1. A driving assistance system that predicts braking of a preceding vehicle, characterized by comprising: drive test equipment and vehicle-mounted equipment; the roadside equipment comprises a millimeter wave radar, an intelligent high-definition camera and an RSU intelligent roadside unit; the RSU intelligent road side unit comprises an edge calculation processing unit; the vehicle-mounted equipment comprises an OBU intelligent vehicle-mounted unit and a vehicle-mounted display screen; the OBU intelligent vehicle-mounted unit comprises a data storage unit, a processing unit and a wireless communication unit;

the millimeter wave radar is installed on the road side or the central isolation belt, faces the lane line direction, and is used for collecting motion data information of vehicles in the lane and transmitting the motion data information to the RSU intelligent road side unit; the motion data information of the vehicle comprises the speed of the vehicle and the distance between the head of the rear vehicle and the tail of the front vehicle;

the intelligent high-definition camera is arranged above the millimeter wave radar and used for shooting images of vehicles in a lane and transmitting the images to the RSU intelligent road side unit;

the RSU intelligent road side unit is installed at the road side of the road and used for transmitting the motion data information and the vehicle image of the vehicle to the edge calculation processing unit;

the edge calculation processing unit is used for extracting the speed of the vehicle, the distance between the head of the rear vehicle and the tail of the front vehicle and the moment when the vehicle brakes in the vehicle image in the motion data information and transmitting the speed, the distance and the moment to the OBU intelligent vehicle-mounted unit;

the OBU intelligent vehicle-mounted unit is arranged on the inner side of a front windshield of the automobile and used for transmitting the speed of the automobile, the distance between the head of a rear automobile and the tail of a front automobile and the moment when the automobile brakes in an automobile image to the data storage unit;

the data storage unit is used for storing the speed of the vehicle, the distance between the head of the rear vehicle and the tail of the front vehicle and the moment when the vehicle brakes in the vehicle image and transmitting the speed, the distance and the moment to the processing unit;

the processing unit is used for processing the speed of the vehicle, the distance between the head of the rear vehicle and the tail of the front vehicle and the moment when the vehicle brakes in the vehicle image and sending the processing result to the wireless communication unit;

the wireless communication unit is used for transmitting the processing result to the vehicle-mounted display screen;

the vehicle-mounted display screen is arranged near a vehicle instrument panel and used for displaying the braking information of the front vehicle to a driver of the vehicle.

2. The driving assistance system for predicting braking of a preceding vehicle according to claim 1, wherein the roadside apparatus is plural, and an interval between two adjacent roadside apparatuses is 200 m.

3. A driving assistance method of predicting a preceding vehicle brake, characterized by comprising the steps of:

step 1, detecting whether a vehicle team with multiple vehicle following behaviors exists on a lane by using a millimeter wave radar and an intelligent high-definition camera, and if so, turning to step 2;

step 2, when the brake lamp of any vehicle in the fleet is on, the braking time t, the speed v of the braking vehicle, the speed v of the rear vehicle of the braking vehicle and the distance l between the braking vehicle and the rear vehicle are obtained, and the braking time t of the ith vehicle is determined_iVelocity v_iSpeed of the (i +1) th vehicle at the time of braking of the ith vehicleDistance l between the i +1 th vehicle and the i-th vehicle_i-(i+1)And braking time t when the (i +1) th vehicle brakes_i+1And velocity v_i+1Calculating the maximum braking deceleration a of the ith vehicle_maxiThe following time interval D of the (i +1) th vehicle_i+1Braking reaction time T_{Reaction (i +1)}And maximum braking deceleration a_max(i+1)(ii) a The total number of the vehicles in the fleet is n, n is more than or equal to 3 and is a positive integer, i is more than or equal to 1 and less than or equal to n, and i is a positive integer;

step 3, acquiring a plurality of historical data, extracting the maximum braking deceleration of the front vehicle, the following time interval of the rear vehicle, the braking reaction time of the rear vehicle and the maximum braking deceleration of the rear vehicle in each historical data, establishing the historical data as a sample, forming a sample set by a plurality of samples, and storing the sample set in a storage unit;

when the brake lamp of the ith vehicle is on, the collected historical data corresponds to the sample (a)_maxi，D_i+1，T_{Reaction (i +1)}，a_max(i+1)) Wherein a is_maxiMaximum braking deceleration of the i-th vehicle, D_i+1The following time interval, T, of the i +1 th vehicle_{Reaction (i +1)}Brake reaction time of i +1 th vehicle, a_max(i+1)Maximum braking deceleration for the (i +1) th vehicle;

