Drunk driving vehicle discrimination system and method
By detecting alcohol concentration, heart rate and speed in real time in the vehicle, using the drunk driving judgment algorithm to judge the drunk driving vehicle, and warning and reporting it to the traffic management department on the on-board display, the problem of low detection accuracy and inability to cover all the coverage in the existing technology is solved. Reduce the incidence of drunk driving accidents and improve road safety.
Patent Information
- Application Number
- CN202411155915.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-22
- Publication Date
- 2024-11-08
AI Technical Summary
The existing drunk driving testing technology has the problems of low detection accuracy, time-consuming and labor-intensive testing, and the inability to achieve full coverage testing, resulting in the still high incidence of drunk driving accidents.
A drunk driving vehicle judgment system is designed, including a detection module, a microcontroller and a drunk driving judgment module. By real-time detection of alcohol concentration, driver's heart rate and vehicle speed in the vehicle, the drunk driving judgment algorithm is used to analyze data to determine whether it is a drunk driving vehicle, and Warn the driver on the on-board monitor and report it to the traffic management department.
Effectively detect and warn of drunk driving behavior, reduce the probability of traffic accidents caused by drunk driving, improve road safety, and achieve timely prevention of drunk driving behavior.
Smart Images

Figure CN118907117A_ABST
Abstract
Description
Technical Field
[0001] The invention relates to a system and method for distinguishing a drunk driving vehicle, and belongs to the technical field of drunk driving detection. Background Art
[0002] When driving after drinking, the driver's reaction ability is reduced due to the stimulation and anesthetic effect of alcohol, and the control over the vehicle's throttle, brake and steering systems is reduced. The driver's judgment, analysis and operation abilities are significantly reduced, which may cause traffic accidents. According to statistics, there are tens of thousands of traffic accidents caused by drunk driving every year, and more than 50% of fatal accidents are related to drunk driving. Drunk driving has been listed as the main cause of death in traffic accidents.
[0003] The existing technologies for detecting drunk driving mainly include breath testing and blood test. First, breath testing requires traffic control personnel to conduct long-term, wide-range inspections on a large number of vehicles on the road. This form of inspection will waste a lot of manpower, material resources and time. In addition, there are disadvantages such as low detection accuracy and the lack of control over the person being tested due to the effect of alcohol. Second, although the blood test has high detection accuracy, it does not have specific real-time detection and has cumbersome detection procedures. These detection methods are time-consuming and labor-intensive, and the detection range is relatively small. It cannot achieve full coverage detection, resulting in a high incidence of drunk driving accidents.
[0004] Existing technologies for preventing drunk driving mostly include mandatory locking of vehicles and mandatory speed limits. The disadvantage is that the mandatory prevention measures cause inconvenience in people's lives and cannot fundamentally solve the problem of drunk driving by drivers. Summary of the invention
[0005] In order to solve the above problems, the present invention proposes a drunk driving vehicle identification system and method, which can effectively detect drunk driving behavior and promptly warn the drunk driving driver.
[0006] The technical solution adopted by the present invention to solve the technical problem is:
[0007] In a first aspect, an embodiment of the present invention provides a drunk driving vehicle identification system, including a detection module, a microcontroller, a communication module and a drunk driving identification module;
[0008] The detection module detects the real-time alcohol concentration in the vehicle cab (driver's surroundings), the driver's real-time heart rate and the vehicle's real-time speed, and sends the detection data to the microcontroller;
[0009] The microcontroller sends the detection data to the drunk driving identification module through the communication module;
[0010] The drunk driving identification module uses a drunk driving identification algorithm to analyze the detection data to determine whether the vehicle is a drunk driving vehicle, and if so, sends the drunk driving behavior information of the driver of the drunk driving vehicle to the microcontroller through the communication module;
[0011] The microcontroller sends the drunk driving behavior information of the drunk driving vehicle driver to the vehicle-mounted display for warning.
[0012] As a possible implementation of this embodiment, the detection module includes an alcohol sensor, a heart rate detector and a vehicle-mounted speedometer. The alcohol sensor and heart rate detector are installed in front of the driver's side in the cab, and are used to detect the real-time alcohol concentration in the vehicle cab and the driver's real-time heart rate; the vehicle-mounted speedometer includes an instrument in the vehicle's existing system for detecting the vehicle's real-time speed, and is used to detect the vehicle's real-time speed.
[0013] As a possible implementation of this embodiment, the detection module further includes a pressure sensor, which is installed directly below the main driving seat of the vehicle and is used to detect whether the driver is sitting in the main driving seat.
[0014] As a possible implementation of this embodiment, a drunk driving vehicle identification system also includes a GPS module, and the GPS module is used to collect real-time location information of the vehicle.
[0015] As a possible implementation of this embodiment, the drunk driving identification module uses a drunk driving identification algorithm to analyze the detection data, and the specific process of determining whether the vehicle is a drunk driving vehicle is as follows:
[0016] In a period of time T n-m The real-time alcohol concentration in the vehicle driving room, the real-time heart rate of the driver and the real-time speed of the vehicle are imported into the database for storage, and the ε group of imported data is selected from the database as the data set:
[0017] A i =(X i ,Y i ,Z i )
[0018] Among them, A is the total set of various data detected by the detection module, X is the real-time alcohol concentration in the cab, Y is the real-time heart rate of the vehicle driver, and Z is the real-time speed of the vehicle;
[0019] Establish the data subset required for classification and convert data set A iThe data are divided into a drunk driving vehicle set B1, a normal vehicle set B2, and a normal vehicle set B3 after being warned, and the centroids of the three data sets B1, B2, and B3 are determined respectively, and drunk driving is judged based on the centroids. The basis for determining the centroids of the three data sets B1, B2, and B3 includes: (1) the real-time alcohol concentration X in the cab is judged based on the GB19522 standard "Blood and breath alcohol content threshold and test for vehicle drivers", (2) the real-time heart rate Y of the vehicle driver is judged based on the heart rate of the drunk driver>100 times / min, and the frequency band is 1.70Hz-2.50Hz; (3) the real-time speed Z of the vehicle is judged as 0m / s, and the drunk driving vehicle includes drunk driving vehicles and drunk driving vehicles.
