A driver distraction warning device and warning method for predicting braking of a preceding vehicle

By combining GPS, millimeter wave radar and image processing technology to predict the braking behavior of the front vehicle in the driving distraction warning device, the problem of the driver's sight in the congested road section is solved, and the early warning of the braking of the front vehicle is realized. Reduced the incidence of traffic accidents.

CN115384403BActive Publication Date: 2024-11-15CHANGAN UNIV
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202210372317.2
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-04-11
Publication Date
2024-11-15
Estimated Expiration
2042-04-11

AI Technical Summary

Technical Problem

In congested sections, drivers use their mobile phones to make their sight no longer in front of the road, which increases the probability of rear-end collision accidents. It is difficult for the existing technology to effectively predict the braking behavior of the vehicle in front and provide early warnings.

Method used

Using driving distraction warning devices that predict front vehicle braking, including vehicle GPS, vehicle motion parameter module, millimeter-wave radar, front vehicle image processing module, line of sight acquisition module and buzzer, predict front vehicle braking through data acquisition and analysis Time and deceleration and alarm when the driver's sight leaves the front of the road.

Benefits of technology

Effective warning of drivers reduces the probability of traffic accidents and improves road safety.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN115384403B_ABST
    Figure CN115384403B_ABST
Patent Text Reader

Abstract

The present invention relates to the technical field of traffic safety facilities, and in particular to a driver distraction warning device and a warning method for predicting the braking of a preceding vehicle. The present invention predicts the braking behavior of the preceding vehicle, and performs collision analysis in combination with the vehicle speed and the distance between the vehicle and the preceding vehicle, and warns the driver when the driver is at high risk of collision and his / her sight is not in front of the road, so as to alert the driver to return his / her sight to the front of the road and take safety measures, thereby greatly reducing the probability of traffic accidents.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of traffic safety facilities, and in particular to a driver distraction warning device and a warning method for predicting braking of a preceding vehicle. Background Art

[0002] On normally congested roads, since the vehicle speed is not high and vehicles are constantly stopping and waiting, it is becoming more and more common for drivers to use their mobile phones, which greatly increases the probability of traffic accidents (especially rear-end collisions). Summary of the invention

[0003] In view of the problems existing in the prior art, the purpose of the present invention is to provide a driver distraction warning device and warning method for predicting the braking of the vehicle in front, predicting the braking behavior of the vehicle in front, and performing collision analysis in combination with the speed of the vehicle itself and the distance between the vehicle itself and the vehicle in front, and warning the driver when the risk of collision is high and the line of sight is not in front of the road, alerting the driver to return the line of sight to the front of the road and take safety measures, thereby greatly reducing the probability of traffic accidents.

[0004] In order to achieve the above object, the present invention adopts the following technical solutions to achieve it.

[0005] Technical solution 1:

[0006] A driver distraction warning device for predicting braking of a preceding vehicle, comprising a vehicle-mounted GPS, a vehicle motion parameter module, a millimeter wave radar, a preceding vehicle image processing module, a sight acquisition module, a buzzer and an industrial computer;

[0007] The vehicle-mounted GPS is used to obtain the vehicle's positioning information, the vehicle motion parameter module is used to obtain the vehicle's speed and braking deceleration, and the millimeter-wave radar is used to collect the relative speed and relative distance between the front vehicle and the vehicle; the front vehicle image processing module is used to collect the image information of the front vehicle and identify the lighting of the front vehicle's brake lights based on the image information; the line of sight acquisition module is used to collect the driver's line of sight and identify whether the driver's line of sight has left the road ahead; the buzzer is used to warn the driver;

[0008] The industrial computer is used to receive and store the vehicle positioning information obtained by the vehicle-mounted GPS to determine whether the vehicle is in a serious congestion condition; the industrial computer is also used to predict the braking time and braking deceleration of the vehicle in front; the industrial computer is also used to control the buzzer alarm;

[0009] The industrial computer is respectively connected with the vehicle-mounted GPS, vehicle motion parameter module, millimeter wave radar, front vehicle image processing module, sight acquisition module and buzzer.

