirectional linked list.

[0105] In a further embodiment, the driving alarm module includes: a real-time acquisition submodule, used to acquire real-time road condition data, and use a preset sudden deceleration evaluation model to determine the driving rationality based on the real-time road condition data, wherein the real-time road condition data includes any one or more of road type information, traffic congestion information, and road slope; a first instruction sending submodule, used to determine an alarm level that matches the driving rationality based on the driving rationality, and call the continuous alarm control instruction corresponding to the alarm level to send it to the target alarm unit.

[0106] In a further embodiment, the driving alarm module includes: a confidence determination submodule, which is used to use a preset habit recognition model to determine the confidence corresponding to the driving data belonging to the driving habit according to the longitudinal axis acceleration set and the vertical axis acceleration set; a second instruction sending submodule, which is used to determine the alarm level that matches it according to the confidence level, and call the continuous alarm control instruction corresponding to the alarm level to send it to the target alarm unit.

[0107] In a further embodiment, the first instruction sending submodule or the second instruction sending submodule includes: a frequent accumulation unit, used to obtain the frequent cumulative number of times corresponding to the sending of continuous alarm control instructions; an instruction sending unit, used to increase the determined alarm level when the frequent cumulative number of times meets the preset conditions, and call the continuous alarm control instruction corresponding to the increased alarm level to send it to the target alarm unit.

[0108] In order to solve the above technical problems, the present application also provides a computer device. Figure 4As shown, a schematic diagram of the internal structure of a computer device. The computer device includes a processor, a computer-readable storage medium, a memory, and a network interface connected via a system bus. Among them, the computer-readable storage medium of the computer device stores an operating system, a database, and computer-readable instructions. The database may store a control information sequence. When the computer-readable instructions are executed by the processor, the processor can implement a vehicle driving alarm method. The processor of the computer device is used to provide computing and control capabilities to support the operation of the entire computer device. The memory of the computer device may store computer-readable instructions. When the computer-readable instructions are executed by the processor, the processor can execute the vehicle driving alarm method of the present application. The network interface of the computer device is used to connect and communicate with a terminal. Those skilled in the art can understand that Figure 4 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0109] In this embodiment, the processor is used to execute Figure 3 The memory stores the program code and various data required to execute the above modules or submodules. The network interface is used to transmit data between user terminals or servers. The memory in this embodiment stores the program code and data required to execute all modules / submodules in the vehicle driving warning device of this application, and the server can call the program code and data of the server to execute the functions of all submodules.

[0110] The present application also provides a storage medium storing computer-readable instructions. When the computer-readable instructions are executed by one or more processors, the one or more processors execute the steps of the vehicle driving warning method of any embodiment of the present application.

[0111] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments of the present application can be implemented by instructing the relevant hardware through a computer program, and the computer program can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, the aforementioned storage medium can be a computer-readable storage medium such as a disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM).

[0112] It will be understood by those skilled in the art that the various operations, methods, steps, measures, and schemes in the processes discussed in this application may be alternated, changed, combined, or deleted. Furthermore, other steps, measures, and schemes in the various operations, methods, and processes discussed in this application may also be alternated, changed, rearranged, decomposed, combined, or deleted. Furthermore, the steps, measures, and schemes in the various operations, methods, and processes in the prior art that are open source and in this application may also be alternated, changed, rearranged, decomposed, combined, or deleted.

[0113] The above description is only a partial implementation method of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. A vehicle driving warning method, characterized in that: The steps include: Obtaining a longitudinal axis acceleration set and a vertical axis acceleration set corresponding to the vehicle's travel within a predetermined time period; When the latest longitudinal acceleration in the longitudinal acceleration set is less than the first preset threshold, obtaining the corresponding longitudinal average absolute value according to the longitudinal acceleration set, and determining whether the longitudinal average absolute value exceeds the second preset threshold; When the longitudinal average absolute value exceeds the second preset threshold, determining the vertical turbulence frequency corresponding to the matching bumpy driving condition according to the vertical axis acceleration set, and judging whether the vertical turbulence frequency exceeds the third preset threshold; When the vertical turbulence frequency is lower than a third preset threshold, a continuous warning control instruction is sent to the target warning unit to continuously control the target warning unit to display that the vehicle is in a driving state of rapid deceleration; The method of sending a continuous warning control instruction to a target warning unit to continuously control the target warning unit to display that the vehicle is in a state of rapid deceleration includes the following steps: Acquire real-time traffic data, and use a preset sudden deceleration evaluation model to determine driving rationality based on the real-time traffic data, wherein the real-time traffic data includes any one or more of road type information, traffic congestion information, and road slope; According to the driving rationality, a warning level that matches the driving rationality is determined, and a continuous warning control instruction corresponding to the warning level is called and sent to a target warning unit.