[0020] As a possible implementation of this embodiment, the specific process of judging drunk driving based on the centroid is as follows:
[0021] The three data sets B1, B2, and B3 are B1(Xa, Ya, Za), B2(Xb, Yb, Zb), and B3(Xc, Yc, Zc), respectively. The centroid of the three data sets B1, B2, and B3 is B(X η ,Y η ,Z η ), according to the Euclidean distance method, we get the following formula:
[0022]
[0023] In the formula, m represents the initial detection time point, n represents the final detection time point, and x n -x m Represents a period of time T n-m The difference of real-time alcohol concentration in the driving cabin, y n -y m Represents a period of time T n-m The difference in the real-time heart rate of the drivers in the vehicle, z n -z m Represents a period of time T n-m The difference in vehicle speed, X i ,Y i ,Z i They represent the real-time alcohol concentration in the driving room, the real-time heart rate of the driver and the vehicle speed data respectively. η ,Y η ,Z η Represents the specific value of the center of mass set in advance based on the judgment range and threshold of the real-time alcohol concentration in the cab, the real-time heart rate of the vehicle driver and the vehicle speed data;
[0024] Select five data sets A1 (X1, Y1, Z1), A2 (X2, Y2, Z2), A3 (X3, Y3, Z3), A4 (X4, Y4, Z4) and A5 (X5, Y5, Z5) from data set A, and calculate the distances between Ai and the centroids of B1, B2 and B3 according to the three-dimensional Euclidean distance method (2), i = 1, 2...5;
[0025] Classify all Ai as follows: ΔAi1=|Ai-B1|, ΔAi2=|Ai-B2|, ΔAi3=|Ai-B3|. After obtaining ΔAi, determine the values of ΔAi1, ΔAi2 and ΔAi3 respectively. Assume that ΔAi2 has the smallest value, that is, ΔAi2 is closest to point B2 in the normal vehicle set. Under normal circumstances, point Ai is classified as part of the normal vehicle set B2, and point Ai is determined to be a normal vehicle.
[0026] As a possible implementation of this embodiment, the specific process of performing drunk driving identification based on the centroid also includes:
[0027] Set the real-time alcohol concentration X in the driving cabin as the "first determining factor", the real-time heart rate Y of the vehicle driver as the "second determining factor", and the vehicle speed Z as the "auxiliary factor". After the classification analysis is completed, the classification situation will be corrected;
[0028] According to the Euclidean distance method (2), we can get n-m The changes of the internal data points in the spatial coordinate system are used to determine the specific situation of the vehicle driver and make another drunk driving judgment.
[0029] In a second aspect, an embodiment of the present invention provides a method for distinguishing a drunk driving vehicle, comprising the following steps:
[0030] Step 1, using a pressure sensor to detect whether the driver is sitting in the main driving seat, if yes, proceed to step 2, otherwise continue to detect whether the driver is sitting in the main driving seat;
[0031] Step 2, using an alcohol sensor and a heart rate detector to detect the real-time alcohol concentration in the vehicle cab and the real-time heart rate of the driver, while the on-board speedometer detects the real-time speed of the vehicle, and the microcontroller transmits the above detection data to the background server through the communication module;
[0032] Step 3: The backend server stores the detection data in the database and uses the drunk driving identification algorithm to analyze the detection data. If the vehicle is identified as a drunk driving vehicle, it proceeds to step 4, otherwise it returns to step 2;
[0033] Step 4, sending the drunk driving behavior information of the drunk driving vehicle driver to the microcontroller through the communication module, and the microcontroller sends the drunk driving behavior information of the drunk driving vehicle driver to the vehicle display screen to play a warning screen to warn the driver of drunk driving behavior, and use the real-time data of the vehicle speedometer to determine whether the driver makes corrections after the warning, if no corrections are made, proceed to step 5, otherwise return to step 2;
[0034] Step 5, using the communication module to report the drunk driving behavior of the drunk driving driver who continues to drive after being warned and the real-time vehicle information to the traffic control and traffic police departments, wherein the real-time vehicle information includes vehicle location information and vehicle speed information.
