CN106274483A - The Vehicular automatic driving switching device differentiated based on driving behavior of diverting one's attention and method - Google Patents

Abstract

The invention discloses a vehicle automatic driving switching device and method based on distracted driving behavior discrimination. The device includes a detection unit and a control unit. The input end of the detection unit is connected with a speed sensor, a three-axis acceleration sensor, a steering wheel angle sensor, and a steering wheel grip force. Sensors, tactile sensors, ranging sensors, heart rate sensors, driving recorders and GPS, the control unit receives significant variable data and time data of the timer, and establishes the dependent variable of the probability of autonomous driving of the vehicle and the independent variable of distracted driving behavior indicators based on the logistic regression model The relationship between the vehicle automatic driving probability is calculated; through the judgment conditions 1 and 2 of the condition judgment program, it is judged whether the given condition is met. If the condition 1 is met, the automatic early warning control switch is triggered. If the condition 2 is met, the vehicle automatically Driving control switch.

Description

Vehicle automatic driving switching device and method based on distracted driving behavior discrimination

technical field

The invention relates to the field of automotive auxiliary safety electronic equipment, in particular to a vehicle automatic driving switching device and method based on distracted driving behavior discrimination.

Background technique

With the rapid development of social economy, the number of motor vehicles has also increased rapidly, the contradiction between traffic supply and demand has become increasingly prominent, and the traffic pressure is also increasing. Various bad driving behaviors lead to more and more traffic accidents. The behavior of "distracted driving" is particularly prominent. "Distracted driving" includes answering and calling while driving, playing mobile phone WeChat, smoking, looking down to pick up things, etc. These behaviors directly lead to the decline of the driver's perception, reaction and operation ability, and indirectly lead to road damage. The decline in traffic safety and efficiency, but these hazards are often ignored by drivers, who do not know that this has become a major "invisible killer" of traffic safety. The lack of concentration caused by the driver's sight deviation or distraction is a common and important cause of traffic accidents, especially in rear-end collisions. Therefore, it is necessary to monitor and monitor distracted driving behaviors in time. control intervention.

For the monitoring and intervention of distracted driving behavior, relevant personnel have done some research before. Toyota Corporation of Japan (2006) developed the Driver Monitoring System (Driver Monitoring System), which monitors the driver's head orientation based on machine vision. , for the same pre-collision scene, different levels of early warning or active braking are carried out according to the driver's attention status; the main function of the driver's attention assistance system (AttentionAssist) developed by Mercedes-Benz in Germany is to monitor driving low concentration and reminded to rest. Mainly use the steering wheel angle as a behavior indicator, integrate vehicle speed, horizontal and vertical acceleration, pedal signals, external environmental information such as road conditions and cross wind, continuous driving time, and the use of the infotainment system in the car to comprehensively judge the driver's current attention state ;Based on some similar studies above, most of them are limited to focusing on the monitoring and identification of distracted driving behaviors, and there are few studies on distracted driving interventions that mainly use control methods to suppress driver distraction or Drivers are reminded of dangerous behaviors through early warning systems, but few studies have proposed specific improvement plans for safety problems caused by distracted driving.

With the continuous reform and innovation of technology, self-driving technology is also developing rapidly. Self-driving cars have become the focus of contemporary car development. Self-driving cars integrate various advanced technologies to allow cars to drive safely without a driver. , increased road safety and improved road efficiency, and reduced the traffic accident rate of driver's risky behavior, but this advanced technology is still not put into use when it comes to safety issues. At present, it is necessary to apply automatic driving technology to the monitoring and prevention of distracted driving, which can effectively monitor and identify the driver's distracted driving situation, and avoid traffic accidents caused by distracted driving behavior by switching the automatic driving mode, effectively Reduce the harm caused by distracted driving.

Contents of the invention

The purpose of the present invention is to provide a vehicle automatic driving switching device and method based on distracted driving behavior discrimination, so as to realize the monitoring and prevention of distracted driving behavior.

