Method, program product and system for evasive maneuvers of a vehicle at an intersection

Abstract

The application relates to a kind of methods for vehicle refuge at intersection, comprising: at intersection, identifying at least one straight vehicle in first lane and at least one left turn vehicle in second lane opposite to first lane;Respectively obtain the driving parameter of straight vehicle and left turn vehicle and the working condition parameter of intersection;Based on driving parameter and working condition parameter, calculate straight risk score by straight risk score model, calculate left turn risk score by left turn risk score model, calculate comprehensive risk score by straight risk score and left turn risk score;According to straight risk score, left turn risk score and comprehensive risk score, the comprehensive risk grade is determined together;According to comprehensive risk grade, corresponding refuge measure is taken.The application also relates to corresponding computer program product and refuge system.The influence factors of intersection can be fully and comprehensively analyzed, and targeted refuge measure is generated, so as to effectively improve the traffic safety of vehicle at intersection.

Description

Technical Field

[0001] This invention relates to the field of vehicles, and more particularly to a method for vehicle hazard avoidance at intersections. The invention also relates to a corresponding computer program product and a corresponding vehicle hazard avoidance system at intersections. Background Technology

[0002] Crossroads, T-junctions, and other intersections are high-risk locations for traffic conflicts and accidents. Due to multiple factors such as intersection traffic rules, vehicle driving status, road conditions, and obstructed visibility, vehicles going straight and turning left in opposite lanes are prone to cutting in and failing to brake in time, which can lead to minor scrapes, collisions, and other traffic accidents, seriously affecting the safety and efficiency of traffic at intersections.

[0003] Currently, collision risk is usually judged based on the detection information of a single vehicle. This risk assessment has a single dimension and is difficult to adapt to the complex and ever-changing real intersection scenarios, which can easily lead to false warnings and missed warnings. Summary of the Invention

[0004] Therefore, the purpose of this invention is to propose an improved method for vehicle hazard avoidance at intersections, which can fully and comprehensively analyze the influencing factors of intersections and generate targeted hazard avoidance measures, thereby effectively improving the safety of vehicle passage at intersections.

[0005] The present invention also aims to provide a corresponding computer program product and a corresponding vehicle avoidance system at intersections.

[0006] According to a first aspect of the present invention, a method for avoiding hazards for vehicles at intersections is provided, wherein the method includes at least the following steps: S1: At the intersection, identify at least one straight-going vehicle in the first lane and at least one left-turning vehicle in the second lane opposite to the first lane; S2: Obtain the driving parameters of the straight-going vehicle and the left-turning vehicle, as well as the operating parameters of the intersection; S3: Based on the driving parameters and the operating parameters, a straight-ahead risk score is calculated using the straight-ahead risk scoring model, a left-turn risk score is calculated using the left-turn risk scoring model, and a comprehensive risk score is calculated using the straight-ahead risk score and the left-turn risk score. S4: Determine the overall risk level based on the straight-ahead risk score, the left-turn risk score, and the comprehensive risk score; S5: Take appropriate risk avoidance measures based on the comprehensive risk level.

[0007] Compared to existing technologies, the method for vehicle risk avoidance at intersections according to the present invention acquires the driving parameters of straight-going and left-turning vehicles in the opposite lanes and the operating parameters of the intersection, respectively. Based on these parameters, a straight-going risk score and a left-turn risk score are calculated, and a comprehensive risk score is calculated from these. The comprehensive risk level is determined by combining the straight-going risk score, the left-turn risk score, and the comprehensive risk score. This fully considers the driving conditions of straight-going and left-turning vehicles at the current intersection and comprehensively analyzes the collision probability of the two in the current scenario, thereby obtaining an accurate comprehensive risk level. This can reliably avoid false warnings and missed warnings and allows for more targeted risk avoidance measures, thereby improving the safety and efficiency of vehicle passage at intersections.

[0008] For example, the driving parameters are selected from the following group: driving speed, driving trajectory, relative distance to other vehicles, steering angle, field of vision obstruction area, brake pedal opening, vehicle type; and / or, the operating condition parameters are selected from the following group: intersection speed limit, intersection size, lane width, traffic light status, intersection slope.

[0009] For example, the straight-ahead risk scoring model includes a speed risk item and a braking distance risk item. The speed risk item is determined by the speed of the straight-ahead vehicle and the speed limit at the intersection, while the braking distance risk item is determined by the braking distance of the straight-ahead vehicle and the remaining braking distance.

