Automobile collision danger estimation method based on external dynamic environment

Abstract

The invention discloses an automobile collision danger estimation method based on an external dynamic environment, which relates to the field of automobile collision and comprises the following steps: acquiring environment information around an automobile so as to obtain a predicted visual distance of the external environment; obtaining a vehicle number change broken line graph, a vehicle speed change broken line graph and a vehicle flow change broken line graph according to the obtained environment information around the vehicle; obtaining a first collision danger coefficient according to the obtained predicted visual distance of the external environment, the vehicle number change broken line graph, the vehicle speed change broken line graph and the vehicle flow change broken line graph; acquiring state data of a driver in the automobile, and acquiring a second collision danger coefficient according to the acquired state data of the driver; obtaining a third collision danger coefficient according to the obtained first collision danger coefficient and the second collision danger coefficient, and obtaining an automobile collision danger level according to the obtained third collision danger coefficient and the second collision danger coefficient; and the probability of automobile collision is further reduced.

Description

Technical Field

[0001] The present invention relates to the field of vehicle collisions, and specifically to a method for estimating vehicle collision risks based on the external dynamic environment. Background Art

[0002] Traffic accidents have always been major public safety issues globally. The casualties and economic losses caused by traffic accidents every year are quite huge. Although traditional passive safety measures (such as seat belts, airbags, etc.) can reduce injuries when an accident occurs, they cannot fundamentally prevent accidents. Therefore, active measures to enhance traffic safety are particularly important, which has promoted the demand for more advanced collision risk prediction and warning systems. How to obtain the first collision risk coefficient and the second collision risk coefficient based on the driver's state data in the vehicle, the predicted visible distance of the external environment, the line graph of vehicle quantity changes, the line graph of vehicle speed changes, and the line graph of traffic flow changes, obtain the third collision risk coefficient based on the obtained first and second collision risk coefficients, and obtain the vehicle collision risk level based on the obtained third collision risk coefficient and the second collision risk coefficient, and further reduce the probability of vehicle collisions is the problem we need to solve. For this purpose, a method for estimating vehicle collision risks based on the external dynamic environment is provided. Summary of the Invention

[0003] In order to solve the above technical problems, the purpose of the present invention is to provide a method for estimating vehicle collision risks based on the external dynamic environment.

[0004] The purpose of the present invention can be achieved through the following technical solutions: A method for estimating vehicle collision risks based on the external dynamic environment, including the following steps: Step S1: Obtain the environmental information around the vehicle, and then obtain the predicted visible distance of the external environment. Step S2: Obtain the line graph of vehicle quantity changes, the line graph of vehicle speed changes, and the line graph of traffic flow changes based on the obtained environmental information around the vehicle. Step S3: Obtain the first collision risk coefficient based on the obtained predicted visible distance of the external environment, the line graph of vehicle quantity changes, the line graph of vehicle speed changes, and the line graph of traffic flow changes. Step S4: Obtain the driver's state data in the vehicle, and obtain the second collision risk coefficient based on the obtained driver's state data. Step S5: Obtain the third collision risk coefficient based on the obtained first and second collision risk coefficients, and obtain the vehicle collision risk level based on the obtained third collision risk coefficient and the second collision risk coefficient.

[0005] Further, the process of obtaining the environmental information around the vehicle includes: The environmental information around the vehicle includes the number and speed of vehicles around the vehicle, the predicted visible distance of the external environment, and the traffic flow of the lane where the vehicle is located; Install a data collection terminal at the corresponding position of the vehicle, configure the installed data collection terminal to obtain the collection area of the data collection terminal. The collection area of the data collection terminal is a circle centered on the vehicle with a radius of k, where k is a positive number. The data collection terminal is used to collect the number and speed of vehicles within the collection area; Obtain the number and speed of vehicles around the vehicle through the data collection terminal; Obtain road monitoring data, and obtain the traffic flow of the lane where the vehicle is located through the obtained road monitoring data; Obtain the predicted visible distance of the external environment at different time periods through weather forecasting. Set up a visible distance collection terminal, and obtain the actual visible distance corresponding to the predicted visible distance at different time periods through the set visible distance collection terminal. Obtain the predicted visible distance of the external environment based on the obtained predicted visible distance and the obtained actual visible distance.