In order to achieve the above object, the technical scheme adopted in the present invention is:

The vehicle automatic driving switching device based on distracted driving behavior discrimination is characterized in that it includes a detection unit and a control unit, and the input end of the detection unit is connected with a speed sensor, a three-axis acceleration sensor, a steering wheel angle sensor, a steering wheel grip sensor, a tactile sensor, The distance measuring sensor, heart rate sensor, driving recorder and GPS, the output end of the detection unit is connected with the input end of the control unit, the output end of the control unit is connected with the vehicle automatic driving control switch and the automatic early warning switch, and the control unit is also connected with a timer, It is triggered by the control unit and the timing time is fed back to the control unit in real time for timing when the distracted driving behavior judgment reaches a given threshold.

The vehicle automatic driving switching device based on distracted driving behavior discrimination is characterized in that: the speed sensor is installed on the accelerator pedal to detect the instantaneous speed of the vehicle, and transmit the measured speed to the detection unit;

The three-axis acceleration sensor is installed at the geometric center of the vehicle body, and is used to detect the actual instantaneous lateral acceleration and longitudinal acceleration of the vehicle, and transmit the measured acceleration to the detection unit;

The steering wheel angle sensor is installed on the steering wheel of the car, and is used to detect the steering wheel angle and the steering wheel speed, and transmit the measured angle and speed to the detection unit;

The steering wheel grip force sensor is installed on the steering wheel of the car, used to detect the grip force of the steering wheel, and transmit the measured grip force to the detection unit;

The tactile sensor is installed on the steering wheel of the car, and is used to detect the time when the driver leaves the steering wheel once and the number of times the driver leaves the steering wheel per unit time, and transmits the measured time and times to the detection unit;

The distance measuring sensor is installed on the front side of the car body, near the bumper, and is used to detect the distance between the vehicle and the adjacent vehicle in front, that is, the following distance, and transmit the measured distance to the detection unit;

The heart rate sensor is installed under the driver's seat and is used to detect changes in the driver's heart rate and transmit the measured heart rate to the detection unit;

The driving recorder and the GPS are located in the car, and are used to detect the distance of the vehicle from the center line of the lane, and transmit the measured distance to the detection unit;

The detection unit is composed of a single-chip microcomputer, the detection unit stores various behavior indicators of the driver in the normal driving state, and receives various behavior indicators of the driver in the distracted driving state through the input terminal, which is different from the various behavior indicators in the normal driving state. Compared with various behavior indicators, a distracted driving detection model is established, and the indicator variable data with significant differences are extracted and sent to the control unit.

The vehicle automatic driving switching device based on distracted driving behavior discrimination is characterized in that the vehicle automatic driving control switch is an integrated manual and automatic control switch.

A vehicle automatic driving switching method based on distracted driving behavior discrimination, characterized in that: the control unit receives indicator variable data and timer time data from the detection unit, and establishes a vehicle automatic driving behavior discrimination based on distracted driving behavior based on a logistic regression model. The probabilistic model of driving, that is, based on distracted driving behavior indicators: vehicle instantaneous speed, vehicle instantaneous acceleration, steering wheel angle, steering wheel speed, steering wheel grip, time to leave the steering wheel once, number of times to leave the steering wheel per unit time, following distance, heart rate and vehicle The distance away from the center line of the lane establishes the vehicle automatic driving probability model, and calculates the vehicle automatic driving probability; then enters the condition judgment program, there are 2 judgment conditions, judgment condition 1: judge whether the real-time automatic driving probability of the vehicle reaches the preset value of the control unit Whether the automatic warning threshold and the duration reach a given time, if the conditions are met, trigger the automatic warning control switch to provide voice warning function; Judgment condition 2: If the driver does not give a response at this time, judge again whether the real-time vehicle automatic driving probability is Whether the vehicle automatic driving threshold preset by the control unit is reached and the duration reaches a given time. If the condition is met, the vehicle automatic driving control switch will be triggered, and the vehicle will drive automatically. The automatic early warning threshold is lower than the automatic driving threshold, and the automatic early warning The level is one level lower than the vehicle's automatic driving level.

