Dangerous driving behavior recognition system, method and device

Abstract

The invention discloses a dangerous driving behavior recognition system, method and device, and relates to the technical field of dangerous driving behavior recognition. According to the technical scheme, sitting posture pressure data transmitted to a seat by a driver and holding pressure data generated when the driver operates a steering wheel are collected; the acceleration and deceleration frequency of the driver is determined based on the sitting posture pressure data, the vehicle speed change condition of the driver during vehicle driving can be represented, the violation operation frequency is determined based on the holding pressure data, and the violation degree of vehicle control by the driver can be represented; and the identification result of the dangerous driving behavior of the driver can be accurately output according to the acceleration and deceleration frequency and the violation operation frequency. According to the technical scheme, the posture characteristics of the driver for controlling the vehicle can be accurately captured through the pressure data, and dangerous driving behaviors of the driver can be accurately recognized.

Description

Technical Field

[0001] The present invention relates to the technical field of dangerous driving behavior identification, and in particular to a dangerous driving behavior identification system, method and device. Background Art

[0002] The driver's driving behavior has a significant impact on the occurrence of automobile accidents. Studies have shown that dangerous behaviors such as speeding, drunk driving, fatigue driving and distracted driving greatly increase the risk of traffic accidents. These behaviors not only directly affect the driver's reaction ability and judgment, but may also pose a threat to other road users. Therefore, the occurrence of automobile accidents is highly correlated with the driver's driving behavior. By identifying the driver's dangerous driving behavior, the safety of vehicle driving can be effectively improved. At present, there is a technical problem of insufficient accuracy in the identification strategy for dangerous driving behavior. Summary of the invention

[0003] In order to solve the technical problem of how to accurately identify dangerous driving behavior, the present invention provides a dangerous driving behavior identification system, method and device, which can accurately identify whether the driver has dangerous driving behavior.

[0004] The embodiment of the present invention provides the following solution:

[0005] In a first aspect, an embodiment of the present invention provides a system for identifying dangerous driving behavior, the system comprising:

[0006] A first acquisition module is installed on the seat of the vehicle, and is used to collect the sitting pressure data transmitted from the driver to the seat, and output it to the outside;

[0007] A second acquisition module is installed on the steering wheel of the vehicle, and is used to collect the grip pressure data of the driver controlling the steering wheel and output it to the outside;

[0008] A processing terminal connected to the first acquisition module and the second acquisition module, the processing terminal is used to obtain the sitting pressure data and the grip pressure data;

[0009] The processing terminal is also used to determine the driver's acceleration and deceleration frequency and illegal operation frequency based on the sitting pressure data and the grip pressure data, and output the recognition result of the driver's dangerous driving behavior based on the acceleration and deceleration frequency and the illegal operation frequency.

[0010] In an optional embodiment, the first acquisition module includes:

[0011] At least four pressure sensors are installed on the seat surface of the seat. The four pressure sensors are used to collect pressure values ​​transmitted to the seat surface by different hip positions of the driver and output them as sitting pressure data.

[0012] In an optional embodiment, the first acquisition module further includes:

[0013] Multiple filters, the input end of each filter is connected to the output end of each pressure sensor in a one-to-one correspondence, the output end of each filter is connected to the input end configured on the processing terminal, and the filter is used to implement a threshold when the fluctuation amount of the pressure value output by the correspondingly connected pressure sensor is less than a first threshold value.

[0014] In a second aspect, an embodiment of the present invention further provides a method for identifying dangerous driving behavior, the method comprising:

[0015] Acquire the sitting posture pressure data transmitted by the driver to the seat, and the grip pressure data of the driver controlling the steering wheel;

[0016] According to the sitting pressure data and the grip pressure data, the driver's acceleration and deceleration frequency and the frequency of illegal operations are determined respectively;

[0017] According to the frequency of acceleration and deceleration and the frequency of illegal operations, the recognition results of the driver's dangerous driving behavior are output.

[0018] In an optional embodiment, before respectively determining the driver's acceleration and deceleration frequency and illegal operation frequency according to the sitting pressure data and the grip pressure data, the method further includes:

[0019] Determining whether the driver is sitting in a target area of ​​the seat based on the sitting posture pressure data, wherein the target area is the area where the driver sits when operating the vehicle in a correct sitting posture;

[0020] When the driver is in the target area, an enabling command for starting the vehicle normally is output;

[0021] When the driver is not sitting in the target area, a prompt instruction is output to remind the driver to correct his sitting posture.

[0022] In an optional embodiment, the sitting posture pressure data is pressure values ​​collected by at least four pressure sensors; judging whether the driver is sitting in the target area of ​​the seat according to the sitting posture pressure data includes:

[0023] Determine the pressure value collected by each pressure sensor in the same period and the fluctuation amount of each pressure value according to the sitting pressure data;

[0024] When all fluctuations are greater than or equal to the first threshold, all pressure values ​​in the same period are summed to obtain a pressure summation result;

[0025] When the pressure summation result is greater than the second threshold, it is determined that the driver is sitting in the target area.

