Method and device for quantifying intelligent vehicle driving risk field size and storage medium

Abstract

The application discloses a kind of intelligent car driving risk field size quantification method, device and storage medium, wherein quantification method includes: the size information of each factor influencing the size of intelligent car driving risk is acquired;With the length-width size of intelligent car, pedestrian, obstacle and lane line, a rectangular model is constructed;The envelope optimization is carried out to the rectangular model, and envelope model is obtained;According to the range of the area range enveloped by envelope model, the range of risk field peak value generated by each influencing factor in the risk field suffered by intelligent car when driving under complex and changeable traffic environment and structured road condition is determined;According to the range of the risk field peak value determined, based on the social force thought, the mathematical model of field strength change is constructed.The application proposes a unified and can accurately reflect the mathematical model of the quantification method of the size of intelligent car driving risk under complex and changeable traffic environment according to the relationship between each factor influencing driving safety that the driving risk suffered by intelligent car originates from.

Description

Technical Field

[0001] This invention relates to the field of intelligent driving. More specifically, it relates to a method, apparatus, and storage medium for quantifying the magnitude of driving risks in intelligent vehicles. Background Technology

[0002] Intelligent vehicles, which integrate vehicle intelligence and connectivity technologies, are the main form of intelligent transportation systems, and the driving safety of intelligent vehicles has always been a key focus in this field.

[0003] Most existing driving risk models only consider simple factors such as road user speed and mass, and cannot comprehensively consider complex factors such as direction, road conditions, and social psychology (obedience to road traffic rules). Summary of the Invention

[0004] To address the shortcomings in this field, the present invention provides a method, device, and storage medium for quantifying the magnitude of driving risks in intelligent vehicles. The purpose is to determine a mathematical model that can reflect the quantitative standard of driving risks in intelligent vehicles under complex and ever-changing traffic conditions, based on the relationship between the sources of driving risks and various factors affecting driving safety.

[0005] To achieve the above-mentioned technical objectives, the present invention adopts the following technical solution:

[0006] A method for quantifying the driving risk of intelligent vehicles, characterized by comprising:

[0007] (1) Obtain the size information of each factor that affects the driving risk of intelligent vehicles, including intelligent vehicles, obstacles, pedestrians and lane lines;

[0008] (2) Construct a rectangular model using the length and width dimensions of the intelligent vehicle, pedestrians, obstacles and lane lines; perform envelope optimization on the constructed rectangular model to obtain the optimized envelope model;

[0009] (3) Based on the range of the region enclosed by the optimized envelope model, determine the range of the peak value of the risk field generated by each influencing factor in the risk field of intelligent vehicles under complex and ever-changing traffic environment and structured road conditions.

[0010] (4) Based on the determined range of the risk field peak, a mathematical model of field strength change is constructed based on the concept of social force.

[0011] This application proposes a method, device, and storage medium for quantifying the magnitude of the driving risk field for intelligent vehicles, establishing a mathematical model for calculating the magnitude of driving risks in intelligent vehicles. The mathematical model analyzes, from the perspective of traditional social forces, the social forces that intelligent vehicles may experience in structured roads, including those from other intelligent vehicles traveling in the opposite or same direction, social forces from stationary obstacles ahead, social forces from pedestrians crossing crosswalks at intersections, and social forces from lane lines. It proposes a concept of an intelligent vehicle driving risk field based on social forces. In comparison, this model can more reasonably and adaptively characterize the magnitude of driving risks for vehicles under complex and ever-changing traffic conditions, providing a new theoretical model and judgment basis for the safe driving of intelligent vehicles.

[0012] Compared with existing technologies, the advantages of this invention are:

[0013] (1) For the first time, the concept of social force theory is applied to the field of intelligent vehicle driving safety. It fully considers the interaction mechanism between people, vehicles, roads and environment, as well as complex factors such as road user types and road conditions. It also introduces social psychology and uses the magnitude of the driving safety field force to characterize the risk level of the vehicle during driving.