The above-mentioned method for switching automatic driving of a vehicle based on the discrimination of distracted driving behavior is characterized in that: in the control unit, the number of times X ₁ of leaving the steering wheel, the number of times of leaving the steering wheel per unit time X ₂ , and the instantaneous speed of the vehicle collected in the period T ₁ X ₃ , vehicle instantaneous acceleration X ₄ , steering wheel angle X ₅ , steering wheel speed X ₆ , steering wheel grip X ₇ , following distance X ₈ , heart rate X ₉ , and vehicle distance from the center line of the lane X ₁₀ are used as independent variables, and the vehicle drives automatically The probability of is used as the dependent variable P, and its value range is (0,1). Based on the logistic regression model, the probability relationship of automatic driving of the vehicle is established as shown in formula (1):

${\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}\mathrm{<span\; class="notranslate">\; j</span>}}<span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; exp</span>}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; \&beta;</span>}}_{\mathrm{<span\; class="notranslate">\; 0</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; \&Sigma;\&beta;</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}\mathrm{<span\; class="notranslate">\; j</span>}}{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}\mathrm{<span\; class="notranslate">\; j</span>}}<span\; class="notranslate">\; )</span>}{\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; exp</span>}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; \&beta;</span>}}_{\mathrm{<span\; class="notranslate">\; 0</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; \&Sigma;\&beta;</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}\mathrm{<span\; class="notranslate">\; j</span>}}{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}\mathrm{<span\; class="notranslate">\; j</span>}}<span\; class="notranslate">\; )</span>}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; ,</span>$

In formula (1), i=1, 2, ..., n; j = 1, 2, ..., 10; P _ij is the automatic driving probability of the i-th driver under j distracted driving behaviors; _β0 is is a constant term, based on multiple experiments calibration; β _ij is the partial regression coefficient of the model.

The above-mentioned method for switching automatic driving of a vehicle based on discrimination of distracted driving behavior is characterized in that: said judgment condition 2: whether the real-time probability of automatic driving of the vehicle reaches the threshold value of automatic driving of the vehicle preset by the control unit and whether the duration is When the given time is reached, the vehicle automatic driving threshold P0 value is shown in formula ( ₂ ):

${\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; 0</span>}}<span\; class="notranslate">\; =</span>\frac{{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}^{\mathrm{<span\; class="notranslate">\; no</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}^{\mathrm{<span\; class="notranslate">\; 10</span>}}{\mathrm{<span\; class="notranslate">\; \&beta;</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}\mathrm{<span\; class="notranslate">\; j</span>}}{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}\mathrm{<span\; class="notranslate">\; j</span>}}}{\mathrm{<span\; class="notranslate">\; 10</span>}\mathrm{<span\; class="notranslate">\; no</span>}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 2</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; .</span>$

The described method for switching automatic driving of a vehicle based on the judgment of distracted driving behavior is characterized in that: the judgment condition 1: whether the real-time probability of automatic driving of the vehicle reaches the automatic warning threshold preset by the control unit and whether the duration reaches In a given time, the automatic warning threshold P1 value is shown in formula ( ₃ ):

${\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; \&alpha;</span>}<span\; class="notranslate">\; \&times;</span>\frac{{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}^{\mathrm{<span\; class="notranslate">\; no</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}^{\mathrm{<span\; class="notranslate">\; 10</span>}}{\mathrm{<span\; class="notranslate">\; \&beta;</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}\mathrm{<span\; class="notranslate">\; j</span>}}{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}\mathrm{<span\; class="notranslate">\; j</span>}}}{\mathrm{<span\; class="notranslate">\; 10</span>}\mathrm{<span\; class="notranslate">\; no</span>}}<span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; \&alpha;P</span>}}_{\mathrm{<span\; class="notranslate">\; 0</span>}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 3</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; .</span>$

In formula (3), α is a proportional coefficient, 0≤α≤1.

The described method for switching automatic driving of a vehicle based on the judgment of distracted driving behavior is characterized in that: in the judgment condition 2, it is judged whether the real-time automatic driving probability of the vehicle reaches the threshold value of the automatic driving of the vehicle preset by the control unit and the duration is Whether the given time is reached, and whether the real-time automatic driving probability of the vehicle in judgment condition 1 reaches the automatic warning threshold preset by the control unit and whether the duration reaches the given time, the given time is t, where t has passed several times The actual vehicle driving experiment is obtained according to the influence degree of the driver's distraction time on the accident.

Compared with the prior art, the beneficial effects of the present invention are reflected in:

1. The present invention applies advanced automatic driving technology to distracted driving. The method is novel and unique, effectively integrates the advanced technology of modern vehicles, and increases traffic safety.