[0026] In an optional embodiment, before respectively determining the driver's acceleration and deceleration frequency and illegal operation frequency according to the sitting pressure data and the grip pressure data, the method further includes:

[0027] Acquiring target pressure data entered by the driver, wherein the target pressure data is data associated with the driver's own riding pressure;

[0028] The driver's identity is verified according to the sitting posture pressure data and the target pressure data; when the identity verification is passed, an enabling instruction for starting the vehicle normally is output.

[0029] In an optional embodiment, the driver's acceleration and deceleration frequency and illegal operation frequency are determined respectively according to the sitting pressure data and the grip pressure data, including:

[0030] Determine the change amount of the pressure value output by each pressure sensor during the identification period according to the sitting posture pressure data;

[0031] When the changes in the continuous recognition periods are all greater than the third threshold, determining that the driver performs a sudden acceleration action or a sudden deceleration action in the continuous recognition period;

[0032] Determine the frequency of acceleration and deceleration according to the cumulative number of executions of the rapid acceleration action and the rapid deceleration action in a preset time period;

[0033] Analyze and process the grip pressure data to obtain the time and number of times the driver's hands leave the steering wheel;

[0034] The frequency of illegal operations is obtained based on the disengagement time and the number of disengagements.

[0035] In an optional embodiment, after outputting the recognition result of the driver's dangerous driving behavior, the method further includes:

[0036] When the current level of the recognition result is less than the level threshold, the driver is prompted to drive dangerously;

[0037] When the current level of the recognition result is greater than or equal to the level threshold, driving control intervention is implemented on the vehicle according to the current level and a preset handling relationship, wherein the preset handling relationship is a correspondence between the danger level and driving intervention strategies of different degrees.

[0038] In a third aspect, an embodiment of the present invention further provides a device for identifying dangerous driving behavior, the device comprising:

[0039] The first acquisition module is used to acquire the sitting posture pressure data transmitted to the seat by the driver on the vehicle, and the grip pressure data of the driver controlling the steering wheel;

[0040] A determination module, used to determine the driver's acceleration and deceleration frequency and illegal operation frequency respectively according to the sitting pressure data and the grip pressure data;

[0041] The first output module is used to output the recognition result of the driver's dangerous driving behavior according to the acceleration and deceleration frequency and the illegal operation frequency.

[0042] Compared with the prior art, the system, method and device for identifying dangerous driving behavior of the present invention have the following advantages:

[0043] The technical solution of the present invention collects the sitting pressure data transmitted by the driver to the seat, and the grip pressure data of the driver controlling the steering wheel; the driver's acceleration and deceleration frequency is determined based on the sitting pressure data, which can characterize the speed change of the driver when driving the vehicle; the frequency of illegal operations is determined based on the grip pressure data, which can characterize the degree of violation of the driver's control of the vehicle; the recognition result of the driver's dangerous driving behavior can be accurately output based on the acceleration and deceleration frequency and the illegal operation frequency. This technical solution can accurately capture the posture characteristics of the driver's control of the vehicle through pressure data. Compared with traditional face image detection, it is not affected by psychological factors such as the driver's facial expression. This recognition method is more scientific and reasonable, and can accurately identify the driver's dangerous driving behavior. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the embodiments of this specification or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this specification. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0045] Figure 1 A schematic diagram of the structure of a dangerous driving behavior recognition system provided by an embodiment of the present invention;

[0046] Figure 2 A flowchart of a method for identifying dangerous driving behavior provided by an embodiment of the present invention;

[0047] Figure 3 A schematic diagram of the structure of a device for identifying dangerous driving behavior provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0048] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention belong to the scope of protection of the embodiments of the present invention.

[0049] See also Figure 1 , Figure 1 A schematic diagram of the structure of a dangerous driving behavior recognition system provided in an embodiment of the present invention, wherein the recognition system comprises a first acquisition module 1, a second acquisition module 2 and a processing terminal 3.

[0050] The first acquisition module 1 is installed on the seat of the vehicle. The first acquisition module 1 is used to collect the sitting posture pressure data transmitted from the driver to the seat and output it to the outside. The first acquisition module 1 can be composed of multiple pressure sensors. For example, at least four sensors are installed on the back and seat surface of the seat. The pressure transmitted from the driver's back to the seat is collected by at least four sensors on the back of the seat to determine whether the driver's back sitting posture is correct; the pressure transmitted from the driver's buttocks to the seat is collected by at least four sensors on the seat surface to determine whether the driver's buttocks sitting posture is accurate. The pressure values ​​collected by all pressure sensors are output as sitting posture pressure data.