[0010] For example, the left-turn risk scoring model includes a field-of-view obstruction risk item and a relative distance risk item. The field-of-view obstruction risk item is determined by the field-of-view obstruction area of ​​the left-turning vehicle and a preset obstruction area threshold. The relative distance risk item is determined by the relative distance between the straight-going vehicle and the left-turning vehicle and a preset relative distance threshold.

[0011] For example, both the straight-ahead risk score and the left-turn risk score are normalized to a preset scoring range, where the minimum value of the scoring range is 0 and the maximum value is 100.

[0012] For example, when the speed of the straight-going vehicle is less than or equal to the speed limit at the intersection, the speed risk item is 0; when the speed of the straight-going vehicle is greater than the speed limit at the intersection, the speed risk item increases exponentially with the increase of the speed.

[0013] For example, the braking distance is calculated from the speed of the straight-going vehicle, and the remaining braking distance is calculated from the intersection size and the travel trajectories of the straight-going vehicle and the left-turning vehicle.

[0014] For example, when the braking distance is less than or equal to the remaining braking distance, the braking distance risk item is 0; when the braking distance is greater than or equal to the sum of the remaining braking distance and the preset safety distance margin, the braking distance risk item is 100; when the braking distance is greater than the remaining braking distance and less than the sum of the remaining braking distance and the safety distance margin, the braking distance risk item increases linearly with the increase of the braking distance.

[0015] For example, when the field of view occlusion area is less than the occlusion area threshold, the field of view occlusion risk item increases linearly with the increase of the field of view occlusion area; when the field of view occlusion area is greater than or equal to the occlusion area threshold, the field of view occlusion risk item is 100.

[0016] For example, when the relative distance is greater than or equal to a first relative distance threshold, the relative distance risk item is 0; when the relative distance is less than or equal to a second relative distance threshold, the relative distance risk item is 100; and when the relative distance is between the first relative distance threshold and the second relative distance threshold, the relative distance risk item decreases linearly as the relative distance increases.

[0017] For example, the speed risk item is assigned a first straight-line weighting coefficient, the braking distance risk item is assigned a second straight-line weighting coefficient, and the sum of the first straight-line weighting coefficient and the second straight-line weighting coefficient is 1; and / or, the visibility obstruction risk item is assigned a first left-turn weighting coefficient, the relative distance risk item is assigned a second left-turn weighting coefficient, and the sum of the first left-turn weighting coefficient and the second left-turn weighting coefficient is 1; and / or, the straight-line risk score is assigned a first comprehensive weighting coefficient, the left-turn risk score is assigned a second comprehensive weighting coefficient, and the sum of the first comprehensive weighting coefficient and the second comprehensive weighting coefficient is 1.

[0018] For example, each weighting coefficient has a reference value, which is pre-calibrated based on traffic accident statistics; and / or, each weighting coefficient has a variable range, which is adaptively adjusted within the variable range according to the driving parameters and / or the operating condition parameters; and / or, the first straight-ahead weighting coefficient is greater than the second straight-ahead weighting coefficient; and / or, the first left-turn weighting coefficient is greater than the second left-turn weighting coefficient; and / or, the first overall weighting coefficient is less than or equal to the second overall weighting coefficient.

[0019] For example, in step S4, multiple risk levels are set, and a corresponding scoring range is assigned to each risk level. The highest risk level among the risk levels corresponding to the straight-ahead risk score, the left-turn risk score, and the comprehensive risk score is determined as the comprehensive risk level.

[0020] For example, when the overall risk level is low, a safe passage prompt is sent to the vehicles going straight and the vehicles turning left; and / or, when the overall risk level is medium, a risk warning and braking guidance information is sent to at least the vehicle with the higher risk level among the vehicles going straight and the vehicles turning left; and / or, when the overall risk level is high, a forced braking command is sent to at least the vehicle with the higher risk level among the vehicles going straight and the vehicles turning left.

[0021] According to a second aspect of the present invention, a computer program product is provided, comprising a computer program, characterized in that, when the computer program is executed by one or more processors, the processors are capable of executing the risk avoidance method according to the present invention.

