Vehicle driving abnormal behavior safety intervention system and method

Abstract

The present invention discloses a vehicle driving abnormal behavior safety intervention system and method, which relates to the field of vehicle safety control; comprising the following steps: determining the evaluation dimension of abnormal driving monitoring; obtaining the threshold range of each dimension under normal driving state through big data; learning according to the driver's driving habits, obtaining the threshold range of each dimension under normal driving state of the driver; comprehensively determining the comparison threshold range according to the threshold range of each dimension obtained from big data and the threshold range of each dimension of the learned driver's habits; in the process of the vehicle being driven, real-time monitoring of the actual value of each dimension, and comparing the actual value with the threshold range; if it does not exceed the threshold range, normal driving is determined. The present invention collects data in multiple dimensions, and then intervenes in management in the form of threshold control, so that detection and safety intervention in abnormal driving state can be achieved, thereby increasing the safety of the vehicle.

Description

Technical Field

[0001] The present invention relates to the field of vehicle safety control, and in particular to a vehicle driving abnormal behavior safety intervention system and method. Background Art

[0002] At present, with the development of economy, logistics, transportation and other fields are also developing at a faster and faster speed. People are increasingly relying on public transportation for shopping, cargo delivery, travel, daily life and other aspects.

[0003] And with the development of artificial intelligence, its application to vehicles can achieve a certain degree of intelligent driving, which can effectively solve the problem of manual driving fatigue.

[0004] For example, after searching, the Chinese patent publication number CN107963085B discloses a vehicle, a vehicle control method and a system, wherein the method comprises: detecting and acquiring a detection signal, wherein the detection signal is a signal emitted by a nearby vehicle within the detection range of the vehicle; and controlling the speed of the vehicle according to the signal strength of the detection signal, wherein the signal strength represents the distance between the vehicle and the nearby vehicle, and the signal strength is inversely correlated with the distance.

[0005] The above patent has the following deficiencies: since the driver may be fatigued during long-distance driving, he cannot analyze abnormal situations and intervene in time when fatigued, thus posing certain safety hazards.

[0006] To this end, the present invention proposes a vehicle driving abnormal behavior safety intervention system and method. Summary of the invention

[0007] The purpose of the present invention is to solve the shortcomings of the prior art and to propose a vehicle driving abnormal behavior safety intervention system and method.

[0008] In order to achieve the above object, the present invention adopts the following technical solutions:

[0009] The method for safely intervening in abnormal vehicle driving behavior includes the following steps:

[0010] S1: Determine the evaluation dimensions of abnormal driving monitoring;

[0011] S2: Obtain the threshold range of each dimension under normal driving conditions through big data;

[0012] S3: learning based on the driver's driving habits to obtain the threshold ranges of various dimensions under the driver's normal driving state;

[0013] S4: comprehensively determining a comparison threshold range based on the threshold ranges of various dimensions obtained from the big data and the threshold ranges of various dimensions of the learned driver habits;

[0014] S5: While the vehicle is being driven, the actual value of each dimension is monitored in real time, and the actual value is compared with the threshold range;

[0015] S51: If the threshold value is not exceeded, it is determined to be normal driving;

[0016] S52: If the threshold value is exceeded, abnormal driving is determined;

[0017] S6: Combined with the vehicle's camera and radar, it takes over the vehicle for safety intervention and simultaneously issues an alarm through the vehicle's audio system.

[0018] Preferably: in the step S1, the evaluation dimensions include the fluctuation size, fluctuation speed and fluctuation frequency of the accelerator pedal, the fluctuation size, fluctuation speed and fluctuation frequency of the brake pedal, the rotation size, rotation speed and rotation frequency of the steering wheel, the vehicle's continuous driving time and the vehicle's driving speed.

[0019] Preferably: in the step S4, the method for comprehensively determining the comparison threshold range includes the following steps:

[0020] S41: Determine the threshold range for big data acquisition as (a1, b1);

[0021] S42: Determine the threshold range of the driver's habits as (a2, b2);

[0022] S43: According to the formula Calculate and obtain the comparison threshold range (a3, b3).

[0023] Preferably, in the step S42, determining the threshold range of the driver's habits includes the following steps:

[0024] S421: Setting a time threshold t1, taking a vehicle start time node within a time interval of at least t1 as a time starting point;

[0025] S422: determining a collection period t2, starting from the time starting point, collecting data of each dimension once every collection period t2, until n time data are collected, forming a driving habit data packet for one start;

[0026] S423: Analyze the data of n time periods to form a time threshold range for a single start;

[0027] S424: Repeat steps S421 to S423 until the driving habit data packets of m startups are collected, and the m startup time threshold ranges are analyzed, and then the average of the m times is taken.