[0010] Technical solution 2:

[0011] A method for predicting the braking of a preceding vehicle while being distracted by driving is provided, based on a device for predicting the braking of a preceding vehicle while being distracted by driving, comprising the following steps:

[0012] Step 1: The vehicle is driving normally, and the vehicle GPS and vehicle motion parameter module continuously collect the positioning information and speed of the vehicle for 100 hours respectively;

[0013] Step 2: The industrial computer processes the speed and positioning information of the vehicle for 100 hours: the vehicle starts and stops four times or more within one minute, and the maximum speed does not exceed 30 kilometers per hour, as a severe congestion condition, and the corresponding road section is regarded as a severe congestion section;

[0014] Step 3: When the vehicle enters a severely congested section of road again and is in a severely congested condition, the vehicle motion parameter module obtains the speed and braking deceleration of the vehicle, the millimeter wave radar collects the relative speed Δv and relative distance between the front vehicle and the vehicle, and the front vehicle image processing module collects the image information of the front vehicle and identifies the moment when the brake lights of the front vehicle are turned on based on the image information;

[0015] Step 4: The industrial computer obtains the front vehicle speed v1=Δv+v according to the vehicle speed v and the front vehicle relative speed Δv; establishes a rectangular coordinate system with the horizontal coordinate being the time t and the vertical coordinate being the front vehicle speed v1; every The corresponding points of the front vehicle speed v1 are marked on the rectangular coordinate system, and all the points represent the changing relationship v1(t) between the front vehicle speed v1 and time t; the discrete points are connected by a smooth curve to obtain the curve of the front vehicle speed and time, and the curve is used to represent the changing relationship v1(t) between the front vehicle speed v1 and time t;

[0016] Step 5, collecting 20 groups of sample data, and establishing a braking moment prediction model and a braking deceleration prediction model under the 20 groups of samples;

[0017] Step 6: When the current vehicle starts again, the braking time prediction model is used to predict the time mt from the start to the braking of the current vehicle, and the braking deceleration prediction model is used to predict the braking deceleration a of the current vehicle;

[0018] After each new set of samples is obtained, all samples are used to re-establish the braking moment prediction model and the braking deceleration prediction model;

[0019] Step 7, establishing a collision model;

[0020] Step 8: When the collision time t′ between the vehicle and the front vehicle is less than or equal to the constant k, and the sight acquisition module recognizes that the driver's sight leaves the road ahead, the industrial computer controls the buzzer to warn the driver.

[0021] Compared with the prior art, the beneficial effects of the present invention are: predicting the braking behavior of the vehicle in front, and performing collision analysis in combination with the speed of the vehicle itself and the distance between the vehicle itself and the vehicle in front, and issuing early warnings to the driver when the risk of collision is high and the driver's line of sight is not in front of the road, alerting the driver to return his line of sight to the front of the road and take safety measures, thereby greatly reducing the probability of traffic accidents. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] The present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0023] Figure 1 A schematic diagram of the structure of a driver distraction warning device for predicting braking of a preceding vehicle according to the present invention;

[0024] Figure 2 This is a schematic diagram of the relationship between the speed v1 of the front vehicle and the time t in the distracted driving warning method for predicting the braking of the front vehicle of the present invention;

[0025] Figure 3 Schematic diagram of the relationship between the speed v1 of the preceding vehicle and time t in an embodiment of the distracted driving warning method for predicting braking of the preceding vehicle of the present invention. DETAILED DESCRIPTION

[0026] The embodiments of the present invention will be described in detail below with reference to examples, but those skilled in the art will understand that the following examples are only used to illustrate the present invention and should not be construed as limiting the scope of the present invention.