taking the maximum braking deceleration of the front vehicle and the following vehicle distance of the rear vehicle in each sample stored in the storage unit as the input of a BP neural network model, taking the braking reaction time of the rear vehicle and the maximum braking deceleration of the rear vehicle as the output of the BP neural network model, and training the BP neural network model to obtain a trained BP neural network model;

and 4, predicting the braking reaction time T of all the following vehicles when the vehicles in the fleet brake according to the trained BP neural network model_{Reaction of}And maximum braking deceleration a_maxAnd then according to the predicted braking reaction time T of all vehicles behind the braking vehicle in the fleet_{Reaction of}Calculating the braking response time T of the front vehicle_{Reaction of}And the brake information of the front vehicle is displayed to remind the self vehicle to make adjustment in time.

4. The driving assistance method for predicting the braking of the preceding vehicle according to claim 3, wherein in the step 1, the detecting whether the vehicle group with the multiple vehicle following behaviors exists on the lane is specifically: when the following distance between every two adjacent vehicles in the motorcade is smaller than 100 meters, judging that the following behavior exists between the vehicles, and judging that a plurality of continuous vehicles with the following behavior can be judged as the motorcade; otherwise, there is no following behavior.

5. The driving assistance method with predicted preceding vehicle braking according to claim 3, characterized in that in step 2, the maximum braking deceleration a of the i-th vehicle_maxiThe following time interval D of the (i +1) th vehicle_i+1Braking reaction time T_{Reaction (i +1)}And maximum braking deceleration a_max(i+1)The calculation processes of (a) are respectively as follows:

T_{reaction (i +1)}＝t_(i+1)-t_i

Where Δ t is set to 0.5 second, t in the present invention_iFor the braking moment of the ith vehicle in the fleet,. DELTA.v_iIs the speed difference of 0.5 second, delta v, of the ith braking vehicle in the fleet_i+1The speed difference of 0.5 second for the i +1 th braking vehicle in the fleet_i+1The braking time of the (i +1) th vehicle in the fleet, i_i-(i+1)The distance between the (i +1) th vehicle and the ith vehicle in the platoon,is the speed of the i +1 th vehicle at the time of braking of the i-th vehicle, a_maxiIs the maximum braking deceleration at the time of braking of the ith vehicle, a_max(i+1)Is the maximum braking deceleration at the time of braking of the (i +1) th vehicle.

6. The driving assistance method for predicting the braking of the preceding vehicle according to claim 3, wherein in the step 3, the BP neural network model is trained, specifically:

acquiring a training sample set and a testing sample set, training a BP neural network model through the training sample to obtain a trained BP neural network model, and testing the trained BP neural network model through the testing sample; when the average value of the difference values between the output of the BP neural network model and the label of the test sample is smaller than a preset error threshold value, determining that the training of the BP neural network model is finished;

the training sample set and the testing sample set are respectively selected from historical data, and each sample consists of the maximum braking deceleration of a front vehicle, the following time interval of a rear vehicle, the braking reaction time of the rear vehicle and the maximum braking deceleration of the rear vehicle; the test specimens are labeled brake reaction time of the following vehicle and maximum brake deceleration of the following vehicle.

7. The driving assistance method with predicted preceding vehicle braking according to claim 3, characterized in that in step 4, the braking reaction time T of the preceding vehicle is calculated_{Reaction of}The method specifically comprises the following steps:

T_{reaction (n-1)}＝T_{Reaction (i +1)}+T_{Reaction (i +2)}+……T_{Reaction (n-1)}

Wherein, T_{Reaction (n-1)}The brake reaction time T of the (n-1) th vehicle when the ith vehicle brakes in the fleet_{Reaction (i +1)}For the braking reaction time of the i +1 th vehicle in the fleet, T_{Reaction (i +2)}For the braking reaction time of the i +2 th vehicle in the fleet, T_{Reaction (n-1)}The brake reaction time of the (n-1) th vehicle in the fleet.

8. The method as claimed in claim 3, wherein the step 4 is performed when the calculated braking response time T of the vehicle ahead in the fleet is determined_{Reaction of}When the time is more than or equal to 5s, the OBU intelligent vehicle-mounted unit temporarily does not issue a prompt to the vehicle-mounted display screen of the vehicle, and the prompt is sent out through t_{Wait for}＝T_{Reaction of}And transmitting the braking information of the front vehicle to a vehicle-mounted display screen of the self vehicle after 5s, and reminding the self vehicle to make adjustment in time.