[0051] The second acquisition module 2 is installed on the steering wheel of the vehicle. The second acquisition module 2 is used to collect the grip pressure data of the driver's control of the steering wheel and output it to the outside. Similarly, the second acquisition module 2 can also be composed of multiple pressure sensors, which can be installed on the steering wheel in a circular array, and the steering wheel is covered with fabric or leather to protect the installed pressure sensors. The grip pressure data collected by the second acquisition module 2 is the pressure transmitted from the driver's hand to the steering wheel when the driver is controlling the vehicle. It can indicate whether the driver's position of holding the steering wheel is correct, whether the hand is off the steering wheel, etc.

[0052] The processing terminal 3 can be a vehicle controller (ECU, Electronic Control Unit) or other equipment capable of performing data calculation and processing. The processing terminal 3 is connected to the first acquisition module 1 and the second acquisition module 2, and can be connected via cables and standard connectors. The processing terminal 3 is used to obtain the sitting pressure data and the grip pressure data; the processing terminal 3 is also used to determine the driver's acceleration and deceleration frequency and the frequency of illegal operations according to the sitting pressure data and the grip pressure data, and output the recognition result of the driver's dangerous driving behavior according to the acceleration and deceleration frequency and the frequency of illegal operations.

[0053] It should be noted that the identification result of dangerous driving behavior can be a dangerous driving level. Different dangerous driving levels are used to characterize the degree of danger of driving behavior. Pushing it to the driver can make the driver understand his or her dangerous driving behavior. The driver can further improve his or her driving behavior through danger prompts, thereby reducing the probability of accidents.

[0054] Exemplarily, the first acquisition module 1 includes at least four pressure sensors, which are installed on the seat surface in a rectangular shape, and each pressure sensor is located at a right angle to the rectangle. Of course, it can also be installed at intervals along the length direction of the driver's right leg. The four pressure sensors are used to collect the pressure values ​​transmitted to the seat surface by the driver's different hip positions, and output them as sitting pressure data. The four pressure sensors can be used to determine whether the driver's sitting posture is correct, and to identify the frequency of the driver's stepping on the accelerator pedal and the brake pedal. Of course, six pressure sensors can also be set to implement sitting pressure data collection. The six pressure sensors are distributed in a hexagon, and each pressure sensor is located at the intersection of the hexagon. A set of parallel sides of the hexagon is parallel to the width direction of the chair back. More pressure values ​​can be collected by six pressure sensors, thereby improving the accuracy of the calculation of the driver's acceleration and deceleration frequency.

[0055] Through experiments, it is found that the deformation of fabrics and leather around the pressure sensor, or other interference factors, can also cause slight changes in the pressure value output by the pressure sensor. In order to improve the accuracy of the sitting pressure data acquisition, it is necessary to filter the pressure value. The filtering process based on software consumes a lot of computing power, which affects the efficiency of identifying dangerous driving behaviors. Based on this, in a specific implementation, the first acquisition module also includes multiple filters.

[0056] The input end of each filter is connected to the output end of each pressure sensor in a one-to-one correspondence, and the output end of each filter is connected to the input end configured on the processing terminal. The filter is used to implement a threshold when the fluctuation of the pressure value output by the corresponding connected pressure sensor is less than the first threshold. The filter can be built as an RC filter based on actual needs, or it can be built through an operational amplifier, and it can filter the fluctuation of the pressure value. The first threshold can be set to 50N. When the fluctuation of the pressure value is less than the first threshold, it indicates that the pressure value fluctuation may be caused by the rebound of the fabric, and the fluctuating pressure value is determined to be invalid data; conversely, when the fluctuation of the pressure value is greater than or equal to the first threshold, the filter outputs the pressure value input by the pressure sensor to the processing terminal.

[0057] It can be understood that the filtering processing of pressure values ​​based on hardware filters has higher processing efficiency, and the filtering processing of all pressure values ​​does not occupy computing resources, which can enable the processing terminal to release more computing power, and apply the released computing power to the identification of dangerous driving behaviors, thereby improving the efficiency of identifying dangerous driving behaviors. The recognition system proposed in the embodiment of the present invention can be equipped on newly developed vehicle models, or it can be modified and upgraded for existing vehicles, and has good applicability. Compared with traditional machine vision recognition solutions, the recognition system of the embodiment of the present invention has better accuracy, which is conducive to the large-scale market application of vehicles.

[0058] Based on the same technical concept as the recognition system, the embodiment of the present invention also provides a method for identifying dangerous driving behavior, please refer to Figure 2 , Figure 2 Flow chart of the identification method. The method can be applied to a processing terminal on a vehicle to implement the identification of dangerous driving behaviors of the driver. The processing terminal can be the vehicle controller of the vehicle or other devices capable of running the identification method, and no specific limitation is made here. The identification method includes:

[0059] S11, acquiring the sitting posture pressure data transmitted from the driver to the seat in the vehicle, and the grip pressure data of the driver controlling the steering wheel.

