Safety Assessment Method for Vehicle Driving State Based on Perceptual Risk Field in the Internet-Connected Environment

Abstract

The present invention discloses a method for safely evaluating the driving state of a vehicle based on a perceptual risk field in a connected environment, belonging to the technical field of automotive intelligent interaction and traffic safety. By collecting the vehicle motion state information in the connected area through the cloud, the historical motion state and real-time motion state data of each vehicle are obtained; by analyzing and processing the historical motion state data of the vehicle, the driving style parameters of the driver are obtained, including the desired time headway and the driver aggressiveness; by combining the driver driving style parameters with the vehicle real-time motion state data, a perceptual risk field model is constructed; by calculating the field strength of the perceptual risk field through the perceptual risk field model and calculating the driving safety factor of the vehicle according to the driving scenario of the vehicle, the safety evaluation result of the vehicle driving state is finally obtained. By considering the differences in driving styles of different drivers, the present invention constructs a perceptual risk field model in a connected environment, which can monitor, quantify, and evaluate the safety state of vehicles in the connected area in real time.

Description

Technical Field

[0001] The present invention belongs to the technical field of vehicle intelligent interaction and traffic safety, and particularly relates to a method for evaluating the safety of vehicle driving states based on a perceptual risk field in a connected environment. Background Technique

[0002] With the increase in the number of automobiles in China, traffic accidents occur frequently, and the safety of vehicle driving states is becoming increasingly important for traffic participants. As the core of vehicle control, the driving decisions made by drivers when facing traffic risks will directly affect the motion state of the vehicle and ultimately affect the driving safety state of the vehicle. Therefore, it is of great significance to quantitatively measure the risk perception of drivers in real time for evaluating the driving safety state of vehicles and reducing traffic accidents.

[0003] In the existing methods for evaluating the safety of vehicle driving states, the artificial potential field theory has been widely used because it can describe various microscopic driving behavior characteristics in more detail. However, the traditional artificial potential field model usually evaluates based on vehicle motion state information (such as relative distance, speed, acceleration, etc.), lacking consideration of driver factors, especially the uncertain factors of driver risk perception. For example, in the same traffic scenario, more conservative drivers have higher perceptual risks.

[0004] Therefore, it is urgent in the prior art to quantitatively measure the risk perception of drivers based on the perceptual risk field theory, and then realize the evaluation of vehicle driving states to solve the problem that it is difficult to grasp the uncertainty of driver risk perception in the existing evaluation methods. Summary of the Invention

[0005] The object of the present invention is to address the technical problems existing in the prior art and propose a method for evaluating the safety of vehicle driving states based on a perceptual risk field in a connected environment, so as to solve the problem that it is difficult to consider the uncertainty of driver risk perception in the existing methods; by considering the differences in driving styles of different drivers, a perceptual risk field model in a connected environment is constructed to monitor, quantitatively measure, and evaluate the safety states of vehicles in the connected area in real time.

[0006] To achieve the above object, the present invention adopts the following technical solutions: A method for evaluating the safety of vehicle driving states based on a perceptual risk field in a connected environment, characterized by including the following steps, and the following steps are carried out sequentially:

[0007] Step S1: Upload the motion state information of all vehicles in the connected environment to the cloud database. The cloud stores data for each vehicle for at least 15 minutes and updates it in real time, so as to obtain the historical motion state data and the real-time motion state data at the current moment of each vehicle; the motion state information of the vehicle includes speed v i , position x i and acceleration a i, where i represents the vehicle number;

[0008] Step S2: Process the vehicle historical motion state data obtained in Step S1, quantify the driving style, and obtain parameters characterizing the vehicle driving style. The driving style parameters include the desired time headway THW i and the driver aggressiveness k i ;

[0009] Step S3: Construct a perceptual risk field model based on the driving style parameters obtained in Step S2 and the vehicle's current position, speed, and acceleration obtained in Step S1, and calculate the perceptual risk field strength E of vehicle j around vehicle i perceived by vehicle i ji ;

[0010] Among them: The contour of the perceptual risk field of vehicle i is an ellipse, vehicle i is located at the center of the ellipse, the major semi-axis a of the ellipse = THW i ×v i , the minor semi-axis b takes the lane width, and v i is the vehicle speed of vehicle i at the current moment;

[0011] Step S4: Divide the influence area of the perceptual risk field in Step S3, and calculate the safety factor S of vehicle i under the influence of vehicle j ij ;

[0012] Step S5: Analyze the impact of vehicle j on the driving safety of vehicle i through the safety factor S ij obtained in Step S4. At the same time, repeat Steps S3 and S4 for each vehicle around vehicle i to obtain the driving safety evaluation result of vehicle i.