2. On the basis of the existing research, the present invention determines effective indicators reflecting the distracted driving behavior by investigating and analyzing the general driving behavior of the driver under the distracted state, thereby ensuring the comprehensiveness and reliability of the monitoring.

3. The present invention monitors the driver's distracted driving behavior in real time, and the monitoring process takes a short time and can transmit the driver's behavior index in time.

4. The detection unit of the present invention utilizes the behavior index under the normal driving state obtained on the basis of the existing research and through multiple real vehicle simulation experiments and investigations, and the real-time monitored index under the distracted driving state For comparison, the indicator variables with significant differences are determined to be input to the control unit to prevent individual differences from affecting the test results.

5. The present invention analyzes the probability of automatic driving of the vehicle based on the logistic regression model, and judges whether the vehicle switches the automatic driving mode by analyzing 10 different distracted driving behavior indicators. In the logistic regression model, the threshold is set through the regression model It is determined by the average value of 1-10 indicators. The method is novel and scientific, and can accurately judge distracted driving behavior in real time.

6. The present invention considers the degree of influence of distracted driving on traffic safety. When distracted driving has a relatively low impact on traffic safety, it triggers an automatic early warning control switch with a relatively low level, provides a voice warning function, and gives the driver a vigilant effect; when distracted driving When the behavior has a serious impact on traffic safety, the higher-level vehicle automatic driving control switch is triggered, and the vehicle drives automatically. This can give distracted drivers a buffer period, not only can warn the driver, but also can give the driver the autonomy to choose the vehicle's automatic driving, and can intervene well in distracted driving behavior.

7. The vehicle automatic driving control switch of the present invention is a manual and automatic control integrated switch, which gives the driver a certain degree of autonomy. When the driver needs it, he can manually turn on the vehicle automatic driving control switch, and the vehicle drives automatically. The design is humanized. It is convenient for drivers and effectively increases traffic safety.

Description of drawings

FIG. 1 is a structural principle diagram of a device for switching to vehicle automatic driving based on distracted driving behavior discrimination according to the present invention.

Fig. 2 is a flow chart of condition judgment in the control unit of the present invention.

detailed description

A device for switching to vehicle automatic driving based on distracted driving behavior discrimination, including a steering wheel grip sensor 1, a three-axis acceleration sensor 2, a speed sensor 3, a steering wheel angle sensor 4, a tactile sensor 5, a distance measuring sensor 6, and a driving recorder And GPS7, heart rate sensor 8, detection unit 9, control unit 10, timer 11, automatic warning control switch 12, vehicle automatic driving control switch 13. The input end of detection unit 9 is connected with steering wheel grip sensor 1, triaxial acceleration sensor 2, speed sensor 3, steering wheel sensor 4, tactile sensor 5, ranging sensor 6, driving recorder and GPS 7, heart rate sensor 8, and the output end of detection unit 9 It is connected with the input end of the control unit 10, the output end of the control unit 10 is connected with the automatic early warning control switch 12 and the vehicle automatic driving control switch 13, and the input end of the control unit 10 is also connected with the timer 11, which is triggered by the control unit and fed back the timing time in real time For the control unit, it is used for timing when the distracted driving behavior judgment reaches a given threshold.

Steering wheel grip force sensor 1 is installed on the automobile steering wheel, is used for detecting the grip force of steering wheel;

The three-axis acceleration sensor 2 is installed at the geometric center of the vehicle body to detect the actual instantaneous lateral acceleration and longitudinal acceleration of the vehicle;

The speed sensor 3 is installed on the accelerator pedal to detect the instantaneous speed of the vehicle; the steering wheel angle sensor 4 is installed on the steering wheel of the vehicle to detect the steering wheel angle and the steering wheel speed;

The tactile sensor 5 is installed on the automobile steering wheel, and is used to detect the time when the driver leaves the steering wheel once and the number of times the driver leaves the steering wheel per unit time;

The ranging sensor 6 is installed on the front side of the car body, near the bumper, and is used to detect the distance between the vehicle and the adjacent vehicle ahead, that is, the following distance;