[0028] Preferably: the step S423 includes the following steps:

[0029] S4231: Calculate the mean σ and variance S of n data respectively 2 , maximum value max and minimum value min, A i is the data value;

[0030] S4232: According to the variance S 2 , the maximum value max and the minimum value min, determine the driving habit threshold range as (min-||S 2 ||σ,max+||S 2 ||σ), where ||S 2 || represents S 2 Scaled to the interval [0,1].

[0031] Preferably: in the step S4232, S 2 The method of scaling to the interval [0,1] is: determine S 2 The number of digits w, S 2 The decimal point is moved forward by w places.

[0032] Preferably: in the step S43, k'1 and k'2 are correlation coefficients respectively,

[0033] Preferably: in the step S43, a' and b' are the error coefficients of the lower limit and the upper limit respectively,

[0034] Preferably: in step S6, the safety intervention includes braking, steering, tightening the seat belt, and automatic parking.

[0035] Abnormal vehicle driving behavior safety intervention system, including:

[0036] A data acquisition module, which is used to collect driving data;

[0037] A data analysis module, which is used to analyze the collected data;

[0038] A control module is used to control the vehicle according to the result of data analysis.

[0039] The beneficial effects of the present invention are:

[0040] 1. The present invention collects data in multiple dimensions and then intervenes in management in the form of threshold control, thereby enabling detection and safe intervention in abnormal driving conditions, thereby increasing the safety of the vehicle.

[0041] 2. The present invention uses big data acquisition and comprehensive evaluation of driving habit learning to control the threshold, so that the threshold can combine the reliability of big data and the collection of driving habits, so as to update the threshold in a targeted manner according to the driving situation of the car owner.

[0042] 3. The present invention adopts comprehensive control of mean, variance, maximum value and minimum value for learning driving habits, so that the deviation of the threshold value can be controlled according to the discrete degree of driver operation fluctuation reflected by the variance. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 This is a flow chart of the method for safely intervening in abnormal vehicle driving behavior proposed by the present invention. DETAILED DESCRIPTION

[0044] The technical solution of the present invention is further described in detail below in conjunction with specific implementation methods.

[0045] In the description of the present invention, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected", "connected", and "set" should be understood in a broad sense, for example, they can be fixedly connected or set, or detachably connected or set, or integrally connected or set. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0046] Embodiment 1:

[0047] The method for safely intervening in abnormal vehicle driving behavior comprises the following steps:

[0048] S1: Determine the evaluation dimensions of abnormal driving monitoring;

[0049] S2: Obtain the threshold range of each dimension under normal driving conditions through big data;

[0050] S3: learning based on the driver's driving habits to obtain the threshold ranges of various dimensions under the driver's normal driving state;

[0051] S4: comprehensively determining a comparison threshold range based on the threshold ranges of various dimensions obtained from the big data and the threshold ranges of various dimensions of the learned driver habits;

[0052] S5: While the vehicle is being driven, the actual value of each dimension is monitored in real time, and the actual value is compared with the threshold range;

[0053] S51: If the threshold value is not exceeded, it is determined to be normal driving;

[0054] S52: If the threshold value is exceeded, abnormal driving is determined;

[0055] S6: Combined with the vehicle's camera and radar, it takes over the vehicle for safety intervention and simultaneously issues an alarm through the vehicle's audio system.

[0056] In the step S1, the evaluation dimensions include the fluctuation size, fluctuation speed and fluctuation frequency of the accelerator pedal, the fluctuation size, fluctuation speed and fluctuation frequency of the brake pedal, the rotation size, rotation speed and rotation frequency of the steering wheel, the vehicle's continuous driving time and the vehicle's driving speed.

[0057] Embodiment 2:

[0058] The method for safely intervening in abnormal vehicle driving behavior comprises the following steps:

[0059] S1: Determine the evaluation dimensions of abnormal driving monitoring;

[0060] S2: Obtain the threshold range of each dimension under normal driving conditions through big data;

[0061] S3: learning based on the driver's driving habits to obtain the threshold ranges of various dimensions under the driver's normal driving state;

[0062] S4: comprehensively determining a comparison threshold range based on the threshold ranges of various dimensions obtained from the big data and the threshold ranges of various dimensions of the learned driver habits;

[0063] S5: While the vehicle is being driven, the actual value of each dimension is monitored in real time, and the actual value is compared with the threshold range;

[0064] S51: If the threshold value is not exceeded, it is determined to be normal driving;

[0065] S52: If the threshold value is exceeded, abnormal driving is determined;

[0066] S6: Combined with the vehicle's camera and radar, it takes over the vehicle for safety intervention and simultaneously issues an alarm through the vehicle's audio system.