[0013] Furthermore, the calculation methods of the desired time headway THW i and the driver aggressiveness k i in Step S2 are as follows:

[0014]

[0015]

[0016]

[0017]

[0018] Among them, n is the number of samples of the vehicle i historical motion state data obtained from the cloud; THW im is the time headway at the m-th sample point of vehicle i; THW i is the desired time headway of vehicle i; x im , v im and a imThey are the position, speed, and acceleration of vehicle i at the m-th sample point, respectively; is the position of the vehicle in front of vehicle i at the m-th sample point; is the variance of the driving acceleration of vehicle i, is the average value of the acceleration of vehicle i.

[0019] Furthermore, the surrounding vehicle j of vehicle i in step S3 is defined as the vehicles in front, behind, front-left, front-right, rear-left, and rear-right within a range of 100 m around vehicle i; when there is no vehicle in the perceptual risk field, vehicle i cannot perceive the driving risk; when there is a vehicle in the perceptual risk field, the perceptual risk field strength E of the surrounding vehicle j perceived by vehicle i ji is calculated in the following two steps:

[0020] Step S301: Consider vehicle i and vehicle j as point charge i and point charge j. The contour of the risk field of point charge i is regarded as an elliptical grounded metal shell, which has a shielding effect on the charges outside the shell. At this time, the calculation method of the electrostatic field strength E ji ' is as follows,

[0021]

[0022] where k is the electrostatic constant, k = 9.0×10 9 N; q j is the electric charge of point charge j; r ij is the distance between point charges i and j; r ij0 is the radius of curvature at the intersection M of the extension line of the line connecting point charges i and j and the ellipse corresponding to point charge i. When the coordinate origin is taken at point charge i, M(x, y) satisfies:

[0023]

[0024] where x and y are the horizontal and vertical coordinates of point M, respectively;

[0025] Step S302: To describe the differences in risk perception of different drivers in the same traffic scenario, the radicalness k ji of the driver of vehicle i and the virtual electric charge q i of vehicle j are introduced into the electrostatic field strength E j '; the virtual electric charge q j is related to the masses of vehicle i and vehicle j. When the mass of vehicle j is larger relative to vehicle i, the virtual electric charge q j is larger, and the risk E ji perceived by vehicle i from vehicle j is larger. Specifically, the calculation method of the perceptual risk field strength E ji is as follows,

[0026]

[0027]

[0028] Among them, r ij represents the minimum distance from the point on the vehicle body contour of vehicle j that has the greatest risk impact on vehicle i to the vehicle body contour of vehicle i, and q j is the virtual charge quantity of vehicle j; m i and m j are the masses of vehicle i and vehicle j respectively; e is the charge quantity of a point charge, e = 1.60218×10 -19 C.

[0029] Furthermore, the calculation process of the safety factor S ij in step S4 is specifically divided into the following steps:

[0030] Step S401: According to the current lane offset of vehicle i, judge the driving scenario of vehicle i, and the driving scenarios include a following scenario and a lane-changing scenario;

[0031] Step S402: According to the driving scenario of vehicle i judged in step S401, divide the area of concern within the perceptual risk field of vehicle i into a primary area of concern and a secondary area of concern;

[0032] Step S403: According to the area of concern of vehicle i divided in step S402 and the perceptual risk field strength E ji obtained in step S3, calculate the driving safety factor S ij ; specifically, the calculation method of the driving safety factor S ij is as follows:

[0033]

[0034] Among them, when the center point of vehicle i does not exceed the left and right extreme value lines of the lane offset, vehicle i is in a following state, and when the center point of vehicle i exceeds the left and right extreme value lines of the lane offset, vehicle i is in a lane-changing state. For the left and right extreme values of the lane offset, take ±0.5 m;

[0035] Divide the perceptual risk field area of vehicle i into two parts, R1 and R2, and define R1 as the primary area of concern and R2 as the secondary area of concern;

[0036] In the following scenario, the primary area of concern R1 is directly in front of vehicle i, in front of the left lane, and in front of the right lane within the perceptual risk field;

[0037] In the lane-changing scenario, vehicle i changes lanes, and the primary area of concern R1 of vehicle i is in front of the current lane, in front of the left lane, and behind the left lane.