[0022] According to a third aspect of the present invention, a hazard avoidance system for vehicles at intersections is provided, characterized in that the hazard avoidance system is configured to implement the hazard avoidance method according to the present invention using a computer program product according to the present invention, wherein the hazard avoidance system comprises at least: - An acquisition module, configured to acquire driving parameters of straight-going vehicles in the first lane and left-turning vehicles in the second lane opposite to the first lane at the intersection, as well as the operating parameters of the intersection. - A calculation module configured to calculate a straight-ahead risk score, a left-turn risk score, and a comprehensive risk score based on the driving parameters and the operating condition parameters, and to determine the comprehensive risk level; and - A risk mitigation execution module, configured to take appropriate risk mitigation measures based on the comprehensive risk level. The avoidance system is at least partially located on the vehicle and / or at least partially integrated into the roadside unit at the intersection and / or at least partially located in the remote service unit. Attached Figure Description

[0023] The invention will now be described in more detail with reference to the accompanying drawings, which will provide a better understanding of its principles, features, and advantages. The drawings include: Figure 1 A schematic flowchart of a method for avoiding hazards at an intersection for vehicles according to an exemplary embodiment of the present invention is shown; Figure 2 A schematic diagram of a vehicle at an intersection according to an exemplary embodiment of the present invention is shown; Figure 3 A schematic connection block diagram of a vehicle avoidance system at an intersection according to an exemplary embodiment of the present invention is shown. Detailed Implementation

[0024] To make the technical problems to be solved, the technical solutions, and the beneficial technical effects of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and several exemplary embodiments. It should be understood that the specific embodiments described herein are for illustrative purposes only and are not intended to limit the scope of protection of this invention.

[0025] This specification provides the operational steps for the methods described in the embodiments or flowcharts, but based on conventional or non-inventive labor, more or fewer operational steps may be included. The order of steps listed in the embodiments is merely one possible execution order among many and does not represent the only possible execution order.

[0026] Figure 1 A schematic flowchart of a method for avoiding hazards for a vehicle at intersection I according to an exemplary embodiment of the present invention is shown. Figure 2 A schematic diagram of a vehicle at intersection I according to an exemplary embodiment of the present invention is shown.

[0027] like Figure 1 and Figure 2 As shown, the risk avoidance method according to the present invention includes at least the following steps: S1: At intersection I, such as a T-junction or a crossroads, identify at least one straight-going vehicle V1 in the first lane L1 and at least one left-turning vehicle V2 in the second lane L2 opposite to the first lane L1. S2: Obtain the driving parameters of straight-going vehicle V1 and left-turning vehicle V2, as well as the operating parameters of intersection I; S3: Based on the obtained driving parameters and operating parameters, the straight-going risk score of straight-going vehicle V1 is calculated by the straight-going risk scoring model, the left-turn risk score of left-turning vehicle V2 is calculated by the left-turn risk scoring model, and the comprehensive risk score for straight-going vehicle V1 and left-turning vehicle V2 is calculated from the calculated straight-going risk score and left-turn risk score. S4: The overall risk level is determined by combining the calculated straight-ahead risk score, left-turn risk score, and comprehensive risk score; S5: Take appropriate risk avoidance measures based on the determined comprehensive risk level.

[0028] According to the present invention, when there are two vehicles traveling in opposite directions at intersection I—one going straight (V1) and the other turning left (V2)—the driving parameters of both vehicles and the operating parameters of intersection I are comprehensively considered. A straight-going risk score, a left-turning risk score, and a comprehensive risk score are calculated using a corresponding risk scoring model. The comprehensive risk level is then determined based on these risk scores. This allows for accurate risk assessment and reliably avoids false or missed warnings. For example, if neither vehicle V1 nor vehicle V2 is driving dangerously, the risk level is low if only vehicle V1 or vehicle V2 is considered, but a collision may still occur. The risk avoidance method according to the present invention can perform a comprehensive analysis and provide targeted warnings and avoidance for such collisions.

[0029] For example, the driving parameters for the straight-going vehicle V1 and the left-turning vehicle V2 are selected from the following group: driving speed, driving trajectory, relative distance to other vehicles, steering angle, field of vision obstruction area, brake pedal opening, and vehicle type. Of course, other parameters that those skilled in the art may also be considered. Here, the driving parameters can be detected by the vehicle's own onboard sensors and transmitted to the roadside unit or cloud platform via a V2X communication link. However, it is also possible that the driving parameters are directly detected by roadside millimeter-wave radar or roadside video detectors deployed at intersection I.

[0030] For example, the operating parameters of intersection I are selected from the following group: intersection speed limit, intersection size, lane width, traffic light status, and intersection slope. Of course, other parameters that those skilled in the art may also be considered meaningful. Here, the operating parameters of intersection I may be pre-stored, for example, in the roadside unit or electronic navigation map at intersection I.

[0031] For example, the straight-ahead risk scoring model includes a speed risk term and a braking distance risk term, as shown in equation (I). (I) in, It is a straight-line risk score. It is a speed-related risk factor. It is a braking distance risk item. It is assigned to the speed risk item The first row weighting coefficient, It is assigned to the braking distance risk item. The second row weighting coefficient, where the first row weighting coefficient Second straight weighting coefficient The sum is 1.

