Vehicle collision risk prediction method, device, electronic device and storage medium

Abstract

Embodiments of the present invention disclose a method, device, electronic device, and storage medium for predicting vehicle collision risk. The method for predicting vehicle collision risk includes: determining a current collision risk value based on current status information of the vehicle to be predicted and the target obstacle; wherein the obstacle includes at least one of the following: a vehicle, a pedestrian, and a stop line; determining a decision influence coefficient of the target obstacle based on the decision information of the target obstacle; and predicting the final collision risk between the target obstacle and the vehicle to be predicted based on the current collision risk value and the decision influence coefficient. The embodiment of the present invention achieves the effect of predicting the final collision risk between the target obstacle and the vehicle to be predicted based on the current information of the vehicle to be predicted and the target obstacle and the decision information of the target obstacle.

Description

Technical Field

[0001] Embodiments of the present invention relate to the field of intelligent vehicle technology, and in particular to a method, device, electronic device, and storage medium for predicting vehicle collision risk. Background Art

[0002] With the development of smart cars, research on vehicle driving in complex application scenarios is becoming more and more in-depth, and research on vehicle collision risk prediction is becoming more and more in-depth.

[0003] The current vehicle collision risk prediction determines whether a collision will occur based on the relative distance, relative speed, and braking distance between the target obstacle and the predicted vehicle.

[0004] This type of risk prediction can only be performed under simple road conditions and cannot effectively predict the various collision risks faced by the predicted vehicle when driving under complex road conditions in reality. Summary of the Invention

[0005] Embodiments of the present invention provide a method, device, electronic device, and storage medium for predicting vehicle collision risk, which can identify potential risk targets, predict the movement of risk targets in advance, and perform risk assessment based on the decision-making behavior of the target obstacle and the vehicle to be predicted, as well as traffic rules.

[0006] In a first aspect, an embodiment of the present invention provides a method for predicting vehicle collision risk, comprising:

[0007] Determining a current collision risk value based on current state information of the vehicle to be predicted and the target obstacle; wherein the obstacle includes at least one of the following: a vehicle, a pedestrian, and a stop line;

[0008] determining a decision influence coefficient of the target obstacle according to the decision information of the target obstacle;

[0009] The final collision risk between the target obstacle and the vehicle to be predicted is predicted according to the current collision risk value and the decision influence coefficient.

[0010] In a second aspect, an embodiment of the present invention further provides a vehicle collision risk prediction device, comprising:

[0011] A current collision risk value determination module is used to determine a current collision risk value based on current state information of the vehicle to be predicted and the target obstacle; wherein the obstacle includes at least one of the following: a vehicle, a pedestrian, and a stop line;

[0012] A decision influence coefficient determination module, configured to determine the decision influence coefficient of the target obstacle based on the decision information of the target obstacle;

[0013] The final collision risk prediction module is used to predict the final collision risk between the target obstacle and the vehicle to be predicted based on the current collision risk value and the decision influence coefficient.

[0014] In a third aspect, an embodiment of the present invention further provides an electronic device, including:

[0015] one or more processors;

[0016] a storage device for storing one or more programs,

[0017] When the one or more programs are executed by the one or more processors, the one or more processors implement the method for predicting vehicle collision risk as described in any embodiment of the present invention.

[0018] In a fourth aspect, an embodiment of the present invention further provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method for predicting vehicle collision risk as described in any embodiment of the present invention.

[0019] The embodiment of the present invention predicts the final collision risk by determining the current state risk value of the target obstacle and the decision influence coefficient, thereby overcoming the defects of traditional methods that cannot effectively predict the behavior of the target obstacle and cause unnecessary braking or untimely braking. It achieves the effect of predicting the collision risk caused by the behavior of the target obstacle in advance, thereby reducing false braking alarms and improving the accuracy of collision prediction, thereby improving the comfort and safety of passengers. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 is a flow chart of a method for predicting vehicle collision risk in embodiment 1 of the present invention;

[0021] Figure 2 is a schematic diagram of the position relationship between the vehicle to be predicted and the target obstacle in the first embodiment of the present invention;