[0027] refer to Figure 1 , a driver distraction warning device for predicting braking of a preceding vehicle, comprising a vehicle-mounted GPS, a vehicle motion parameter module, a millimeter wave radar, a preceding vehicle image processing module, a sight acquisition module, a buzzer and an industrial computer;

[0028] The vehicle-mounted GPS is used to obtain the vehicle's positioning information, the vehicle motion parameter module is used to obtain the vehicle's speed and braking deceleration, and the millimeter-wave radar is used to collect the relative speed and relative distance between the front vehicle and the vehicle; the front vehicle image processing module is used to collect the image information of the front vehicle and identify the lighting of the front vehicle's brake lights based on the image information; the line of sight acquisition module is used to collect the driver's line of sight and identify whether the driver's line of sight has left the road ahead; the buzzer is used to warn the driver;

[0029] The industrial computer is used to receive and store the vehicle positioning information obtained by the vehicle-mounted GPS to determine whether the vehicle is in a serious congestion condition; the industrial computer is also used to predict the braking time and braking deceleration of the vehicle in front; the industrial computer is also used to control the buzzer alarm;

[0030] The industrial computer is respectively connected with the vehicle-mounted GPS, vehicle motion parameter module, millimeter wave radar, front vehicle image processing module, sight acquisition module and buzzer.

[0031] The preceding vehicle refers to the vehicle in front of the vehicle, closest to the vehicle and in the same lane; when the distance between the vehicle and the vehicle closest to the vehicle in front of the vehicle in the direction perpendicular to the lane is less than 2m, the two vehicles are in the same lane.

[0032] A method for predicting the braking of a preceding vehicle while being distracted by driving is provided, based on a device for predicting the braking of a preceding vehicle while being distracted by driving, comprising the following steps:

[0033] Step 1: The vehicle is driving normally, and the vehicle GPS and vehicle motion parameter module continuously collect the positioning information and speed of the vehicle for 100 hours respectively;

[0034] Step 2: The industrial computer processes the speed and positioning information of the vehicle for 100 hours: the vehicle starts and stops four times or more within one minute, and the maximum speed does not exceed 30 kilometers per hour, as a severe congestion condition, and the corresponding road section is regarded as a severe congestion section;

[0035] Step 3: When the vehicle enters a severely congested section of road again and is in a severely congested condition, the vehicle motion parameter module obtains the speed and braking deceleration of the vehicle, the millimeter wave radar collects the relative speed Δv and relative distance between the front vehicle and the vehicle, and the front vehicle image processing module collects the image information of the front vehicle and identifies the moment when the brake lights of the front vehicle are turned on based on the image information;

[0036] Identify whether the brake lights of the vehicle in front are on. Specifically, (1) segment the image into a binary image using the HSV color space and perform a Gaussian transform to obtain a dispersed grayscale value; (2) normalize the grayscale value and use it as a weighted proportional coefficient. Since this coefficient is relatively large in the taillight area detected by the HSV color threshold and is smaller in other areas, the area where the taillights are located is determined; (3) After the taillight area is detected in the image, the maximum inter-class variance method is used to segment the original image into foreground and background parts; (4) Morphological transformation is used to eliminate the internal holes in the taillight area in the image to obtain the contour of the taillight; (5) The number of contours is used to exclude the situation where the turn signal is on. When the number of contours obtained is 2, the situation where the turn signal is on can be excluded; (6) Double flash light status recognition: The physical characteristics of the taillight are used to eliminate the interfering contours to determine whether the vehicle has turned on the double flash lights, brake lights and fog lights. (7) For vehicles whose taillights are identified, a vehicle tracking algorithm is used to track the taillights and record the time when the taillights are turned on and off. The duration of the taillights being on can be used to exclude the possibility that the fog lights are on. The pattern of the taillights turning on and off can then be used to ultimately determine the condition that the brake lights are on.

[0037] Step 4: The industrial computer obtains the front vehicle speed v1 = Δv + v according to the vehicle speed v and the front vehicle relative speed Δv; Figure 2 , establish a rectangular coordinate system with the horizontal axis as time t and the vertical axis as the front vehicle speed v1; every The corresponding points of the front vehicle speed v1 are marked on the rectangular coordinate system, and all the points represent the changing relationship v1(t) between the front vehicle speed v1 and time t; the discrete points are connected by a smooth curve to obtain the curve of the front vehicle speed and time, and the curve is used to represent the changing relationship v1(t) between the front vehicle speed v1 and time t;

[0038] Step 5, collecting 20 groups of sample data, and establishing a braking moment prediction model and a braking deceleration prediction model under the 20 groups of samples;