[0067] In the step S1, the evaluation dimensions include the fluctuation size, fluctuation speed and fluctuation frequency of the accelerator pedal, the fluctuation size, fluctuation speed and fluctuation frequency of the brake pedal, the rotation size, rotation speed and rotation frequency of the steering wheel, the vehicle's continuous driving time and the vehicle's driving speed.

[0068] In the step S4, the method for comprehensively determining the comparison threshold range includes the following steps:

[0069] S41: Determine the threshold range for big data acquisition as (a1, b1);

[0070] S42: Determine the threshold range of the driver's habits as (a2, b2);

[0071] S43: According to the formula Calculate and obtain the comparison threshold range (a3, b3). In the step S42, determining the threshold range of the driver's habit includes the following steps:

[0072] S421: Setting a time threshold t1, taking a vehicle start time node within a time interval of at least t1 as a time starting point;

[0073] S422: determining a collection period t2, starting from the time starting point, collecting data of each dimension once every collection period t2, until n time data are collected, forming a driving habit data packet for one start;

[0074] S423: Analyze the data of n time periods to form a time threshold range for a single start;

[0075] S424: Repeat steps S421 to S423 until the driving habit data packets of m startups are collected, and the m startup time threshold ranges are analyzed, and then the average of the m times is taken.

[0076] The step S423 includes the following steps:

[0077] S4231: Calculate the mean σ and variance S of n data respectively 2 , maximum value max and minimum value min, A i is the data value;

[0078] S4232: According to the variance S 2 , the maximum value max and the minimum value min, determine the driving habit threshold range as (min-||S 2 ||σ,max+||S 2 ||σ), where ||S 2 || represents S 2 Scaled to the interval [0,1].

[0079] In the step S4232, S 2 The method of scaling to the interval [0,1] is: determine S 2 The number of digits w, S 2 The decimal point is moved forward by w places.

[0080] Embodiment 3:

[0081] Vehicle driving abnormal behavior safety intervention system and method.

[0082] Embodiment 4:

[0083] The method for safely intervening in abnormal vehicle driving behavior comprises the following steps:

[0084] S1: Determine the evaluation dimensions of abnormal driving monitoring;

[0085] S2: Obtain the threshold range of each dimension under normal driving conditions through big data;

[0086] S3: learning based on the driver's driving habits to obtain the threshold ranges of various dimensions under the driver's normal driving state;

[0087] S4: comprehensively determining a comparison threshold range based on the threshold ranges of various dimensions obtained from the big data and the threshold ranges of various dimensions of the learned driver habits;

[0088] S5: While the vehicle is being driven, the actual value of each dimension is monitored in real time, and the actual value is compared with the threshold range;

[0089] S51: If the threshold value is not exceeded, it is determined to be normal driving;

[0090] S52: If the threshold value is exceeded, abnormal driving is determined;

[0091] S6: Combined with the vehicle's camera and radar, it takes over the vehicle for safety intervention and simultaneously issues an alarm through the vehicle's audio system.

[0092] In the step S1, the evaluation dimensions include the fluctuation size, fluctuation speed and fluctuation frequency of the accelerator pedal, the fluctuation size, fluctuation speed and fluctuation frequency of the brake pedal, the rotation size, rotation speed and rotation frequency of the steering wheel, the vehicle's continuous driving time and the vehicle's driving speed.

[0093] In the step S4, the method for comprehensively determining the comparison threshold range includes the following steps:

[0094] S41: Determine the threshold range for big data acquisition as (a1, b1);

[0095] S42: Determine the threshold range of the driver's habits as (a2, b2);

[0096] S43: According to the formula Calculate and obtain the comparison threshold range (a3, b3). In the step S42, determining the threshold range of the driver's habit includes the following steps:

[0097] S421: Setting a time threshold t1, taking a vehicle start time node within a time interval of at least t1 as a time starting point;

[0098] S422: determining a collection period t2, starting from the time starting point, collecting data of each dimension once every collection period t2, until n time data are collected, forming a driving habit data packet for one start;

[0099] S423: Analyze the data of n time periods to form a time threshold range for a single start;

[0100] S424: Repeat steps S421 to S423 until the driving habit data packets of m startups are collected, and the m startup time threshold ranges are analyzed, and then the average of the m times is taken.