[0060] Specifically, multiple pressure sensors can be installed on the seat to obtain the pressure values ​​at the corresponding positions, and all the pressure values ​​can be used as sitting pressure data. Similarly, a pressure sensor is set on the steering wheel to collect the pressure value applied by the driver's hand to the steering wheel, which is used as holding pressure data. The sitting pressure data and holding pressure data can be collected and obtained based on a preset period, and the preset period can be set based on actual needs, such as 0.1s. The obtained sitting pressure data and holding pressure data are stored based on the corresponding period tags.

[0061] In practical applications, since the sitting posture pressure data can represent the identity characteristics of the driver, based on this, after obtaining the sitting posture pressure data, the embodiment of the present invention can further verify the identity of the driver. Specifically, it includes:

[0062] The first step is to obtain the target pressure data entered by the driver. The target pressure data is data related to the driver's own riding pressure. The target pressure data can be directly set to the driver's weight, or the target pressure data can be set to the hip riding pressure calibrated by the historical riding record. The target pressure data can be entered through the vehicle system, or through the driver's mobile terminal (such as a mobile phone or watch), and then input into the vehicle system through the mobile terminal. The entry method is not specifically limited here.

[0063] The second step is to verify the identity of the driver based on the sitting pressure data and the target pressure data. The sitting pressure data is the pressure value applied to the seat when the current person sits on the seat. The sitting pressure data is consistent with the actual driver of the vehicle, and there may be differences. Identity verification is based on the sitting pressure data to determine whether the person sitting on the seat is the driver himself, that is, the actual driver. The target pressure data can be used to determine the pressure data corresponding to the driver himself, and compare it with the sitting pressure data to implement identity verification. If the pressure applied to the seat by the person represented by the sitting pressure data and the target pressure data is consistent, it is determined that the identity verification is passed, the person on the seat is the actual driver of the vehicle, and has the authority to start the vehicle, then the enable command for starting the vehicle normally is output; on the contrary, when the identity verification fails, it means that the vehicle may be stolen or driven by an illegal user, and the illegal user may have dangerous driving behavior, then an alarm is implemented, and the alarm method can be an intermittent horn of the vehicle or a message is sent to remind the vehicle owner to be vigilant against the illegal occupation of the vehicle.

[0064] Furthermore, when performing real-time identity verification based on the sitting pressure data and the target pressure data, a target interval can be set based on the target pressure data. If the sitting pressure data is within the target interval, the verification passes; otherwise, the identity verification fails.

[0065] Of course, identity verification can also be implemented through other data processing methods. For example, the sitting pressure data is the pressure value collected by four pressure sensors. All pressure values ​​in the same collection period are summed to obtain the total pressure AP. Based on the sum of the pressure values ​​collected by the pressure sensors at adjacent positions, four summation results of the pressure values ​​are obtained. The summation result of the pressure values ​​of the two front pressure sensors is recorded as the first summation result, the summation result of the pressure values ​​of the two rear pressure sensors is recorded as the second summation result, the summation result of the pressure values ​​of the two left pressure sensors is recorded as the third summation result, and the summation result of the pressure values ​​of the two right pressure sensors is recorded as the fourth summation result. The ratio of the first summation result to the total pressure AP is recorded as FP, the ratio of the second summation result to the total pressure AP is recorded as RP1, the ratio of the third summation result to the total pressure AP is recorded as LP, and the ratio of the fourth summation result to the total pressure AP is recorded as RP2. Because each person has a different weight and a different sitting posture, the five characteristic values ​​AP, FP, RP1, LP and RP2 are different from the recorded target pressure data. This comparison and verification strategy has high efficiency and accuracy. When the identity verification is passed using the sitting posture pressure data, step S12 is entered.

[0066] S12. Determine the driver's acceleration and deceleration frequency and illegal operation frequency respectively according to the sitting pressure data and the grip pressure data.

[0067] Specifically, when the driver steps on the accelerator pedal or the brake pedal, the leg muscles will contract or relax synchronously. This type of action will cause a corresponding change in the sitting pressure data. Therefore, the frequency of acceleration and deceleration of the driver's vehicle control can be obtained after analysis based on the sitting pressure data. The frequency of acceleration and deceleration can characterize the change in vehicle speed when the driver is driving the vehicle. The frequency of acceleration and deceleration directly reflects the change in vehicle speed caused by the driver's control factors and is not affected by road conditions. Therefore, this determination method has better accuracy than the speed sensor collecting vehicle speed to implement dangerous driving judgment. Similarly, the grip pressure data is the pressure applied by the driver's hands to the steering wheel when the driver is controlling the vehicle. If both hands are off the steering wheel, the grip pressure data is zero. When this type of data is collected, it means that the driver has violated the regulations. Therefore, based on the analysis and statistics of all grip pressure data, the frequency of the driver's illegal operations can be obtained.