[0038] Furthermore, in step S5, through the safety factor Sij The driving safety assessment of vehicle i is specifically divided into the following steps:

[0039] Step S501: According to the safety factor S ij Judge the impact of vehicle j on the safety of vehicle i. Specifically,

[0040] S ij ∈[-1, 1], which is used to characterize the safety of vehicle i under the influence of vehicle j. The larger S ij is, the safer vehicle j is for vehicle i; S ij =-1 indicates that vehicle j has collided with vehicle i, and S ij =1 indicates that vehicle j does not pose a risk to vehicle i;

[0041] Step S502: Calculate the safety factor S ij for each vehicle around vehicle i, find the safety factor of the most dangerous vehicle for vehicle i, and output the safety evaluation result of vehicle i. Specifically,

[0042] S i =minS ij

[0043] S i ∈[-1, 1], which is used to represent the driving safety of vehicle i. The larger S i is, the safer vehicle i is. S i =-1 indicates that vehicle i has collided with other vehicles, and S i =1 indicates that there is no driving risk for vehicle i.

[0044] Through the above design scheme, the present invention can bring the following beneficial effects:

[0045] 1. The vehicle driving state safety assessment method based on the perceptual risk field solves the problem that it is difficult to consider the uncertainty of the driver's perception of driving risk in the traditional driving safety assessment method. The present invention assesses the safety state of the vehicle according to the differences in the perceptual risks of different drivers, which is of great significance for the risk warning of vehicles in the networked environment and the safety decision-making of autonomous vehicles;

[0046] 2. The perceptual risk field model proposed by the present invention can reflect the differences in risk measurement of different drivers in the same traffic scenario; the proposed method for dividing the area of concern reflects the different sources of risk of drivers in different traffic scenarios; the proposed safety factor calculation method can fully consider the different directions of surrounding vehicles relative to the self-vehicle to obtain the driving safety assessment result compared with the traditional artificial potential field risk assessment;

[0047] 3. The present invention can be applied to traffic control by establishing a connected cloud database. The present invention fully considers the impact of drivers on traffic safety. The method for evaluating the safety state of vehicle driving conditions based on the perceptual risk field in a connected environment proposed by the present invention can monitor, quantify, and evaluate the safety state and risk sources of vehicles in the connected area in real time, which is of great significance for analyzing the causes of traffic accidents in the area. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] The following drawings are used to provide a further understanding of the present invention and form a part of the present invention application. The schematic embodiments and descriptions of the present invention are used to understand the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0049] Figure 1 is a schematic flowchart of the method for evaluating the safety state of vehicle driving conditions based on the perceptual risk field in a connected environment of the present invention;

[0050] Figure 2 is a schematic diagram of the perceptual risk field model provided by the present invention;

[0051] Figure 3 is a schematic diagram of the main attention area division in the following - following scenario provided by the present invention;

[0052] Figure 4 is a schematic diagram of the main attention area division in the lane - changing scenario provided by the present invention;

[0053] Figure 5 is a graph of the safety factor function provided by the present invention;

[0054] Figure 6 is a schematic diagram of the traffic scenario corresponding to the safety factor provided by the present invention, where Figure 6 in (a) - (c) is the traffic scenario corresponding to S ij = 1; Figure 6 in (d) - (f) is the traffic scenario corresponding to - 1 < S ij < 1; Figure 6 in (g) - (h) is the traffic scenario corresponding to S ij = - 1. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0055] The present invention applies the concept of "perceptual risk" to the field of traffic safety technology to describe the uncertainty of different drivers' perception of driving risks.