[0039] refer to Figure 3 , the sample specifically refers to: the front vehicle stops at time t Ei-1 Stop, then start at time t Bi To stop time t Ei A start-drive-stop process was performed. At the start time t Bi+1 To stop time t Ei+1 The start-drive-stop process was repeated again;

[0040] When the following three conditions are met:

[0041] ①t Bi -t Ei ≥3s, t Bi+1 -t Ei+1 ≥3s;

[0042] ②t∈(t Ei-1 ,t Bi )∪(t Ei ,t Bi+1 ), v1(t)=0km / h;

[0043] ③t∈(t Bi , t Ei ), v1(t)≤30km / h;

[0044] Then t∈(t Bi , t Ei ) as a group of samples, where i is the sequence number, i = 1, 2,…, n.

[0045] Specifically, a braking moment prediction model under group i samples is established, and the sub-steps are:

[0046] Sub-step 5.11, let the braking time of the front vehicle in the i-th sample be t li , then the time from the start to the braking of the front vehicle is mt i =t li -t Bi ; Get the time mt1, mt2, ..., mt from the start to the braking of the preceding vehicle of the first to the i-th group of samples respectively i ;

[0047] Sub-step 5.12, get the average time from starting to braking of the front vehicle under group i samples

[0048]

[0049] Sub-step 5.13, respectively obtain the first group of samples in front of the car at t B1 +0.01mt′ time, t B1 +0.02mt′, …, t B1 The speed corresponding to the moment of +0.85mt′

[0050] Get the number of samples in the second group at t B2 +0.01mt′ time, t B2 +0.02mt′, …, t B2 The speed corresponding to the moment of +0.85mt′

[0051] …

[0052] Get the front car in the i-th group of samples at t Bi +0.01mt′ time, t Bi +0.02mt′, …, t Bi The speed corresponding to the moment of +0.85mt′

[0053] Sub-step 5.14, with the samples from group 1 to group i The input is the time mt from the start to the braking of the front vehicle, and the output is the time mt. The BP neural network is used for multivariate nonlinear fitting to establish a braking moment prediction model.

[0054] Specifically, a braking deceleration prediction model under group i samples is established, and the sub-steps are:

[0055] Sub-step 5.21, let the braking time of the front vehicle in the i-th sample be t li The front vehicle's braking speed is The time from the start to the braking of the front vehicle is mt i =t li -t Bi ; Get the time mt1, mt2, ..., mt from the start to the braking of the preceding vehicle of the first to the i-th group of samples respectively i ;

[0056] Sub-step 5.22, obtain the braking deceleration of the front vehicle in the first group of samples

[0057] Get the braking deceleration of the front vehicle in the second group of samples

[0058] …

[0059] Get the braking deceleration of the front vehicle in the i-th group of samples

[0060] Sub-step 5.23, the time mt from the start to the braking of the front vehicle of the first group to the i-th group of samples and the speed of the front vehicle at the braking moment The braking deceleration prediction model is established by using BP neural network for multivariate nonlinear fitting.

[0061] Step 6: When the current vehicle starts again, the braking time prediction model is used to predict the time mt from the start to the braking of the current vehicle, and the braking deceleration prediction model is used to predict the braking deceleration a of the current vehicle;

[0062] After each new set of samples is obtained, all samples are used to re-establish the braking moment prediction model and the braking deceleration prediction model;

[0063] After the model is re-established, when the front vehicle starts again, the re-established braking moment prediction model and braking deceleration prediction model are used to re-predict the braking moment and braking deceleration of the front vehicle respectively.

[0064] Step 7, establishing a collision model;

[0065] Specifically, let the collision time between the self-vehicle and the front vehicle be t′. When t′<mt, the front vehicle has not braked yet, then the front vehicle travels a distance d1=v1t′, and the self-vehicle travels a distance

[0066] When t′≥mt, the front vehicle brakes at time mt, then the distance traveled by the front vehicle is Distance travelled by car

[0067] Where v1 is the speed of the front vehicle, v is the speed of the own vehicle, mt is the time from the start of the front vehicle to the braking, a′ is the braking deceleration of the own vehicle, and a is the braking deceleration of the front vehicle;

[0068] When d1+Δd=d, the vehicle collides with the vehicle in front, and the simultaneous equations are used to solve t′; where Δd is the relative distance between the vehicle and the vehicle in front.