[0006] Further, the process of obtaining the predicted visible distance of the external environment includes: Construct a visible distance prediction model based on the corresponding relationship between the obtained predicted visible distance and the actual visible distance. Use the obtained predicted visible distance as the input value of the visible distance prediction model, and use the output value of the visible distance prediction model as the predicted visible distance of the input predicted visible distance; Train the visible distance prediction model using the actual visible distance corresponding to the predicted visible distance. Obtain the predicted visible distance of the subsequent time period through weather forecasting, input the obtained predicted visible distance into the visible distance prediction model, and obtain the predicted visible distance of the subsequent time period.

[0007] Further, the process of obtaining the vehicle quantity change line chart, vehicle speed change line chart, and traffic flow change line chart based on the obtained environmental information around the vehicle includes: Construct a two-dimensional rectangular coordinate system with time regarding the number of vehicles around the vehicle. Map the obtained number of vehicles around the vehicle into the constructed two-dimensional rectangular coordinate system with time regarding the number of vehicles around the vehicle to generate corresponding vehicle quantity data points. Label the generated vehicle quantity data points, and connect each vehicle quantity data point in sequence by a straight line according to the time sequence to obtain the vehicle quantity change line chart; Obtain the average speed of the vehicles around the car, construct a two-dimensional rectangular coordinate system with time on the x-axis and the average speed of the vehicles around the car on the y-axis. Map the obtained average speed of the vehicles around the car into the constructed two-dimensional rectangular coordinate system to generate corresponding vehicle speed data points. Label the generated vehicle speed data points and connect each vehicle speed data point in sequence by a straight line according to the time order to obtain a vehicle speed change line graph. Construct a two-dimensional rectangular coordinate system with time on the x-axis and the traffic flow of the lane where the car is located on the y-axis. Map the obtained traffic flow of the lane where the car is located into the constructed two-dimensional rectangular coordinate system to generate corresponding traffic flow data points. Label the generated traffic flow data points and connect each traffic flow data point in sequence by a straight line according to the time order to obtain a traffic flow change line graph.

[0008] Further, the process of obtaining the first collision risk coefficient based on the predicted visible distance of the external environment, the vehicle quantity change line graph, the vehicle speed change line graph, and the traffic flow change line graph includes: Obtain the number of vehicles around the car corresponding to the vehicle quantity data point. Draw a perpendicular line from the vehicle quantity data point to the time axis to obtain the intersection point of the perpendicular line and the time axis. Draw a perpendicular line from the previous vehicle quantity data point to the time axis to obtain the intersection point of the perpendicular line and the time axis. Obtain the area of the right trapezoid formed by the vehicle quantity data point, the previous vehicle quantity data point, and the two intersection points of the two perpendicular lines and the time axis. Obtain the change in the number of vehicles around the car corresponding to the vehicle quantity data point based on the number of vehicles around the car corresponding to the vehicle quantity data point and the area of the right trapezoid. Obtain the average speed of the vehicles around the car corresponding to the vehicle speed data point. Draw a perpendicular line from the vehicle speed data point to the time axis to obtain the intersection point of the perpendicular line and the time axis. Draw a perpendicular line from the previous vehicle speed data point to the time axis to obtain the intersection point of the perpendicular line and the time axis. Obtain the area of the right trapezoid formed by the vehicle speed data point, the previous vehicle speed data point, and the two intersection points of the two perpendicular lines and the time axis. Obtain the change in the average speed of the vehicles around the car corresponding to the vehicle speed data point based on the average speed of the vehicles around the car corresponding to the vehicle speed data point and the area of the right trapezoid. By analogy, obtain the change in the traffic flow of the lane where the car is located corresponding to the traffic flow data point. Obtain the time corresponding to the traffic flow data point and obtain the predicted visible distance corresponding to the time corresponding to the traffic flow data point. Obtain the first collision risk coefficient based on the obtained change in the number of vehicles around the car, the change in the average speed of the vehicles around the car, the change in the traffic flow of the lane where the car is located, and the predicted visible distance.