[0101] The step S423 includes the following steps:

[0102] S4231: Calculate the mean σ and variance S of n data respectively 2 , maximum value max and minimum value min, A i is the data value;

[0103] S4232: According to the variance S 2 , the maximum value max and the minimum value min, determine the driving habit threshold range as (min-||S 2 ||σ,max+||S 2 ||σ), where ||S 2 || represents S 2 Scaled to the interval [0,1].

[0104] In the step S4232, S 2 The method of scaling to the interval [0,1] is: determine S 2 The number of digits w, S 2 The decimal point is moved forward by w places.

[0105] In the step S43, k'1 and k'2 are correlation coefficients, respectively.

[0106] Embodiment 5:

[0107] The method for safely intervening in abnormal vehicle driving behavior comprises the following steps:

[0108] S1: Determine the evaluation dimensions of abnormal driving monitoring;

[0109] S2: Obtain the threshold range of each dimension under normal driving conditions through big data;

[0110] S3: learning based on the driver's driving habits to obtain the threshold ranges of various dimensions under the driver's normal driving state;

[0111] S4: comprehensively determining a comparison threshold range based on the threshold ranges of various dimensions obtained from the big data and the threshold ranges of various dimensions of the learned driver habits;

[0112] S5: While the vehicle is being driven, the actual value of each dimension is monitored in real time, and the actual value is compared with the threshold range;

[0113] S51: If the threshold value is not exceeded, it is determined to be normal driving;

[0114] S52: If the threshold value is exceeded, abnormal driving is determined;

[0115] S6: Combined with the vehicle's camera and radar, it takes over the vehicle for safety intervention and simultaneously issues an alarm through the vehicle's audio system.

[0116] In the step S1, the evaluation dimensions include the fluctuation size, fluctuation speed and fluctuation frequency of the accelerator pedal, the fluctuation size, fluctuation speed and fluctuation frequency of the brake pedal, the rotation size, rotation speed and rotation frequency of the steering wheel, the vehicle's continuous driving time and the vehicle's driving speed.

[0117] In the step S4, the method for comprehensively determining the comparison threshold range includes the following steps:

[0118] S41: Determine the threshold range for big data acquisition as (a1, b1);

[0119] S42: Determine the threshold range of the driver's habits as (a2, b2);

[0120] S43: According to the formula Calculate and obtain the comparison threshold range (a3, b3). In the step S42, determining the threshold range of the driver's habit includes the following steps:

[0121] S421: Setting a time threshold t1, taking a vehicle start time node within a time interval of at least t1 as a time starting point;

[0122] S422: determining a collection period t2, starting from the time starting point, collecting data of each dimension once every collection period t2, until n time data are collected, forming a driving habit data packet for one start;

[0123] S423: Analyze the data of n time periods to form a time threshold range for a single start;

[0124] S424: Repeat steps S421 to S423 until the driving habit data packets of m startups are collected, and the m startup time threshold ranges are analyzed, and then the average of the m times is taken.

[0125] The step S423 includes the following steps:

[0126] S4231: Calculate the mean σ and variance S of n data respectively 2 , maximum value max and minimum value min, A i is the data value;

[0127] S4232: According to the variance S 2 , the maximum value max and the minimum value min, determine the driving habit threshold range as (min-||S 2 ||σ,max+||S 2 ||σ), where ||S 2 || represents S 2 Scaled to the interval [0,1].

[0128] In the step S4232, S 2 The method of scaling to the interval [0,1] is: determine S 2 The number of digits w, S 2 The decimal point is moved forward by w places.

[0129] In the step S43, k'1 and k'2 are correlation coefficients, respectively.

[0130] Embodiment 6:

[0131] The method for safely intervening in abnormal vehicle driving behavior comprises the following steps:

[0132] S1: Determine the evaluation dimensions of abnormal driving monitoring;

[0133] S2: Obtain the threshold range of each dimension under normal driving conditions through big data;

[0134] S3: learning based on the driver's driving habits to obtain the threshold ranges of various dimensions under the driver's normal driving state;

[0135] S4: comprehensively determining a comparison threshold range based on the threshold ranges of various dimensions obtained from the big data and the threshold ranges of various dimensions of the learned driver habits;

[0136] S5: While the vehicle is being driven, the actual value of each dimension is monitored in real time, and the actual value is compared with the threshold range;

[0137] S51: If the threshold value is not exceeded, it is determined to be normal driving;

[0138] S52: If the threshold value is exceeded, abnormal driving is determined;

[0139] S6: Combined with the vehicle's camera and radar, it takes over the vehicle for safety intervention and simultaneously issues an alarm through the vehicle's audio system.