[0014] (2) Compared with existing horizontal and vertical models based on speed and distance, as well as potential field models, the social force model for intelligent vehicle driving safety established in this invention, which considers social psychology, is applicable to complex driving scenarios and provides new ideas and methods for active safety of intelligent vehicles in complex traffic scenarios. Attached Figure Description

[0015] Figure 1 A flowchart for constructing a risk field;

[0016] Figure 2 Here is an example of an envelope model for optimizing the rectangular model of each influencing factor, where n = 3. Detailed Implementation

[0017] Example 1

[0018] A method for quantifying the driving risk of intelligent vehicles, comprising the following steps:

[0019] (1) Obtain the size information of each factor that affects the driving risk of intelligent vehicles, including intelligent vehicles, obstacles, pedestrians and lane lines;

[0020] (2) Construct a rectangular model using the length and width dimensions of the intelligent vehicle, pedestrians, obstacles and lane lines; perform envelope optimization on the constructed rectangular model to obtain the optimized envelope model;

[0021] (3) Based on the range of the region enclosed by the optimized envelope model, determine the range of the peak value of the risk field generated by each influencing factor in the risk field of intelligent vehicles under complex and ever-changing traffic environment and structured road conditions.

[0022] (4) Based on the determined range of the risk field peak, a mathematical model of field strength change is constructed based on the concept of social force.

[0023] In step (2), the method for performing envelope optimization on the constructed rectangular model to obtain the optimized envelope model is as follows:

[0024] (21) Construct a simplified rectangular model using the length and width dimensions of the intelligent car, pedestrian, and obstacles. The length of the rectangular model is L, and the width of the rectangular model is W;

[0025] (22) Select n circles of equal size and equal spacing to enclose the simplified rectangular model, where n≥3.

[0026] In step (22), the method of selecting n equally sized and equally spaced circles to enclose the simplified rectangular model is as follows:

[0027] (221) Based on the length and width dimensions of the simplified rectangular model, the number of enclosing circles is generally chosen to be n = 3;

[0028] (222) Determine the radius r of the envelope circles based on the number of circles in the envelope:

[0029]

[0030] Where r is the radius of the envelope circle, L is the length of the simplified rectangular model, w is the width of the simplified rectangular model, and n is the number of circles in the envelope;

[0031] (223) Determine the optimized envelope model based on the radius of the circle and the number of circles in the envelope.

[0032] In step (3), the determined risk field peak range is the range enclosed by the circle used for envelopment in the optimized envelope model.

[0033] In step (4), the determined peak range of the risk field is:

[0034]

[0035] in, A represents the peak magnitude of the driving risk field posed by other intelligent vehicles in the same lane to the vehicle. b r is the amplitude coefficient of the field strength effect on the safety field of vehicles traveling in the same direction in the same lane. ab Let d be the sum of the radii of car a and car b. ab Let R be the distance between car a and car b.b The sensitivity coefficient for the distance of the field strength applied in the same direction of traffic safety is Δv. ab Cosθ represents the difference in relative speed between the two vehicles. b H is the cosine of the angle between the direction of the relative velocity of the two vehicles and the direction of the line connecting the two vehicles. b M is the relative speed sensitivity coefficient for vehicles traveling in the same direction. a M b K represents the equivalent mass of car a and car b, respectively. b Let R represent the degree of compliance of vehicle b with road traffic rules, and let R be the road condition factor.

[0036] Equivalent mass M i It should be expressed as:

[0037]

[0038] Where i = a, b; α, β, ε, ζ are constant coefficients; m i The quality of intelligent vehicles; T i For the type of intelligent vehicle; v i For the speed of intelligent vehicles;

[0039] Road traffic users' compliance with road traffic rules (K) i for:

[0040]

[0041] Where, η * η represents the standard value for compliance with road traffic rules. i Let κ be the road traffic rule compliance value for road traffic participants, and let κ be an undetermined constant.

[0042] The road condition factor R is:

[0043]

[0044] In the formula: δ is road visibility; μ is road surface adhesion coefficient; ρ is road curvature; τ is road slope; γ1, γ2, γ3, and γ4 are all undetermined constants, and γ1, γ2 < 0, γ3, γ4 > 0; μ * δ is the standard road surface adhesion coefficient. * Standard road visibility; ρ * For standard road curvature; τ * This is the standard road slope.

[0045] Based on the optimization model, and using the concept of social forces, a mathematical model is determined to account for the changes in risk field strength caused by other influencing factors such as intelligent vehicles, obstacles, pedestrians, and lane lines in a complex and ever-changing traffic environment. The steps include:

[0046] Based on the concept of social forces, and combined with other characteristic factors, the mathematical model for the changes in risk field strength caused by other influencing factors such as intelligent vehicles, obstacles, pedestrians, and lane lines in a complex and ever-changing traffic environment is as follows:

[0047]

[0048] In the formula, E b′ A mathematical model for the change in field strength of factor b in the driving risk field; r ab′ Let D be the sum of the radii of any point within the field and the radius of vehicle a, which are influenced by factor b; ab′ θ represents the distance from any point within the driving environment to vehicle a, which is the driving risk caused by factor b in the driving environment; b′ R is the angle between any point within the risk field generated by factor b and the direction of the line connecting point a (vehicle a) and factor b; b The field strength sensitivity coefficient for factor b is the distance sensitivity coefficient.