[0035] As a possible implementation of this embodiment, the process of analyzing the detection data using the drunk driving discrimination algorithm includes the following steps:
[0036] Step 31, after a period of time T m-n The real-time alcohol concentration in the vehicle driving room, the real-time heart rate of the driver and the real-time speed of the vehicle are imported into the database for storage, and the ε group of imported data is selected from the database as the data set:
[0037] A i =(X i ,Y i ,Z i )
[0038] Among them, A is the total set of various data detected by the detection module, X is the real-time alcohol concentration in the cab, Y is the real-time heart rate of the vehicle driver, and Z is the real-time speed of the vehicle;
[0039] Step 32: Create a data subset required for classification. i The data are divided into a drunk driving vehicle set B1, a normal vehicle set B2, and a normal vehicle set B3 after being warned, and the centroids of the three data sets B1, B2, and B3 are determined respectively. The centroids of the three data sets B1, B2, and B3 are determined based on: (1) the real-time alcohol concentration X in the cab is determined based on the GB19522 standard "Blood and breath alcohol content threshold and test for vehicle drivers", (2) the real-time heart rate Y of the vehicle driver is determined based on the heart rate of the drunk driver being >100 beats / min and the frequency band being 1.70Hz-2.50Hz; (3) the real-time speed Z of the vehicle is determined as 0m / s. The drunk driving vehicle includes drunk driving vehicles and drunk driving vehicles;
[0040] Step 33, the three data sets B1, B2, and B3 are B1(Xa, Ya, Za), B2(Xb, Yb, Zb), and B3(Xc, Yc, Zc), respectively. The centroid of the three data sets B1, B2, and B3 is B(X η ,Y η,Z η ), according to the Euclidean distance method, we get the following formula:
[0041]
[0042]
[0043] In the formula, m represents the initial detection time point, n represents the final detection time point, and x n -x m Represents a period of time T n-m The difference of real-time alcohol concentration in the driving cabin, y n -y m Represents a period of time T n-m The difference in the real-time heart rate of the drivers in the vehicle, z n -z m Represents a period of time T n-m The difference in vehicle speed, X i ,Y i ,Z i They represent the real-time alcohol concentration in the driving room, the real-time heart rate of the driver and the vehicle speed data respectively. η ,Y η ,Z η Represents the specific value of the center of mass set in advance based on the judgment range and threshold of the real-time alcohol concentration in the cab, the real-time heart rate of the vehicle driver and the vehicle speed data;
[0044] Step 34, select five data sets A1 (X1, Y1, Z1), A2 (X2, Y2, Z2), A3 (X3, Y3, Z3), A4 (X4, Y4, Z4) and A5 (X5, Y5, Z5) from the data set A, and calculate the distances between Ai and the three centroids of B1, B2 and B3 respectively according to the three-dimensional Euclidean distance method (2), i = 1, 2...5;
[0045] Step 35, classify all Ai: ΔAi1 = |Ai-B1|, ΔAi2 = |Ai-B2|, ΔAi3 = |Ai-B3|. After obtaining ΔAi, determine the values of ΔAi1, ΔAi2 and ΔAi3 respectively. Assume that ΔAi2 has the smallest value, that is, ΔAi2 is closest to point B2 of the normal vehicle set. Under normal circumstances, point Ai is classified as the normal vehicle set B2, and point Ai is determined to be a normal vehicle, thereby completing the drunk driving judgment.
[0046] As a possible implementation of this embodiment, in step 4, the process of analyzing the detection data using the drunk driving discrimination algorithm further includes the following steps:
[0047] Step 36, setting the real-time alcohol concentration X in the driving room as the "first determining factor", the real-time heart rate Y of the vehicle driver as the "second determining factor", and the vehicle speed Z as the "auxiliary factor", after the classification analysis is completed, the classification situation will be corrected;
[0048] Step 37, according to the Euclidean distance method (2), obtain the n-m The changes of the internal data points in the spatial coordinate system are used to determine the specific situation of the vehicle driver and make another drunk driving judgment.
[0049] The technical solution of the embodiment of the present invention may have the following beneficial effects:
[0050] The present invention provides a drunk driving vehicle identification system that detects relevant parameters of the vehicle and the driver and processes and analyzes the detection data to further identify the drunk driving vehicle. The drunk driving behavior is prevented by warning the drunk driving driver and reporting the real-time location of the vehicle to the traffic management department, thereby reducing the probability of traffic accidents caused by drunk driving and improving road safety.
[0051] A drunk driving vehicle identification method of the present invention is based on a drunk driving identification algorithm, which determines whether the vehicle is a drunk driving vehicle by analyzing the status of the vehicle and the driver over a period of time, including the driver's heart rate, the alcohol concentration in the vehicle environment and the vehicle speed, and then warns the driver and reports the vehicle status to the traffic control department, forming a double insurance mechanism.
[0052] The present invention can effectively detect drunk driving behavior, solves the shortcomings of the existing technology, can not only subjectively remind and warn drunk driving drivers in a timely manner, but also provide real-time information and basic guarantees for traffic management departments, so that preventive measures can be taken in time, avoid the occurrence of drunk driving behavior, and improve road travel safety. BRIEF DESCRIPTION OF THE DRAWINGS
[0053] Figure 1 is a structural schematic diagram of a drunk driving vehicle identification system according to an exemplary embodiment;
[0054] Figure 2 is a logic diagram of elements in a drunk driving discrimination algorithm setting according to an exemplary embodiment;
[0055] Figure 3 This is a schematic diagram showing the installation distribution of various modules of a drunk driving vehicle identification system in a vehicle according to an exemplary embodiment;
[0056] Figure 4 The present invention is a flow chart of a method for distinguishing a drunk driving vehicle according to an exemplary embodiment. DETAILED DESCRIPTION
[0057] The present invention will be further described below in conjunction with the accompanying drawings and embodiments:
[0058] In order to clearly illustrate the technical features of the present solution, the present invention is described in detail below through specific implementation methods and in conjunction with the accompanying drawings. The disclosure below provides many different embodiments or examples for realizing different structures of the present invention. In order to simplify the disclosure of the present invention, the components and settings of specific examples are described below. In addition, the present invention may repeat reference numbers and / or letters in different examples. This repetition is for the purpose of simplification and clarity, and does not itself indicate the relationship between the various embodiments and / or settings discussed. It should be noted that the components illustrated in the accompanying drawings are not necessarily drawn to scale. The present invention omits the description of known components and processing techniques and processes to avoid unnecessary limitations on the present invention.
[0059] like Figure 1 As shown, a drunk driving vehicle identification system provided by an embodiment of the present invention includes a detection module, a microcontroller, a communication module and a drunk driving identification module;
[0060] The detection module detects the real-time alcohol concentration in the vehicle driving room, the real-time heart rate of the driver and the real-time speed of the vehicle, and sends the detection data to the microcontroller;
[0061] The microcontroller sends the detection data to the drunk driving identification module through the communication module;
[0062] The drunk driving identification module uses a drunk driving identification algorithm to analyze the detection data to determine whether the vehicle is a drunk driving vehicle, and if so, sends the drunk driving behavior information of the driver of the drunk driving vehicle to the microcontroller through the communication module;
[0063] The microcontroller sends the drunk driving behavior information of the drunk driving vehicle driver to the vehicle on-board display for warning. The on-board display is a display in the existing system of the vehicle, which is connected to both the vehicle system and the drunk driving vehicle identification system.