[0140] In the step S1, the evaluation dimensions include the fluctuation size, fluctuation speed and fluctuation frequency of the accelerator pedal, the fluctuation size, fluctuation speed and fluctuation frequency of the brake pedal, the rotation size, rotation speed and rotation frequency of the steering wheel, the vehicle's continuous driving time and the vehicle's driving speed.

[0141] In the step S4, the method for comprehensively determining the comparison threshold range includes the following steps:

[0142] S41: Determine the threshold range for big data acquisition as (a1, b1);

[0143] S42: Determine the threshold range of the driver's habits as (a2, b2);

[0144] S43: According to the formula Calculate and obtain the comparison threshold range (a3, b3). In the step S42, determining the threshold range of the driver's habit includes the following steps:

[0145] S421: Setting a time threshold t1, taking a vehicle start time node within a time interval of at least t1 as a time starting point;

[0146] S422: determining a collection period t2, starting from the time starting point, collecting data of each dimension once every collection period t2, until n time data are collected, forming a driving habit data packet for one start;

[0147] S423: Analyze the data of n time periods to form a time threshold range for a single start;

[0148] S424: Repeat steps S421 to S423 until the driving habit data packets of m startups are collected, and the m startup time threshold ranges are analyzed, and then the average of the m times is taken.

[0149] The step S423 includes the following steps:

[0150] S4231: Calculate the mean σ and variance S of n data respectively 2 , maximum value max and minimum value min, A i is the data value;

[0151] S4232: According to the variance S 2 , the maximum value max and the minimum value min, determine the driving habit threshold range as (min-||S 2 ||σ,max+||S 2 ||σ), where ||S 2 || represents S 2 Scaled to the interval [0,1].

[0152] In the step S4232, S 2 The method of scaling to the interval [0,1] is: determine S 2 The number of digits w, S 2 The decimal point is moved forward by w places.

[0153] In the step S43, k'1 and k'2 are correlation coefficients, respectively.

[0154] Embodiment 7:

[0155] The method for safely intervening in abnormal vehicle driving behavior comprises the following steps:

[0156] S1: Determine the evaluation dimensions of abnormal driving monitoring;

[0157] S2: Obtain the threshold range of each dimension under normal driving conditions through big data;

[0158] S3: learning based on the driver's driving habits to obtain the threshold ranges of various dimensions under the driver's normal driving state;

[0159] S4: comprehensively determining a comparison threshold range based on the threshold ranges of various dimensions obtained from the big data and the threshold ranges of various dimensions of the learned driver habits;

[0160] S5: While the vehicle is being driven, the actual value of each dimension is monitored in real time, and the actual value is compared with the threshold range;

[0161] S51: If the threshold value is not exceeded, it is determined to be normal driving;

[0162] S52: If the threshold value is exceeded, abnormal driving is determined;

[0163] S6: Combined with the vehicle's camera and radar, it takes over the vehicle for safety intervention and simultaneously issues an alarm through the vehicle's audio system.

[0164] In the step S1, the evaluation dimensions include the fluctuation size, fluctuation speed and fluctuation frequency of the accelerator pedal, the fluctuation size, fluctuation speed and fluctuation frequency of the brake pedal, the rotation size, rotation speed and rotation frequency of the steering wheel, the vehicle's continuous driving time and the vehicle's driving speed.

[0165] In the step S4, the method for comprehensively determining the comparison threshold range includes the following steps:

[0166] S41: Determine the threshold range for big data acquisition as (a1, b1);

[0167] S42: Determine the threshold range of the driver's habits as (a2, b2);

[0168] S43: According to the formula Calculate and obtain the comparison threshold range (a3, b3). In the step S42, determining the threshold range of the driver's habit includes the following steps:

[0169] S421: Setting a time threshold t1, taking a vehicle start time node within a time interval of at least t1 as a time starting point;

[0170] S422: determining a collection period t2, starting from the time starting point, collecting data of each dimension once every collection period t2, until n time data are collected, forming a driving habit data packet for one start;

[0171] S423: Analyze the data of n time periods to form a time threshold range for a single start;

[0172] S424: Repeat steps S421 to S423 until the driving habit data packets of m startups are collected, and the m startup time threshold ranges are analyzed, and then the average of the m times is taken.