[0069] Step 8: When the collision time t′ between the vehicle and the front vehicle is less than or equal to the constant k, and the sight acquisition module recognizes that the driver's sight leaves the road ahead, the industrial computer controls the buzzer to warn the driver.

[0070] Specifically, k is a constant, which is determined by the maximum braking deceleration of the vehicle, the driver's reaction time and the driving style. The larger the value, the more conservative the driving style.

[0071] If the driver's reaction time is taken as 1s, it takes at least 1s for the car to slow down from the maximum speed of 30km / h on a congested road to 0, so k=2s. k is also affected by the driving style of the driver.

[0072] Although the present invention has been described in detail in general terms and in specific embodiments in this specification, it is obvious to those skilled in the art that some modifications or improvements may be made to the present invention. Therefore, these modifications or improvements made without departing from the spirit of the present invention all fall within the scope of protection claimed by the present invention.

Claims

1. A driver distraction warning method for predicting braking of the vehicle ahead, It is characterized in that A driver distraction warning device for predicting braking of a preceding vehicle is provided, wherein the driver distraction warning device comprises a vehicle-mounted GPS, a vehicle motion parameter module, a millimeter wave radar, a preceding vehicle image processing module, a sight acquisition module, a buzzer and an industrial computer; The vehicle-mounted GPS is used to obtain the positioning information of the vehicle, the vehicle motion parameter module is used to obtain the speed and braking deceleration of the vehicle, and the millimeter wave radar is used to collect the relative speed and relative distance between the front vehicle and the vehicle; the front vehicle image processing module is used to collect the image information of the front vehicle and identify the lighting of the front vehicle's brake lights based on the image information; the line of sight acquisition module is used to collect the driver's line of sight and identify whether the driver's line of sight has left the road ahead; The buzzer is used to warn the driver; The industrial computer is used to receive and store the vehicle positioning information obtained by the vehicle-mounted GPS to determine whether the vehicle is in a serious congestion condition; the industrial computer is also used to predict the braking time and braking deceleration of the vehicle in front; the industrial computer is also used to control the buzzer alarm; The industrial computer is connected to the vehicle GPS, vehicle motion parameter module, millimeter wave radar, front vehicle image processing module, sight acquisition module and buzzer respectively; The driving distraction warning method comprises the following steps: Step 1: The vehicle is driving normally, and the vehicle GPS and vehicle motion parameter module continuously collect the positioning information and speed of the vehicle for 100 hours respectively; Step 2: The industrial computer processes the speed and positioning information of the vehicle for 100 hours: the vehicle starts and stops four times or more within one minute, and the maximum speed does not exceed 30 kilometers per hour, as a severe congestion condition, and the corresponding road section is regarded as a severe congestion section; Step 3: When the vehicle enters a severely congested section of road again and is in a severely congested condition, the vehicle motion parameter module obtains the speed and braking deceleration of the vehicle, the millimeter wave radar collects the relative speed Δv and relative distance between the front vehicle and the vehicle, and the front vehicle image processing module collects the image information of the front vehicle and identifies the moment when the brake lights of the front vehicle are turned on based on the image information; Step 4: The industrial computer obtains the front vehicle speed v according to the vehicle speed v and the front vehicle relative speed Δv 1 =Δv+v; establish a system with the horizontal axis as time t and the vertical axis as the speed v of the preceding vehicle 1 The rectangular coordinate system; every Mark the speed v of the preceding vehicle on this rectangular coordinate system 1 The corresponding points of all points represent the speed v of the front vehicle. 