The driving recorder and GPS 7 are located in the car, and are used to detect the distance of the vehicle from the center line of the lane; the heart rate sensor 8 is installed under the driver's seat, and is used to detect the change of the driver's heart rate;

The detection unit 9 is connected with the above-mentioned sensors and vehicle-mounted devices, and receives various behavior indicators under the distracted driving state through the input terminal. The detection unit 9 internally stores various behavior indicators under the normal driving state of the driver. By comparing distracted driving with normal driving behavior indicators, a distracted driving detection model is established, and index variable data with significant differences are extracted and sent to the control unit. The control unit 10 receives the indicator variable data and the time data of the timer from the detection unit, and establishes a probability model of vehicle automatic driving based on the discrimination of distracted driving behavior based on the logistic regression model, according to the distracted driving behavior indicators: vehicle instantaneous speed, vehicle instantaneous Acceleration, steering wheel angle, steering wheel speed, steering wheel grip, time to leave the steering wheel for a single time, number of times to leave the steering wheel per unit time, following distance, heart rate, and distance from the centerline of the vehicle to establish a vehicle automatic driving probability model and calculate the vehicle automatic driving probability , and then enter the condition judgment program, judgment condition 1: judge whether the real-time vehicle automatic driving probability reaches the automatic warning threshold preset by the control unit and whether the duration reaches a given time, if the condition is met, trigger the automatic warning control switch and provide voice Warning function; Judgment condition 2: If the driver does not give a response at this time, judge again whether the real-time vehicle automatic driving probability reaches the vehicle automatic driving threshold preset by the control unit and whether the duration reaches a given time. If the condition is met, then Trigger the vehicle's automatic driving control switch, and the vehicle will drive automatically. The automatic warning threshold here is lower than the automatic driving threshold, and the automatic warning level is one level lower than the vehicle's automatic driving level.

In the control unit 10 of the present invention, the times of leaving the steering wheel X ₁ , the number of times of leaving the steering wheel per unit time X ₂ , the vehicle instantaneous speed X ₃ , the vehicle instantaneous acceleration X ₄ , the steering wheel rotation angle X ₅ , the steering wheel rotation speed X _{6 ,} and Steering wheel grip strength X ₇ , following distance X ₈ , heart rate X ₉ , and distance from the vehicle lane centerline X ₁₀ are used as independent variables, and the probability of automatic driving of the vehicle is used as the dependent variable P, whose value range is (0, 1), based on The logistic regression model establishes the probability relational expression of vehicle automatic driving as shown in formula (1):

${\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}\mathrm{<span\; class="notranslate">\; j</span>}}<span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; exp</span>}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; \&beta;</span>}}_{\mathrm{<span\; class="notranslate">\; 0</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; \&Sigma;\&beta;</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}\mathrm{<span\; class="notranslate">\; j</span>}}{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}\mathrm{<span\; class="notranslate">\; j</span>}}<span\; class="notranslate">\; )</span>}{\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; exp</span>}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; \&beta;</span>}}_{\mathrm{<span\; class="notranslate">\; 0</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; \&Sigma;\&beta;</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}\mathrm{<span\; class="notranslate">\; j</span>}}{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}\mathrm{<span\; class="notranslate">\; j</span>}}<span\; class="notranslate">\; )</span>}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span>$

In the formula, i=1, 2, ..., n; j = 1, 2, ..., 10; P _ij is the automatic driving probability of the i-th driver under j kinds of distracted driving behaviors; _β0 is a constant term , based on multiple experimental calibrations; β _ij is the partial regression coefficient of the model;

Judgment condition 2 Whether the real-time probability of vehicle automatic driving reaches the threshold value of vehicle automatic driving preset by the control unit and whether the duration reaches a given time. The value of the vehicle automatic driving threshold P ₀ is repeatedly simulated and verified based on several real vehicle experiments The obtained vehicle automatic driving threshold value P0 is shown in formula ( ₂ ):

${\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; 0</span>}}<span\; class="notranslate">\; =</span>\frac{{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}^{\mathrm{<span\; class="notranslate">\; no</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}^{\mathrm{<span\; class="notranslate">\; 10</span>}}{\mathrm{<span\; class="notranslate">\; \&beta;</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}\mathrm{<span\; class="notranslate">\; j</span>}}{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}\mathrm{<span\; class="notranslate">\; j</span>}}}{\mathrm{<span\; class="notranslate">\; 10</span>}\mathrm{<span\; class="notranslate">\; no</span>}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 2</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; .</span>$