[0173] The step S423 includes the following steps:

[0174] S4231: Calculate the mean σ and variance S of n data respectively 2 , maximum value max and minimum value min, A i is the data value;

[0175] S4232: According to the variance S 2 , the maximum value max and the minimum value min, determine the driving habit threshold range as (min-||S 2 ||σ,max+||S 2 ||σ), where ||S 2 || represents S 2 Scaled to the interval [0,1].

[0176] In the step S4232, S 2 The method of scaling to the interval [0,1] is: determine S 2 The number of digits w, S 2 The decimal point is moved forward by w places.

[0177] In the step S43, k'1 and k'2 are correlation coefficients, respectively.

[0178] In the step S43, a' and b' are the error coefficients of the lower limit and the upper limit respectively.

[0179] Embodiment 8:

[0180] The method for safely intervening in abnormal vehicle driving behavior comprises the following steps:

[0181] S1: Determine the evaluation dimensions of abnormal driving monitoring;

[0182] S2: Obtain the threshold range of each dimension under normal driving conditions through big data;

[0183] S3: learning based on the driver's driving habits to obtain the threshold ranges of various dimensions under the driver's normal driving state;

[0184] S4: comprehensively determining a comparison threshold range based on the threshold ranges of various dimensions obtained from the big data and the threshold ranges of various dimensions of the learned driver habits;

[0185] S5: While the vehicle is being driven, the actual value of each dimension is monitored in real time, and the actual value is compared with the threshold range;

[0186] S51: If the threshold value is not exceeded, it is determined to be normal driving;

[0187] S52: If the threshold value is exceeded, abnormal driving is determined;

[0188] S6: Combined with the vehicle's camera and radar, it takes over the vehicle for safety intervention and simultaneously issues an alarm through the vehicle's audio system.

[0189] In the step S1, the evaluation dimensions include the fluctuation size, fluctuation speed and fluctuation frequency of the accelerator pedal, the fluctuation size, fluctuation speed and fluctuation frequency of the brake pedal, the rotation size, rotation speed and rotation frequency of the steering wheel, the vehicle's continuous driving time and the vehicle's driving speed.

[0190] In the step S4, the method for comprehensively determining the comparison threshold range includes the following steps:

[0191] S41: Determine the threshold range for big data acquisition as (a1, b1);

[0192] S42: Determine the threshold range of the driver's habits as (a2, b2);

[0193] S43: According to the formula Calculate and obtain the comparison threshold range (a3, b3). In the step S42, determining the threshold range of the driver's habit includes the following steps:

[0194] S421: Setting a time threshold t1, taking a vehicle start time node within a time interval of at least t1 as a time starting point;

[0195] S422: determining a collection period t2, starting from the time starting point, collecting data of each dimension once every collection period t2, until n time data are collected, forming a driving habit data packet for one start;

[0196] S423: Analyze the data of n time periods to form a time threshold range for a single start;

[0197] S424: Repeat steps S421 to S423 until the driving habit data packets of m startups are collected, and the m startup time threshold ranges are analyzed, and then the average of the m times is taken.

[0198] The step S423 includes the following steps:

[0199] S4231: Calculate the mean σ and variance S of n data respectively 2 , maximum value max and minimum value min, A i is the data value;

[0200] S4232: According to the variance S 2 , the maximum value max and the minimum value min, determine the driving habit threshold range as (min-||S 2 ||σ,max+||S 2 ||σ), where ||S 2 || represents S 2 Scaled to the interval [0,1].

[0201] In the step S4232, S 2 The method of scaling to the interval [0,1] is: determine S 2 The number of digits w, S 2 The decimal point is moved forward by w places.

[0202] In the step S43, k'1 and k'2 are correlation coefficients, respectively.

[0203] In the step S43, a' and b' are the error coefficients of the lower limit and the upper limit respectively.

[0204] Embodiment 9:

[0205] The method for safely intervening in abnormal vehicle driving behavior comprises the following steps:

[0206] S1: Determine the evaluation dimensions of abnormal driving monitoring;

[0207] S2: Obtain the threshold range of each dimension under normal driving conditions through big data;

[0208] S3: learning based on the driver's driving habits to obtain the threshold ranges of various dimensions under the driver's normal driving state;

[0209] S4: comprehensively determining a comparison threshold range based on the threshold ranges of various dimensions obtained from the big data and the threshold ranges of various dimensions of the learned driver habits;

[0210] S5: While the vehicle is being driven, the actual value of each dimension is monitored in real time, and the actual value is compared with the threshold range;

[0211] S51: If the threshold value is not exceeded, it is determined to be normal driving;

[0212] S52: If the threshold value is exceeded, abnormal driving is determined;

[0213] S6: Combined with the vehicle's camera and radar, it takes over the vehicle for safety intervention and simultaneously issues an alarm through the vehicle's audio system.