[0022] Figure 3 1 is a schematic diagram of quantifying collision risk determined according to relative speed in the first embodiment of the present invention;

[0023] Figure 4 is a flow chart of a method for predicting vehicle collision risk when the target obstacle is a vehicle in the second embodiment of the present invention;

[0024] Figure 5 is a schematic diagram of the positional relationship between the target vehicle and the vehicle to be predicted when the target obstacle is the target vehicle in the method for predicting vehicle collision risk in the second embodiment of the present invention;

[0025] Figure 6 is a flow chart of a method for predicting vehicle collision risk when the target obstacle is a pedestrian in Example 3 of the present invention;

[0026] Figure 7 is a schematic diagram of the positional relationship between a target pedestrian and a vehicle to be predicted when the target obstacle is a target pedestrian in the method for predicting vehicle collision risk in the third embodiment of the present invention;

[0027] Figure 8 This is a quantitative schematic diagram of determining an individual collision risk coefficient based on the pedestrian's area and speed in the third embodiment of the present invention;

[0028] Figure 9 This is a flow chart of a method for predicting vehicle collision risk when the target obstacle is a stop line in Example 3 of the present invention;

[0029] Figure 10 Schematic diagram of the relationship between the target stop line and the position of the vehicle to be predicted when the target obstacle is the target stop line in the fourth embodiment of the present invention;

[0030] Figure 11 2 is a schematic structural diagram of a vehicle collision risk prediction device in a fifth embodiment of the present invention;

[0031] Figure 12 It is a structural diagram of an electronic device in Embodiment 6 of the present invention. DETAILED DESCRIPTION

[0032] The present invention will be further described in detail below with reference to the accompanying drawings and examples. It will be understood that the specific embodiments described herein are intended only to illustrate the present invention and are not intended to limit the present invention. It should also be noted that, for ease of description, the accompanying drawings only illustrate portions relevant to the present invention, not all structures.

[0033] Example 1

[0034] Figure 1 This is a flow chart of the method for predicting vehicle collision risk in the first embodiment of the present invention. This embodiment is applicable to the case of predicting vehicle collision risk in complex scenarios. The method can be executed by a vehicle collision risk prediction device, which can be implemented in software or hardware and can be configured in an electronic device, such as a computer or other device with communication and computing capabilities. Figure 1 As shown, the method specifically includes:

[0035] S101. Determine a current collision risk value based on current state information of a vehicle to be predicted and a target obstacle; wherein the obstacle includes at least one of the following: a vehicle, a pedestrian, and a stop line.

[0036] Among them, the vehicle to be predicted is the vehicle to be protected by the vehicle collision risk prediction method; the target obstacle can be an object that may collide with the vehicle to be predicted or restrict the behavior of the vehicle to be predicted; the current collision risk value is used to characterize the possibility of collision between the vehicle to be predicted and the target obstacle.

[0037] Specifically, the current status information can be identified and obtained through identification devices such as cameras and sensing devices such as distance sensors.

[0038] Based on the above technical solution, optionally, the current state information includes at least the relative distance, relative speed and relative acceleration between the vehicle to be predicted and the target obstacle; accordingly, the current collision risk value is determined based on the current state information between the vehicle to be predicted and the target obstacle, including: determining the collision time based on the relative distance, relative speed and relative acceleration between the vehicle to be predicted and the target obstacle; and determining the current collision risk value based on the collision time.

[0039] Among them, the relative speed can be obtained according to the following formula: r =v 目标障碍物 -v 待预测车辆 ;v 待预测车辆 is the speed of the vehicle to be predicted, v 目标障碍物 is the speed of the target obstacle, v r is the relative speed between the vehicle to be predicted and the target obstacle;

[0040] The relative acceleration a can be obtained according to the following formula:

[0041]

[0042] The formula for collision time TTC is: Among them, TTC is the collision time, L r is the relative distance between the vehicle to be predicted and the target obstacle; the relative acceleration can be used to calculate the collision time more comprehensively.