Background

While the human and vehicle roads are rapidly developed, traffic accidents bring huge casualties and property losses to the nation and people. Wherein the proportion of rear-end accidents in the total number of accidents is very high. The main reason for such accidents is that the distance between the front vehicle and the rear vehicle is too close, so that the driver cannot react in time when the front vehicle suddenly brakes, and the rear-end collision accident occurs.

At present, an anti-collision early warning system is arranged on part of vehicles, when the self vehicle has rear-end collision risk, the early warning system can remind a driver to keep the distance between the vehicles in an acoustic, optical and electric mode, and the occurrence of rear-end collision accidents is reduced to a certain extent. However, the existing anti-collision early warning system only considers the relative motion relationship between the own vehicle and the preceding vehicle, and when the current vehicle has an emergency, the own vehicle has the risk of being rear-ended if taking emergency braking, which easily causes the occurrence of a chain rear-end accident.

Disclosure of Invention

In view of the problems in the prior art, an object of the present invention is to provide a driving assistance system and method for predicting the braking of a leading vehicle, which can predict the braking response time and the maximum braking deceleration of each vehicle in a fleet when the first vehicle in the fleet brakes, and display the predicted braking information of the leading vehicle to the driver of the leading vehicle, so that the driver can control the vehicle in advance, thereby avoiding the occurrence of rear-end collision.

In order to achieve the purpose, the invention adopts the following technical scheme:

a driving assistance system predicting braking of a preceding vehicle, comprising: drive test equipment and vehicle-mounted equipment; the roadside equipment comprises a millimeter wave radar, an intelligent high-definition camera and an RSU intelligent roadside unit; the intelligent road side unit comprises an edge calculation processing unit; the vehicle-mounted equipment comprises an OBU intelligent vehicle-mounted unit and a vehicle-mounted display screen; the OBU intelligent vehicle-mounted unit comprises a data storage unit, a processing unit and a wireless communication unit;

the millimeter wave radar is installed on the road side or the central isolation belt, faces the lane line direction, and is used for collecting motion data information of vehicles in the lane and transmitting the motion data information to the RSU intelligent road side unit; the motion data information of the vehicle comprises the speed of the vehicle and the distance between the head of the rear vehicle and the tail of the front vehicle;

the intelligent high-definition camera is arranged above the millimeter wave radar and used for shooting images of vehicles in a lane and transmitting the images to the RSU intelligent road side unit;

the RSU intelligent road side unit is installed at the road side of the road and used for transmitting the motion data information and the vehicle image of the vehicle to the edge calculation processing unit;

the edge calculation processing unit is used for extracting the speed of the vehicle, the distance between the head of the rear vehicle and the tail of the front vehicle and the moment when the vehicle brakes in the vehicle image in the motion data information and transmitting the speed, the distance and the moment to the OBU intelligent vehicle-mounted unit;

the OBU intelligent vehicle-mounted unit is arranged on the inner side of a front windshield of the automobile and used for transmitting the speed of the automobile, the distance between the head of a rear automobile and the tail of a front automobile and the moment when the automobile brakes in an automobile image to the data storage unit;

the data storage unit is used for storing the speed of the vehicle, the distance between the head of the rear vehicle and the tail of the front vehicle and the moment when the vehicle brakes in the vehicle image and transmitting the speed, the distance and the moment to the processing unit;

the processing unit is used for processing the speed of the vehicle, the distance between the head of the rear vehicle and the tail of the front vehicle and the moment when the vehicle brakes in the vehicle image and sending the processing result to the wireless communication unit;

the wireless communication unit is used for transmitting the processing result to the vehicle-mounted display screen;

the vehicle-mounted display screen is arranged near a vehicle instrument panel and used for displaying the braking information of the front vehicle to a driver of the vehicle.

Further, there are a plurality of roadside devices, and the interval between two adjacent roadside devices is 200 m.