[0009] Further, the process of obtaining the second collision risk coefficient based on the acquired driver's status data includes: The driver's status data includes the driver's blink frequency. Obtain the time corresponding to all traffic flow data points, label the time corresponding to all the acquired traffic flow data points, where the time corresponding to the traffic flow data point is the same as the label of the traffic flow data point. Obtain the driver's blink frequency corresponding to each time, and obtain the second collision risk coefficient based on the acquired driver's blink frequency.

[0010] Further, the process of obtaining the third collision risk coefficient based on the obtained first collision risk coefficient and second collision risk coefficient includes: Set the weights of the first collision risk coefficient and the second collision risk coefficient according to the actual situation, and denote the set weights of the first collision risk coefficient and the second collision risk coefficient as QZ_y and QZ_e respectively; Multiply the first collision risk coefficient by its weight, multiply the second collision risk coefficient by its weight, and add the two multiplication results to obtain the third collision risk coefficient.

[0011] Further, the process of obtaining the vehicle collision risk level based on the obtained third collision risk coefficient and the second collision risk coefficient includes: Set the threshold range of the collision risk coefficient according to the actual situation; Obtain the vehicle collision risk level based on the comparison result between the obtained third collision risk coefficient and the set threshold range of the collision risk coefficient and the value of the second collision risk coefficient; The vehicle collision risk level includes special - level collision risk, level - three collision risk, level - two collision risk, and level - one collision risk.

[0012] Compared with the prior art, the beneficial effects of the present invention are as follows: Obtain the environmental information around the vehicle, and then obtain the predicted visible distance of the external environment. Obtain the vehicle quantity change line graph, vehicle speed change line graph, and traffic flow change line graph based on the acquired environmental information around the vehicle. Obtain the first collision risk coefficient based on the obtained predicted visible distance of the external environment, vehicle quantity change line graph, vehicle speed change line graph, and traffic flow change line graph. Obtain the driver's status data inside the vehicle, obtain the second collision risk coefficient based on the acquired driver's status data, obtain the third collision risk coefficient based on the obtained first collision risk coefficient and second collision risk coefficient, and obtain the vehicle collision risk level based on the obtained third collision risk coefficient and the second collision risk coefficient, further reducing the probability of vehicle collision, increasing the safety of the driver inside the vehicle, and enhancing the driver's awareness of civilized and safe driving. Description of the Drawings

[0013] Figure 1 This is the schematic diagram of the present invention. Detailed implementation manners

