Driving Behavior Monitoring Method, Device and Vehicle-mounted Equipment

Abstract

The present invention provides a driving behavior monitoring method, device and vehicle-mounted device, relating to the field of communication technologies. The driving behavior monitoring method is executed by the vehicle-mounted device and includes: monitoring the first driving behavior of a first vehicle within a preset time period, where the absolute value of the acceleration of the first vehicle in the first driving behavior is greater than or equal to a first preset value; obtaining the driving state information of at least two first driving behaviors of the first vehicle; determining a dangerous driving probability value of the first vehicle according to the driving state information of the at least two first driving behaviors; judging whether the first vehicle meets the dangerous driving reporting condition according to the dangerous driving probability value; and if the first vehicle meets the dangerous driving reporting condition, sending a reporting message to a first device. The above solution can accurately detect dangerous driving behaviors without missed detection. At the same time, this method is carried out separately for the vehicle where the vehicle-mounted device is located, which is easy to implement and reduces the detection cost.

Description

Technical Field

[0001] The present invention relates to the field of communication technologies, and particularly to a driving behavior monitoring method, apparatus, and vehicle-mounted device. Background Art

[0002] The law of large numbers derived from probability accumulation events states that the accumulation of small-probability events one by one makes the occurrence of the event an inevitable event. Each dangerous driving behavior of a vehicle driver can be abstracted into a two-point distribution of probability. As the number of dangerous driving behaviors increases, it gradually approaches the binomial distribution of probability, and finally forms the Bernoulli law of large numbers, resulting in a car accident caused by dangerous driving behavior, or in other words, the occurrence of a car accident has stability. Further, the occurrence of a car accident becomes an inevitable event. Based on the above logical analysis, it can be seen that early and continuous detection of dangerous driving behaviors is of great practical significance for eliminating potential car accident hazards.

[0003] In the prior art, based on the detection of acceleration data, steering wheel turning data, etc., a functional relationship with the data at a specified historical moment is established, and the detection threshold formed for the historical data is used to determine the current dangerous driving behavior. Or the dangerous driving data is uploaded to the cloud platform for processing. It is feasible to perform cloud-side calculations for a single device, but with a large amount of data, there will be a blocking delay after calculation, which is not conducive to quickly issuing a warning for dangerous driving behaviors. In addition, for a large number of concurrent devices, it is not economical in terms of cost.

[0004] In the process of implementing the present application, the inventors found that there are at least the following problems in the prior art:

[0005] Currently, the detection of dangerous driving behaviors is divided into two types. The first is bottom-layer detection and then verification at specific locations. This method lacks intelligence and only detects dangerous driving behaviors by comparing thresholds. The system does not have the function of advanced machine learning, and the detection of dangerous driving behaviors requires verification at specific locations with cameras. However, for the sections with specific camera monitoring, the driving behaviors of the drivers of dangerous driving vehicles have moral risks. Specifically, they drive safely where there are cameras. Therefore, this verification method is likely to lead to missed detection of dangerous driving behaviors. The second is detection for the cloud, which requires high concurrency and high costs. In addition, delays, etc., cannot issue a dangerous warning immediately. Summary of the Invention

[0006] Embodiments of the present invention provide a driving behavior monitoring method, apparatus, and vehicle-mounted device to solve the problem that the existing methods for detecting dangerous driving behaviors are prone to missed detection or have relatively high detection costs.

[0007] To solve the above technical problems, an embodiment of the present invention provides a driving behavior monitoring method, which is executed by a vehicle-mounted device and includes:

[0008] Monitor the first driving behavior of the first vehicle within a preset time period, where the absolute value of the acceleration of the first vehicle in the first driving behavior is greater than or equal to a first preset value;

[0009] Obtain the driving state information of at least two first driving behaviors of the first vehicle;

[0010] Determine the dangerous driving probability value of the first vehicle according to the driving state information of the at least two first driving behaviors;

[0011] Judge whether the first vehicle meets the dangerous driving reporting condition according to the dangerous driving probability value;

[0012] If the first vehicle meets the dangerous driving reporting condition, send a reporting message to the first device.

[0013] Optionally, the determining the dangerous driving probability value of the first vehicle according to the driving state information of the at least two first driving behaviors includes:

[0014] Determine the first probability value of each first driving behavior of the first vehicle according to the driving state information of the at least two first driving behaviors and in combination with a probability calculation coefficient;

[0015] Determine the dangerous driving probability value of the first vehicle according to the first probability value.

[0016] Optionally, the determining the first probability value of each first driving behavior of the first vehicle includes:

[0017] According to the formula: φ(z) = 1 / (1 + e -z ), obtain the first probability value of each driving behavior;

[0018] where φ(z) is the first probability value corresponding to one first driving behavior; z = θ0AF + θ1AB + θ2AR + θ3AL + θ4VF + θ5VB + θ6AD, AF is the forward acceleration, AB is the backward acceleration, AL is the leftward acceleration, AR is the rightward acceleration, VF is the forward instantaneous speed, VB is the backward instantaneous speed, AD is the probability calculation coefficient; θ0 - θ6 are the regression coefficients of the regression classification model.

[0019] Optionally, after monitoring the first driving behavior of the first vehicle within the preset time period, the driving behavior monitoring method further includes:

[0020] After detecting the first first driving behavior within the preset time period, record the number of first driving behaviors within the preset time period, and record the time interval between every two first driving behaviors.