Judgment condition 1 Whether the real-time probability of automatic driving of the vehicle reaches the automatic warning threshold preset by the control unit and whether the duration reaches a given time, the value of the automatic warning threshold P1 is shown in formula ( ₃ )

${\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; \&alpha;</span>}<span\; class="notranslate">\; \&times;</span>\frac{{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}^{\mathrm{<span\; class="notranslate">\; no</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}^{\mathrm{<span\; class="notranslate">\; 10</span>}}{\mathrm{<span\; class="notranslate">\; \&beta;</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}\mathrm{<span\; class="notranslate">\; j</span>}}{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}\mathrm{<span\; class="notranslate">\; j</span>}}}{\mathrm{<span\; class="notranslate">\; 10</span>}\mathrm{<span\; class="notranslate">\; no</span>}}<span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; \&alpha;P</span>}}_{\mathrm{<span\; class="notranslate">\; 0</span>}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 3</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; ,</span>$

In the formula, α is a proportional coefficient, 0≤α≤1, and α is obtained through repeated fitting and verification based on the logistic regression model through several real vehicle driving experiments.

Judging whether the real-time vehicle automatic driving probability reaches the threshold value preset by the control unit and whether the duration reaches a given time, the given time is t, where t has been tested for several times in actual vehicle driving, according to the driver's distraction time. The extent of the impact of the accident is obtained.

In the present invention, the vehicle automatic driving control switch 13 is an integrated manual and automatic control switch.

The specific implementation process of the example of the present invention is as follows: Referring to Fig. 1 and Fig. 2, when the driver is in the normal driving state, the vehicle's speed sensor, triaxial acceleration sensor, steering wheel angle sensor, steering wheel grip sensor, tactile sensor, range sensor, The heart rate sensor, driving recorder and GPS device will collect the index data of the driver during normal driving and store it in the detection unit. When the driver is distracted, the collected distracted driving behavior index data will be input into the detection unit and The index data under normal driving is compared, a distracted driving detection model is established, and index variable data with significant differences are extracted and sent to the control unit. The automatic warning threshold and vehicle automatic driving threshold in the control unit are reliable values obtained through several real vehicle experiments, and the establishment of a logistic regression model for repeated simulation verification. The distracted driving behavior index is filtered by the detection unit to obtain the significant index variable input To the control unit, the control unit calculates the probability of automatic driving of the vehicle based on the logistic regression model, enters the condition judgment program, judgment condition 1: judge whether the real-time probability of automatic driving of the vehicle reaches the automatic warning threshold preset by the control unit and whether the duration reaches the given value Time, if the condition is satisfied, trigger the automatic warning control switch to provide voice warning function; Judgment condition 2: If the driver does not give a response at this time, judge again whether the real-time vehicle automatic driving probability reaches the vehicle automatic driving preset by the control unit Whether the threshold and the duration reach the given time, if the condition is satisfied, the vehicle automatic driving control switch will be triggered, and the vehicle will drive automatically. The vehicle driving control switch here can also be manually selected to open according to the needs of the driver.

The core of the present invention lies in: first: the combination of advanced vehicle automatic driving technology and distracted driving, the method is novel and unique, and has great prospects for future development; second, in the detection unit, the distracted driving behavior index data through real-time detection Compared with the normal driving behavior index data, a driving detection model is established to screen out significant index variables with relatively high accuracy; third, in the control unit, the driver’s distracted driving behavior index variable is used as an independent variable, and the probability of automatic driving of the vehicle is The dependent variable establishes a logistic regression model, and calculates the probability of automatic driving of the vehicle through the significant index variables output in the detection unit. The method is innovative; the fourth is to determine the automatic early warning described in the two judgment conditions of the present invention through a large number of real vehicle experiments The threshold and the vehicle automatic driving threshold have high reliability and credibility. Fifth, the use of the timer takes into account the impact of the driver's distraction time on traffic safety, which is relatively comprehensive.