[0014] As Figure 1 shown, the method for estimating the risk of vehicle collision based on the external dynamic environment includes the following steps: Step S1: Obtain the environmental information around the vehicle, and then obtain the predicted visible distance of the external environment. Step S2: Obtain the line graph of the change in the number of vehicles, the line graph of the change in vehicle speed, and the line graph of the change in traffic flow according to the obtained environmental information around the vehicle. Step S3: Obtain the first collision risk coefficient according to the obtained predicted visible distance of the external environment, the line graph of the change in the number of vehicles, the line graph of the change in vehicle speed, and the line graph of the change in traffic flow. Step S4: Obtain the status data of the driver in the vehicle, and obtain the second collision risk coefficient according to the obtained status data of the driver. Step S5: Obtain the third collision risk coefficient according to the obtained first collision risk coefficient and the second collision risk coefficient, and obtain the vehicle collision risk level according to the obtained third collision risk coefficient and the second collision risk coefficient. It should be further noted that in the specific implementation process, the process of obtaining the environmental information around the vehicle includes: The environmental information around the vehicle includes the number and speed of the vehicles around the vehicle, the predicted visible distance of the external environment, and the traffic flow of the lane where the vehicle is located. Install a data acquisition terminal at the corresponding position of the vehicle, configure the installed data acquisition terminal to obtain the acquisition area of the data acquisition terminal. The acquisition area of the data acquisition terminal is a circle centered on the vehicle with a radius of k, where k is a positive number. The data acquisition terminal is used to collect the number and speed of the vehicles in the acquisition area. Obtain the number and speed of the vehicles around the vehicle through the data acquisition terminal. Obtain the road monitoring data, and obtain the traffic flow of the lane where the vehicle is located through the obtained road monitoring data. Obtain the predicted visible distance of the external environment at different time periods through weather forecasting, set up a visible distance acquisition terminal, and obtain the actual visible distance corresponding to the predicted visible distance at different time periods through the set visible distance acquisition terminal. Construct a visible distance prediction model according to the corresponding relationship between the obtained predicted visible distance and the actual visible distance, use the obtained predicted visible distance as the input value of the visible distance prediction model, and use the output value of the visible distance prediction model as the predicted visible distance of the input predicted visible distance. The visual range prediction model is trained using the actual visual range corresponding to the predicted visual range to continuously optimize the visual range prediction model. The predicted visual range for a subsequent time period is obtained through weather forecasting and input into the visual range prediction model to obtain the predicted visual range for the subsequent time period. In the embodiments of the present invention, instead of directly using the predicted visual range in weather forecasting as a reference, a visual range prediction model is first constructed and the predicted visual range output by the visual range prediction model is used as a reference. This is because the predicted visual range provided by weather forecasting often has too large a coverage range and the values are not specific, unable to provide a more accurate reference. Therefore, the present invention uses the visual range prediction model to more accurately predict the visual range of the external environment, and at the same time continuously trains and optimizes the visual range prediction model with the actual visual range to obtain a more targeted predicted visual range.

[0015] It should be further noted that in the specific implementation process, the process of obtaining the vehicle quantity change line graph, vehicle speed change line graph, and traffic flow change line graph based on the acquired environmental information around the vehicle includes: A two-dimensional rectangular coordinate system of time with respect to the quantity of vehicles around the vehicle is constructed. The quantity of vehicles around the vehicle obtained is mapped into the constructed two-dimensional rectangular coordinate system of time with respect to the quantity of vehicles around the vehicle to generate corresponding vehicle quantity data points. The generated vehicle quantity data points are numbered, denoted as i, where i = 1, 2, 3, ……, n, and n is a positive integer. Each vehicle quantity data point is sequentially connected by a straight line in chronological order to obtain the vehicle quantity change line graph. The average speed of the vehicles around the vehicle is obtained. A two-dimensional rectangular coordinate system of time with respect to the average speed of the vehicles around the vehicle is constructed. The obtained average speed of the vehicles around the vehicle is mapped into the constructed two-dimensional rectangular coordinate system of time with respect to the average speed of the vehicles around the vehicle to generate corresponding vehicle speed data points. The generated vehicle speed data points are numbered, denoted as i, and each vehicle speed data point is sequentially connected by a straight line in chronological order to obtain the vehicle speed change line graph. A two-dimensional rectangular coordinate system of time with respect to the traffic flow of the lane where the vehicle is located is constructed. The traffic flow of the lane where the vehicle is located obtained is mapped into the constructed two-dimensional rectangular coordinate system of time with respect to the traffic flow of the lane where the vehicle is located to generate corresponding traffic flow data points. The generated traffic flow data points are numbered, denoted as i, and each traffic flow data point is sequentially connected by a straight line in chronological order to obtain the traffic flow change line graph.