[0032] For example, the speed risk term is determined by the speed of the straight-going vehicle V1 and the intersection speed limit at intersection I. Specifically, when the speed of the straight-going vehicle V1 is less than or equal to the intersection speed limit, the speed risk term... The value is 0 when the speed of the straight-going vehicle V1 exceeds the speed limit at the intersection. As the driving speed increases, the increase is exponential, as shown in equation (II). (II) in, It is the speed of vehicle V1 traveling in a straight line. It's a speed limit at the intersection. This is the speed risk constant. In this case, when the driving speed... Exceeding the speed limit at the intersection At that time, with the driving speed The increase in speed risk item As speed increases, the risk increase caused by the speed increase gradually decreases, indicating that the impact of speeding on the accident rate has a diminishing marginal effect, and it has a greater impact on the severity of the collision.

[0033] In particular, = In this situation, when the speeding exceeds 100%, that is... Speed ​​risk item The value is 50. This parameter setting aligns with actual accident patterns. Of course, other parameter settings that those skilled in the art deem meaningful can also be considered.

[0034] For example, braking distance risk item The braking distance is determined by the braking distance of the straight-going vehicle V1 and the remaining braking distance. Specifically, when the braking distance is less than or equal to the remaining braking distance, the braking distance risk item... The value is 0. When the braking distance is greater than or equal to the sum of the remaining braking distance and the preset safety distance margin, the braking distance risk item is zero. The value is 100. When the braking distance is greater than the remaining braking distance but less than the sum of the remaining braking distance and the safety distance margin, the braking distance risk item is... The braking distance increases linearly with increasing distance, as shown in equation (III). (III) in, It is the braking distance of the straight-going vehicle V1. It is the remaining braking distance of the straight-going vehicle V1, and This is the safety distance margin, which is determined by empirical data and / or statistical data, for example, 15m.

[0035] Here, the braking distance of the straight-going vehicle V1. It can be based on the speed of the straight-going vehicle V1. The calculation is shown in equation (IV). in, It is the driver's reaction time. It is the road surface adhesion coefficient. It is the acceleration due to gravity. From this, the braking distance can be determined. It consists of two parts: reaction distance and actual braking distance, and is related to driving speed. Strongly correlated. Additionally, braking distance... It can also adaptively adjust according to the brake pedal opening.

[0036] Here, driver reaction time This includes the time it takes for the driver to perceive a hazard, make a decision, and take action, and can be determined by experimental and / or statistical data, for example, 1.2 seconds. In particular, driver reaction time... It can adaptively adjust within a preset range based on the driver's condition and historical driving behavior. This is easily understood as the driver's reaction time decreasing when the driver is fatigued. It increases accordingly.

[0037] In addition, the road surface adhesion coefficient The road surface adhesion coefficient depends on the road surface condition and weather conditions. Specifically, when the road surface is dry asphalt, the road surface adhesion coefficient... For example, the coefficient of friction is 0.75, while in rainy, snowy, or icy weather, the coefficient of friction is... The corresponding decrease.

[0038] For example, such as Figure 2 As shown, the remaining braking distance The remaining braking distance is calculated by combining the intersection dimensions and the trajectories of the straight-going vehicle V1 and the left-turning vehicle V2. Based on their trajectories, a possible point of overlap is determined, and the distance of the current position of the straight-going vehicle V1 relative to this point of overlap is determined as the remaining braking distance. .

[0039] For example, the first row weighting coefficient Second straight weighting coefficient Each has a reference value, which can be pre-calibrated based on traffic accident statistics. Due to the speed of the straight-going vehicle V1... For straight-line risk scoring It has a significant impact, making the first row weighting coefficient... Greater than the second vertical weighting coefficient For example, the weighting coefficient of the first row. The baseline value is 0.6, while the second straight-line weighting coefficient... The baseline value is 0.4.

[0040] In particular, the first row weighting coefficient Second straight weighting coefficient Each factor has a variable range, allowing the weighting coefficients to adaptively adjust within their respective ranges based on driving parameters and / or operating condition parameters, thus calculating a straight-ahead risk score that more realistically reflects actual conditions. For example, the first row weighting coefficient The value ranges from 0.55 to 0.65, while the second vertical weighting coefficient... The value is taken in the range of 0.35 to 0.45, where the speed of the straight-going vehicle V1 is... The larger the value, the higher the weighting coefficient of the first row. The larger it is, the greater the weighting coefficient of the second straight line. The smaller the corresponding value, the better. It's easy to understand that the weighting coefficients in the first row... Second straight weighting coefficient The sum always remains 1.