[0021] Optionally, determining whether the first vehicle meets the dangerous driving reporting condition according to the dangerous driving probability value includes:

[0022] Determining a correction value of the dangerous driving probability according to the number of the first driving behaviors and the time interval between every two first driving behaviors;

[0023] Determining a comprehensive judgment coefficient according to the correction value and the dangerous driving probability value;

[0024] Determining whether the first vehicle meets the dangerous driving reporting condition according to the comprehensive judgment coefficient.

[0025] Optionally, determining whether the first vehicle meets the dangerous driving reporting condition according to the comprehensive judgment coefficient includes:

[0026] If the comprehensive judgment coefficient is greater than or equal to a second preset value, determining that the first vehicle meets the dangerous driving reporting condition;

[0027] If the comprehensive judgment coefficient is less than the second preset value, determining that the first vehicle does not meet the dangerous driving reporting condition.

[0028] Optionally, after determining whether the first vehicle meets the dangerous driving reporting condition according to the dangerous driving probability value, it further includes:

[0029] If the first vehicle does not meet the dangerous driving reporting condition, adjusting the probability calculation coefficient by using the comprehensive judgment coefficient.

[0030] Optionally, the driving state information includes at least one of the following:

[0031] Forward acceleration, backward acceleration, leftward acceleration, rightward acceleration, forward instantaneous speed, backward instantaneous speed, leftward instantaneous speed, and rightward instantaneous speed.

[0032] Optionally, the reported information includes: dangerous driving data of the first vehicle and the position of the first vehicle.

[0033] Optionally, the position of the first vehicle includes: the position where the first vehicle is located when each first driving behavior occurs.

[0034] An embodiment of the present invention further provides an in-vehicle device, including: a processor, a memory, and a computer program stored on the memory and executable on the processor, where when the computer program is executed by the processor, the steps of the above-mentioned driving behavior monitoring method are implemented.

[0035] An embodiment of the present invention further provides a driving behavior monitoring device, which is applied to an in-vehicle device and includes:

[0036] A monitoring module, configured to monitor the first driving behavior of a first vehicle within a preset time period, where the absolute value of the acceleration of the first vehicle in the first driving behavior is greater than or equal to a first preset value;

[0037] An acquisition module, configured to acquire the driving state information of at least two first driving behaviors of the first vehicle;

[0038] A determination module, configured to determine a dangerous driving probability value of the first vehicle according to the driving state information of the at least two first driving behaviors;

[0039] A judgment module, configured to judge whether the first vehicle meets the dangerous driving reporting condition according to the dangerous driving probability value;

[0040] A sending module, configured to send a reporting message to a first device if the first vehicle meets the dangerous driving reporting condition.

[0041] An embodiment of the present invention further provides a readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps of the above-mentioned driving behavior monitoring method are implemented.

[0042] The beneficial effects of the present invention are:

[0043] In the above solution, by monitoring the first driving behavior by an in-vehicle device, determining the dangerous driving probability value of the first vehicle according to the driving state information of at least two first driving behaviors, and sending a reporting message when the reporting condition is met, the detection of dangerous driving behaviors can be accurately performed without missed detection. At the same time, this method is carried out separately for the vehicle where the in-vehicle device is located, which is easy to implement and reduces the detection cost. Description of the Drawings

[0044] Figure 1 It represents a schematic flowchart of the driving behavior monitoring method according to an embodiment of the present invention;

[0045] Figure 2 It represents a schematic diagram of the functional modules of the in-vehicle device according to an embodiment of the present invention;

[0046] Figure 3 It represents one of the schematic flowcharts according to an embodiment of the present invention;

[0047] Figure 4 It represents another schematic flowchart according to an embodiment of the present invention;

[0048] Figure 5 It represents a schematic diagram of the modules of the driving behavior monitoring device according to an embodiment of the present invention;

[0049] Figure 6 It represents the structural diagram of the in-vehicle device according to an embodiment of the present invention. Detailed implementation manners

[0050] To make the objectives, technical solutions and advantages of the present invention clearer, the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0051] Aiming at the problems that the existing methods for detecting dangerous driving behaviors are prone to missed detections or have relatively high detection costs, the present invention provides a driving behavior monitoring method, device and vehicle-mounted device.

[0052] As Figure 1 shown, the driving behavior monitoring method according to an embodiment of the present invention is executed by a vehicle-mounted device and includes:

[0053] Step 11, monitoring a first driving behavior of a first vehicle within a preset time period, where the absolute value of the acceleration of the first vehicle in the first driving behavior is greater than or equal to a first preset value;

[0054] It should be noted that the first driving behavior mentioned in the embodiments of the present application is recognized as a dangerous driving behavior. That is to say, when the first vehicle suddenly accelerates or decelerates, and the acceleration of the acceleration or deceleration exceeds the first preset value, it is considered that the first vehicle has a dangerous driving behavior.

[0055] Step 12, obtaining driving state information of at least two first driving behaviors of the first vehicle;

[0056] It should be noted that the driving state information includes at least one of the following:

[0057] Forward acceleration, backward acceleration, leftward acceleration, rightward acceleration, forward instantaneous speed, backward instantaneous speed, leftward instantaneous speed and rightward instantaneous speed.

[0058] It should be noted that the instantaneous speed refers to the speed after a preset time (for example, 3 seconds of acceleration) in the acceleration or deceleration direction at each acceleration or deceleration moment. For example, the forward instantaneous speed refers to the forward speed of the first vehicle after a preset time at the acceleration moment when the first vehicle suddenly accelerates.