[0056] The vehicle driving state safety assessment method based on the perceptual risk field in the networked environment proposed by the present invention collects the vehicle motion state information in the networked area through the cloud, and obtains the historical motion state and real-time motion state data of each vehicle; by analyzing and processing the historical motion state data of the vehicle, the driving style parameters of the driver are obtained, and the driving style parameters include the expected time headway and the driver aggressiveness; by combining the driver driving style parameters with the vehicle real-time motion state data, a perceptual risk field model is constructed; by calculating the perceptual risk field intensity through the perceptual risk field model and calculating the driving safety factor of the vehicle according to the driving scenario of the vehicle, the vehicle driving state safety evaluation result is finally obtained. The present invention solves the problem that it is difficult to consider the uncertainty of the driver's perception of driving risks in the existing methods; by considering the differences in driving styles of different drivers, a perceptual risk field model in the networked environment is constructed, which can monitor, quantify, and evaluate the safety state of vehicles in the networked area in real time.

[0057] In order to make the objectives, features, and advantages of the present invention more obvious and understandable, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the present invention is not limited by the following embodiments, and the specific implementation manners can be determined according to the technical solutions of the present invention and the actual situation. In order to avoid confusing the essence of the present invention, well-known methods, processes, and procedures are not described in detail.

[0058] As Figure 1 shown, the vehicle driving state safety assessment method based on the perceptual risk field in the networked environment includes:

[0059] Step S1: The motion state information of all vehicles in the networked environment (including speed v i , position x i , acceleration a i , where i represents the vehicle number) is uploaded to the cloud database, and the cloud stores at least 15 minutes of data for each vehicle and updates it in real time, so as to obtain the historical motion state data of each vehicle and the real-time motion state data at the current moment;

[0060] Step S2: Process the vehicle historical motion state data obtained in Step S1, perform driving style quantification, and obtain parameters representing the vehicle driving style, including the expected time headway THW i and the driver aggressiveness k i , specifically:

[0061]

[0062]

[0063]

[0064]

[0065] where n is the number of samples of the historical motion state data of vehicle i obtained by the cloud; THW im is the time headway at the m-th sample point of vehicle i; THW i is the desired time headway of vehicle i. Existing research has shown that when the time headway is greater than three seconds, the vehicle is in a free driving state; x im , v im , a im are the position, speed, and acceleration of vehicle i at the m-th sample point; is the position of the vehicle in front of vehicle i at the m-th sample point; is the variance of the driving acceleration of vehicle i, is the average value of the acceleration of vehicle i.

[0066] Step S3. Construct a perceptual risk field model based on the driving style parameters obtained in step S2 and the current position, speed, and acceleration of the vehicle obtained in step S1. The contour of the perceptual risk field of vehicle i is an ellipse, and vehicle i is located at the center of the ellipse. The major semi-axis a of the ellipse = THW i ×v i , and the minor semi-axis b is taken as the lane width of 3.5 m. v i is the vehicle speed of vehicle i at the current moment. As shown in Figure 2 , and calculate the perceptual risk field strength E ji perceived by vehicle i from the surrounding vehicle j to vehicle i. The center point of vehicle i in the figure is the coordinate origin, j is the lane-changing vehicle, and r ij represents the minimum distance from the point on the body contour of vehicle j that has the greatest impact on the risk of vehicle i to the body contour of vehicle i. Point M is the intersection point of the extension line of the connection line between the center points of vehicle i and vehicle j and the elliptical contour of the perceptual risk field of vehicle i. For the surrounding vehicle j of vehicle i, it is specifically defined as the vehicles in front, behind, left front, right front, left rear, and right rear within 100 m around vehicle i.