[0041] For example, the left-turn risk scoring model includes a field-of-view occlusion risk term and a relative distance risk term, as shown in equation (V). (V) in, It is a left-turn risk score. It is a risk item related to obstructed vision. It is a relative distance risk item. It is assigned to the risk item of field of vision obstruction. The first left-turn weighting coefficient, It is assigned to the relative distance risk item The second left-turn weighting coefficient, where the first left-turn weighting coefficient... Second left turn weighting coefficient The sum is 1.

[0042] For example, the risk of field of vision obstruction The obstruction area of ​​the field of vision of the left-turning vehicle V2 is determined by the preset obstruction area threshold. Specifically, when the obstruction area is less than the obstruction area threshold, the field of vision obstruction risk item... The visual field occlusion risk increases linearly with the increase of the occlusion area. When the occlusion area is greater than or equal to the occlusion area threshold, the visual field occlusion risk item increases. The value is 100, as shown in equation (VI). (VI) in, It refers to the area of ​​the field of vision obstructed by the left-turning vehicle V2, and This is the shading area threshold, which is determined by empirical data and / or statistical data, for example, 20m. 2 It's easy to understand that the field of vision obstructed by a left-turning vehicle (V2) is... The larger the value, the smaller the driver's field of vision for oncoming vehicles when turning left (V2), the lower the probability of taking timely evasive action, and the greater the risk of collision.

[0043] Here, as Figure 2 As shown, the field of vision obstruction area of ​​left-turning vehicle V2 For example, it can be estimated based on semantic segmentation algorithms using the turning angle of the left-turning vehicle V2 combined with real-time images of the front view captured by the onboard camera. However, the area of ​​the left-turning vehicle V2's field of vision obstruction can also be considered. The global vehicle positions were estimated from the roadside units located at intersection I. Additionally, the field of vision obstruction area of ​​left-turning vehicle V2 was also considered. It also depends on the vehicle type of the vehicle and surrounding vehicles. For example, when there is a large vehicle between a straight-going vehicle V1 and a left-turning vehicle V2, the obstruction area of ​​the left-turning vehicle V2's field of vision... The corresponding increase.

[0044] For example, relative distance risk item The relative distance is determined by the relative distance between the straight-going vehicle V1 and the left-turning vehicle V2 and a preset relative distance threshold. Specifically, when the relative distance is greater than or equal to the first relative distance threshold, the relative distance risk item... The relative distance risk item is 0 when the relative distance is less than or equal to the second relative distance threshold. The relative distance risk item is set to 100 when the relative distance is between the first and second relative distance thresholds. As the relative distance increases, it decreases linearly, as shown in equation (VII). (VII) Where R is the relative distance between the straight-going vehicle V1 and the left-turning vehicle V2. It is the first relative distance threshold. This is the second relative distance threshold. Here, the first relative distance threshold... Second relative distance threshold Determined by empirical data and / or statistical data, wherein the first relative distance threshold The baseline value is, for example, 25m, while the second relative distance threshold... The baseline value is, for example, 5m. Furthermore, the first relative distance threshold... Second relative distance threshold It can also adaptively adjust within a preset range according to actual working conditions. For example, at night or in rainy or snowy weather, both relative distance thresholds are reduced accordingly.

[0045] For example, the first left-turn weighting coefficient Second left turn weighting coefficient Each has a baseline value, which can be pre-calibrated based on traffic accident statistics. Since obstruction of the view by left-turning vehicle V2 is a major contributing factor to left-turn accidents, the first left-turn weighting coefficient... Greater than the second left turn weighting coefficient For example, the first left-turn weighting coefficient. The baseline value is 0.7, while the second left-turn weighting coefficient... The baseline value is 0.3.

[0046] In particular, the first left-turn weighting coefficient Second left turn weighting coefficient Each factor has a variable range, allowing the weighting coefficients to adaptively adjust within their respective ranges based on driving and / or operating condition parameters, thus calculating a more realistic left-turn risk score. For example, the first left-turn weighting coefficient. The value ranges from 0.65 to 0.75, while the second left-turn weighting coefficient... It takes values ​​between 0.25 and 0.35. It's easy to understand that the first left-turn weighting coefficient... Second left turn weighting coefficient The sum always remains 1.

[0047] For example, straight-line risk scoring and left turn risk score All scores are normalized to a preset scoring interval, where the minimum value is 0 and the maximum value is 100. This results in a straight-ahead risk score. and left turn risk score They have the same dimensions and can be directly weighted and summed.