[0059] Step 13, determining a dangerous driving probability value of the first vehicle according to the driving state information of the at least two first driving behaviors;

[0060] Step 14, judging whether the first vehicle meets the dangerous driving reporting condition according to the dangerous driving probability value;

[0061] Step 15, if the first vehicle meets the dangerous driving reporting condition, sending a reporting message to a first device;

[0062] It should be noted that the reported information includes: the dangerous driving data of the first vehicle and the location of the first vehicle; optionally, the location of the first vehicle includes: the location of the first vehicle when each first driving behavior occurs, and optionally, the location of the first vehicle includes: the location of the first vehicle when the most recent dangerous driving behavior occurs.

[0063] It should be noted that the execution subject of the embodiment of the present invention is an in-vehicle device installed in a vehicle. By detecting the dangerous driving behavior of the vehicle through the in-vehicle device, real-time detection of the vehicle is achieved, avoiding missed detection. At the same time, this method is carried out separately for the vehicle where the in-vehicle device is located, which is easy to implement and reduces the detection cost.

[0064] Furthermore, the first device mentioned in this application can be a roadside device or a base station. After the in-vehicle device sends the reported information to the roadside device, the roadside device can send the dangerous driving behavior of the first vehicle to other vehicles, and other vehicles can then know the location of the first vehicle, thereby avoiding the occurrence of accidents; after the in-vehicle device sends the reported information to the base station, the base station can send the dangerous driving behavior of the first vehicle to the platform (such as the highway supervision department) to facilitate the platform to monitor the first vehicle and avoid the occurrence of accidents.

[0065] It should be further noted that the specific implementation method of step 13 is as follows:

[0066] Step 131, determine the first probability value of each first driving behavior of the first vehicle according to the driving state information of the at least two first driving behaviors and in combination with the probability calculation coefficient;

[0067] It should be noted that the specific implementation method of this step is as follows:

[0068] According to formula one: φ(z) = 1 / (1 + e -z ), obtain the first probability value of each driving behavior;

[0069] where φ(z) is the first probability value corresponding to a first driving behavior; z = θ0AF + θ1AB + θ2AR + θ3AL + θ4VF + θ5VB + θ6AD, AF is the forward acceleration, AB is the backward acceleration, AL is the leftward acceleration, AR is the rightward acceleration, VF is the forward instantaneous speed, VB is the backward instantaneous speed, and AD is the probability calculation coefficient; θ0 - θ6 are the regression coefficients of the regression classification model.

[0070] Step 132, determine the dangerous driving probability value of the first vehicle according to the first probability value;

[0071] It should be noted that after obtaining the first probability value of each dangerous driving behavior, the dangerous driving probability value of the first vehicle can be obtained based on the first probability values of multiple dangerous driving behaviors. Specifically, the method for obtaining the dangerous driving probability value can be: taking the mean of the first probability values of multiple dangerous driving behaviors as the dangerous driving probability value.

[0072] Further, it should be noted that after step 11, the driving behavior monitoring method according to the embodiment of the present invention further includes:

[0073] After detecting the first first driving behavior within the preset time period, record the number of first driving behaviors within the preset time period, and record the time interval between every two first driving behaviors.

[0074] It should be noted that here, the first driving behavior is counted, that is, the number of times the first driving behavior appears within the preset time period is statistically counted. At the same time, the time interval between every two first driving behaviors is statistically counted. Further, the number of times of the first driving behavior and the time interval between every two first driving behaviors can be used to subsequently determine whether the first vehicle meets the dangerous driving reporting condition.

[0075] It should also be noted that in specific implementation, the implementation process of step 14 is: determining whether the dangerous driving probability value is greater than or equal to the initial danger coefficient. If the dangerous driving probability value is greater than or equal to the initial danger coefficient, then determine whether the first vehicle meets the dangerous driving reporting condition according to the dangerous driving probability value.

[0076] Further, the specific implementation method for determining whether the first vehicle meets the dangerous driving reporting condition according to the dangerous driving probability value includes:

[0077] Step 141, determining a correction value of the dangerous driving probability according to the number of first driving behaviors and the time interval between every two first driving behaviors;

[0078] It should be noted that one way to obtain the correction value is: pre-set a correspondence table between the number of first driving behaviors, the time interval between every two first driving behaviors, and the correction value. This table contains multiple groups of corresponding relationships between the number of first driving behaviors, the time interval between every two first driving behaviors, and the correction value. Just look up the table according to the currently obtained number of first driving behaviors and the time interval between every two first driving behaviors to obtain the corresponding correction value. The correction value can be positive or negative.

[0079] Step 142, determining a comprehensive judgment coefficient according to the correction value and the dangerous driving probability value;

[0080] It should be noted that after obtaining the correction value, adding the correction value to the obtained dangerous driving probability value can obtain the comprehensive judgment coefficient.

[0081] Step 143, according to the comprehensive judgment coefficient, determine whether the first vehicle meets the dangerous driving reporting condition;

[0082] Specifically, the implementation method of this step is: if the comprehensive judgment coefficient is greater than or equal to the second preset value, it is determined that the first vehicle meets the dangerous driving reporting condition; if the comprehensive judgment coefficient is less than the second preset value, it is determined that the first vehicle does not meet the dangerous driving reporting condition.

[0083] It should be further noted that after determining whether the first vehicle meets the dangerous driving reporting condition according to the dangerous driving probability value, it further includes:

[0084] If the first vehicle does not meet the dangerous driving reporting condition, adjust the probability calculation coefficient using the comprehensive judgment coefficient.