1 The relationship between the change v and time t 1 (t); Connect the discrete points through a smooth curve to obtain the curve of the front vehicle speed and time, and use this curve to represent the front vehicle speed v 1 The relationship between the change v and time t 1 (t); Step 5, collect 20 sets of sample data, and establish a braking moment prediction model and a braking deceleration prediction model under 20 sets of samples; the samples in step 5, specifically, the front vehicle stops at time t Ei-1 Stop, then start at time t Bi To stop time t Ei A start-drive-stop process was performed. At the start time t Bi+1 To stop time t Ei+1 The start-drive-stop process was repeated again; When the following three conditions are met: ①t Bi -t Ei ≥3s,t Bi+1 -t Ei+1 ≥3s; ②When t ∈ (t Ei-1 , t Bi ) ∪ (t Ei , t Bi+1 ), v 1 (t) = 0 km / h; ③t∈(t Bi , t Ei ), v 1 (t)≤30km / h; Then t∈(t Bi , t Ei ) 1 (t) as a set of samples, where i is the sequence number, i = 1, 2, ..., n; In step 5, the sub-steps of establishing the braking moment prediction model are as follows: Sub-step 5.11, let the braking time of the front vehicle in the i-th sample be t li , then the time from the start to the braking of the front vehicle is mt i =t li -t Bi ; Get the time mt from starting to braking of the front vehicle of the first to the i-th group of samples respectively 1 、mt 2 ,…,mt i ; Sub-step 5.12, get the average time from starting to braking of the front vehicle under group i samples Sub-step 5.13, respectively obtain the first group of samples in front of the car at t B1 +0.01mt′ time, t B1 +0.02mt′, …, t B1 The speed corresponding to the moment of +0.85mt′ Get the number of samples in the second group at t B2 +0.01mt′ time, t B2 +0.02mt′, …, t B2 The speed corresponding to the moment of +0.85mt′ …… Get the front car in the i-th group of samples at t Bi +0.01mt′ time, t Bi +0.02mt′, …, t Bi The speed corresponding to the moment of +0.85mt′ Sub-step 5.14, with the samples from group 1 to group i The input is the time mt from the start to the braking of the preceding vehicle, and the output is the multivariate nonlinear fitting by BP neural network to establish the braking moment prediction model. In step 5, the sub-steps of establishing the braking deceleration prediction model are as follows: Sub-step 5.21, let the braking time of the front vehicle in the i-th sample be t li The speed of the front vehicle at the time of braking is The time from the start to the braking of the front vehicle is mt i =t li -t Bi ; Get the time mt from starting to braking of the front vehicle of the first to the i-th group of samples respectively 1 、mt 2 ,…,mt i ; Sub-step 5.22, obtain the braking deceleration of the front vehicle in the first group of samples Get the braking deceleration of the front vehicle in the second group of samples …… Get the braking deceleration of the front vehicle in the i-th group of samples Sub-step 5.23, the time mt from the start to the braking of the front vehicle of the first group to the i-th group of samples and the speed of the front vehicle at the braking moment The input is the braking deceleration a of the preceding vehicle, and the output is the multivariate nonlinear fitting using BP neural network to establish the braking deceleration prediction model. Step 6: When the current vehicle starts again, the braking time prediction model is used to predict the time mt from the start to the braking of the current vehicle, and the braking deceleration prediction model is used to predict the braking deceleration a of the current vehicle; After each new set of samples is obtained, all samples are used to re-establish the braking moment prediction model and the braking deceleration prediction model; Step 7, establish a collision model; specifically, assume that the collision time between the vehicle and the front vehicle is t′, when t′<mt, the front vehicle has not braked yet, then the front vehicle travels a distance d 1 =v 1 t′, the distance traveled by the vehicle When t′≥mt, the front vehicle brakes at time mt, then the distance traveled by the front vehicle is Distance travelled by car Among them, v 1 is the speed of the leading vehicle, v is the speed of the own vehicle, mt is the time from the leading vehicle starting to braking, a′ is the braking deceleration of the own vehicle, and a is the braking deceleration of the leading vehicle; When 1 +Δd=d, the ego vehicle collides with the front vehicle, and the simultaneous equations are solved for t′; where Δd is the relative distance between the ego vehicle and the front vehicle; Step 8: When the collision time t′ between the vehicle and the front vehicle is less than or equal to the constant k, and the sight acquisition module recognizes that the driver's sight has left the road ahead, the industrial computer controls the buzzer to alert the driver; k is a constant, which is determined by the maximum braking deceleration of the vehicle, the driver's reaction time and the driving style. The larger the value, the more conservative the driving style.