Claims (8)

1. A vehicle automatic driving switching device based on distracted driving behavior discrimination, characterized in that: it includes a detection unit and a control unit, and the input end of the detection unit is connected with a speed sensor, a three-axis acceleration sensor, a steering wheel angle sensor, a steering wheel grip sensor, a tactile Sensors, ranging sensors, heart rate sensors, driving recorders and GPS, the output of the detection unit is connected to the input of the control unit, the output of the control unit is connected to the vehicle automatic driving control switch and the automatic early warning switch, and the control unit is also connected to the timing The controller is triggered by the control unit and feeds back the timing time to the control unit in real time for timing when the distracted driving behavior judgment reaches a given threshold. 2. The vehicle automatic driving switching device based on distracted driving behavior discrimination according to claim 1, characterized in that: the speed sensor is installed on the accelerator pedal for detecting the instantaneous speed of the vehicle and transmitting the measured speed to the detection unit; The three-axis acceleration sensor is installed at the geometric center of the vehicle body, and is used to detect the actual instantaneous lateral acceleration and longitudinal acceleration of the vehicle, and transmit the measured acceleration to the detection unit; The steering wheel angle sensor is installed on the steering wheel of the car, and is used to detect the steering wheel angle and the steering wheel speed, and transmit the measured angle and speed to the detection unit; The steering wheel grip force sensor is installed on the steering wheel of the car, used to detect the grip force of the steering wheel, and transmit the measured grip force to the detection unit; The tactile sensor is installed on the steering wheel of the car, and is used to detect the time when the driver leaves the steering wheel once and the number of times the driver leaves the steering wheel per unit time, and transmits the measured time and times to the detection unit; The distance measuring sensor is installed on the front side of the car body, near the bumper, and is used to detect the distance between the vehicle and the adjacent vehicle in front, that is, the following distance, and transmit the measured distance to the detection unit; The heart rate sensor is installed under the driver's seat and is used to detect changes in the driver's heart rate and transmit the measured heart rate to the detection unit; The driving recorder and the GPS are located in the car, and are used to detect the distance of the vehicle from the center line of the lane, and transmit the measured distance to the detection unit; The detection unit is composed of a single-chip microcomputer, the detection unit stores various behavior indicators of the driver in the normal driving state, and receives various behavior indicators of the driver in the distracted driving state through the input terminal, which is different from the various behavior indicators in the normal driving state. Compared with various behavior indicators, a distracted driving detection model is established, and the indicator variable data with significant differences are extracted and sent to the control unit. 3 . The vehicle automatic driving switching device based on distracted driving behavior discrimination according to claim 1 , wherein the vehicle automatic driving control switch is an integrated manual and automatic control switch. 4 . 4. A vehicle automatic driving switching method based on distracted driving behavior discrimination, characterized in that: the control unit receives the index variable data and the time data of the timer from the detection unit, and establishes a method based on the distracted driving behavior discrimination based on a logistic regression model. The probability model of vehicle automatic driving, that is, based on distracted driving behavior indicators: vehicle instantaneous speed, vehicle instantaneous acceleration, steering wheel angle, steering wheel speed, steering wheel grip, time to leave the steering wheel for a single time, number of times to leave the steering wheel per unit time, following distance, heart rate Establish a vehicle automatic driving probability model based on the distance from the vehicle’s deviation from the center line of the lane, and calculate the vehicle automatic driving probability; then enter the condition judgment program, there are 2 judgment conditions, judgment condition 1: judge whether the real-time automatic driving probability of the vehicle reaches the preset value of the control unit Whether the specified automatic warning threshold and the duration reach the given time, if the conditions are met, the automatic warning control switch will be triggered to provide the voice warning function; Judgment condition 2: If the driver does not give a response at this time, judge the real-time vehicle automatic driving again Whether the probability reaches the vehicle automatic driving threshold preset by the control unit and whether the duration reaches a given time. If the condition is met, the vehicle automatic driving control switch will be triggered, and the vehicle will drive automatically. The automatic warning threshold is lower than the automatic driving threshold. The automatic warning level is one level lower than the vehicle's automatic driving level. 