[0068] It should be noted that when determining the frequency of acceleration and deceleration and the frequency of illegal operations, a data processing model can be constructed to implement data analysis and statistics, or the sitting pressure data and the holding pressure data can be processed into a first pressure curve and a second pressure curve respectively, and the frequency of acceleration and deceleration and the frequency of illegal operations can be determined based on the change characteristics of the first pressure curve and the second pressure curve. Among them, the first pressure curve is a curve of the sitting pressure data changing with the collection time, and the second pressure curve is a curve of the holding pressure data changing with the collection time. The first pressure curve and the second pressure curve can clearly reflect the changes in the pressure data, and then accurately derive the frequency of acceleration and deceleration and the frequency of illegal operations.

[0069] Exemplarily, step S12 includes sub-steps S12-1 to S12-5, which are specifically described as follows:

[0070] S12-1. Determine the change in the pressure value output by each pressure sensor during the identification period based on the sitting pressure data. The identification period can be set based on actual needs. For example, the collection period of a single pressure value can be set as the identification period. To reduce data processing redundancy, it can also be set to 2-5 collection periods as the identification period. Taking the calculation of the change in the pressure value output by a certain pressure sensor as an example, the characteristic values ​​ΔS0.1, ΔS0.2, ΔS0.3, ΔS0.4, and ΔS0.5 are calculated based on the pressure value change between 0.1s, where ΔS0.j represents the change within 0.1 seconds, 0.2 seconds, 0.3 seconds, 0.4 seconds, and 0.5 seconds, respectively, and j = 1, 2, 3, 4, 5.

[0071] S12-2, when the changes in the continuous recognition period are all greater than the third threshold, it is determined that the driver performs a sudden acceleration action or a sudden deceleration action in the continuous recognition period. The third threshold is set to 100N, according to the formula:

[0072]

[0073] t represents different acquisition moments. This variable corresponds to different acquisition cycles. NUM ΔS is the change in the continuous recognition period, and the continuous recognition period is set to 3 recognition periods. When more than 3 ΔS0.j is greater than 100, it means that a sudden acceleration or deceleration operation has occurred, and it is determined that the driver has performed a sudden acceleration or deceleration action in the continuous recognition period.

[0074] S12-3. Determine the frequency of acceleration and deceleration based on the cumulative number of executions of sudden acceleration and deceleration in a preset period. The preset period can be set to 30 minutes. If multiple sudden acceleration and deceleration actions occur within 30 minutes, the frequency of acceleration and deceleration is obtained based on the cumulative number of executions in the preset period. The frequency of acceleration and deceleration can be represented as the ratio of the cumulative number of executions to the preset period. When the frequency of acceleration and deceleration is too large, it indicates that the driver has dangerous driving behavior. The size of this value corresponds to the severity of the dangerous driving behavior and serves as an intervention decision parameter for subsequent fuzzy logic reasoning.

[0075] S12-4, analyzing and processing the grip pressure data to obtain the time and number of times the driver's hands are separated from the steering wheel. Statistical analysis can be performed based on the change characteristics of the grip pressure data. If the grip pressure data is less than a first preset value and the duration is greater than a second preset value, it indicates that the driver's hands are separated from the steering wheel, and the time and number of times the driver's hands are separated from the steering wheel are recorded accordingly.

[0076] The grip pressure data is obtained based on two pressure sensors installed on the steering wheel. The length of each pressure sensor is 20 cm. The two pressure sensors are arranged at the 3 o'clock and 9 o'clock directions of the steering wheel respectively. The pressure values ​​extracted by the two sensors are f k (t). Through experiments, it is found that the deformation of the pressure sensor or other interference factors will also cause the pressure value it outputs to change accordingly, and the change range is approximately between 0-50N. Therefore, when the pressure value output by the pressure sensor is greater than 50N, the pressure value is valid data. The calculation formula is as follows:

[0077]

[0078] In the above formula, f k When (t) is 0, the data is invalid. k (t) is 1 when the data is valid, the number of pressure sensors on the steering wheel is 2, k = 1 or 2; t represents different acquisition moments, and this variable corresponds to different acquisition cycles; NUM h It is the grip pressure data. It reads the data of the two pressure sensors and starts recording the disengagement time HOT and the number of disengagements HC at the same time.

[0079] S12-5. Obtain the frequency of illegal operations according to the disengagement time and the number of disengagements. The frequency of illegal operations is obtained by counting the disengagement time and the number of disengagements when the hands leave the steering wheel. The frequency of illegal operations can be directly represented by the disengagement time and the number of disengagements, or converted into a single value, which is not specifically limited here.

[0080] Furthermore, the grip pressure data can be used to identify the time and frequency of the driver's leaving the steering wheel with one hand, as well as the time and frequency of the driver leaving the steering wheel with both hands. When the vehicle speed is lower than the speed threshold, the time and frequency of the driver leaving the steering wheel with one hand are recorded as the frequency of illegal operations; when the vehicle speed is higher than or equal to the speed threshold, the time and frequency of the driver leaving the steering wheel with both hands are recorded as the frequency of illegal operations, thereby obtaining an accurate frequency of illegal operations.