[0064] As a possible implementation of this embodiment, the detection module includes an alcohol sensor, a heart rate detector and a vehicle-mounted speedometer. The alcohol sensor and heart rate detector are installed in front of the driver's side in the cab, and are used to detect the real-time alcohol concentration in the vehicle cab and the driver's real-time heart rate; the vehicle-mounted speedometer includes an instrument in the vehicle's existing system for detecting the vehicle's real-time speed, and is used to detect the vehicle's real-time speed.
[0065] The alcohol sensor is an electrochemical alcohol tester. There is a catalyst inside it that can oxidize alcohol into carbon dioxide (acetaldehyde) and release chemical energy through water. The chemical energy is converted into electrical energy and then converted into current by a voltmeter to obtain the real-time alcohol concentration in the car.
[0066] The heart rate detector will collect the driver's vital signs, heart rate signals, and pulse pressure signals when he is sitting, and analyze the pressure signals to finally obtain the heartbeat signal. Under normal circumstances, the heartbeat of a normal adult after drinking will be higher than 100 beats per minute, so the frequency of the heartbeat signal should be higher than 1.67Hz. In the present invention, in order to reduce the misjudgment rate, the heartbeat frequency after drinking is set to be higher than 1.70Hz, and a filtered signal is obtained through a Butterworth filter, and the selected frequency band is 1.70Hz-2.50Hz.
[0067] As a possible implementation of this embodiment, the detection module further includes a pressure sensor, which is installed directly below the main driving seat of the vehicle and is used to detect whether the driver is sitting in the main driving seat. The pressure sensor is a semiconductor piezoresistive intake pressure sensor, which is made using the piezoresistive effect of a semiconductor.
[0068] As a possible implementation of this embodiment, a drunk driving vehicle identification system also includes a GPS module, and the GPS module is used to collect real-time location information of the vehicle.
[0069] As a possible implementation manner of this embodiment, the communication module is a NB-IoT module.
[0070] As a possible implementation of this embodiment, the microcontroller establishes a connection with each module, and the microcontroller has the following functions:
[0071] The signal that the pressure sensor recognizes that the driver has sat in the main driving seat is transmitted to the alcohol sensor and the heart rate detector, and the alcohol sensor and the heart rate detector are activated to start detecting corresponding values;
[0072] Transmitting the values detected by the alcohol sensor, the heart rate detector and the vehicle speedometer to the backend server for analysis and processing and transmitting them to the database for storage;
[0073] The signal of whether to warn the vehicle driver is transmitted to the vehicle display screen according to the result of the judgment by the background server, and the signal of whether to report the vehicle information to the traffic control department is transmitted to the communication module;
[0074] The real-time location information of the vehicle obtained by the GPS module is transmitted to a database for storage;
[0075] Establish a connection between the vehicle system and the drunk driving vehicle identification system.
[0076] As a possible implementation of this embodiment, the drunk driving identification module includes a database and a backend server, as well as a drunk driving identification algorithm installed on the backend server. The database is used to store various data of the detection module and the functional module, and the backend server is used for data processing and analysis.
[0077] As a possible implementation of this embodiment, the drunk driving identification module uses a drunk driving identification algorithm to analyze the detection data, and the specific process of determining whether the vehicle is a drunk driving vehicle is as follows:
[0078] In a period of time T n-m The real-time alcohol concentration in the vehicle driving room, the real-time heart rate of the driver and the real-time speed of the vehicle are imported into the database for storage, and the ε group of imported data is selected from the database as the data set:
[0079] A i =(X i ,Y i ,Z i )
[0080] Among them, A is the total set of various data detected by the detection module, X is the real-time alcohol concentration in the cab, Y is the real-time heart rate of the vehicle driver, and Z is the real-time speed of the vehicle;
[0081] Establish the data subset required for classification and convert data set A i The data are divided into a drunk driving vehicle set B1, a normal vehicle set B2, and a normal vehicle set B3 after being warned, and the centroids of the three data sets B1, B2, and B3 are determined respectively, and drunk driving is judged based on the centroids. The basis for determining the centroids of the three data sets B1, B2, and B3 includes: (1) the real-time alcohol concentration X in the cab is judged based on the GB19522 standard "Blood and breath alcohol content threshold and test for vehicle drivers", (2) the real-time heart rate Y of the vehicle driver is judged based on the heart rate of the drunk driver>100 times / min, and the frequency band is 1.70Hz-2.50Hz; (3) the real-time speed Z of the vehicle is judged as 0m / s, and the drunk driving vehicle includes drunk driving vehicles and drunk driving vehicles.
[0082] The determination of the center of mass is based on the judgment range and threshold of the quantities represented by X, Y, and Z. The judgment range of the real-time alcohol concentration X in the cab is set according to the GB19522 standard, as shown in Table 1. The judgment threshold of the real-time heart rate Y of the vehicle driver is based on the setting of the algorithm in the driver's heart rate detection method: the heart rate of a drunk driver is >100 times / min, and the frequency band is 1.70Hz-2.50Hz; the real-time speed Z of the vehicle is bounded by 0m / s, which is used as the standard for the system to judge whether the driver continues to drive after warning the driver.
[0083] Table 1: GB19522 standard
[0084] Driving status Alcohol concentration / (mg / 100mL) Normal driving <20 Drinking and driving ≥20,<80 Drunk driving ≥80
[0085] In the present invention, driving under the influence of alcohol and drunk driving are uniformly classified as drunk driving, and are uniformly identified as drunk driving vehicles in the identification module, and have the same prevention and control measures.