[0043] The beneficial effects of such a setting are as follows: based on the relative distance, relative speed and relative acceleration, the time required for a collision to occur based on the current state of the predicted vehicle and the target obstacle can be accurately judged, and the current collision risk value is determined based on the collision time to improve the accuracy of the risk value determination. For example, a mapping relationship between the collision time and the current collision risk value is suggested in advance, and the current collision risk value is determined based on the mapping relationship, where the mapping relationship can be determined based on the actual collision determination accuracy, which is not limited here.

[0044] Specifically, taking a vehicle as an example of a target obstacle, pedestrians and stop lines can be determined by reference, and will not be described in detail here. Figure 2This is a schematic diagram showing the positional relationship between the vehicle to be predicted and the target obstacle. The vehicle to be predicted is vehicle 100, and the target obstacle is vehicle 300. Vehicle 300 is directly in front of vehicle 100 in its lane. Vehicle 100 performs risk assessment by measuring the relative speed and relative acceleration between vehicle 100 and vehicle 300. To simplify the representation, this example sets the relative acceleration to 0. A collision risk quantification diagram is pre-established based on the relative distance and relative speed, as shown in the following example: Figure 3 The figure shows a schematic diagram of collision risk quantification based on relative speed, where the horizontal axis represents the relative distance between the vehicle to be predicted and the target obstacle, and the vertical axis represents the relative speed between the vehicle to be predicted and the target obstacle. 本车，目标 The following methods can be used for quantification:

[0045] 1)v r =v 目标障碍物 -v 待预测车辆 ; According to the calculation formula of TTC The collision time under the current state can be calculated;

[0046] 2) Pre-set collision time safety value TTC safe , used to represent the safe collision time without collision risk, which can be 4 seconds for example (this value can be adjusted according to the actual collision accuracy and is not limited here);

[0047] 3)TTC is less than TTC safe , then the current collision risk value is considered to be the first risk value, and TTC is greater than 2*TTC safe , then the current collision risk value can be considered to be the second risk value, and TTC is greater than TTC safe and less than 2*TTC safe When , the value can be taken linearly, where the first risk value is greater than the second risk value. For example, the first risk value is 1 and the second risk value is 0, then the expression for quantifying the current collision risk value according to the collision time is as follows:

[0048]

[0049] Among them, when the target is a vehicle, R is the vehicle, and vehicle is the current collision risk value between the vehicle to be predicted and the target obstacle.

[0050] Risk assessment can also be performed using the braking distance of the vehicle at its maximum braking capacity, the deceleration of the preceding vehicle, and the current relative speed and relative distance.

[0051] S102: Determine a decision influence coefficient of the target obstacle based on the decision information of the target obstacle.

[0052] Among them, the decision information can be behavioral information that may affect the collision risk exhibited by the target obstacle, such as the braking behavior information or steering behavior information of the target vehicle, the walking behavior information of the target pedestrian, and traffic lights; the decision information can be obtained through information acquisition devices such as cameras and radars installed on the vehicle to be predicted.

[0053] Because the behavior of a target obstacle may affect the collision risk between the vehicles being predicted, the impact coefficient of the target obstacle's decision on the collision risk needs to be determined based on this decision information. For example, braking or changing lanes into the vehicle ahead can increase the collision risk with the vehicle being predicted, while pedestrians may stop walking at any time, which can also reduce the collision risk with the vehicle being predicted.

[0054] Specifically, the decision influence coefficient can be determined by a mapping relationship between preset decision information and the decision influence coefficient.

[0055] S103 : Predicting the final collision risk between the target obstacle and the vehicle to be predicted based on the current collision risk value and the decision influence coefficient.

[0056] The final collision risk is calculated according to the following formula:

[0057] R 系统，目标障碍物 =R 本车，目标障碍物 *R 目标障碍物

[0058] Among them, R 系统，目标障碍物 is the final collision risk, R 目标障碍物 is the decision-making influence coefficient.

[0059] The technical solution of the embodiment of the present application solves the problem that the existing technology cannot identify the different risks brought about by the decision-making of potential risk targets by identifying the impact of the decision-making behavior of potential collision risk target obstacles on the collision risk in advance, and achieves the effect of early warning of potential risk targets and correcting the collision risk according to the target obstacle decision information.