[0081] It is understandable that since the sitting pressure data needs to be collected based on the pressure sensors, if the driver's sitting position is offset, his buttocks may not cover all the pressure sensors, resulting in insufficient accuracy of the sitting pressure data collection. Based on this, in a specific implementation, before determining the driver's acceleration and deceleration frequency and illegal operation frequency respectively based on the sitting pressure data and the grip pressure data, the method also includes:

[0082] The first step is to determine whether the driver is sitting in the target area of ​​the seat based on the sitting posture pressure data. The target area is the area where the driver sits when operating the vehicle in the correct sitting posture. The target area can be determined based on the calibration experiment. When the driver sits in the target area, the pressure applied by the driver to the seat can cover all pressure sensors and the vehicle can be driven normally and safely. Based on the pressure value represented by the sitting posture pressure data, it can be determined whether the driver's sitting posture is offset, including sitting forward, left, right, or back.

[0083] For example, the pressure values ​​collected by each pressure sensor in the same period and the fluctuation amount of each pressure value are determined based on the sitting pressure data. When all fluctuation amounts are greater than or equal to the first threshold, it means that each pressure value is valid, and all pressure values ​​in the same period are summed to obtain the pressure summation result; when the pressure summation result is greater than the second threshold, it is determined that the driver is sitting in the target area.

[0084] The first threshold may be set to 50N, and the determination of the target area may be implemented according to the following formula.

[0085]

[0086] t represents different acquisition moments, and this variable corresponds to different acquisition cycles; i = 1, 2, 3, 4, the initial pressure value collected by the four pressure sensors is n i (t), judge its effectiveness, if n i(t) is greater than 50N, then determine the pressure value N i (t) Valid, sum the four pressure values ​​and obtain the pressure sum result NUM S . NUM S When it is greater than the second threshold, it indicates that the pressure applied by the driver to the seat can cover all the pressure sensors, and it is determined that the driver is sitting in the target area.

[0087] In the second step, when the driver is in the target area, it indicates that the vehicle is being controlled in the correct sitting posture, and an enabling instruction for starting the vehicle normally is output. When the driver is not in the target area, a prompt instruction is output to remind the driver to correct the sitting posture until the driver adjusts the sitting posture to the target area. This strategy can perform sitting posture correction based on the pressure value collected by each pressure sensor to obtain accurate sitting posture pressure data and grip pressure data, and further derive the acceleration and deceleration frequency and the frequency of illegal operations. After obtaining the acceleration and deceleration frequency and the frequency of illegal operations, step S13 is entered.

[0088] S13. Output the recognition result of the driver's dangerous driving behavior according to the acceleration and deceleration frequency and the illegal operation frequency.

[0089] Specifically, the recognition results can be used to determine whether the driver has engaged in dangerous driving behavior and the severity of the dangerous driving behavior. Based on the frequency of illegal operations, the time HOT and the number of times HC that the driver's hands leave the steering wheel can be obtained. The frequency of acceleration and deceleration is AT. The fuzzy logic membership function is determined based on HOT, HC and AT through the three-part method, thereby forming a fuzzy logic judgment model. The identification result of the dangerous driving behavior is obtained based on the output result of the fuzzy logic judgment model. The fuzzy logic judgment model is a structure that uses three input parameters and three output conclusions. In the embodiments of the present invention, the fuzzy output values ​​are: information notification (CV), partial intervention (PV), and forced intervention (SV). Different reminder methods are used for different degrees of dangerous driving behaviors to achieve the purpose of intelligent reminder.

[0090] Furthermore, different levels of reminders and intervention controls are implemented based on the recognition results. After outputting the recognition results of the driver's dangerous driving behavior, the recognition method further includes:

[0091] S14. When the current level of the recognition result is less than the level threshold, it indicates that the dangerous driving behavior does not affect driving safety, and the driver is reminded of dangerous driving. The dangerous driving reminder includes sound reminders and instrument light reminders, which can remind the driver to drive safely. It should be noted that the implementation of dangerous driving reminders can also be implemented based on different levels. For example, when the current level of the recognition result is low, a lower decibel sound and lower brightness light reminder are implemented. When the current level increases, the decibel of the reminder sound and the brightness of the reminder light are increased.

[0092] S15. When the current level of the recognition result is greater than or equal to the level threshold, it means that the driver's dangerous driving behavior is serious and has affected the driving safety of the vehicle. Then, the vehicle is subject to driving control intervention according to the current level and the preset disposal relationship. The preset disposal relationship is the correspondence between the danger level and the driving intervention strategy of different degrees. When the danger level is low, the vehicle can be partially intervened, such as low-intensity braking; when the danger level is high, the vehicle is forced to intervene, including braking control and steering control, or the vehicle can be directly controlled to pull over to ensure driving safety.

[0093] Based on the identification method of the embodiment of the present invention, data statistics can also be implemented to generate a driving report for the driver, and the driving report can be sent to the driver's or car owner's mobile terminal or other devices to remind dangerous driving behaviors and display driving behavior analysis reports, so that the driver or car owner can know their dangerous driving behaviors and make targeted changes.