[0085] Specifically, when making the adjustment, the comprehensive judgment coefficient can be assigned to the probability calculation coefficient to achieve the adjustment of the probability calculation coefficient.

[0086] The specific implementation method of the embodiments of the present application is described as follows.

[0087] As Figure 2 shown, the in-vehicle device of the embodiments of the present invention mainly includes the following functional modules:

[0088] A wireless communication module, used for information sending and receiving, mainly for sending reporting information.

[0089] A lane-level positioning module, used to locate the position of the vehicle when a dangerous driving behavior occurs, and its positioning accuracy is at the lane level, that is, it can accurately locate which lane the vehicle is in on the road.

[0090] A vehicle condition information perception module, used to sense the driving state of the vehicle. As the main information collection module for dangerous driving behaviors of the present invention, it mainly collects the accelerations in the front, rear, left, and right directions during the vehicle driving process, mainly including: forward acceleration AF, backward acceleration AB, leftward acceleration AL, and rightward acceleration AR; and quantified data such as the instantaneous speed information in a certain acceleration direction calculated from the above accelerations.

[0091] A timing and counting module, timing is used to record the time interval between two times before and after a dangerous driving behavior, and counting is to accumulate the number of dangerous driving behaviors; at the same time, it can determine the continuously recorded time T after one trigger, that is, timing and counting are performed within the time T.

[0092] A logic classification module is used to logically classify the data of the vehicle condition information perception module and output the probability data of dangerous driving behaviors. It mainly calculates the probability of each dangerous driving behavior using the above formula (1). It should be noted that this logic classification module has the ability of machine learning.

[0093] A random statistics module is used to accumulate and determine dangerous driving behaviors and output trigger warning data.

[0094] It should be noted that for the statistical model involved in the random statistics module, the initial risk coefficient is set to D0, for example, 0.8. The risk coefficient is not fixed and will be adjusted downwards or upwards during the later learning process. This parameter will be adjusted by the system's machine learning itself.

[0095] A trigger module is used to fuse the output data of the timing and counting module and the random statistics module to obtain the result of whether to report information.

[0096] A silent module, if no information reporting is performed after passing through the trigger module, will trigger the self-learning of the logic classification module.

[0097] A warning reporting module reports the dangerous driving behaviors of the vehicle to roadside devices or base stations, and reports the location of the vehicle and the dangerous driving data. The dangerous driving data here refers to the output data of the trigger module, that is, the above comprehensive judgment coefficient.

[0098] The main implementation process of the embodiment of the present invention is as follows:

[0099] First, after a dangerous driving behavior occurs, the vehicle condition information perception module is triggered to work. For example, the vehicle condition information perception module outputs the forward acceleration, and then the in-vehicle device will automatically calculate the speed of the vehicle after t seconds. After the various detection data of the vehicle condition information perception module are quantified, they enter the logic classification module. At the same time, after the vehicle condition information perception module is triggered to start, it will also trigger the timing and counting module. The timing and counting module will record the time interval between two dangerous behaviors. For example, t n , which is the time interval between the (n - 1)-th and n-th dangerous driving occurrences. In addition, for the timing and counting module, the continuous working time T (i.e., the above-mentioned preset time period) can be set after it starts working.

[0100] Then, the quantified data of the vehicle condition information perception module enters the logic classification module. The data output by the logic classification module is the probability value p of a suspected traffic accident for a specific dangerous driving behavior (i.e., the above-mentioned first probability value). For example, the probability of a certain dangerous driving behavior accident is p = 0.7. For each dangerous driving behavior, the logic classification module will output a specific probability value, for example, p1, p2, p3...

[0101] Subsequently, the data output by the logical classification module enters the random statistics module, which randomly statistics the logical classification output data N times. For the N - time logical classification output, the average value A is taken, and the average value output is defined as the dangerous data of traffic accident occurrence (i.e., the probability value of dangerous driving); this data is compared with the preset danger coefficient D (for example, D = 0.8). If the average value A ≥ D, the data of the random statistics module enters the trigger module.

[0102] Finally, based on the dangerous data output by the random statistics module, the trigger module totals the correction values output by the timing and counting module. After superposition, a comprehensive judgment coefficient is obtained. If the comprehensive judgment coefficient is greater than or equal to the trigger value C0 (i.e., the second preset value, for example, C0 = 0.8), the vehicle with dangerous driving behavior is reported, and the data is reported to a certain platform (for example, the highway supervision department), including the dangerous driving data of the vehicle (i.e., the cumulative information of the vehicle's dangerous driving data, mainly referring to the comprehensive judgment coefficient) and the occurrence location of the most recent dangerous driving behavior. For example, the dangerous driving probability value D = 0.81 output by the random statistics module, after timing and counting superposition, the comprehensive judgment coefficient C = 0.82, C > C0, and dangerous data reporting is required; if C < C0, the data of the trigger module enters the silent module, and the quantization value of the data of the silent module will be input into the logical classification module as a parameter for machine learning, and the silent module will trigger the logical classification module to learn to improve the detection sensitivity of dangerous driving behavior.

[0103] The following is an example of the specific application situation of the embodiments of the present invention as follows.