2. The vehicle driving warning method according to claim 1, characterized in that: When the vertical turbulence frequency is lower than the third preset threshold, after sending a continuous alarm control instruction to the target alarm unit, the method includes the following steps: When the predetermined time period has passed again, obtaining the corresponding longitudinal average absolute value according to the longitudinal axis acceleration set within the predetermined time period, and determining whether the longitudinal average absolute value exceeds the second preset threshold; When the longitudinal average absolute value is lower than or equal to a second preset threshold, an end warning control instruction is sent to the target warning unit to end controlling the target warning unit to display that the vehicle is in a driving state of rapid deceleration.

3. The vehicle driving warning method according to claim 1, characterized in that: Obtaining a longitudinal axis acceleration set and a vertical axis acceleration set corresponding to vehicle travel within a predetermined time period includes the following steps: From a preset first bidirectional linked list, a first preset number of longitudinal accelerations within a preset time length is obtained according to a last-to-first time sequence to construct a longitudinal acceleration set; From a preset second bidirectional linked list, a second preset number of vertical axis accelerations within a preset time length are acquired according to a last-to-first time sequence to construct a vertical axis acceleration set.

4. The vehicle driving warning method according to claim 3, characterized in that: From a preset first bidirectional linked list, a first preset number of longitudinal accelerations within a preset time length is obtained according to a time sequence from last to first, and before a longitudinal acceleration set is constructed, the following steps are included: Obtaining the original three-axis acceleration set measured by the inertial measurement device within a preset time period; Performing low-pass filtering on the original three-axis acceleration set to obtain a corresponding target three-axis acceleration set; Each longitudinal axis acceleration in the target three-axis acceleration set is stored in a first bidirectional linked list, and each vertical axis acceleration in the target three-axis acceleration set is stored in a second bidirectional linked list.

5. The vehicle driving warning method according to claim 1, characterized in that: Sending a continuous warning control instruction to a target warning unit to continuously control the target warning unit to display that the vehicle is in a state of rapid deceleration, comprises the following steps: Using a preset habit recognition model to determine the confidence level corresponding to the driving data belonging to the driving habit based on the longitudinal axis acceleration set and the vertical axis acceleration set; An alarm level that matches the confidence level is determined according to the confidence level, and a continuous alarm control instruction corresponding to the alarm level is called and sent to a target alarm unit.

6. The vehicle driving warning method according to claim 1 or 5, characterized in that: Calling a continuous alarm control instruction corresponding to the alarm level and sending it to the target alarm unit includes the following steps: Obtain the cumulative number of frequent transmissions corresponding to the continuous alarm control instruction; When the accumulated number of frequent occurrences meets a preset condition, the determined alarm level is increased, and a continuous alarm control instruction corresponding to the increased alarm level is called and sent to a target alarm unit.

7. A vehicle driving warning device, characterized in that: include: A data acquisition module, used to acquire a longitudinal axis acceleration set and a vertical axis acceleration set corresponding to the vehicle's travel within a predetermined time period; a first judgment module, configured to obtain a corresponding longitudinal average absolute value according to the longitudinal acceleration set when the latest longitudinal acceleration in the longitudinal acceleration set is less than a first preset threshold, and to judge whether the longitudinal average absolute value exceeds a second preset threshold; a second judgment module, configured to determine, when the longitudinal average absolute value exceeds a second preset threshold, a vertical turbulence frequency corresponding to a matching bumpy driving condition according to the vertical axis acceleration set, and to judge whether the vertical turbulence frequency exceeds a third preset threshold; A driving warning module, configured to send a continuous warning control instruction to a target warning unit when the vertical turbulence frequency is lower than a third preset threshold value, so as to continuously control the target warning unit to display that the vehicle is in a driving state of rapid deceleration; The method of sending a continuous warning control instruction to a target warning unit to continuously control the target warning unit to display that the vehicle is in a state of rapid deceleration includes the following steps: Acquire real-time traffic data, and use a preset sudden deceleration evaluation model to determine driving rationality based on the real-time traffic data, wherein the real-time traffic data includes any one or more of road type information, traffic congestion information, and road slope; According to the driving rationality, a warning level that matches the driving rationality is determined, and a continuous warning control instruction corresponding to the warning level is called and sent to a target warning unit.

8. A vehicle driving warning device, comprising a central processing unit and a memory, characterized in that: The central processing unit is used to call and run the computer program stored in the memory to execute the steps of the method according to any one of claims 1 to 6.

9. A non-volatile readable storage medium, characterized in that: It stores a computer program implemented according to the method described in any one of claims 1 to 6 in the form of computer-readable instructions, and when the computer program is called and executed by a computer, the steps included in the corresponding method are executed.

Images