[0048] For example, the comprehensive risk score S is composed of the straight-line risk score. and left turn risk score The weighted summation yields the result shown in formula (VIII). (VIII) Among them, straight-line risk score The first comprehensive weighting coefficient is assigned. Left turn risk score The allocation has a second comprehensive weighting coefficient. First comprehensive weighting coefficient Second comprehensive weighting coefficient The sum is 1.

[0049] For example, the first comprehensive weighting coefficient Second comprehensive weighting coefficient Each has a baseline value, which can be pre-calibrated based on traffic accident statistics. Analysis of traffic accident statistics shows that the number of accidents in which left-turning vehicles bear primary responsibility is significantly greater than the number of accidents in which straight-going vehicles bear primary responsibility, thus affecting the first comprehensive weighting coefficient. Less than or equal to the second comprehensive weighting coefficient For example, the first comprehensive weighting coefficient. The baseline value is 0.4, while the second comprehensive weighting coefficient... The baseline value is 0.6.

[0050] In particular, the first comprehensive weighting coefficient Second comprehensive weighting coefficient Each weighting coefficient has a variable range, allowing it to adaptively adjust based on driving and / or operating condition parameters within its respective range to more realistically calculate the overall risk score S. For example, the first overall weighting coefficient... The value ranges from 0.3 to 0.5, while the second comprehensive weighting coefficient... It takes values ​​between 0.5 and 0.7. It's easy to understand that the first comprehensive weighting coefficient... Second comprehensive weighting coefficient The sum always remains 1.

[0051] For example, in step S4, multiple risk levels are set, and a corresponding scoring range is assigned to each risk level, wherein the straight-line risk score is... Left turn risk score The comprehensive risk score S corresponds to a specific risk level, and the highest of these three risk levels is determined as the comprehensive risk level. For example, there are low-risk, medium-risk, and high-risk levels. The low-risk level corresponds to a score range of 0 to 30, the medium-risk level to a score range of 30 to 60, and the high-risk level to a score range of 60 to 100. If the direct risk score... Corresponding to the medium risk level, the risk score for turning left is... Corresponding to the medium risk level, the comprehensive risk score S corresponds to the high risk level, so the comprehensive risk level is determined to be high risk. Of course, other risk level settings and their corresponding scoring ranges that are considered meaningful by those skilled in the art can also be considered.

[0052] For example, in step S5, when the overall risk level is low, a safe passage prompt is sent to the straight-going vehicle V1 and the left-turning vehicle V2, so that both the straight-going vehicle V1 and the left-turning vehicle V2 can maintain their current driving strategy and continue to pass through intersection I.

[0053] For example, in step S5, when the overall risk level is medium risk, a risk warning and braking guidance message is sent to at least the vehicle with the higher risk level (i.e., the risk-dominant vehicle) between the straight-going vehicle V1 and the left-turning vehicle V2 to remind its driver of the potential collision risk and to apply the brakes. The risk warning and braking guidance message is presented to the driver of the relevant vehicle, for example, in voice or text form. Alternatively, risk warnings and braking guidance messages may be sent to both the straight-going vehicle V1 and the left-turning vehicle V2 simultaneously to minimize the risk of a collision.

[0054] For example, in step S5, when the overall risk level is high, a forced braking command is sent to at least the vehicle with the higher risk level, i.e., the risk-dominant vehicle, between the straight-going vehicle V1 and the left-turning vehicle V2, to actively intervene in vehicle control, thereby avoiding a collision due to the driver's untimely reaction and improving driving safety. Alternatively, forced braking commands may be sent to both the straight-going vehicle V1 and the left-turning vehicle V2 simultaneously to minimize the possibility of a collision.

[0055] According to the present invention, a computer program product is provided, comprising a computer program that, when executed by one or more processors, enables the processors to perform the risk avoidance method according to the present invention.

[0056] Figure 3 A schematic connection block diagram of a vehicle avoidance system 100 at intersection I according to an exemplary embodiment of the present invention is shown. Here, the avoidance system 100 is capable of implementing the avoidance method according to the present invention using a computer program product according to the present invention.

[0057] like Figure 3 As shown, the avoidance system 100 includes an acquisition module 10, which is configured to acquire the driving parameters of a straight-going vehicle V1 in the first lane L1 and a left-turning vehicle V2 in the second lane L2 opposite to the first lane L1 at intersection I, as well as the operating parameters of intersection I.