[0104] As Figure 3 shown, taking the example that the vehicle driver has performed 10 times of forward sudden acceleration and speeding driving behavior, a detailed process of the driving behavior monitoring method of the embodiments of the present invention is as follows:

[0105] Step S11: The vehicle condition information perception module detects the driving behavior data of each forward sudden acceleration and quantifies and outputs the data to the logical classification module;

[0106] Specifically, the vehicle condition information perception module detects the forward acceleration AF each time, and at the same time calculates the speed VF after 3 seconds of sudden acceleration each time, as well as other data. It should be noted that this other data refers to the data required for the operation of Formula 1. It should be noted that data quantization refers to normalizing data at different data levels to obtain data with the same magnitude.

[0107] Step S12: The logical classification module performs calculations and outputs the first probability value of the driving behavior of each forward sudden acceleration to the random statistics module;

[0108] Specifically, for example, in the case of 10 speeding behaviors, the first probability values output by the logic classification module to the random statistics module are: 0.84, 0.73, 0.86, 0.87, 0.87, 0.63, 0.82, 0.77, 0.84, 0.86. At the same time, the first hard acceleration driving behavior will trigger the timing and counting module to start working, start timing and counting; specifically, timing is to record the interval between two hard acceleration driving, and counting is to record the number of dangerous driving behaviors within a specified time.

[0109] Step S13: The random statistics module stores N first probability values output from the logic classification module, obtains the dangerous driving probability value, and when the dangerous driving probability value is greater than the danger coefficient, inputs the dangerous driving probability value into the trigger module;

[0110] Specifically, the random statistics module will store N first probability values output from the logic classification module. In this example, N = 10. A danger coefficient D is set in the random statistics module. In this example, D = 0.8. The random statistics module takes the average value: D’=(0.84 + 0.73 + 0.86 + 0.87 + 0.87 + 0.63 + 0.82 + 0.77 + 0.84 + 0.86) / 10 = 0.809, D’> D, and the data of the random statistics module will enter the trigger module.

[0111] Step S14: The trigger module determines whether the dangerous driving reporting condition is met according to the dangerous driving probability value and the correction value input by the timing and counting module. When the dangerous driving reporting condition is met, the reporting information is sent to the highway supervision department.

[0112] Specifically, the calculated value C’ of the trigger module is 0.82 > C0, which meets the dangerous driving reporting condition. Therefore, the in-vehicle device defines the driving behavior of this vehicle as dangerous driving and reports the dangerous driving data of the vehicle and the location of the vehicle to the highway supervision department.

[0113] As Figure 4 shown, taking the example of the vehicle driver performing 2 forward hard acceleration speeding driving behaviors, another detailed process of the driving behavior monitoring method according to the embodiment of the present invention is:

[0114] Step S21: The vehicle condition information perception module detects the driving behavior data of each forward hard acceleration and quantifies and outputs the data to the logic classification module;

[0115] Specifically, the vehicle condition information perception module detects the forward acceleration AF each time, and at the same time calculates the speed VF after 3 seconds of each hard acceleration, as well as other data. It should be noted that this other data refers to the data required for the operation of Formula 1. It should be noted that data quantization refers to normalizing data at different data levels to obtain data with the same magnitude.

[0116] Step S22: The logic classification module performs calculations and outputs the first probability value of each forward rapid acceleration driving behavior to the random statistics module;

[0117] Specifically, for example, for 2 speeding behaviors, the first probability values output by the logic classification module to the random statistics module are: 0.84, 0.83. At the same time, the first rapid acceleration driving behavior will trigger the timing and counting module to start working, start timing and counting; specifically, timing is to record the interval between two rapid acceleration driving behaviors, and counting is to record the number of dangerous driving behaviors within a specified time.

[0118] Step S23: The random statistics module stores N first probability values output from the logic classification module, obtains the dangerous driving probability value, and when the dangerous driving probability value is greater than the danger coefficient, inputs the dangerous driving probability value into the trigger module;

[0119] Specifically, the random statistics module will store N first probability values output from the logic classification module. In this example, N = 2. A danger coefficient D is set in the random statistics module. In this example, D = 0.8. The random statistics module takes the average value: D’=(0.84 + 0.83) / 2 = 0.835, D’>D, and the data of the random statistics module will enter the trigger module.

[0120] Step S24: The trigger module determines whether the dangerous driving reporting condition is met based on the dangerous driving probability value and the correction value input by the timing and counting module. When the dangerous driving reporting condition is not met, the data of the trigger module enters the silent module;

[0121] Specifically, the calculated value of the trigger module is C’ = 0.72 < C0 = 0.8, that is, the vehicle driver only performs two occasional overtaking behaviors, the interval time between the two is long, and the number of times is small. Therefore, the on-vehicle device does not define this behavior as a dangerous driving behavior, and the data will enter the silent module.

[0122] Step S25: The silent module processes the data and then enters the logic classification module to enable the logic classification module to perform self-learning;

[0123] It should be noted that the silent module mainly performs normalization processing on the data input by the trigger module, so that the processed data can meet the requirements for self-learning and adjustment of the logic classification module.

[0124] Specifically, what the logic classification module performs self-learning and adjustment on is the probability calculation coefficient used for formula one calculation; in this way, the on-vehicle device will reduce the detection sensitivity of the driver of this vehicle next time.

[0125] In summary, the embodiments of the present invention can achieve the following beneficial effects:

[0126] 1. The hardware cost is extremely low. Only two-axis acceleration statistics and data communication are required to digitalize the driving behavior of the driver, and it can be quantified into standard data for statistical machine learning to output and give early warnings for dangerous driving behaviors.

[0127] 2. It can monitor the whole process and accurately determine the dangerous behaviors of the vehicle driver.