[0086] As a possible implementation of this embodiment, the specific process of judging drunk driving based on the centroid is as follows:
[0087] The three data sets B1, B2, and B3 are B1(Xa, Ya, Za), B2(Xb, Yb, Zb), and B3(Xc, Yc, Zc), respectively. The centroid of the three data sets B1, B2, and B3 is B(X η ,Y η ,Z η ), according to the Euclidean distance method, we get the following formula:
[0088]
[0089] In the formula, m represents the initial detection time point, n represents the final detection time point, and x n -x m Represents a period of time T n-m The difference of real-time alcohol concentration in the driving cabin, y n -y m Represents a period of time T n-m The difference in the real-time heart rate of the drivers in the vehicle, z n -z m Represents a period of time T n-m The difference in vehicle speed, X i ,Y i ,Z i They represent the real-time alcohol concentration in the driving room, the real-time heart rate of the driver and the vehicle speed data respectively. η ,Y η ,Z η Represents the specific value of the center of mass set in advance based on the judgment range and threshold of the real-time alcohol concentration in the cab, the real-time heart rate of the vehicle driver and the vehicle speed data;
[0090] Select five data sets A1 (X1, Y1, Z1), A2 (X2, Y2, Z2), A3 (X3, Y3, Z3), A4 (X4, Y4, Z4) and A5 (X5, Y5, Z5) from data set A, and calculate the distances between Ai and the centroids of B1, B2 and B3 according to the three-dimensional Euclidean distance method (2), i = 1, 2...5;
[0091] Classify all Ai: ΔAi1=|Ai-B1|, ΔAi2=|Ai-B2|, ΔAi3=|Ai-B3|. After obtaining ΔAi, determine the values of ΔAi1, ΔAi2 and ΔAi3 respectively. Assume that ΔAi2 has the smallest value, that is, ΔAi2 is closest to point B2 of the normal vehicle set. Under normal circumstances, point Ai is classified as the normal vehicle set B2, and point Ai is determined to be a normal vehicle. Classify all Ai according to the above classification analysis method. Taking point 1 as an example: △A11=|A1-B1|, △A12=|A1-B2|, △A13=|A1-B3|. After obtaining △Ai, each situation is classified. Similarly, taking point 1 as an example, the values of △A11, △A12 and △A13 are judged respectively. Assume that △A12 has the smallest value, that is, △A12 is closest to point B2 in the normal vehicle set. Under normal circumstances, point A1 is classified as the normal vehicle set B2, and point A1 is judged as a normal vehicle.
[0092] In particular, the above method steps are not fully applicable to all actual situations, such as preparing to drive a vehicle after strenuous exercise and not drinking alcohol, or other passengers in the vehicle drinking alcohol, which increases the alcohol concentration in the vehicle environment, etc. Therefore, it is necessary to correct the drunk driving judgment results.
[0093] As a possible implementation of this embodiment, the specific process of performing drunk driving identification based on the centroid also includes a process of correcting the drunk driving identification result:
[0094] Set the real-time alcohol concentration X in the driving cabin as the "first determining factor", the real-time heart rate Y of the vehicle driver as the "second determining factor", and the vehicle speed Z as the "auxiliary factor". After the classification analysis is completed, the classification situation will be corrected;
[0095] According to the Euclidean distance method (2), we can get n-m The changes of the internal data points in the spatial coordinate system are used to determine the specific situation of the vehicle driver and make another drunk driving judgment.
[0096] like Figure 2As shown, X is the alcohol concentration in the car - "the first determining factor", Y is the driver's heart rate - "the second determining factor", Z is the vehicle speed - "the auxiliary factor", W is the warning measure for the driver, and E is the measure for reporting to the relevant departments; among them, X0-the input of the alcohol concentration in the system that does not meet the drunk driving conditions, X1-the input of the alcohol concentration in the system that meets the drunk driving conditions, Y0-the input of the driver's heart rate in the system that does not meet the drunk driving conditions, Y1-the input of the driver's heart rate in the system that meets the drunk driving conditions, Z0-the input of the vehicle's final speed in the system that does not meet the drunk driving conditions, Z1-the input of the vehicle's final speed in the system that meets the drunk driving conditions, W0-the output of the prevention and control measures in the system that do not warn the driver of the drunk driving vehicle, W1-the output of the prevention and control measures in the system that warn the driver of the drunk driving vehicle, E0-the output of the prevention and control measures in the system that do not report to the relevant departments, E1-the output of the prevention and control measures in the system that report to the relevant departments.
[0097] Specifically, the process of correcting the drunk driving judgment result is as follows:
[0098] In the backend server, the alcohol concentration in the vehicle X is set as the "first determining factor", the driver's heart rate Y is set as the "second determining factor", and the vehicle speed Z is set as the "auxiliary factor";
[0099] The speed of the Z vehicle as an "auxiliary factor" is not involved in the discussion for the time being;
[0100] When A* is classified as the drunk driving vehicle set B1, A*=(X*, Y*, Z*) is specifically analyzed:
[0101] If X is greater than or equal to 20 mg / 100 mL and Y is greater than 100 times / min, the classification result remains unchanged;
[0102] If X is less than 20 mg / 100 mL and Y is greater than 100 times / min, it is classified as B2;
[0103] If X is greater than or equal to 20 mg / 100 mL and Y is less than 100 times / min, the classification result remains unchanged;
[0104] The following situations are derived from the above-mentioned element rules and the corresponding preventive measures taken by the drunk driving identification system are shown in decision table 2.
[0105] Table 2: Preventive measures for various drunk driving judgment results
[0106]
[0107] In Table 2, yes means that the precondition is met, and no means that the precondition is not met;
[0108] The above method and the detection distance limit of the alcohol sensor are used to correct the judgment result and improve the accuracy of the data point classification data subset.
[0109] The modules of the drunk driving vehicle identification system of the present invention are installed and distributed in the vehicle as follows: Figure 3 As shown, Figure 3 In the figure, 1 is a pressure sensor, 2 is a heart rate detector, 3 is an alcohol sensor, 4 is a vehicle speedometer, 5 is a vehicle display screen, 6 is a GPS module, 7 is a communication module, and 8 is a microcontroller. The above modules are all installed inside the vehicle. The pressure sensor 1 is installed just below the main driver's seat; the alcohol sensor 3 and the heart rate detector 2 are installed side by side and slightly spaced to the right of the vehicle speedometer 4, wherein the heart rate detector 2 is installed on the left side of the alcohol sensor 3; the GPS module 6, the communication module 7 and the microcontroller 8 are installed side by side in sequence to the right of the vehicle display screen 5.