[0094] Based on the same technical concept as the recognition system, the embodiment of the present invention also provides a device for identifying dangerous driving behavior, please refer to Figure 3 , Figure 3 This is a schematic diagram of the structure of the identification device, which includes:

[0095] The first acquisition module 301 is used to acquire the sitting posture pressure data transmitted to the seat by the driver in the vehicle, and the grip pressure data of the driver controlling the steering wheel;

[0096] A determination module 302, for determining the driver's acceleration and deceleration frequency and illegal operation frequency respectively according to the sitting pressure data and the grip pressure data;

[0097] The first output module 303 is used to output the recognition result of the driver's dangerous driving behavior according to the acceleration and deceleration frequency and the illegal operation frequency.

[0098] In an optional embodiment, the identification device further includes:

[0099] A judgment module, used to judge whether the driver is sitting in a target area of ​​the seat according to the sitting posture pressure data, wherein the target area is an area where the driver sits when operating the vehicle in a correct sitting posture;

[0100] The second output module is used to output an enabling instruction for normally starting the vehicle when the driver is sitting in the target area;

[0101] The third output module is used for outputting a prompt instruction to remind the driver to correct the sitting posture when the driver is not sitting in the target area.

[0102] In an optional embodiment, the sitting posture pressure data is pressure values ​​collected by at least four pressure sensors; and the judgment module includes:

[0103] A first determination submodule is used to determine the pressure value collected by each pressure sensor in the same period and the fluctuation amount of each pressure value according to the sitting posture pressure data;

[0104] A processing submodule, used for summing all pressure values ​​in the same period when all fluctuation amounts are greater than or equal to a first threshold value to obtain a pressure summation result;

[0105] The second determination submodule is used to determine that the driver is sitting in the target area when the pressure summation result is greater than a second threshold.

[0106] In an optional embodiment, the identification device further includes:

[0107] A second acquisition module is used to acquire target pressure data entered by the driver, wherein the target pressure data is data associated with the driver's own riding pressure;

[0108] A verification module, used to verify the identity of the driver based on the sitting posture pressure data and the target pressure data;

[0109] The fourth output module is used to output an enabling instruction for normally starting the vehicle when the identity verification is passed.

[0110] In an optional embodiment, the determining module includes:

[0111] A third determination submodule is used to determine the change amount of the pressure value output by each pressure sensor during the identification period according to the sitting posture pressure data;

[0112] A fourth determination submodule, configured to determine that the driver performs a sudden acceleration action or a sudden deceleration action in the continuous identification period when the change amount in the continuous identification period is greater than a third threshold;

[0113] A fifth determination submodule, used to determine the acceleration and deceleration frequency according to the cumulative number of executions of the rapid acceleration action and the rapid deceleration action in a preset time period;

[0114] The first acquisition submodule is used to analyze and process the grip pressure data to obtain the time and number of times the driver's hands leave the steering wheel;

[0115] The second obtaining submodule is used to obtain the frequency of illegal operations according to the disengagement time and the number of disengagements.

[0116] In an optional embodiment, the identification device further includes:

[0117] A prompting module, used to prompt the driver of dangerous driving when the current level of the recognition result is less than the level threshold;

[0118] A control module is used to implement driving control intervention on the vehicle according to the current level and a preset handling relationship when the current level of the identification result is greater than or equal to the level threshold, wherein the preset handling relationship is the correspondence between the danger level and different degrees of driving intervention strategies.

[0119] The technical solution provided in the embodiments of the present invention has at least the following technical effects or advantages:

[0120] By collecting the sitting pressure data transmitted by the driver to the seat and the grip pressure data of the driver controlling the steering wheel; determining the driver's acceleration and deceleration frequency based on the sitting pressure data, it can characterize the speed change of the driver when driving the vehicle; determining the frequency of illegal operations based on the grip pressure data can characterize the degree of violation of the driver's operation of the vehicle; according to the acceleration and deceleration frequency and the frequency of illegal operations, the recognition result of the driver's dangerous driving behavior can be accurately output. This technical solution can accurately capture the posture characteristics of the driver's control of the vehicle through pressure data. Compared with traditional face image detection, it is not affected by psychological factors such as the driver's facial expression. This recognition method is more scientific and reasonable, and can accurately identify the driver's dangerous driving behavior.

[0121] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0122] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (modules, systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded computer, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0123] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0124] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0125] Although the preferred embodiments of the present invention have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0126] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A dangerous driving behavior recognition system, characterized in that: The system comprises: A first acquisition module is installed on a seat of the vehicle, and is used to collect sitting posture pressure data transmitted from the driver to the seat, and output the data to the outside; A second collection module is installed on the steering wheel of the vehicle, and the second collection module is used to collect the grip pressure data of the driver manipulating the steering wheel and output it to the outside; a processing terminal, connected to the first acquisition module and the second acquisition module, and used for acquiring the sitting pressure data and the holding pressure data; The processing terminal is also used to determine the driver's acceleration and deceleration frequency and illegal operation frequency according to the sitting pressure data and the holding pressure data, and output the recognition result of the driver's dangerous driving behavior according to the acceleration and deceleration frequency and the illegal operation frequency.