[0016] It should be further noted that in the specific implementation process, the process of obtaining the first collision risk coefficient based on the predicted visible distance of the external environment, the line graph of vehicle quantity change, the line graph of vehicle speed change, and the line graph of traffic flow change obtained includes: Obtain the number of vehicles around the vehicle corresponding to the vehicle quantity data point, and denote the number of vehicles around the vehicle corresponding to the obtained vehicle quantity data point as Li. Draw a perpendicular line from the vehicle quantity data point to the time axis to obtain the intersection point of the perpendicular line and the time axis. Draw a perpendicular line from the previous vehicle quantity data point to the time axis to obtain the intersection point of the perpendicular line and the time axis. Obtain the area of the right trapezoid formed by the vehicle quantity data point, the previous vehicle quantity data point, and the two intersection points of the two perpendicular lines and the time axis, and denote the area of the obtained right trapezoid as SLi. If the vehicle quantity data point is the first vehicle quantity data point, then denote the area of the right trapezoid formed by the vehicle quantity data point, the previous vehicle quantity data point, and the two intersection points of the two perpendicular lines and the time axis as SL0, that is, SL1 = SL0, and SL0 is a positive number; Obtain the change quantity of the vehicles around the vehicle corresponding to the vehicle quantity data point based on the number of vehicles around the vehicle corresponding to the obtained vehicle quantity data point and the area of the right trapezoid, and denote the obtained change quantity of the vehicles around the vehicle corresponding to the vehicle quantity data point as L_changei; Among them, L_changei = Li + KP_s × 〖SL〗_i × Li; KP_s is an adjustment coefficient; Obtain the average speed of the vehicles around the vehicle corresponding to the vehicle speed data point, and denote the average speed of the vehicles around the vehicle corresponding to the obtained vehicle speed data point as Di. Draw a perpendicular line from the vehicle speed data point to the time axis to obtain the intersection point of the perpendicular line and the time axis. Draw a perpendicular line from the previous vehicle speed data point to the time axis to obtain the intersection point of the perpendicular line and the time axis. Obtain the area of the right trapezoid formed by the vehicle speed data point, the previous vehicle speed data point, and the two intersection points of the two perpendicular lines and the time axis, and denote the area of the obtained right trapezoid as SDi. If the vehicle speed data point is the first vehicle speed data point, then denote the area of the right trapezoid formed by the vehicle speed data point, the previous vehicle speed data point, and the two intersection points of the two perpendicular lines and the time axis as SD0, that is, SD1 = SD0, and SD0 is a positive number; Obtain the change average speed of the vehicles around the vehicle corresponding to the vehicle speed data point based on the average speed of the vehicles around the vehicle corresponding to the obtained vehicle speed data point and the area of the right trapezoid, and denote the obtained change average speed of the vehicles around the vehicle corresponding to the vehicle speed data point as D 变i ; Among them, ; KP_d is an adjustment coefficient; obtain the traffic flow of the lane where the vehicle corresponding to the traffic flow data point is located, and denote the traffic flow of the lane where the vehicle corresponding to the obtained traffic flow data point is located as Ui. Draw a perpendicular line from the traffic flow data point to the time axis to obtain the intersection point of the perpendicular line and the time axis. Draw a perpendicular line from the previous traffic flow data point to the time axis to obtain the intersection point of the perpendicular line and the time axis. Obtain the area of the right trapezoid formed by the traffic flow data point, the previous traffic flow data point, and the two intersection points of the two perpendicular lines and the time axis, and denote the area of the obtained right trapezoid as Sui. If the traffic flow data point is the first traffic flow data point, then denote the area of the right trapezoid formed by the traffic flow data point, the previous traffic flow data point, and the two intersection points of the two perpendicular lines and the time axis as SU0, that is, SU1 = SU0, and SU0 is a positive number; Obtain the changing traffic flow of the lane where the vehicle corresponding to the obtained traffic flow data point is located according to the traffic flow of the lane where the vehicle corresponding to the traffic flow data point is located and the area of the right trapezoid, and denote the changing traffic flow of the lane where the vehicle corresponding to the obtained traffic flow data point is located as U_change_i; Among them, U_i; KP_u is an adjustment coefficient; Obtain the time corresponding to the traffic flow data point, and obtain the predicted visible distance corresponding to this time, and denote the predicted visible distance corresponding to the obtained time as Ji; Obtain the first collision risk coefficient according to the obtained change quantity of the vehicles around the vehicle, the average change speed of the vehicles around the vehicle, the changing traffic flow of the lane where the vehicle is located, and the predicted visible distance, and denote the obtained first collision risk coefficient as PZ i ; Among them, ; TJ_s is an adjustment coefficient for the change quantity of the vehicles around the vehicle, TJ_d is an adjustment coefficient for the average change speed of the vehicles around the vehicle, TJ_u is an adjustment coefficient for the changing traffic flow of the lane where the vehicle is located, and TJ_j is an adjustment coefficient for the predicted visible distance.