(II) a driving assistance method for predicting braking of a preceding vehicle, comprising the steps of:

step 1, detecting whether a vehicle team with multiple vehicle following behaviors exists on a lane by using a millimeter wave radar and an intelligent high-definition camera, and if so, turning to step 2;

step 2, when the brake lamp of any vehicle in the fleet is on, the braking time t, the speed v of the braking vehicle, the speed v of the rear vehicle of the braking vehicle and the distance l between the braking vehicle and the rear vehicle are obtained, and the braking time t of the ith vehicle is determined_iVelocity v_iSpeed of the (i +1) th vehicle at the time of braking of the ith vehicleDistance l between the i +1 th vehicle and the i-th vehicle_i-(i+1)And braking time t when the (i +1) th vehicle brakes_i+1And velocity v_i+1Calculating the maximum braking deceleration a of the ith vehicle_maxiThe following time interval D of the (i +1) th vehicle_i+1Braking reaction time T_{Reaction (i +1)}And maximum braking deceleration a_max(i+1)(ii) a The total number of the vehicles in the fleet is n, n is more than or equal to 3 and is a positive integer, i is more than or equal to 1 and less than or equal to n, and i is a positive integer;

step 3, acquiring a plurality of historical data, extracting the maximum braking deceleration of the front vehicle, the following time interval of the rear vehicle, the braking reaction time of the rear vehicle and the maximum braking deceleration of the rear vehicle in each historical data, establishing the historical data as a sample, forming a sample set by a plurality of samples, and storing the sample set in a storage unit;

when the brake lamp of the ith vehicle is on, the collected historical data corresponds to the sample (a)_maxi，D_i+1， T_{Reaction (i +1)}，a_max(i+1)) Wherein a is_maxiMaximum braking deceleration of the i-th vehicle, D_i+1The following time interval, T, of the i +1 th vehicle_{Reaction (i +1)}Brake reaction time of i +1 th vehicle, a_max(i+1)Is the maximum of the (i +1) th vehicleBraking deceleration;

taking the maximum braking deceleration of the front vehicle and the following vehicle distance of the rear vehicle in each sample stored in the storage unit as the input of a BP neural network model, taking the braking reaction time of the rear vehicle and the maximum braking deceleration of the rear vehicle as the output of the BP neural network model, and training the BP neural network model to obtain a trained BP neural network model;

and 4, predicting the braking reaction time T of all the following vehicles when the vehicles in the fleet brake according to the trained BP neural network model_{Reaction of}And maximum braking deceleration a_maxAnd then according to the predicted braking reaction time T of all vehicles behind the braking vehicle in the fleet_{Reaction of}Calculating the braking response time T of the front vehicle_{Reaction of}And the brake information of the front vehicle is displayed to remind the self vehicle to make adjustment in time.

Further, in step 1, the detecting whether there is a fleet with multiple following behaviors on the lane specifically includes: when the following distance between every two adjacent vehicles in the motorcade is smaller than 100 meters, judging that the following behavior exists between the vehicles, and judging that a plurality of continuous vehicles with the following behavior can be judged as the motorcade; otherwise, there is no following behavior.

Further, in step 2, the maximum braking deceleration a of the i-th vehicle_maxiThe following time interval D of the (i +1) th vehicle_i+1Braking reaction time T_{Reaction (i +1)}And maximum braking deceleration a_max(i+1)The calculation processes of (a) are respectively as follows:

T_{reaction (i +1)}＝t_(i+1)-t_i

Where Δ t is set to 0.5 second, t in the present invention_iFor the braking moment of the ith vehicle in the fleet, Δ v_iIs the speed difference of 0.5 second, Deltav, of the ith braking vehicle in the fleet_i+1The speed difference of 0.5 second for the i +1 th braking vehicle in the fleet_i+1The braking time of the (i +1) th vehicle in the fleet, i_i-(i+1)The distance between the (i +1) th vehicle and the ith vehicle in the platoon,is the speed of the i +1 th vehicle at the time of braking of the i-th vehicle, a_maxiIs the maximum braking deceleration at the time of braking of the ith vehicle, a_max(i+1)Is the maximum braking deceleration at the time of braking of the (i +1) th vehicle.

Further, in step 3, training the BP neural network model specifically comprises:

acquiring a training sample set and a testing sample set, training a BP neural network model through the training sample to obtain a trained BP neural network model, and testing the trained BP neural network model through the testing sample; when the average value of the difference values between the output of the BP neural network model and the label of the test sample is smaller than a preset error threshold value, determining that the training of the BP neural network model is finished;

the training sample set and the testing sample set are respectively selected from historical data, and each sample consists of the maximum braking deceleration of a front vehicle, the following time interval of a rear vehicle, the braking reaction time of the rear vehicle and the maximum braking deceleration of the rear vehicle; the test specimens are labeled brake reaction time of the following vehicle and maximum brake deceleration of the following vehicle.