[0058] like Figure 3As shown, the risk avoidance system 100 includes a calculation module 20. The calculation module is configured to calculate a straight-going risk score, a left-turn risk score, and a comprehensive risk score based on the driving parameters and operating parameters obtained by the acquisition module 10, and to determine the comprehensive risk level by using these risk scores. The calculation module 20 stores a straight-going risk score model for calculating the straight-going risk score and a left-turn risk score model for calculating the left-turn risk score.

[0059] like Figure 3 As shown, the risk avoidance system 100 includes a risk avoidance execution module 30, which is configured to take corresponding risk avoidance measures according to the determined comprehensive risk level. The risk avoidance measures may include sending risk warnings and braking guidance information or sending forced braking commands.

[0060] Here, the hazard avoidance system 100 can be at least partially deployed on the vehicle and / or at least partially integrated into the roadside unit at intersection I and / or at least partially deployed in the remote service unit. For example, the acquisition module 10 can be deployed on the vehicle to directly detect the vehicle's driving parameters and acquire the driving parameters of other vehicles and the operating parameters of intersection I via wireless communication. Alternatively, the acquisition module 10 may be integrated into the roadside unit at intersection I to directly detect the driving parameters of each vehicle or acquire the driving parameters of each vehicle via wireless communication, while the operating parameters of intersection I can be pre-stored in the roadside unit. Furthermore, the computing module 20 and / or the hazard avoidance execution module 30 can be integrated into the roadside unit or deployed in the remote service unit, such as a cloud platform, as needed. This can improve the deployment flexibility and compatibility of the hazard avoidance system 100, thereby adapting to different architectural designs.

[0061] The foregoing description of the embodiments is limited to the framework of the examples given. Of course, the various features of the embodiments can be freely combined with each other without departing from the framework of the invention, as long as it is technically meaningful.

[0062] Other advantages and alternative embodiments of the present invention will be apparent to those skilled in the art. Therefore, the present invention is not, in its broader sense, limited to the specific details, representative structures, and exemplary embodiments shown and described. Rather, those skilled in the art can make various modifications and substitutions without departing from the basic spirit and scope of the invention.

Claims

1. A method for avoiding hazards for vehicles at an intersection (I), characterized in that, The risk avoidance method includes at least the following steps: S1: At the intersection (I), identify at least one straight-going vehicle (V1) in the first lane (L1) and at least one left-turning vehicle (V2) in the second lane (L2) opposite to the first lane (L1). S2: Obtain the driving parameters of the straight-going vehicle (V1) and the left-turning vehicle (V2) and the operating parameters of the intersection (I) respectively; S3: Based on the driving parameters and the operating parameters, a straight-ahead risk score is calculated using the straight-ahead risk scoring model, a left-turn risk score is calculated using the left-turn risk scoring model, and a comprehensive risk score is calculated using the straight-ahead risk score and the left-turn risk score. S4: Determine the overall risk level based on the straight-ahead risk score, the left-turn risk score, and the comprehensive risk score; S5: Take appropriate risk avoidance measures based on the comprehensive risk level.

2. The risk avoidance method according to claim 1, characterized in that, The driving parameters are selected from the following group: driving speed, driving trajectory, relative distance to other vehicles, steering angle, field of vision obstruction area (A), brake pedal opening, vehicle type; and / or The operating parameters are selected from the following group: intersection speed limit, intersection size, lane width, traffic light status, and intersection slope.

3. The risk avoidance method according to claim 2, characterized in that, The straight-ahead risk scoring model includes a speed risk item and a braking distance risk item. The speed risk item is determined by the speed of the straight-ahead vehicle (V1) and the speed limit at the intersection, while the braking distance risk item is determined by the braking distance of the straight-ahead vehicle (V1) and the remaining braking distance (L); and / or The left-turn risk scoring model includes a view obstruction risk item and a relative distance risk item. The view obstruction risk item is determined by the view obstruction area (A) of the left-turning vehicle (V2) and a preset obstruction area threshold. The relative distance risk item is determined by the relative distance between the straight-going vehicle (V1) and the left-turning vehicle (V2) and a preset relative distance threshold; and / or Both the straight-ahead risk score and the left-turn risk score are normalized to a preset scoring range, where the minimum value of the scoring range is 0 and the maximum value is 100.