[0060] Example 2

[0061] Figure 4 This is a flow chart of a method for predicting vehicle collision risk when the target obstacle is a vehicle in Example 2 of the present invention. In this embodiment, based on the above embodiment, optionally, the target obstacle is a target vehicle, and the decision information includes taillight information; wherein the taillight information includes at least one of the following: brake light information and turn signal information; and accordingly, determining the decision influence coefficient of the target obstacle based on the decision information of the target obstacle includes: obtaining lane information of the target vehicle; and determining the decision influence coefficient of the target vehicle based on the taillight information and the lane information.

[0062] like Figure 4 As shown, the method includes the following steps:

[0063] S301. Determine a current collision risk value based on current state information of the vehicle to be predicted and the target vehicle.

[0064] Among them, the target vehicle can be a vehicle in the lane where the predicted vehicle is located or in an adjacent lane; when a vehicle in the lane of the predicted vehicle decelerates or a vehicle in an adjacent lane changes lanes to this lane, there may be a collision risk for the normal driving of the predicted vehicle. Therefore, vehicles in the same lane or adjacent lane are regarded as target obstacles.

[0065] S302. Acquire lane information of the target vehicle, and determine a decision influence coefficient of the target vehicle based on the taillight information and the lane information; wherein the taillight information includes at least one of the following: brake light information and turn signal information.

[0066] During the period between detecting the target vehicle's taillights and detecting its movements, the target vehicle's behavior can be predicted based on the taillight information. Changes in the target vehicle's behavior can lead to changes in the collision risk with the vehicle being predicted. Therefore, a decision-making influence coefficient is determined based on the target vehicle's behavior as represented by its taillights, resulting in a more accurate final collision risk.

[0067] Based on the above embodiment, the lane information includes the same lane and adjacent lanes as the vehicle to be predicted; accordingly, the decision influence coefficient of the target vehicle is determined based on the taillight information and the lane information, including: if the lane information is in the same lane as the vehicle to be predicted, the decision influence coefficient of the target vehicle is determined based on the brake light information and the deceleration state of the target vehicle; if the lane information is in an adjacent lane as the vehicle to be predicted, the decision influence coefficient of the target vehicle is determined based on the turn signal information and the motion trajectory of the target vehicle.

[0068] Figure 5 A schematic diagram illustrating the positional relationship between the target vehicle and the vehicle to be predicted, when the target obstacle is a target vehicle in the vehicle collision risk prediction method provided in Example 2 of the present invention. Target vehicle 300 is directly in front of the vehicle to be predicted 100 in the lane, and target vehicle 200 is in the adjacent lane.

[0069] Specifically, if target vehicle 300, which is in the same lane as vehicle 100 to be predicted, does not brake to slow down, and thus has no impact on the entire current collision risk system, the decision influence coefficient is the first decision influence coefficient. When target vehicle 300, which is potentially in the same lane, begins braking, i.e., the brake lights illuminate, the current collision risk value determined at this time does not change due to delays in the braking system of target vehicle 300 and the measurement system of vehicle 100 to be predicted. However, the collision risk has increased, and the decision influence coefficient is determined to be the second decision influence coefficient. If target vehicle 300 brakes and vehicle 100 to be predicted has detected its deceleration, the collision risk can be determined based on the current deceleration state of target vehicle 300. The decision influence coefficient determined at this time is the third decision influence coefficient. The first decision influence coefficient is smaller than the second decision influence coefficient, and the third decision influence coefficient is smaller than the second decision influence coefficient.