Further, in step 4, calculating the brake reaction time T of the vehicle in front of the bicycle_{Reaction of}The method specifically comprises the following steps:

T_{reaction (n-1)}＝T_{Reaction (i +1)}+T_{Reaction (i +2)}+……T_{Reaction (n-1)}

Wherein, T_{Reaction (n-1)}Braking the ith vehicle in the fleetTime of brake reaction time, T, of the (n-1) th vehicle_{Reaction (i +1)}For the braking reaction time of the i +1 th vehicle in the fleet, T_{Reaction (i +2)}For the braking reaction time of the i +2 th vehicle in the fleet, T_{Reaction (n-1)}The brake reaction time of the (n-1) th vehicle in the fleet.

Further, in step 4, when the calculated braking reaction time T of the bicycle in front of the motorcade_{Reaction of}When the time is more than or equal to 5s, the OBU intelligent vehicle-mounted unit temporarily does not issue a prompt to the vehicle-mounted display screen of the vehicle, and the prompt is sent out through t_{Wait for}＝T_{Reaction of}And transmitting the braking information of the front vehicle to a vehicle-mounted display screen of the self vehicle after 5s, and reminding the self vehicle to make adjustment in time.

Compared with the prior art, the invention has the beneficial effects that:

(1) the driving auxiliary system for predicting the braking of the front vehicle provided by the invention mainly utilizes the millimeter wave radar, the intelligent high-definition camera and the intelligent road side unit RSU to acquire the vehicle information on the road in real time, then processes the vehicle information through the intelligent vehicle-mounted unit OBU to obtain the braking information of the vehicles in the fleet, and finally displays the braking information of the front vehicle through the vehicle-mounted display screen to remind the driver of the rear vehicle of timely decelerating, so that the rear-end collision is avoided.

(2) The driving auxiliary method for predicting the braking of the front vehicle can predict the braking behavior of the front vehicle of the bicycle in advance and give a safety prompt to the driver, so that the driver can control the vehicle in advance, and the occurrence of rear-end accidents is effectively avoided.

Drawings

Fig. 1 is a schematic diagram of a driving assistance apparatus for predicting braking of a preceding vehicle according to an embodiment of the present invention.

Wherein, 1: the 1 st vehicle in the fleet; 2: a 2 nd vehicle within the fleet; 3: the 3 rd vehicle in the platoon; 4: the 4 th vehicle in the fleet; 5: an OBU intelligent vehicle-mounted unit; 6: a vehicle-mounted display screen; 7: a millimeter wave radar; 8: an intelligent high-definition camera; 9: RSU intelligence road side unit.

Detailed Description

In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention will be described in further detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention. Test methods in which specific conditions are not specified in the following examples are generally carried out under conventional conditions or under conditions recommended by the respective manufacturers.

Example 1

Referring to fig. 1, an embodiment of the present invention provides a driving assistance system for predicting braking of a preceding vehicle, including: drive test equipment and vehicle-mounted equipment; the system comprises a plurality of road side devices, wherein the interval between every two adjacent road side devices is 200m, and each road side device comprises a millimeter wave radar 7, an intelligent high-definition camera 8 and an RSU intelligent road side unit 9; the RSU intelligent roadside unit 9 includes an edge calculation processing unit; the vehicle-mounted equipment comprises an OBU intelligent vehicle-mounted unit 5 and a vehicle-mounted display screen 6; the OBU intelligent vehicle-mounted unit 5 comprises a data storage unit, a processing unit and a wireless communication unit;

the millimeter wave radar 7 is installed on the road side or the central isolation belt, faces the lane line direction, and is used for acquiring the motion data information of the vehicles in the lane and transmitting the motion data information to the RSU intelligent road side unit 9; the motion data information of the vehicle comprises the speed of the vehicle and the distance between the head of the rear vehicle and the tail of the front vehicle;

the intelligent high-definition camera 8 is arranged above the millimeter-wave radar 7 and used for shooting images of vehicles in the lane and transmitting the images to the RSU intelligent road side unit 9;

the RSU intelligent road side unit 9 is installed at the road side of the road and used for transmitting the motion data information and the vehicle image of the vehicle to the edge calculation processing unit;

the edge calculation processing unit is used for extracting the speed of the vehicle, the distance between the head of the rear vehicle and the tail of the front vehicle and the moment when the vehicle brakes in the vehicle image in the motion data information and transmitting the speed, the distance and the moment to the OBU intelligent vehicle-mounted unit 5;