[0110] like Figure 4 As shown, a method for distinguishing drunk driving vehicles provided by an embodiment of the present invention includes the following steps:
[0111] Step 1, using a pressure sensor to detect whether the driver is sitting in the main driving seat, if yes, proceed to step 2, otherwise continue to detect whether the driver is sitting in the main driving seat;
[0112] Step 2, using an alcohol sensor and a heart rate detector to detect the real-time alcohol concentration in the vehicle cab and the real-time heart rate of the driver, while the on-board speedometer detects the real-time speed of the vehicle, and the microcontroller transmits the above detection data to the background server through the communication module;
[0113] Step 3: The backend server stores the detection data in the database and uses the drunk driving identification algorithm to analyze the detection data. If the vehicle is identified as a drunk driving vehicle, it proceeds to step 4, otherwise it returns to step 2;
[0114] Step 4, sending the drunk driving behavior information of the drunk driving vehicle driver to the microcontroller through the communication module, and the microcontroller sends the drunk driving behavior information of the drunk driving vehicle driver to the vehicle display screen to play a warning screen to warn the driver of drunk driving behavior, and use the real-time data of the vehicle speedometer to determine whether the driver makes corrections after the warning, if no corrections are made, proceed to step 5, otherwise return to step 2;
[0115] Step 5, using the communication module to report the drunk driving behavior of the drunk driving driver who continues to drive after being warned and the real-time vehicle information to the traffic control and traffic police departments, wherein the real-time vehicle information includes vehicle location information and vehicle speed information.
[0116] As a possible implementation of this embodiment, the process of analyzing the detection data using the drunk driving discrimination algorithm includes the following steps:
[0117] Step 31, after a period of time T m-nThe real-time alcohol concentration in the vehicle driving room, the real-time heart rate of the driver and the real-time speed of the vehicle are imported into the database for storage, and the ε group of imported data is selected from the database as the data set:
[0118] A i =(X i ,Y i ,Z i )
[0119] Among them, A is the total set of various data detected by the detection module, X is the real-time alcohol concentration in the cab, Y is the real-time heart rate of the vehicle driver, and Z is the real-time speed of the vehicle;
[0120] Step 32: Create a data subset required for classification. i The data are divided into a drunk driving vehicle set B1, a normal vehicle set B2, and a normal vehicle set B3 after being warned, and the centroids of the three data sets B1, B2, and B3 are determined respectively. The centroids of the three data sets B1, B2, and B3 are determined based on: (1) the real-time alcohol concentration X in the cab is determined based on the GB19522 standard "Blood and breath alcohol content threshold and test for vehicle drivers", (2) the real-time heart rate Y of the vehicle driver is determined based on the heart rate of the drunk driver being >100 beats / min and the frequency band being 1.70Hz-2.50Hz; (3) the real-time speed Z of the vehicle is determined as 0m / s. The drunk driving vehicle includes drunk driving vehicles and drunk driving vehicles;
[0121] Step 33, the three data sets B1, B2, and B3 are B1(Xa, Ya, Za), B2(Xb, Yb, Zb), and B3(Xc, Yc, Zc), respectively. The centroid of the three data sets B1, B2, and B3 is B(X η ,Y η ,Z η ), according to the Euclidean distance method, we get the following formula:
[0122]
[0123] In the formula, m represents the initial detection time point, n represents the final detection time point, and x n -x m Represents a period of time T n-m The difference of real-time alcohol concentration in the driving cabin, y n -y m Represents a period of time T n-m The difference in the real-time heart rate of the drivers in the vehicle, z n -z m Represents a period of time T n-m The difference in vehicle speed, X i ,Y i ,Zi They represent the real-time alcohol concentration in the driving room, the real-time heart rate of the driver and the vehicle speed data respectively. η ,Y η ,Z η Represents the specific value of the center of mass set in advance based on the judgment range and threshold of the real-time alcohol concentration in the cab, the real-time heart rate of the vehicle driver and the vehicle speed data;
[0124] Step 34, select five data sets A1 (X1, Y1, Z1), A2 (X2, Y2, Z2), A3 (X3, Y3, Z3), A4 (X4, Y4, Z4) and A5 (X5, Y5, Z5) from the data set A, and calculate the distances between Ai and the three centroids of B1, B2 and B3 respectively according to the three-dimensional Euclidean distance method (2), i = 1, 2...5;
[0125] Step 35, classify all Ai: ΔAi1 = |Ai-B1|, ΔAi2 = |Ai-B2|, ΔAi3 = |Ai-B3|. After obtaining ΔAi, determine the values of ΔAi1, ΔAi2 and ΔAi3 respectively. Assume that ΔAi2 has the smallest value, that is, ΔAi2 is closest to point B2 of the normal vehicle set. Under normal circumstances, point Ai is classified as the normal vehicle set B2, and point Ai is determined to be a normal vehicle. All Ai are classified according to the above classification and analysis method, thereby completing the drunk driving judgment.
[0126] As a possible implementation of this embodiment, in step 4, the process of analyzing the detection data using the drunk driving discrimination algorithm further includes the following steps:
[0127] Step 36, setting the real-time alcohol concentration X in the driving room as the "first determining factor", the real-time heart rate Y of the vehicle driver as the "second determining factor", and the vehicle speed Z as the "auxiliary factor", after the classification analysis is completed, the classification situation will be corrected;
[0128] Step 37, according to the Euclidean distance method (2), obtain the n-m The changes of the internal data points in the spatial coordinate system are used to determine the specific situation of the vehicle driver and make another drunk driving judgment.