[0067] Specifically, the calculation method of E ji is divided into the following two steps:

[0068] Step S301. Regard vehicle i and vehicle j as point charge i and point charge j. The contour of the risk field of point charge i is regarded as the grounded metal shell of the ellipse, which has a shielding effect on the charges outside the shell. At this time, the calculation method of the electrostatic field strength E ji ' is,

[0069]

[0070] where k is the electrostatic force constant, k = 9.0×10 9 N; q j is the electric charge of point charge j; rij is the distance between point charges i and j; r ij0 is the radius of curvature at the intersection M of the extension line of the line connecting point charges i and j and the ellipse formed by i. When the coordinate origin is at point charge i, M(x, y) satisfies:

[0071]

[0072] where x and y are the abscissa and ordinate of point M, respectively;

[0073] Step S302: To describe the differences in risk perception among different drivers in the same traffic scenario, the aggressiveness k of driver of vehicle i and the virtual charge quantity q of vehicle j are introduced into the electrostatic field strength E ji '. Since the aggressiveness k of the driver i is quantified by the historical acceleration fluctuation of vehicle i, when dealing with risks, the driver with a greater aggressiveness is more likely to take a greater braking deceleration. The virtual charge quantity q j is related to the masses of vehicle i and vehicle j. When the mass of vehicle j is larger relative to vehicle i, the virtual charge quantity q i is larger, and the risk E j perceived by vehicle i from vehicle j to vehicle i is greater. Specifically, the calculation method of the perceptual risk field strength E j is ji ji ij j

[0074]

[0075]

[0076] where r ij represents the minimum distance from the point on the body contour of vehicle j that has the greatest impact on the risk of vehicle i to the body contour of vehicle i, q j is the virtual charge quantity of vehicle j; m i , m j are the masses of vehicle i and vehicle j, respectively; e is the charge quantity of a point charge, e = 1.60218×10 -19 C.

[0077] Step S4: Divide the area affected by the perceptual risk field in Step S3, and calculate the safety factor S ij of the impact of vehicle j on the safety of vehicle i. Specifically:

[0078] Step S401: According to the current lane offset of vehicle i, judge the driving scenario of vehicle i. The following is a following scenario as Figure 3 shown, and the following is a lane-changing scenario as Figure 4 shown. In the figures, vehicle i is located in the middle lane. Specifically, Figure 3Among them, L1 and L5 are lane lines, L3 is the lane center line, and L2 and L4 are the left and right extreme value lines of the lane offset when vehicle i is driving in the lane keeping state. When the center point of vehicle i does not exceed the left and right extreme value lines of the lane offset, vehicle i is considered to be in a following state, as Figure 3 shown; when the center point of vehicle i exceeds the left and right extreme value lines of the lane offset, vehicle i is considered to be in a lane-changing state, as Figure 4 shown. Further, the left and right extreme values of the lane offset are taken as ±0.5 m.

[0079] Step S402: According to the driving scenario of vehicle i determined in step S401, divide the area of concern within the perceptual risk field of vehicle i into a primary area of concern and a secondary area of concern. The area of concern division for the following scenario and the lane-changing scenario is as Figure 3 and Figure 4 , and the shaded part R1 is the primary area of concern. Specifically, Figure 3 shows that when the leading vehicle of vehicle i enters the perceptual risk field of vehicle i in the following scenario, vehicle i will sense the risk of the leading vehicle. If the risk is too high, vehicle i will decelerate and yield or change to another lane; the following vehicle of vehicle i is also within the perceptual risk field of vehicle i, but generally vehicle i will not react to the behavior of the following vehicle, that is, the risk measurement of vehicle i for the vehicles in front and behind is different. The present invention divides the perceptual risk field area of vehicle i into two parts, R1 and R2, and defines R1 as the primary area of concern and R2 as the secondary area of concern. In Figure 3 , R1 is directly in front of vehicle i, in front of the left lane, and in front of the right lane within the perceptual risk field.

[0080] Figure 4 For the lane-changing scenario, when vehicle i changes lanes, the primary area of concern of vehicle i will change. The primary area of concern R1 of vehicle i will change from in front of the current lane, in front of the left lane, and in front of the right lane to in front of the current lane, in front of the left lane, and behind the left lane. Figure 4 shows that the vehicle behind the left lane of vehicle i is within the primary area of concern of vehicle i in the lane-changing scenario.

[0081] Step S403: According to the area of concern of vehicle i divided in step S402 and the perceptual risk field strength E ji obtained in step S3, calculate the driving safety factor S ij . Specifically, the calculation method of S ij is as follows:

[0082]

[0083] Among them, when vehicle j is within the primary area of concern of vehicle i, the function curve of S ij with respect to r ij is as Figure 5 shown.