the OBU intelligent vehicle-mounted unit 5 is arranged on the inner side of a front windshield of the automobile and used for transmitting the speed of the automobile, the distance between the head of a rear automobile and the tail of a front automobile and the moment when the automobile brakes in an automobile image to the data storage unit;

the data storage unit is used for storing the speed of the vehicle, the distance between the head of the rear vehicle and the tail of the front vehicle and the moment when the vehicle brakes in the vehicle image and transmitting the speed and the distance to the processing unit;

the processing unit is used for processing the speed of the vehicle, the distance between the head of the rear vehicle and the tail of the front vehicle and the moment when the vehicle brakes in the vehicle image and sending the processing result to the wireless communication unit;

the wireless communication unit is used for transmitting the processing result to the vehicle-mounted display screen 6;

the vehicle-mounted display screen 6 is arranged near a vehicle instrument panel and is used for displaying the braking information of the front vehicle to a driver of the vehicle;

in the embodiment of the invention, the millimeter wave radar 7 adopts a 77GHz high-frequency-band millimeter wave radar, the effective detection distance is 200m, and the detection angle is plus or minus 45 degrees.

The driving auxiliary system for predicting the braking of the front vehicle provided by the invention mainly utilizes the millimeter wave radar, the intelligent high-definition camera and the intelligent road side unit RSU to acquire the vehicle information on the road in real time, then processes the vehicle information through the intelligent vehicle-mounted unit OBU to obtain the braking information of the vehicles in the fleet, and finally displays the braking information of the front vehicle through the vehicle-mounted display screen to remind the driver of the rear vehicle of timely decelerating, so that the rear-end collision is avoided.

Example 2

The embodiment of the invention provides a driving assistance method for predicting braking of a front vehicle, which comprises the following steps:

step 1, detecting whether a vehicle team with multiple vehicle following behaviors exists on a lane by using a millimeter wave radar and an intelligent high-definition camera, and if so, turning to step 2;

specifically, when the following distance between every two adjacent vehicles in the vehicle fleet is less than 100 meters, the following behavior exists between the vehicles, and a plurality of continuous vehicles with the following behavior can be judged as one vehicle fleet; otherwise, no following behavior exists; wherein the total number of vehicles in the fleet is n, n is more than or equal to 3, and n is a positive integer;

referring to fig. 1, when the distance l between the 1 st vehicle and the 2 nd vehicle_1-2When the distance between the 2 nd vehicle and the 3 rd vehicle is less than or equal to 100m, the 2 nd vehicle is judged to drive along with the 1 st vehicle_2-3When the distance between the 3 rd vehicle and the 4 th vehicle is less than or equal to 100m, the 3 rd vehicle is judged to drive along with the 2 nd vehicle_3-4When the distance is less than or equal to 100m, judging that the 4 th vehicle runs along with the 3 rd vehicle, and further judging that the 1 st vehicle, the 2 nd vehicle, the 3 rd vehicle and the 4 th vehicle form a fleet;

step 2, when the brake lamp of any vehicle in the fleet is on, the braking time t, the speed v of the braking vehicle, the speed v of the rear vehicle of the braking vehicle and the distance l between the braking vehicle and the rear vehicle are obtained, and the braking time t of the ith vehicle is determined_iVelocity v_iSpeed of the (i +1) th vehicle at the time of braking of the ith vehicleDistance l between the i +1 th vehicle and the i-th vehicle_i-(i+1)And braking time t when the (i +1) th vehicle brakes_i+1And velocity v_i+1Calculating the maximum braking deceleration a of the ith vehicle_maxiThe following time interval D of the (i +1) th vehicle_i+1Braking reaction time T_{Reaction (i +1)}And maximum braking deceleration a_max(i+1)(ii) a The total number of the vehicles in the fleet is n, n is more than or equal to 3 and is a positive integer, i is more than or equal to 1 and less than or equal to n, and i is a positive integer;

specifically, referring to fig. 1, when the number of vehicles in the fleet is 4 and the brake light of the 1 st vehicle 1 is on, the braking time t of the 1 st vehicle 1 is obtained₁Velocity v₁Speed of 2 nd vehicle 2Distance l between 2 nd vehicle 2 and 1 st vehicle 1_1-2(ii) a When the brake lamp of the 2 nd vehicle 2 is onAt the time of starting, the braking time t of the 2 nd vehicle 2 is acquired₂Speed v of the 2 nd vehicle 2₂Speed of 3 rd vehicle 3Distance l between 3 rd vehicle 3 and 2 nd vehicle 2_2-3(ii) a When the brake lamp of the 3 rd vehicle 3 is turned on, the braking time t of the 3 rd vehicle 3 is acquired₃Speed v of the 3 rd vehicle 3₃Speed of 4 th vehicle 4Distance l between 4 th vehicle 4 and 3 rd vehicle 3_3-4(ii) a When the brake lamp of the 4 th vehicle 4 is turned on, the braking time t of the 4 th vehicle 4 is acquired₄And the speed v of the 4 th vehicle 4₄；