[0128] As Figure 5 shown, an embodiment of the present invention further provides a driving behavior monitoring device 50, which is applied to an in-vehicle device and includes:

[0129] A monitoring module 51, configured to monitor the first driving behavior of the first vehicle within a preset time period, where the absolute value of the acceleration of the first vehicle in the first driving behavior is greater than or equal to a first preset value;

[0130] An acquisition module 52, configured to acquire the driving state information of at least two first driving behaviors of the first vehicle;

[0131] A determination module 53, configured to determine the dangerous driving probability value of the first vehicle according to the driving state information of the at least two first driving behaviors;

[0132] A judgment module 54, configured to judge whether the first vehicle meets the dangerous driving reporting condition according to the dangerous driving probability value;

[0133] A sending module 55, configured to send a reporting message to the first device if the first vehicle meets the dangerous driving reporting condition.

[0134] Optionally, the determination module 53 includes:

[0135] A first determination unit, configured to determine the first probability value of each first driving behavior of the first vehicle according to the driving state information of the at least two first driving behaviors and in combination with a probability calculation coefficient;

[0136] A second determination unit, configured to determine the dangerous driving probability value of the first vehicle according to the first probability value.

[0137] Optionally, the first determination unit is configured to:

[0138] According to the formula: φ(z) = 1 / (1 + e -z ), obtain the first probability value of each driving behavior;

[0139] Among them, φ(z) is the first probability value corresponding to the first driving behavior; z = θ0AF + θ1AB + θ2AR + θ3AL + θ4VF + θ5VB + θ6AD, where AF is the forward acceleration, AB is the backward acceleration, AL is the leftward acceleration, AR is the rightward acceleration, VF is the instantaneous forward speed, VB is the instantaneous backward speed, and AD is the probability calculation coefficient; θ0-θ6 are the regression coefficients of the regression classification model.

[0140] Optionally, after the monitoring module 51 monitors the first driving behavior of the first vehicle within a preset time period, the driving behavior monitoring device further includes:

[0141] A recording module, configured to record the number of times of the first driving behavior within the preset time period and the time interval between every two first driving behaviors after the first first driving behavior within the preset time period is monitored.

[0142] Optionally, the judgment module 54 includes:

[0143] A third determination unit, configured to determine a correction value of the dangerous driving probability according to the number of times of the first driving behavior and the time interval between every two first driving behaviors;

[0144] A fourth determination unit, configured to determine a comprehensive judgment coefficient according to the correction value and the dangerous driving probability value;

[0145] A judgment unit, configured to judge whether the first vehicle meets the dangerous driving reporting condition according to the comprehensive judgment coefficient.

[0146] Optionally, the judgment unit is configured to:

[0147] If the comprehensive judgment coefficient is greater than or equal to a second preset value, it is determined that the first vehicle meets the dangerous driving reporting condition;

[0148] If the comprehensive judgment coefficient is less than the second preset value, it is determined that the first vehicle does not meet the dangerous driving reporting condition.

[0149] Optionally, after the judgment module 54 judges whether the first vehicle meets the dangerous driving reporting condition according to the dangerous driving probability value, it further includes:

[0150] An adjustment module, configured to adjust the probability calculation coefficient by using the comprehensive judgment coefficient if the first vehicle does not meet the dangerous driving reporting condition.

[0151] Optionally, the driving state information includes at least one of the following:

[0152] Forward acceleration, backward acceleration, leftward acceleration, rightward acceleration, forward instantaneous velocity, backward instantaneous velocity, leftward instantaneous velocity, and rightward instantaneous velocity.

[0153] Optionally, the reported information includes: the dangerous driving data of the first vehicle and the position of the first vehicle.

[0154] Optionally, the position of the first vehicle includes: the position where the first vehicle is located each time the first driving behavior occurs.

[0155] It should be noted here that although the module division method in the embodiment of the driving behavior monitoring device is different from that in the above embodiment, both can implement the driving behavior monitoring method.

[0156] It should be noted that the embodiment of the driving behavior monitoring device is a device corresponding one by one to the above method embodiment. All implementation manners in the above method embodiment are applicable to the embodiment of this device and can also achieve the same technical effects.

[0157] As Figure 6 shown, an embodiment of the present invention further provides an in-vehicle device 60, including a processor 61, a transceiver 62, a memory 63, and a program stored in the memory 63 and executable on the processor 61; wherein, the transceiver 62 is connected to the processor 61 and the memory 63 through a bus interface, and wherein, the processor 61 is configured to read the program in the memory and execute the following processes:

[0158] Monitor the first driving behavior of the first vehicle within a preset time period, where the absolute value of the acceleration of the first vehicle in the first driving behavior is greater than or equal to a first preset value;

[0159] Obtain the driving state information of at least two first driving behaviors of the first vehicle;

[0160] Determine the dangerous driving probability value of the first vehicle according to the driving state information of the at least two first driving behaviors;

[0161] Judge whether the first vehicle meets the dangerous driving reporting condition according to the dangerous driving probability value;

[0162] If the first vehicle meets the dangerous driving reporting condition, send the reported information to the first device.