[0017] It should be further noted that in the specific implementation process, the process of obtaining the second collision risk coefficient according to the obtained driver's state data includes: The driver's state data includes the driver's blink frequency. Obtain the time corresponding to all traffic flow data points, number the time corresponding to all obtained traffic flow data points, the time corresponding to the traffic flow data point is the same as the label of the traffic flow data point, obtain the driver's blink frequency corresponding to each time, and denote the driver's blink frequency corresponding to each obtained time as Z i ; Obtain the second collision risk coefficient according to the obtained driver's blink frequency, and denote the obtained second collision risk coefficient as PZE i ; Among them, , Zi ≠ 0; when Zi = 0, PZEi = PZEmax, and PZEmax is a positive number.

[0018] It should be further noted that in the specific implementation process, the process of obtaining the third collision risk coefficient based on the obtained first collision risk coefficient and second collision risk coefficient, and obtaining the vehicle collision risk level based on the obtained third collision risk coefficient and second collision risk coefficient includes: Set the weights of the first collision risk coefficient and the second collision risk coefficient according to the actual situation, and denote the set weights of the first collision risk coefficient and the second collision risk coefficient as QZ_y and QZ_e respectively; Multiply the first collision risk coefficient by its weight, multiply the second collision risk coefficient by its weight, add the two multiplication results to obtain the third collision risk coefficient, and denote the obtained third collision risk coefficient as PZSi; Among them, ; The vehicle collision risk levels include special - level collision risk, level - three collision risk, level - two collision risk, and level - one collision risk; It should be further noted that in the specific implementation process, the collision probability of the special - level collision risk is higher than that of the level - three collision risk, which is higher than that of the level - two collision risk, which is higher than that of the level - one collision risk; Set the threshold range of the collision risk coefficient according to the actual situation, and denote the set threshold range of the collision risk coefficient as (PZS0, PZS1); When the value of the second collision risk coefficient is PZEmax, the vehicle collision risk level is the special - level collision risk, and a red alarm is issued and the driver in the vehicle is immediately reminded; When PZSi ≤ PZS0, the vehicle collision risk level is the level - one collision risk; When PZS0 < PZSi < PZS1, the vehicle collision risk level is the level - two collision risk, and an orange alarm is issued and the driver in the vehicle is immediately reminded; When PZSi ≥ PZS1, the vehicle collision risk level is the level - three collision risk, and a yellow alarm is issued and the driver in the vehicle is immediately reminded.

[0019] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical method of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical method of the present invention.