According to the speed v of the first vehicle when braking₁Calculating a_max1(ii) a According to l_1-2Andcalculating D₂Wherein, in the step (A),according to t₁And t₂Calculating T_{Reaction 2}Wherein, T_{Reaction 2}＝t₂-t₁(ii) a According to the speed v of the second vehicle when braking₂Calculating a_max2(ii) a According to l_2-3Andcalculating D₃Wherein, in the step (A),according to t₂And t₃Calculating T_{Reaction 3}Wherein, T_{Reaction 3}＝t₃-t₂(ii) a According to the speed v of the third vehicle when braking₃Calculating a_max3(ii) a According to l_3-4Andcalculating D₄Wherein, in the step (A),according to t₃And t₄Calculating T_{Reaction 4}Wherein, T_{Reaction 4}＝t₄-t₃；

Step 3, acquiring a plurality of historical data, extracting the maximum braking deceleration of the front vehicle, the following time interval of the rear vehicle, the braking reaction time of the rear vehicle and the maximum braking deceleration of the rear vehicle in each historical data, establishing the historical data as a sample, forming a sample set by a plurality of samples, and storing the sample set in a storage unit;

taking the ith vehicle in the fleet as an example, when the brake lamp of the ith vehicle is on, the collected historical data corresponds to the sample (a)_maxi，D_i+1，T_{Reaction (i +1)}，a_max(i+1)) Wherein a is_maxiMaximum braking deceleration of the i-th vehicle, D_i+1The following time interval, T, of the i +1 th vehicle_{Reaction (i +1)}Brake reaction time of i +1 th vehicle, a_max(i+1)Maximum braking deceleration for the (i +1) th vehicle;

taking the maximum braking deceleration of the front vehicle and the following vehicle distance of the rear vehicle in each sample stored in the storage unit as the input of a BP neural network model, taking the braking reaction time of the rear vehicle and the maximum braking deceleration of the rear vehicle as the output of the BP neural network model, and training the BP neural network model to obtain a trained BP neural network model;

in order to ensure the prediction result of the BP neural network model, enough braking information of the vehicle needs to be acquired, and the initial sample data size is 1000 groups;

in the embodiment of the invention, the training of the BP neural network model specifically comprises the following steps:

acquiring a training sample set and a testing sample set, training a BP neural network model through the training sample to obtain a trained BP neural network model, and testing the trained BP neural network model through the testing sample; when the average value of the difference values between the output of the BP neural network model and the label of the test sample is smaller than a preset error threshold value, determining that the training of the BP neural network model is finished;

the training sample set and the testing sample set are respectively selected from historical data, and each sample consists of the maximum braking deceleration of a front vehicle, the following time interval of a rear vehicle, the braking reaction time of the rear vehicle and the maximum braking deceleration of the rear vehicle; the labels of the test samples are the brake reaction time of the rear vehicle and the maximum brake deceleration of the rear vehicle;

specifically, in the embodiment of the present invention, 100 samples are selected as test samples, the maximum braking deceleration of the front vehicle and the following vehicle distance of the rear vehicle in each sample are input into a trained BP neural network model, the trained BP neural network model outputs the braking response time of the rear vehicle and the maximum braking deceleration of the rear vehicle, and when an average value of a difference value between the predicted braking response time of the rear vehicle output by the BP neural network model and the following vehicle response time in the sample is smaller than a preset error threshold of the braking response time, and an average value of a difference value between the predicted maximum braking deceleration of the rear vehicle output by the BP neural network model and the maximum braking deceleration of the rear vehicle in the sample is smaller than a preset error threshold of the maximum braking deceleration, it is determined that the training of the BP neural network model is completed;

if the average value of the difference value between the predicted rear vehicle braking reaction time output by the BP neural network model and the rear vehicle reaction time in the sample is greater than or equal to the preset error threshold of braking reaction time or the average value of the difference value between the predicted maximum braking deceleration of the rear vehicle output by the BP neural network model and the maximum braking deceleration of the rear vehicle in the sample is greater than or equal to the preset error threshold of maximum braking deceleration, continuously acquiring the braking information of the vehicle fleet on the road, expanding the number of samples, and continuously training the BP neural network model until the precision requirement is met;

the preset error threshold in the embodiment of the present invention includes a preset error threshold of the brake response time and a preset error threshold of the maximum brake deceleration, and the preset error threshold of the brake response time and the preset error threshold of the maximum brake deceleration may be set according to the training precision of the BP neural network model, which is not limited specifically herein.