[0163] It should be noted that in Figure 6Among them, the bus architecture may include any number of interconnected buses and bridges, and various circuits represented by one or more processors represented by processor 61 and a memory represented by memory 63 are specifically linked together. The bus architecture can also link together various other circuits such as peripheral devices, voltage regulators, and power management circuits, which are well known in the art, and thus will not be further described herein. The bus interface provides an interface. The transceiver 62 may be multiple elements, that is, including a transmitter and a transceiver, and provides a unit for communicating with various other devices on the transmission medium. For different terminals, the user interface 64 may also be an interface capable of externally connecting and internally connecting required devices, and the connected devices include but are not limited to a keypad, a display, a speaker, a microphone, a joystick, etc. The processor 61 is responsible for managing the bus architecture and general processing, and the memory 63 can store the data used by the processor 61 when performing operations.

[0164] Optionally, when the processor executes the program for determining the dangerous driving probability value of the first vehicle based on the driving state information of the at least two first driving behaviors, the following steps are implemented:

[0165] Based on the driving state information of the at least two first driving behaviors and in combination with a probability calculation coefficient, determine the first probability value of each first driving behavior of the first vehicle;

[0166] Based on the first probability value, determine the dangerous driving probability value of the first vehicle.

[0167] Optionally, when the processor executes the program for determining the first probability value of each first driving behavior of the first vehicle, the following steps are implemented:

[0168] According to the formula: φ(z) = 1 / (1 + e -z ), obtain the first probability value of each driving behavior;

[0169] where φ(z) is the first probability value corresponding to a first driving behavior; z = θ0AF + θ1AB + θ2AR + θ3AL + θ4VF + θ5VB + θ6AD, AF is the forward acceleration, AB is the backward acceleration, AL is the leftward acceleration, AR is the rightward acceleration, VF is the forward instantaneous speed, VB is the backward instantaneous speed, AD is the probability calculation coefficient; θ0 - θ6 are the regression coefficients of the regression classification model.

[0170] Optionally, after the processor executes the program for monitoring the first driving behavior of the first vehicle within a preset time period, the following steps are also implemented when the processor executes the program:

[0171] After detecting the first first driving behavior within the preset time period, record the number of first driving behaviors within the preset time period, and record the time interval between every two first driving behaviors.

[0172] Optionally, when the processor executes the program of determining whether the first vehicle meets the dangerous driving reporting condition according to the dangerous driving probability value, the following steps are implemented:

[0173] Determine a correction value of the dangerous driving probability according to the number of first driving behaviors and the time interval between every two first driving behaviors;

[0174] Determine a comprehensive judgment coefficient according to the correction value and the dangerous driving probability value;

[0175] Judge whether the first vehicle meets the dangerous driving reporting condition according to the comprehensive judgment coefficient.

[0176] Optionally, when the processor executes the program of determining whether the first vehicle meets the dangerous driving reporting condition according to the comprehensive judgment coefficient, the following steps are implemented:

[0177] If the comprehensive judgment coefficient is greater than or equal to a second preset value, determine that the first vehicle meets the dangerous driving reporting condition;

[0178] If the comprehensive judgment coefficient is less than the second preset value, determine that the first vehicle does not meet the dangerous driving reporting condition.

[0179] Optionally, after the processor executes the program of determining whether the first vehicle meets the dangerous driving reporting condition according to the dangerous driving probability value, the processor further implements the following steps:

[0180] If the first vehicle does not meet the dangerous driving reporting condition, adjust the probability calculation coefficient by using the comprehensive judgment coefficient.

[0181] Optionally, the driving state information includes at least one of the following:

[0182] Forward acceleration, backward acceleration, leftward acceleration, rightward acceleration, forward instantaneous speed, backward instantaneous speed, leftward instantaneous speed, and rightward instantaneous speed.

[0183] Optionally, the reported information includes: dangerous driving data of the first vehicle and the position of the first vehicle.

[0184] Optionally, the position of the first vehicle includes: the position where the first vehicle is located when each first driving behavior occurs.

[0185] An embodiment of the present invention also provides a readable storage medium, on which a computer program is stored. When the program is executed by a processor, the steps of the driving behavior monitoring method applied to a vehicle-mounted device are implemented.

[0186] Another embodiment of the present invention provides a chip, which includes a processor and a communication interface. The communication interface is coupled to the processor. The processor is used to run a program or an instruction to implement each process of the above-mentioned embodiment of the driving behavior monitoring method and can achieve the same technical effect. To avoid repetition, it will not be elaborated here.

[0187] It should be understood that the chip mentioned in the embodiments of the present application can also be referred to as a system-on-chip, system chip, chip system, or system-on-chip.

[0188] It should be noted that in this article, the term "comprising", "including" or any other variant thereof is intended to cover a non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements, but also includes other elements not expressly listed, or also includes elements inherent to such process, method, article or device. Without further limitation, an element defined by the statement "including one..." does not exclude the existence of another identical element in the process, method, article or device including the element. In addition, it should be pointed out that the methods and devices in the embodiments of the present application are not limited to performing functions in the order shown or discussed, and may also include performing functions in a substantially simultaneous manner or in a reverse order according to the functions involved. For example, the described methods may be performed in an order different from that described, and various steps may be added, omitted, or combined. In addition, the features described with reference to certain examples may be combined in other examples.

[0189] Through the description of the above embodiments, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation. Based on such an understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. The computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions for causing a terminal (which can be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in the various embodiments of the present application.

[0190] The above are the preferred embodiments of the present invention. It should be noted that for those of ordinary skill in the art, several improvements and refinements can be made without departing from the principle described in the present invention, and these improvements and refinements are also within the protection scope of the present invention.