Claims

1. A method for estimating the risk of vehicle collision based on an external dynamic environment, characterized in that, It includes the following steps: Step S1: Obtain the environmental information around the vehicle, and then obtain the predicted visible distance of the external environment; Step S2: Obtain the line graph of vehicle quantity change, the line graph of vehicle speed change, and the line graph of traffic flow change according to the obtained environmental information around the vehicle; Step S3: Obtain the first collision risk coefficient according to the obtained predicted visible distance of the external environment, the line graph of vehicle quantity change, the line graph of vehicle speed change, and the line graph of traffic flow change; Step S4: Obtain the state data of the driver in the vehicle, and obtain the second collision risk coefficient according to the obtained state data of the driver; Step S5: Obtain the third collision risk coefficient according to the obtained first collision risk coefficient and the second collision risk coefficient, and obtain the vehicle collision risk level according to the obtained third collision risk coefficient and the second collision risk coefficient.

2. The method for estimating the risk of vehicle collision based on an external dynamic environment according to claim 1, wherein The process of obtaining the environmental information around the vehicle includes: The environmental information around the vehicle includes the quantity and speed of the vehicles around the vehicle, the predicted visible distance of the external environment, and the traffic flow of the lane where the vehicle is located; Install a data acquisition terminal at the corresponding position of the vehicle, configure the installed data acquisition terminal to obtain the acquisition area of the data acquisition terminal. The acquisition area of the data acquisition terminal is a circle centered on the vehicle with a radius of k, where k is a positive number. The data acquisition terminal is used to collect the quantity and speed of the vehicles in the acquisition area; Obtain the quantity and speed of the vehicles around the vehicle through the data acquisition terminal; Obtain the road monitoring data, and obtain the traffic flow of the lane where the vehicle is located through the obtained road monitoring data; Obtain the predicted visible distance of the external environment at different time periods through weather forecasting, set up a visible distance acquisition terminal, obtain the actual visible distance corresponding to the predicted visible distance at different time periods through the set visible distance acquisition terminal, and obtain the predicted visible distance of the external environment according to the obtained predicted visible distance and the obtained actual visible distance.

3. The method for estimating the risk of vehicle collision based on an external dynamic environment according to claim 2, characterized in that, The process of obtaining the predicted visible distance of the external environment includes: Construct a visible distance prediction model according to the corresponding relationship between the obtained predicted visible distance and the actual visible distance, use the obtained predicted visible distance as the input value of the visible distance prediction model, and use the output value of the visible distance prediction model as the predicted visible distance of the input predicted visible distance; Train the visible distance prediction model with the actual visible distance corresponding to the predicted visible distance, obtain the predicted visible distance of the subsequent time period through weather forecasting, input the obtained predicted visible distance into the visible distance prediction model, and obtain the predicted visible distance of the subsequent time period.

4. The method for estimating the risk of vehicle collision based on an external dynamic environment according to claim 3, wherein The process of obtaining the line graph of vehicle quantity change, the line graph of vehicle speed change, and the line graph of traffic flow change according to the obtained environmental information around the vehicle includes: Construct a two-dimensional rectangular coordinate system of time with respect to the number of vehicles around the vehicle, map the obtained number of vehicles around the vehicle into the constructed two-dimensional rectangular coordinate system of time with respect to the number of vehicles around the vehicle, generate corresponding vehicle number data points, label the generated vehicle number data points, and connect each vehicle number data point in sequence by a straight line according to the time sequence to obtain a vehicle number change line graph; Obtain the average speed of the vehicles around the vehicle, construct a two-dimensional rectangular coordinate system of time with respect to the average speed of the vehicles around the vehicle, map the obtained average speed of the vehicles around the vehicle into the constructed two-dimensional rectangular coordinate system of time with respect to the average speed of the vehicles around the vehicle, generate corresponding vehicle speed data points, label the generated vehicle speed data points, and connect each vehicle speed data point in sequence by a straight line according to the time sequence to obtain a vehicle speed change line graph; Construct a two-dimensional rectangular coordinate system of time with respect to the traffic flow of the lane where the vehicle is located, map the obtained traffic flow of the lane where the vehicle is located into the constructed two-dimensional rectangular coordinate system of time with respect to the traffic flow of the lane where the vehicle is located, generate corresponding traffic flow data points, label the generated traffic flow data points, and connect each traffic flow data point in sequence by a straight line according to the time sequence to obtain a traffic flow change line graph.