4. The risk avoidance method according to claim 3, characterized in that, When the speed of the straight-going vehicle (V1) is less than or equal to the intersection speed limit, the speed risk term is 0; when the speed of the straight-going vehicle (V1) is greater than the intersection speed limit, the speed risk term increases exponentially with the increase of the speed; and / or The braking distance is calculated from the speed of the straight-going vehicle (V1), and the remaining braking distance (L) is calculated from the intersection dimensions and the trajectories of the straight-going vehicle (V1) and the left-turning vehicle (V2); and / or When the braking distance is less than or equal to the remaining braking distance (L), the braking distance risk item is 0; when the braking distance is greater than or equal to the sum of the remaining braking distance (L) and the preset safety distance margin, the braking distance risk item is 100; when the braking distance is greater than the remaining braking distance (L) and less than the sum of the remaining braking distance (L) and the safety distance margin, the braking distance risk item increases linearly with the increase of the braking distance; and / or When the field of view occlusion area (A) is less than the occlusion area threshold, the field of view occlusion risk item increases linearly with the increase of the field of view occlusion area (A); when the field of view occlusion area (A) is greater than or equal to the occlusion area threshold, the field of view occlusion risk item is 100; and / or When the relative distance is greater than or equal to the first relative distance threshold, the relative distance risk item is 0; when the relative distance is less than or equal to the second relative distance threshold, the relative distance risk item is 100; when the relative distance is between the first relative distance threshold and the second relative distance threshold, the relative distance risk item decreases linearly as the relative distance increases.

5. The risk avoidance method according to claim 3, characterized in that, The speed risk term is assigned a first straight-line weighting coefficient, the braking distance risk term is assigned a second straight-line weighting coefficient, and the sum of the first straight-line weighting coefficient and the second straight-line weighting coefficient is 1; and / or The field-of-view obstruction risk item is assigned a first left-turn weighting coefficient, and the relative distance risk item is assigned a second left-turn weighting coefficient. The sum of the first left-turn weighting coefficient and the second left-turn weighting coefficient is 1; and / or The straight-ahead risk score is assigned a first comprehensive weighting coefficient, and the left-turn risk score is assigned a second comprehensive weighting coefficient. The sum of the first comprehensive weighting coefficient and the second comprehensive weighting coefficient is 1.

6. The risk avoidance method according to claim 5, characterized in that, Each weighting coefficient has a reference value, which is pre-calibrated based on traffic accident statistics; and / or Each weighting coefficient has a variable range, and the weighting coefficient is adaptively adjusted within the variable range according to the driving parameters and / or the operating condition parameters; and / or The first straight-line weighting coefficient is greater than the second straight-line weighting coefficient; and / or The first left-turn weighting coefficient is greater than the second left-turn weighting coefficient; and / or The first comprehensive weighting coefficient is less than or equal to the second comprehensive weighting coefficient.

7. The risk avoidance method according to any one of the preceding claims, characterized in that, In step S4, multiple risk levels are set, and a corresponding scoring range is assigned to each risk level. The highest risk level among the risk levels corresponding to the straight-ahead risk score, the left-turn risk score, and the comprehensive risk score is determined as the comprehensive risk level.

8. The risk avoidance method according to claim 7, characterized in that, When the overall risk level is low, a safe passage prompt is sent to the straight-going vehicle (V1) and the left-turning vehicle (V2); and / or When the overall risk level is medium risk, at least the vehicle with the higher risk level among the straight-going vehicles (V1) and the left-turning vehicles (V2) shall be sent risk warnings and braking guidance information. and / or When the overall risk level is high, a forced braking command shall be sent to at least the vehicle with the higher risk level among the straight-going vehicles (V1) and the left-turning vehicles (V2).

9. A computer program product comprising a computer program, characterized in that, When the computer program is executed by one or more processors, the processors are capable of performing the risk avoidance method according to any one of claims 1 to 8.

10. A hazard avoidance system (100) for vehicles at an intersection (I), characterized in that, The risk avoidance system (100) is configured to implement the risk avoidance method according to any one of claims 1 to 8 using the computer program product according to claim 9, wherein the risk avoidance system (100) comprises at least: - Acquisition module (10), the acquisition module (10) is configured to acquire the driving parameters of a straight-going vehicle (V1) in the first lane (L1) and a left-turning vehicle (V2) in the second lane (L2) opposite to the first lane (L1) at the intersection (I), as well as the operating parameters of the intersection (I); - A calculation module (20), configured to calculate a straight-ahead risk score, a left-turn risk score, and a comprehensive risk score based on the driving parameters and the operating condition parameters, and to determine the comprehensive risk level; and - Risk mitigation execution module (30), which is configured to take corresponding risk mitigation measures according to the comprehensive risk level. The avoidance system (100) is at least partially located on the vehicle and / or at least partially integrated into the roadside unit at the intersection (I) and / or at least partially located in the remote service unit.

Images