[0084] Step S5, the safety factor S obtained through step S4 ij Analyze the impact of vehicle j on the driving safety of vehicle i. Specifically, as Figure 6 shown, S ij ∈[-1, 1], which is used to characterize the safety of vehicle i under the influence of vehicle j. The larger S ij is, the safer vehicle j is for vehicle i. S ij = 1 means that vehicle j does not pose a risk to vehicle i, as shown in Figure 6 (a)-(c) in; -1 < S ij < 1 indicates that vehicle j has an impact on vehicle i but no collision has occurred, as shown in Figure 6 (d)-(f) in; S ij = -1 means that vehicle j has collided with vehicle i, as shown in Figure 6 (g)-(h) in.

[0085] Furthermore, calculate the safety factor S for each vehicle around vehicle i ij , find the safety factor of the most dangerous vehicle for vehicle i, and output the safety evaluation result of vehicle i. Specifically,

[0086] S i = min S ij

[0087] S i ∈[-1, 1], which is used to represent the driving safety of vehicle i. The larger S i is, the safer vehicle i is. S i = -1 means that vehicle i has collided with other vehicles. S i = 1 means that there is no driving risk for vehicle i.

[0088] Furthermore, through the above steps, the safety factor S of each vehicle in the connected area can be calculated i , so as to evaluate the driving safety of each vehicle.

[0089] In summary, the vehicle driving state safety assessment method based on the perceptual risk field in the connected environment proposed by the present invention quantifies the driver's perceived risk based on the perceptual risk field theory, and then realizes the vehicle driving state assessment, which can solve the problem of the uncertainty of the driver's risk perception that is difficult to grasp in the existing assessment methods.