[0049] Example 2

[0050] The present invention also provides a device for quantifying the driving risk of an intelligent vehicle, comprising a processor and a memory; the memory stores a program or instructions, which are loaded by the processor and execute the steps of the method for quantifying the driving risk of an intelligent vehicle in Embodiment 1.

[0051] Example 3

[0052] The present invention also provides a computer-readable storage medium storing a program or instructions, which, when executed by a processor, implement the steps of the method for quantifying the magnitude of driving risk of an intelligent vehicle as described in Embodiment 1.

Claims

1. A method for quantifying the driving risk of intelligent vehicles, characterized in that, include: (1) Obtain the size information of each factor that affects the driving risk of intelligent vehicles, including intelligent vehicles, obstacles, pedestrians and lane lines; (2) Construct a rectangular model using the length and width dimensions of the intelligent vehicle, pedestrian, obstacle and lane line; perform envelope optimization on the constructed rectangular model to obtain the optimized envelope model; (3) Based on the range of the region enclosed by the optimized envelope model, determine the range of the peak value of the risk field generated by each influencing factor in the risk field of intelligent vehicles under complex and ever-changing traffic environment and structured road conditions. (4) Based on the determined range of the risk field peak, construct a mathematical model of field strength change based on the concept of social force; In step (2), the method for performing envelope optimization on the constructed rectangular model to obtain the optimized envelope model is as follows: (21) Determine the radius of the enveloping circles based on the length and width dimensions of the rectangular model and the number of circles in the enveloping circle. : in, The length of the rectangular model; The width of the rectangular model; The number of circles in the envelope. ≥1; (22) Based on the radius of the circle and the number of envelope circles in the envelope model, the optimized envelope model is obtained; In step (4), based on the determined range of the risk field peak, and based on the concept of social forces, for autonomous vehicles... Constructing influencing factors The mathematical model for the change in electric field strength is: In the formula, Factors affecting driving risks Mathematical model of electric field change; Influencing factors The resulting driving risks occur at any point within the site and for the vehicle itself. The radius and; Factors affecting driving conditions The resulting driving risk is that the vehicle can travel from any point within the parking area to its own vehicle. The distance; Influencing factors Any point within the risk field generated and the vehicle and influencing factors The angle between the directions of the lines connecting two points; Influencing factors The field strength sensitivity coefficient at the distance of action; In step (4), based on the established mathematical model, the quantization model of the peak size within the peak range of the optimized envelope model is obtained as follows: In the formula, Influencing factors In the car A model of field strength variation in the driving risk field generated during driving; To comprehensively consider the peak size of other characteristic factors; Influencing factors In the car A quantitative model of the driving risk field generated during driving.

2. The method for quantifying the driving risk of an intelligent vehicle according to claim 1, characterized in that, In step (3), the determined risk field peak range is the range encompassed by all circles used for the envelope in the optimized envelope model.

3. The method for quantifying the driving risk of an intelligent vehicle according to claim 1, characterized in that, Peak size for: in, Factors affecting the complex and ever-changing driving environment The field strength intensity influence coefficient; For bicycle and influencing factors The radius and; For bicycle and influencing factors The distance between them; For bicycle and influencing factors Relative speed difference; For bicycle and influencing factors The cosine of the angle between the direction of relative velocity and the direction of the line connecting the two points; The relative velocity sensitivity coefficient; Bicycles and influencing factors The quality; Influencing factors The degree of obedience to road traffic rules; For road condition factors.

4. The method for quantifying the driving risk of an intelligent vehicle according to claim 3, characterized in that, Influencing factors Obedience to road traffic rules : in, The standard value for compliance with road traffic rules; The value for vehicle b's compliance with road traffic rules; is an undetermined constant.

5. A device for quantifying the magnitude of driving risk in intelligent vehicles, characterized in that, It includes a processor and a memory; the memory stores a program or instructions which are loaded and executed by the processor to implement the steps of the method for quantifying the magnitude of driving risk of an intelligent vehicle as described in any one of claims 1 to 4.

6. A computer-readable storage medium storing a program or instructions that, when executed by a processor, implement the steps of the method for quantifying the magnitude of driving risk of an intelligent vehicle as described in any one of claims 1 to 4.

Images