2. The dangerous driving behavior recognition system according to claim 1, characterized in that: The first acquisition module includes: At least four pressure sensors are installed on the seat surface of the seat, and the four pressure sensors are used to collect pressure values ​​transmitted to the seat surface from different hip positions of the driver, and output them as the sitting pressure data.

3. The dangerous driving behavior recognition system according to claim 2, characterized in that: The first acquisition module also includes: Multiple filters, the input end of each filter is connected to the output end of each pressure sensor in a one-to-one correspondence, the output end of each filter is connected to the input end configured on the processing terminal, and the filter is used to implement a threshold when the fluctuation amount of the pressure value output by the correspondingly connected pressure sensor is less than a first threshold value.

4. A method for identifying dangerous driving behavior, characterized in that: The method comprises: Acquiring sitting posture pressure data transmitted to the seat by the driver on the vehicle, and grip pressure data of the driver controlling the steering wheel; Determining the driver's acceleration and deceleration frequency and illegal operation frequency respectively according to the sitting posture pressure data and the grip pressure data; The recognition result of the dangerous driving behavior of the driver is outputted according to the acceleration / deceleration frequency and the illegal operation frequency.

5. The method for identifying dangerous driving behavior according to claim 4, characterized in that: Before respectively determining the acceleration / deceleration frequency and the illegal operation frequency of the driver according to the sitting posture pressure data and the grip pressure data, the method further includes: Determining whether the driver is sitting in a target area of ​​the seat according to the sitting posture pressure data, wherein the target area is an area where the driver sits when operating the vehicle in a correct sitting posture; When the driver is in the target area, outputting an enabling instruction for normally starting the vehicle; When the driver is not sitting in the target area, a prompt instruction is outputted to remind the driver to correct the sitting posture.

6. The method for identifying dangerous driving behavior according to claim 5, characterized in that: The sitting posture pressure data is pressure values ​​collected by at least four pressure sensors; judging whether the driver is sitting in the target area of ​​the seat according to the sitting posture pressure data includes: Determine the pressure value collected by each pressure sensor in the same period and the fluctuation amount of each pressure value according to the sitting pressure data; When all fluctuations are greater than or equal to the first threshold, all pressure values ​​in the same period are summed to obtain a pressure summation result; When the pressure summation result is greater than a second threshold, it is determined that the driver is sitting in the target area.

7. The method for identifying dangerous driving behavior according to claim 4, characterized in that: Before respectively determining the acceleration / deceleration frequency and the illegal operation frequency of the driver according to the sitting posture pressure data and the grip pressure data, the method further includes: Acquiring target pressure data entered by the driver, wherein the target pressure data is data associated with the driver's own riding pressure; performing identity verification on the driver according to the sitting posture pressure data and the target pressure data; When the identity verification is passed, an enabling instruction for normally starting the vehicle is output.

8. The method for identifying dangerous driving behavior according to claim 4, characterized in that: The determining, according to the sitting posture pressure data and the grip pressure data, respectively, the acceleration and deceleration frequency and the illegal operation frequency of the driver includes: Determine the change amount of the pressure value output by each pressure sensor during the identification period according to the sitting posture pressure data; When the changes in the continuous recognition periods are all greater than a third threshold, determining that the driver performs a sudden acceleration action or a sudden deceleration action in the continuous recognition period; Determining the acceleration and deceleration frequency according to the cumulative number of executions of the rapid acceleration action and the rapid deceleration action in a preset time period; Analyzing and processing the grip pressure data to obtain the time and number of times the driver's hands are released from the steering wheel; The frequency of the illegal operation is obtained according to the disengagement time and the disengagement number.

9. The method for identifying dangerous driving behavior according to claim 4, characterized in that: After outputting the recognition result of the driver's dangerous driving behavior, the method further includes: When the current level of the recognition result is less than the level threshold, prompting the driver for dangerous driving; When the current level of the recognition result is greater than or equal to the level threshold, driving control intervention is implemented on the vehicle according to the current level and a preset handling relationship, wherein the preset handling relationship is the correspondence between the danger level and different degrees of driving intervention strategies.

10. A device for identifying dangerous driving behavior, characterized in that: The device comprises: A first acquisition module is used to acquire sitting posture pressure data transmitted to the seat by the driver on the vehicle, and grip pressure data of the driver controlling the steering wheel; A determination module, used to determine the driver's acceleration and deceleration frequency and illegal operation frequency respectively according to the sitting pressure data and the grip pressure data; The first output module is used to output the recognition result of the dangerous driving behavior of the driver according to the acceleration and deceleration frequency and the illegal operation frequency.

Images