Claims

1. A driving behavior monitoring method, characterized in that, Executed by a vehicle-mounted device, including: Monitoring the first driving behavior of the first vehicle within a preset time period, where the absolute value of the acceleration of the first vehicle in the first driving behavior is greater than or equal to a first preset value; Obtaining the driving state information of at least two first driving behaviors of the first vehicle; Determining the dangerous driving probability value of the first vehicle according to the driving state information of the at least two first driving behaviors; Judging whether the first vehicle meets the dangerous driving reporting condition according to the dangerous driving probability value; If the first vehicle meets the dangerous driving reporting condition, sending a reporting message to the first device, where the first device is used to send the dangerous driving behavior of the first vehicle to other vehicles; Wherein, after monitoring the first driving behavior of the first vehicle within the preset time period, the driving behavior monitoring method further includes: After detecting the first first driving behavior within the preset time period, recording the number of first driving behaviors within the preset time period and the time interval between every two first driving behaviors; Wherein, judging whether the first vehicle meets the dangerous driving reporting condition according to the dangerous driving probability value includes: Determining a correction value of the dangerous driving probability according to the number of first driving behaviors and the time interval between every two first driving behaviors; Determining a comprehensive judgment coefficient according to the correction value and the dangerous driving probability value; Judging whether the first vehicle meets the dangerous driving reporting condition according to the comprehensive judgment coefficient.

2. The driving behavior monitoring method according to claim 1, wherein Determining the dangerous driving probability value of the first vehicle according to the driving state information of the at least two first driving behaviors includes: Determining the first probability value of each first driving behavior of the first vehicle according to the driving state information of the at least two first driving behaviors and in combination with a probability calculation coefficient; Determining the dangerous driving probability value of the first vehicle according to the first probability value.

3. The driving behavior monitoring method according to claim 2, characterized in that Determining the first probability value of each first driving behavior of the first vehicle includes: According to the formula: φ(z) = 1 / (1 + e -z ), obtain the first probability value of each driving behavior; Wherein, φ(z) is the first probability value corresponding to a first driving behavior; z = θ0AF + θ1AB + θ2AR + θ3AL + θ4VF + θ5VB + θ6AD, AF is the forward acceleration, AB is the backward acceleration, AL is the leftward acceleration, AR is the rightward acceleration, VF is the forward instantaneous speed, VB is the backward instantaneous speed, AD is the probability calculation coefficient; θ0-θ6 are the regression coefficients of the regression classification model.

4. The driving behavior monitoring method according to claim 1, wherein Judging whether the first vehicle meets the dangerous driving reporting condition according to the comprehensive judgment coefficient includes: If the comprehensive judgment coefficient is greater than or equal to a second preset value, determining that the first vehicle meets the dangerous driving reporting condition; If the comprehensive judgment coefficient is less than the second preset value, determining that the first vehicle does not meet the dangerous driving reporting condition.

5. The driving behavior monitoring method according to claim 4, characterized in that, After judging whether the first vehicle meets the dangerous driving reporting condition according to the dangerous driving probability value, it further includes: If the first vehicle does not meet the dangerous driving reporting condition, adjusting the probability calculation coefficient by using the comprehensive judgment coefficient.

6. The driving behavior monitoring method according to claim 1, characterized in that, The driving state information includes at least one of the following: Forward acceleration, backward acceleration, leftward acceleration, rightward acceleration, forward instantaneous velocity, backward instantaneous velocity, leftward instantaneous velocity, and rightward instantaneous velocity.

7. The driving behavior monitoring method according to claim 1, wherein The reported information includes: the dangerous driving data of the first vehicle and the position of the first vehicle.

8. The driving behavior monitoring method according to claim 7, wherein The position of the first vehicle includes: the position where the first vehicle is located when each first driving behavior occurs.

9. A vehicle-mounted device, characterized in that, Includes: A processor, a memory, and a computer program stored on the memory and executable on the processor. When the computer program is executed by the processor, it implements the steps of the driving behavior monitoring method according to any one of claims 1 to 8.

10. A driving behavior monitoring device, applied to in-vehicle equipment, characterized in that, Includes: A monitoring module for monitoring the first driving behavior of the first vehicle within a preset time period, where the absolute value of the acceleration of the first vehicle in the first driving behavior is greater than or equal to a first preset value; An acquisition module for acquiring the driving state information of at least two first driving behaviors of the first vehicle; A determination module for determining the dangerous driving probability value of the first vehicle according to the driving state information of the at least two first driving behaviors; A judgment module for judging whether the first vehicle meets the dangerous driving reporting condition according to the dangerous driving probability value; A sending module for sending the reported information to the first device if the first vehicle meets the dangerous driving reporting condition, where the first device is used to send the dangerous driving behavior of the first vehicle to other vehicles; Wherein, after the monitoring module monitors the first driving behavior of the first vehicle within a preset time period, the driving behavior monitoring device further includes: A recording module for recording the number of first driving behaviors within the preset time period and the time interval between every two first driving behaviors after monitoring the first first driving behavior within the preset time period; Wherein, the judgment module includes: A third determination unit for determining the correction value of the dangerous driving probability according to the number of first driving behaviors and the time interval between every two first driving behaviors; A fourth determination unit for determining the comprehensive judgment coefficient according to the correction value and the dangerous driving probability value; A judgment unit for judging whether the first vehicle meets the dangerous driving reporting condition according to the comprehensive judgment coefficient.

11. A readable storage medium, characterized in that, A computer program is stored on the readable storage medium. When the computer program is executed by the processor, it implements the steps of the driving behavior monitoring method according to any one of claims 1 to 8.

Images