[0129] Different from the current technology, the present invention is based on sensor technology, microcontroller technology, algorithm identification technology, GPS positioning technology, communication technology, etc., and has the function of real-time detection of the driver's vehicle and status. The effective measures taken are uploading real-time vehicle information, warning the driver of drunk driving, and reporting the real-time vehicle situation to the traffic management department and the traffic police department. These three measures can fundamentally prevent the occurrence of drunk driving, avoid traffic accidents caused by drunk driving to a large extent, and improve road safety.
[0130] The present invention provides a drunk driving identification algorithm, which detects the alcohol concentration in the vehicle (driver's surrounding environment), the driver's heart rate and the real-time speed of the vehicle to determine in real time whether the detected vehicle is a drunk driving vehicle.
[0131] The present invention also provides a method for correcting a drunk driving identification algorithm. According to the actual situation and the importance of various parameters in determining whether a vehicle is a drunk driving vehicle, a "first factor", "second factor" and "auxiliary factor" are set. After the classification analysis algorithm is completed, the background server will correct the classification according to the "factor level", thereby greatly improving the accuracy of distinguishing drunk driving vehicles.
[0132] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.
Claims
1. A drunk driving vehicle identification system, characterized in that: It includes a detection module, a microcontroller, a communication module and a drunk driving identification module; The detection module detects the real-time alcohol concentration in the vehicle driving room, the real-time heart rate of the driver and the real-time speed of the vehicle, and sends the detection data to the microcontroller; The microcontroller sends the detection data to the drunk driving identification module through the communication module; The drunk driving identification module uses a drunk driving identification algorithm to analyze the detection data to determine whether the vehicle is a drunk driving vehicle. If so, the drunk driving behavior information of the driver of the drunk driving vehicle is sent to the microcontroller through the communication module; The microcontroller sends the drunk driving behavior information of the drunk driving vehicle driver to the vehicle-mounted display for warning.
2. A drunk driving identification system according to claim 1, characterized in that: The detection module includes an alcohol sensor, a heart rate detector and a vehicle-mounted speedometer. The alcohol sensor and heart rate detector are installed in the front side of the driver in the cab and are used to detect the real-time alcohol concentration in the vehicle cab and the real-time heart rate of the driver; the vehicle-mounted speedometer includes an instrument in the vehicle's existing system for detecting the vehicle's real-time speed, which is used to detect the vehicle's real-time speed.
3. A drunk driving identification system according to claim 2, characterized in that: The detection module also includes a pressure sensor, which is installed just below the main driving seat of the vehicle and is used to detect whether the driver is sitting in the main driving seat.
4. A drunk driving identification system according to claim 1, characterized in that: It also includes a GPS module, which is used to collect real-time location information of the vehicle.
5. A drunk driving identification system according to any one of claims 1 to 4, characterized in that: The drunk driving identification module uses a drunk driving identification algorithm to analyze the detection data. The specific process of determining whether a vehicle is a drunk driving vehicle is as follows: In a period of time T n-m The real-time alcohol concentration in the vehicle driving room, the real-time heart rate of the driver and the real-time speed of the vehicle are imported into the database for storage, and the ε group of imported data is selected from the database as the data set: A i =(X i ,Y i ,Z i ) Among them, A is the total set of various data detected by the detection module, X is the real-time alcohol concentration in the cab, Y is the real-time heart rate of the vehicle driver, and Z is the real-time speed of the vehicle; Establish the data subset required for classification and convert data set A i The data are divided into a drunk driving vehicle set B1, a normal vehicle set B2, and a normal vehicle set B3 after being warned, and the centroids of the three data sets B1, B2, and B3 are determined respectively, and drunk driving is judged based on the centroids. The basis for determining the centroids of the three data sets B1, B2, and B3 includes: (1) the real-time alcohol concentration X in the cab is judged based on the GB19522 standard "Blood and breath alcohol content threshold and test for vehicle drivers", (2) the real-time heart rate Y of the vehicle driver is judged based on the heart rate of the drunk driver>100 times / min, and the frequency band is 1.70Hz-2.50Hz; (3) the real-time speed Z of the vehicle is judged as 0m / s, and the drunk driving vehicle includes drunk driving vehicles and drunk driving vehicles.
6. A drunk driving identification system according to claim 5, characterized in that: The specific process of judging drunk driving based on the centroid is as follows: The three data sets B1, B2, and B3 are B1(Xa, Ya, Za), B2(Xb, Yb, Zb), and B3(Xc, Yc, Zc), respectively. The centroid of the three data sets B1, B2, and B3 is B(X η ,Y η ,Z η ), according to the Euclidean distance method, we get the following formula: In the formula, m represents the initial detection time point, n represents the final detection time point, and x n -x m Represents a period of time T n-m The difference in real-time alcohol concentration inside the driving cabin, y n -y m Represents a period of time T n-m The difference in the real-time heart rate of the drivers in the vehicles, z n -z m Represents a period of time T n-m The difference in vehicle speed, X i ,Y i ,Z i They represent the real-time alcohol concentration in the driving room, the real-time heart rate of the driver and the vehicle speed data respectively. η ,Y η ,Z η Represents the specific value of the center of mass set in advance based on the judgment range and threshold of the real-time alcohol concentration in the cab, the real-time heart rate of the vehicle driver and the vehicle speed data; Select five data sets A1 (X1, Y1, Z1), A2 (X2, Y2, Z2), A3 (X3, Y3, Z3), A4 (X4, Y4, Z4) and A5 (X5, Y5, Z5) from data set A, and calculate the distances between Ai and the centroids of B1, B2 and B3 according to the three-dimensional Euclidean distance method (2), i = 1, 2...5; Classify all Ai as follows: ΔAi1=|Ai-B1|, ΔAi2=|Ai-B2|, ΔAi3=|Ai-B3|. After obtaining ΔAi, determine the values of ΔAi1, ΔAi2 and ΔAi3 respectively. Assume that ΔAi2 has the smallest value, that is, ΔAi2 is closest to point B2 in the normal vehicle set. Under normal circumstances, point Ai is classified as part of the normal vehicle set B2, and point Ai is determined to be a normal vehicle.