[0214] In the step S1, the evaluation dimensions include the fluctuation size, fluctuation speed and fluctuation frequency of the accelerator pedal, the fluctuation size, fluctuation speed and fluctuation frequency of the brake pedal, the rotation size, rotation speed and rotation frequency of the steering wheel, the vehicle's continuous driving time and the vehicle's driving speed.

[0215] In the step S4, the method for comprehensively determining the comparison threshold range includes the following steps:

[0216] S41: Determine the threshold range for big data acquisition as (a1, b1);

[0217] S42: Determine the threshold range of the driver's habits as (a2, b2);

[0218] S43: According to the formula Calculate and obtain the comparison threshold range (a3, b3). In the step S42, determining the threshold range of the driver's habit includes the following steps:

[0219] S421: Setting a time threshold t1, taking a vehicle start time node within a time interval of at least t1 as a time starting point;

[0220] S422: determining a collection period t2, starting from the time starting point, collecting data of each dimension once every collection period t2, until n time data are collected, forming a driving habit data packet for one start;

[0221] S423: Analyze the data of n time periods to form a time threshold range for a single start;

[0222] S424: Repeat steps S421 to S423 until the driving habit data packets of m startups are collected, and the m startup time threshold ranges are analyzed, and then the average of the m times is taken.

[0223] The step S423 includes the following steps:

[0224] S4231: Calculate the mean σ and variance S of n data respectively 2 , maximum value max and minimum value min, A i is the data value;

[0225] S4232: According to the variance S 2 , the maximum value max and the minimum value min, determine the driving habit threshold range as (min-||S 2 ||σ,max+||S 2 ||σ), where ||S 2 || represents S 2 Scaled to the interval [0,1].

[0226] In the step S4232, S 2 The method of scaling to the interval [0,1] is: determine S 2 The number of digits w, S 2 The decimal point is moved forward by w places.

[0227] In the step S43, k'1 and k'2 are correlation coefficients, respectively.

[0228] In the step S43, a' and b' are the error coefficients of the lower limit and the upper limit respectively.

[0229] Embodiment 10:

[0230] The method for safely intervening in abnormal vehicle driving behavior comprises the following steps:

[0231] S1: Determine the evaluation dimensions of abnormal driving monitoring;

[0232] S2: Obtain the threshold range of each dimension under normal driving conditions through big data;

[0233] S3: learning based on the driver's driving habits to obtain the threshold ranges of various dimensions under the driver's normal driving state;

[0234] S4: comprehensively determining a comparison threshold range based on the threshold ranges of various dimensions obtained from the big data and the threshold ranges of various dimensions of the learned driver habits;

[0235] S5: While the vehicle is being driven, the actual value of each dimension is monitored in real time, and the actual value is compared with the threshold range;

[0236] S51: If the threshold value is not exceeded, it is determined to be normal driving;

[0237] S52: If the threshold value is exceeded, abnormal driving is determined;

[0238] S6: Combined with the vehicle's camera and radar, it takes over the vehicle for safety intervention and simultaneously issues an alarm through the vehicle's audio system.

[0239] In the step S1, the evaluation dimensions include the fluctuation size, fluctuation speed and fluctuation frequency of the accelerator pedal, the fluctuation size, fluctuation speed and fluctuation frequency of the brake pedal, the rotation size, rotation speed and rotation frequency of the steering wheel, the vehicle's continuous driving time and the vehicle's driving speed.

[0240] In the step S4, the method for comprehensively determining the comparison threshold range includes the following steps:

[0241] S41: Determine the threshold range for big data acquisition as (a1, b1);

[0242] S42: Determine the threshold range of the driver's habits as (a2, b2);

[0243] S43: According to the formula Calculate and obtain the comparison threshold range (a3, b3). In the step S42, determining the threshold range of the driver's habit includes the following steps:

[0244] S421: Setting a time threshold t1, taking a vehicle start time node within a time interval of at least t1 as a time starting point;

[0245] S422: determining a collection period t2, starting from the time starting point, collecting data of each dimension once every collection period t2, until n time data are collected, forming a driving habit data packet for one start;

[0246] S423: Analyze the data of n time periods to form a time threshold range for a single start;

[0247] S424: Repeat steps S421 to S423 until the driving habit data packets of m startups are collected, and the m startup time threshold ranges are analyzed, and then the average of the m times is taken.