Further, in the embodiment of the invention, the maximum braking deceleration a of the ith vehicle is used_maxiAnd the following time distance D of the (i +1) th vehicle_i+1As an input layer of the BP neural network model, the number of nodes of the input layer is 2; the brake reaction time T of the i +1 th vehicle_{Reaction (i +1)}And maximum braking deceleration a_max(i+1)As an output layer of the BP neural network model, the number of nodes of the output layer is 2, wherein the BP neural network model only comprises 1 hidden layer, and the number of nodes of the hidden layer is set to be 10;

for facilitating the training of the BP neural network model, sample data is input into the BP neural network model after being normalized, an excitation function of a hidden layer of the BP neural network model is set to be a tan Sig function, a log Sig function is selected as the excitation function of an output layer of the BP neural network model, the learning rate is set to be 0.01, the maximum network iteration frequency is 10000, and the training target is set to be 0.01. In the embodiment of the present invention, the BP neural network model is taken as an example, but not limited to the BP neural network model.

And 4, predicting the braking reaction time T of all the following vehicles when the vehicles in the fleet brake according to the trained BP neural network model_{Reaction of}And maximum braking deceleration a_maxAnd then according to the predicted braking reaction time T of all vehicles behind the braking vehicle in the fleet_{Reaction of}Calculating the braking response time T of the front vehicle_{Reaction of}And the brake information of the front vehicle is displayed to remind the self vehicle to make adjustment in time;

wherein, the braking response time T of the front vehicle of the bicycle is calculated_{Reaction of}The method specifically comprises the following steps:

T_{reaction (n-1)}＝T_{Reaction (i +1)}+T_{Reaction (i +2)}+……T_{Reaction (n-1)}

Wherein, T_{Reaction (n-1)}The brake reaction time T of the (n-1) th vehicle when the ith vehicle brakes in the fleet_{Reaction (i +1)}For the braking reaction time of the i +1 th vehicle in the fleet, T_{Reaction of(i+2)}For the braking reaction time of the i +2 th vehicle in the fleet, T_{Reaction (n-1)}The braking reaction time of the (n-1) th vehicle in the fleet is obtained;

specifically, when the ith vehicle in the fleet brakes, and the preceding vehicle is the (n-1) th vehicle in the fleet, for example, the vehicle-mounted display screen displays the braking information of the preceding vehicle, specifically, the preceding vehicle will be at T_{Reaction (n-1)}s rear with maximum brake deceleration a_max(n-1)Carrying out deceleration' and reminding the driver of the bicycle to pay attention to deceleration;

further, in step 4, when the calculated braking reaction time T of the bicycle in front of the motorcade_{Reaction of}When the time is more than or equal to 5s, the OBU intelligent vehicle-mounted unit temporarily does not issue a prompt to the vehicle-mounted display screen of the vehicle, and the prompt is sent out through t_{Wait for}＝T_{Reaction of}Transmitting the braking information of the front vehicle to a vehicle-mounted display screen of the self vehicle after 5s, and reminding the self vehicle to make adjustment in time;

specifically, when the ith vehicle in the fleet brakes, for example, the nth-1 vehicle in the fleet, the vehicle-mounted display screen displays the braking information of the front vehicle of the vehicle, specifically, "the front vehicle will brake at the maximum braking deceleration a after 5s_max(n-1)And carrying out deceleration ", and reminding the driver of the bicycle of paying attention to the deceleration.

The driving auxiliary method for predicting the braking of the front vehicle can predict the braking behavior of the front vehicle of the bicycle in advance and give a safety prompt to the driver, so that the driver can control the vehicle in advance, and the occurrence of rear-end accidents is effectively avoided.

It will be apparent to those skilled in the art that various changes and modifications may be made in the present invention without departing from the spirit and scope of the invention. Thus, if such modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include such modifications and variations.