Claims

1. A method for evaluating the safety of a vehicle's driving state based on a perceptual risk field in a networked environment, characterized in that, including the following steps, and the following steps are carried out sequentially: Step S1: The motion state information of all vehicles in the connected environment is uploaded to the cloud database. The cloud stores the data of each vehicle for at least 15 minutes and updates it in real time, so as to obtain the historical motion state data and the real-time motion state data at the current moment of each vehicle. The motion state information of the vehicle includes speed v i , position x i and acceleration a i , where i represents the vehicle number; Step S2: Process the vehicle historical motion state data obtained in Step S1, quantify the driving style, and obtain parameters characterizing the vehicle driving style. The driving style parameters include the desired time headway THW i and the driver aggressiveness k i ; Step S3: Construct a perceptual risk field model based on the driving style parameters obtained in step S2 and the current position, speed, and acceleration of the vehicle obtained in step S1, and calculate the perceptual risk field strength E of vehicle j around vehicle i perceived by vehicle i ji ; Wherein: The contour of the perceived risk field of vehicle i is an ellipse, and vehicle i is located at the center of the ellipse. The major semi-axis a of the ellipse is THW i × v i , the minor semi-axis b is taken as the lane width, and v i is the vehicle speed of vehicle i at the current moment; Step S4: Divide the perceived risk field influence area in Step S3, and calculate the safety factor S of vehicle i under the influence of vehicle j ij ; Step S5: Safety factor S obtained through step S4 ij Analyze the impact of vehicle j on the driving safety of vehicle i, and at the same time repeat steps S3 and S4 for each vehicle around vehicle i to obtain the driving safety evaluation result of vehicle i; The desired time headway THW in step S2 i and the driver aggressiveness k i are calculated as follows: where n is the number of samples of the historical motion state data of vehicle i obtained by the cloud; THW im is the time headway of vehicle i at the m-th sample point; THW i is the desired time headway of vehicle i; x im , v im and a im are the position, speed, and acceleration of vehicle i at the m-th sample point, respectively; is the position of the vehicle in front of vehicle i at the m-th sample point; is the variance of the driving acceleration of vehicle i, is the average value of the acceleration of vehicle i; The surrounding vehicle j of vehicle i in step S3 is defined as the vehicles in front of, behind, in the front left, in the front right, in the rear left, and in the rear right within 100 m around vehicle i; when there is no vehicle in the perceptual risk field, vehicle i cannot perceive the driving risk; when there is a vehicle in the perceptual risk field, the perceptual risk field strength E ji of the surrounding vehicle j of vehicle i perceived by vehicle i is calculated in the following two steps: Step S301: Consider vehicle i and vehicle j as point charge i and point charge j respectively. The contour of the risk field of point charge i is regarded as an elliptical grounded metal shell, which has a shielding effect on the charges outside the shell. At this time, the electrostatic field strength E ji The calculation method is as follows: Among them, k is the electrostatic constant, k = 9.0×10 9 N; q j is the electric charge quantity of point charge j; r ij is the distance between point charges i and j; r ij0 is the radius of curvature at the intersection point M of the extension line of the line connecting point charges i and j and the ellipse corresponding to point charge i. When the origin of coordinates is taken at point charge i, M(x, y) satisfies: where x and y are the horizontal and vertical coordinates of point M respectively; Step S302. To describe the differences in risk perception of different drivers in the same traffic scenario, the aggressiveness k of the driver of vehicle i and the virtual electric charge q of vehicle j are introduced into the electrostatic field strength E ji '. i The virtual electric charge q j is related to the masses of vehicle i and vehicle j. When the mass of vehicle j is larger than that of vehicle i, the virtual electric charge q j is larger, and the risk E j perceived by vehicle i from vehicle j is larger. Specifically, the calculation method of the perceptual risk field strength E ji is as follows: ji ​ Among them, r ij represents the minimum distance from the point on the vehicle body contour of vehicle j that has the greatest risk impact on vehicle i to the vehicle body contour of vehicle i, and q j is the virtual charge quantity of vehicle j; m i and m j are the masses of vehicle i and vehicle j respectively; e is the charge quantity of a point charge, e = 1.60218×10 -19 C; The safety factor S in step S4 ij is calculated through the following steps: Step S401: According to the current lane offset of vehicle i, judge the driving scenario of vehicle i, and the driving scenario includes a following scenario and a lane-changing scenario; Step S402: According to the driving scenario of vehicle i judged in step S401, divide the attention area in the perceptual risk field of vehicle i into two areas, namely the main attention area and the secondary attention area; Step S403: Based on the area of vehicle i's concern divided in step S402 and the perceived risk field strength E obtained in step S3 ji , calculate the driving safety factor S ij ; Specifically, the calculation method of the driving safety factor S ij is as follows: In the step S5, the driving safety of vehicle i is evaluated through a safety factor S ij which specifically includes the following steps: Step S501: Determine the impact of vehicle j on the safety of vehicle i according to the safety factor S ij Specifically, S ij ∈[-1, 1], which is used to characterize the safety of vehicle i under the influence of vehicle j. The larger S ij is, the safer vehicle j is for vehicle i; S ij =-1 indicates that vehicle j has collided with vehicle i, S ij =1 indicates that vehicle j poses no risk to vehicle i; Step S502: Calculate the safety factor S for each vehicle around vehicle i ij , find out the safety factor of the most dangerous vehicle for vehicle i, and output the safety evaluation result of vehicle i. Specifically, S i = minS ij S i ∈[-1, 1], which is used to represent the driving safety of vehicle i. The larger the S i is, the safer vehicle i is. When S i = -1, it means that vehicle i has collided with other vehicles. When S i = 1, it means that there is no driving risk for vehicle i.

2. The method for evaluating the safety of a vehicle driving state based on a perceptual risk field in a connected environment according to claim 1, wherein: When the center point of vehicle i does not exceed the left and right extreme lines of lane offset, vehicle i is in a following state. When the center point of vehicle i exceeds the left and right extreme lines of lane offset, vehicle i is in a lane-changing state. The left and right extreme values of lane offset are taken as ±0.5 m.

3. The vehicle driving state safety assessment method based on the perceptual risk field in the networked environment according to claim 1, characterized in that: Divide the perceptual risk field area of vehicle i into two parts, R1 and R2, and define R1 as the main attention area and R2 as the secondary attention area; In the following scenario, the main attention area R1 is the front, the front of the left lane, and the front of the right lane of vehicle i in the perceptual risk field; In the lane-changing scenario, when vehicle i changes lanes, the main attention area R1 of vehicle i is the front of the current lane, the front of the left lane, and the rear of the left lane.

Images