5. A vehicle automatic driving switching method based on distracted driving behavior discrimination according to claim 4, characterized in that: in the control unit, the number of times X ₁ of leaving the steering wheel collected in the period T ₁ , the number of times of leaving the steering wheel per unit time X ₂ , vehicle instantaneous speed X ₃ , vehicle instantaneous acceleration X ₄ , steering wheel rotation angle X ₅ , steering wheel rotation speed X ₆ , steering wheel grip force X ₇ , following distance X ₈ , heart rate X ₉ and vehicle distance from the lane center line X ₁₀ as The independent variable, the probability of automatic driving of the vehicle is used as the dependent variable P, and its value range is (0,1). Based on the logistic regression model, the probability relationship of automatic driving of the vehicle is established as shown in formula (1): ${\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}\mathrm{<span\; class="notranslate">\; j</span>}}<span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; exp</span>}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; \&beta;</span>}}_{\mathrm{<span\; class="notranslate">\; 0</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; \&Sigma;\&beta;</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}\mathrm{<span\; class="notranslate">\; j</span>}}{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}\mathrm{<span\; class="notranslate">\; j</span>}}<span\; class="notranslate">\; )</span>}{\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; exp</span>}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; \&beta;</span>}}_{\mathrm{<span\; class="notranslate">\; 0</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; \&Sigma;\&beta;</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}\mathrm{<span\; class="notranslate">\; j</span>}}{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}\mathrm{<span\; class="notranslate">\; j</span>}}<span\; class="notranslate">\; )</span>}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; ,</span>$ In formula (1), i=1, 2, ..., n; j = 1, 2, ..., 10; P _ij is the automatic driving probability of the i-th driver under j distracted driving behaviors; _β0 is The constant term is calibrated based on multiple experiments; β _ij is the partial regression coefficient of the model. 6. A vehicle automatic driving switching method based on distracted driving behavior discrimination according to claim 4, characterized in that: said judgment condition 2: whether the real-time probability of vehicle automatic driving reaches the vehicle automatic driving preset preset by the control unit. Whether the driving threshold and the duration reach a given time, the vehicle automatic driving threshold P0 value is shown in formula ( ₂ ): ${\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; 0</span>}}<span\; class="notranslate">\; =</span>\frac{{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}^{\mathrm{<span\; class="notranslate">\; no</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}^{\mathrm{<span\; class="notranslate">\; 10</span>}}{\mathrm{<span\; class="notranslate">\; \&beta;</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}\mathrm{<span\; class="notranslate">\; j</span>}}{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}\mathrm{<span\; class="notranslate">\; j</span>}}}{\mathrm{<span\; class="notranslate">\; 10</span>}\mathrm{<span\; class="notranslate">\; no</span>}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 2</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; .</span>$ 7. A vehicle automatic driving switching method based on distracted driving behavior discrimination according to claim 4, characterized in that: said judgment condition 1: whether the real-time automatic driving probability of the vehicle reaches the automatic early warning preset by the control unit Threshold and whether the duration reaches a given time, the automatic warning threshold P1 value is shown in formula ( ₃ ): ${\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; \&alpha;</span>}<span\; class="notranslate">\; \&times;</span>\frac{{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}^{\mathrm{<span\; class="notranslate">\; no</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}^{\mathrm{<span\; class="notranslate">\; 10</span>}}{\mathrm{<span\; class="notranslate">\; \&beta;</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}\mathrm{<span\; class="notranslate">\; j</span>}}{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}\mathrm{<span\; class="notranslate">\; j</span>}}}{\mathrm{<span\; class="notranslate">\; 10</span>}\mathrm{<span\; class="notranslate">\; no</span>}}<span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; \&alpha;P</span>}}_{\mathrm{<span\; class="notranslate">\; 0</span>}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 3</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; .</span>$ In formula (3), α is a proportional coefficient, 0≤α≤1. 8. A vehicle automatic driving switching method based on distracted driving behavior discrimination according to claim 4, characterized in that: in said judging condition 2, it is judged whether the real-time automatic driving probability of the vehicle reaches the vehicle preset by the control unit. Whether the automatic driving threshold and the duration reach a given time, and whether the real-time automatic driving probability of the vehicle in judgment condition 1 reaches the automatic warning threshold preset by the control unit and whether the duration reaches a given time, the given time is t, The t here is obtained through several real vehicle driving experiments, and is obtained according to the degree of influence of the driver's distraction time on the accident.