7. A drunk driving identification system according to claim 6, characterized in that: The specific process of judging drunk driving according to the centroid also includes: Set the real-time alcohol concentration X in the driving cabin as the "first determining factor", the real-time heart rate Y of the vehicle driver as the "second determining factor", and the vehicle speed Z as the "auxiliary factor". After the classification analysis is completed, the classification situation will be corrected; According to the Euclidean distance method (2), we can get n-m The changes of the internal data points in the spatial coordinate system are used to determine the specific situation of the vehicle driver and make another drunk driving judgment.
8. A method for identifying drunk driving vehicles, characterized in that: The following steps are involved: Step 1, using a pressure sensor to detect whether the driver is sitting in the main driving seat, if yes, proceed to step 2, otherwise continue to detect whether the driver is sitting in the main driving seat; Step 2, using an alcohol sensor and a heart rate detector to detect the real-time alcohol concentration in the vehicle cab and the real-time heart rate of the driver, while the on-board speedometer detects the real-time speed of the vehicle, and the microcontroller transmits the above detection data to the background server through the communication module; Step 3: The backend server stores the detection data in the database and uses the drunk driving identification algorithm to analyze the detection data. If the vehicle is identified as a drunk driving vehicle, it proceeds to step 4, otherwise it returns to step 2; Step 4, sending the drunk driving behavior information of the drunk driving vehicle driver to the microcontroller through the communication module, and the microcontroller sends the drunk driving behavior information of the drunk driving vehicle driver to the vehicle display screen to play a warning screen to warn the driver of drunk driving behavior, and use the real-time data of the vehicle speedometer to determine whether the driver makes corrections after the warning, if no corrections are made, proceed to step 5, otherwise return to step 2; Step 5, using the communication module to report the drunk driving behavior of the drunk driving driver who continues to drive after being warned and the real-time vehicle information to the traffic control and traffic police departments, wherein the real-time vehicle information includes vehicle location information and vehicle speed information.
9. A drunk driving identification method according to claim 8, characterized in that: In step 4, the process of analyzing the detection data using the drunk driving discrimination algorithm includes the following steps: Step 31, after a period of time T m-n The real-time alcohol concentration in the vehicle driving room, the real-time heart rate of the driver and the real-time speed of the vehicle are imported into the database for storage, and the ε group of imported data is selected from the database as the data set: A i =(X i ,Y i ,Z i ) Among them, A is the total set of various data detected by the detection module, X is the real-time alcohol concentration in the cab, Y is the real-time heart rate of the vehicle driver, and Z is the real-time speed of the vehicle; Step 32: Create a data subset required for classification. i The data are divided into a drunk driving vehicle set B1, a normal vehicle set B2, and a normal vehicle set B3 after being warned, and the centroids of the three data sets B1, B2, and B3 are determined respectively. The centroids of the three data sets B1, B2, and B3 are determined based on: (1) the real-time alcohol concentration X in the cab is determined based on the GB19522 standard "Blood and breath alcohol content threshold and test for vehicle drivers", (2) the real-time heart rate Y of the vehicle driver is determined based on the heart rate of the drunk driver being >100 beats / min and the frequency band being 1.70Hz-2.50Hz; (3) the real-time speed Z of the vehicle is determined as 0m / s. The drunk driving vehicle includes drunk driving vehicles and drunk driving vehicles; Step 33, the three data sets B1, B2, and B3 are B1(Xa, Ya, Za), B2(Xb, Yb, Zb), and B3(Xc, Yc, Zc), respectively. The centroid of the three data sets B1, B2, and B3 is B(X η ,Y η ,Z η ), according to the Euclidean distance method, we get the following formula: In the formula, m represents the initial detection time point, n represents the final detection time point, and x n -x m Represents a period of time T n-m The difference in real-time alcohol concentration inside the driving cabin, y n -y m Represents a period of time T n-m The difference in the real-time heart rate of the drivers in the vehicles, z n -z m Represents a period of time T n-m The difference in vehicle speed, X i ,Y i ,Z i They represent the real-time alcohol concentration in the driving room, the real-time heart rate of the driver and the vehicle speed data respectively. η ,Y η ,Z η Represents the specific value of the center of mass set in advance based on the judgment range and threshold of the real-time alcohol concentration in the cab, the real-time heart rate of the vehicle driver and the vehicle speed data; Step 34, select five data sets A1 (X1, Y1, Z1), A2 (X2, Y2, Z2), A3 (X3, Y3, Z3), A4 (X4, Y4, Z4) and A5 (X5, Y5, Z5) from the data set A, and calculate the distances between Ai and the three centroids of B1, B2 and B3 respectively according to the three-dimensional Euclidean distance method (2), i = 1, 2...5; Step 35, classify all Ai: ΔAi1 = |Ai-B1|, ΔAi2 = |Ai-B2|, ΔAi3 = |Ai-B3|. After obtaining ΔAi, determine the values of ΔAi1, ΔAi2 and ΔAi3 respectively. Assume that ΔAi2 has the smallest value, that is, ΔAi2 is closest to point B2 of the normal vehicle set. Under normal circumstances, point Ai is classified as the normal vehicle set B2, and point Ai is determined to be a normal vehicle, thereby completing the drunk driving judgment.
10. A drunk driving identification method according to claim 9, characterized in that: In step 4, the process of analyzing the detection data using the drunk driving discrimination algorithm also includes the following steps: Step 36, setting the real-time alcohol concentration X in the driving room as the "first determining factor", the real-time heart rate Y of the vehicle driver as the "second determining factor", and the vehicle speed Z as the "auxiliary factor", after the classification analysis is completed, the classification situation will be corrected; Step 37, according to the Euclidean distance method (2), obtain the n-m The changes of the internal data points in the spatial coordinate system are used to determine the specific situation of the vehicle driver and make another drunk driving judgment.