[0248] The step S423 includes the following steps:

[0249] S4231: Calculate the mean σ and variance S of n data respectively 2 , maximum value max and minimum value min, A i is the data value;

[0250] S4232: According to the variance S 2 , the maximum value max and the minimum value min, determine the driving habit threshold range as (min-||S 2 ||σ,max+||S 2 ||σ), where ||S 2 || represents S 2 Scaled to the interval [0,1].

[0251] In the step S4232, S 2 The method of scaling to the interval [0,1] is: determine S 2 The number of digits w, S 2 The decimal point is moved forward by w places.

[0252] In the step S43, k'1 and k'2 are correlation coefficients, respectively.

[0253] In the step S43, a' and b' are the error coefficients of the lower limit and the upper limit respectively.

[0254] In the step S6, safety intervention includes braking, steering, tightening seat belts, and automatic parking.

[0255] Embodiment 11:

[0256] Vehicle driving abnormal behavior safety intervention system, which includes:

[0257] A data acquisition module, which is used to collect driving data;

[0258] A data analysis module, which is used to analyze the collected data;

[0259] A control module is used to control the vehicle according to the result of data analysis.

[0260] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A method for safely intervening in abnormal vehicle driving behavior, characterized in that: The following steps are involved: S1: Determine the evaluation dimensions of abnormal driving monitoring; S2: Obtain the threshold range of each dimension under normal driving conditions through big data; S3: learning based on the driver's driving habits to obtain the threshold ranges of various dimensions under the driver's normal driving state; S4: comprehensively determining a comparison threshold range based on the threshold ranges of various dimensions obtained from the big data and the threshold ranges of various dimensions of the learned driver habits; S5: While the vehicle is being driven, the actual value of each dimension is monitored in real time, and the actual value is compared with the threshold range; S51: If the threshold value is not exceeded, it is determined to be normal driving; S52: If the threshold value is exceeded, abnormal driving is determined; S6: Combines the vehicle's camera and radar to take over the vehicle and issue an alarm through the vehicle's audio system. In the step S1, the evaluation dimensions include the fluctuation size, fluctuation speed and fluctuation frequency of the accelerator pedal, the fluctuation size, fluctuation speed and fluctuation frequency of the brake pedal, the rotation size, rotation speed and rotation frequency of the steering wheel, the continuous driving time of the vehicle and the driving speed of the vehicle; In the step S4, the method for comprehensively determining the comparison threshold range includes the following steps: S41: Determine the threshold range for big data acquisition ; S42: Determine the threshold range of the driver's habit ; S43: According to the formula Calculate and get the comparison threshold range ; In the step S43, , are the correlation coefficients, ; In the step S43, , are the error coefficients for the lower and upper limits, respectively, .

2. The method for safe intervention in abnormal vehicle driving behavior according to claim 1, characterized in that: In the step S42, determining the threshold range of the driver's habits includes the following steps: S421: Setting a time threshold t1, taking a vehicle start time node within a time interval of at least t1 as a time starting point; S422: determining a collection period t2, starting from the time starting point, collecting data of each dimension once every collection period t2, until n time data are collected, forming a driving habit data packet for one start; S423: Analyze the data of n time periods to form a time threshold range for a single start; S424: Repeat steps S421 to S423 until the driving habit data packets of m startups are collected, and the m startup time threshold ranges are analyzed, and then the average of the m times is taken.

3. The method for safe intervention in abnormal vehicle driving behavior according to claim 2, characterized in that: The step S423 includes the following steps: S4231: Calculate the mean of n data separately , variance S 2 , maximum value max and minimum value min, ; is the data value; S4232: According to the variance S 2 , the maximum value max and the minimum value min, determine the driving habit threshold range ,in Representatives of S 2 Scaled to the interval [0,1].

4. The method for safe intervention in abnormal vehicle driving behavior according to claim 3, characterized in that: In the step S4232, S 2 The method of scaling to the interval [0,1] is: determine S 2 The number of digits w, S 2 The decimal point is moved forward by w places.

5. The method for safe intervention in abnormal vehicle driving behavior according to claim 1, characterized in that: In the step S6, safety intervention includes braking, steering, tightening seat belts, and automatic parking.

6. A vehicle driving abnormal behavior safety intervention system, which is used to implement the vehicle driving abnormal behavior safety intervention method according to any one of claims 1 to 5, characterized in that: include: A data acquisition module, which is used to collect driving data; A data analysis module, which is used to analyze the collected data; A control module is used to control the vehicle according to the result of data analysis.

Images