5. The method for estimating the risk of vehicle collision based on an external dynamic environment according to claim 4, wherein The process of obtaining the first collision risk coefficient according to the predicted visible distance of the external environment, the vehicle number change line graph, the vehicle speed change line graph, and the traffic flow change line graph obtained includes: Obtain the number of vehicles around the vehicle corresponding to the vehicle number data point, draw a perpendicular line from the vehicle number data point to the time axis to obtain the intersection point of the perpendicular line and the time axis, draw a perpendicular line from the previous vehicle number data point to the time axis to obtain the intersection point of the perpendicular line and the time axis, and obtain the area of the right trapezoid formed by the vehicle number data point, the previous vehicle number data point, and the two intersection points of the two perpendicular lines and the time axis; Obtain the change number of vehicles around the vehicle corresponding to the vehicle number data point according to the number of vehicles around the vehicle corresponding to the obtained vehicle number data point and the area of the right trapezoid; Obtain the average speed of the vehicles around the vehicle corresponding to the vehicle speed data point, draw a perpendicular line from the vehicle speed data point to the time axis to obtain the intersection point of the perpendicular line and the time axis, draw a perpendicular line from the previous vehicle speed data point to the time axis to obtain the intersection point of the perpendicular line and the time axis, and obtain the area of the right trapezoid formed by the vehicle speed data point, the previous vehicle speed data point, and the two intersection points of the two perpendicular lines and the time axis; Obtain the change average speed of the vehicles around the vehicle corresponding to the vehicle speed data point according to the average speed of the vehicles around the vehicle corresponding to the obtained vehicle speed data point and the area of the right trapezoid; And so on, obtain the changed traffic flow of the lane where the vehicle is located corresponding to the traffic flow data point; Obtain the time corresponding to the traffic flow data point, and obtain the predicted visible distance corresponding to the time corresponding to the traffic flow data point; Obtain the first collision risk coefficient based on the number of changes in vehicles around the vehicle obtained, the average speed of changes in vehicles around the vehicle, the traffic flow of changes in the lane where the vehicle is located, and the predicted visible distance.

6. The method for estimating the risk of vehicle collision based on an external dynamic environment according to claim 5, wherein, The process of obtaining the second collision risk coefficient based on the obtained driver's status data includes: The driver's status data includes the driver's blink frequency. Obtain the time corresponding to all traffic flow data points, number the time corresponding to all the obtained traffic flow data points, the time corresponding to the traffic flow data point is the same as the label of the traffic flow data point, obtain the driver's blink frequency corresponding to each time, and obtain the second collision risk coefficient based on the obtained driver's blink frequency.

7. The method for estimating the vehicle collision risk based on an external dynamic environment according to claim 6, characterized in that The process of obtaining the third collision risk coefficient based on the obtained first collision risk coefficient and second collision risk coefficient includes: Set the weights of the first collision risk coefficient and the second collision risk coefficient according to the actual situation, and record the set weights of the first collision risk coefficient and the second collision risk coefficient as QZ_y and QZ_e respectively; Multiply the first collision risk coefficient by the weight of the first collision risk coefficient, multiply the second collision risk coefficient by the weight of the second collision risk coefficient, and add the two multiplied results to obtain the third collision risk coefficient.

8. The method for estimating the risk of vehicle collision based on an external dynamic environment according to claim 7, characterized in that, The process of obtaining the vehicle collision risk level based on the obtained third collision risk coefficient and the second collision risk coefficient includes: Set the collision risk coefficient threshold range according to the actual situation; Obtain the vehicle collision risk level based on the comparison result between the obtained third collision risk coefficient and the set collision risk coefficient threshold range and the value of the second collision risk coefficient; The vehicle collision risk level includes special-level collision risk, third-level collision risk, second-level collision risk, and first-level collision risk.

Images