[0070] Exemplarily, the first decision influence coefficient is 1, the second decision influence coefficient is 1.2, and the third decision influence coefficient is 1;

[0071] The decision influence coefficient R of the target vehicle 300 in the same lane as the vehicle 100 to be predicted is determined according to the following formula: 车 :

[0072]

[0073] Specifically, if target vehicle 200 is in an adjacent lane to vehicle 100 and has no turning requirements (i.e., its turn signal is off), the collision risk between target vehicle 200 and vehicle 100 will not change. The fourth decision influence coefficient is determined to be used at this time. If target vehicle 200 has its turn signal on in the adjacent lane but has not yet cut in laterally, it must move laterally to complete the cut-in before a collision risk arises. During this lateral movement time, the collision risk is lower than that of vehicles already in the lane of vehicle 100. The fifth decision influence coefficient is used. If target vehicle 200 changes lanes, the sixth decision influence coefficient is used for target vehicle 200 in the adjacent lane to vehicle 100. The fourth decision influence coefficient is smaller than the fifth decision influence coefficient, and the fifth decision influence coefficient is smaller than the sixth decision influence coefficient.

[0074] Exemplarily, the fourth decision influence coefficient is 0, the fifth decision influence coefficient is 0.9, and the sixth decision influence coefficient is 1;

[0075] The decision influence coefficient R of the target vehicle 200 in the adjacent lane to the vehicle to be predicted 100 is determined according to the following formula: 车 :

[0076]

[0077] S303: Predicting the final collision risk between the target obstacle and the vehicle to be predicted based on the current collision risk value and the decision influence coefficient.

[0078] The final collision risk of the predicted vehicle relative to the target vehicle can be calculated according to the following formula:

[0079] R 系统，目标车辆 =R 待预测车辆，目标车辆 *R 车

[0080] Among them, R 系统，目标车辆 is the final collision risk between the vehicle to be predicted and the target vehicle, R 待预测车辆，目标车辆 is the current collision risk value between the vehicle to be predicted and the target vehicle, R 车 is the decision influence coefficient of the target vehicle.

[0081] Specifically, the final collision risk of the predicted vehicle 100 relative to the target vehicle 300 can be calculated according to the following formula:

[0082] R 系统，目标车辆三 =R 待预测车辆，目标车辆三 *R 车

[0083] Among them, R 系统，目标车辆三 is the final collision risk between the vehicle 100 to be predicted and the target vehicle 300, R 待预测车辆，目标车辆三 is the current collision risk value between the vehicle to be predicted 100 and the target vehicle 300, R 车 is the decision influence coefficient of target vehicle three 300.

[0084] Specifically, the final collision risk of the predicted vehicle 100 relative to the target vehicle 200 can be calculated according to the following formula:

[0085] R 系统，目标车辆二 =R 待预测车辆，目标车辆二 *R 车

[0086] Among them, R 系统，目标车辆二 is the final collision risk between the vehicle 100 to be predicted and the target vehicle 200, R 待预测车辆，目标车辆二 is the current collision risk value between the vehicle to be predicted 100 and the target vehicle 200, R 车 is the decision influence coefficient of target vehicle 200.

[0087] The technical solution of the embodiment of the present application solves the problem of judgment delay and inaccuracy caused by the traditional method of judging the collision risk based solely on the motion state of the target vehicle and the motion state of the vehicle to be predicted, by identifying the taillight information of the target vehicle and determining the decision influence coefficient according to the different behaviors of the target vehicle. It achieves the effect of predicting the collision risk with the vehicle to be predicted before the target vehicle brakes, changes lanes, etc., and reasonably adjusts the decision influence coefficient to ensure that the system makes correct decisions, such as: early alarm or braking to avoid.

[0088] Example 3

[0089] Figure 6 This is a flowchart of a method for predicting vehicle collision risk when the target obstacle is a pedestrian in Example 3 of the present invention. In this embodiment, based on the first embodiment described above, the target obstacle is a target pedestrian, and the decision information includes behavioral information of pedestrians in the same pedestrian crossing area as the target pedestrian. The behavioral information includes at least position information and speed information. Accordingly, a decision influence coefficient of the target obstacle is determined based on the decision information of the target obstacle, including: determining an individual collision risk coefficient of the target pedestrian based on the position information and speed information of the target pedestrian; if the target pedestrian is within a lane line in the pedestrian crossing area, using the individual collision risk coefficient as the decision influence coefficient of the target pedestrian; if the target pedestrian is within a first area of ​​the pedestrian crossing area, updating the individual collision risk coefficient based on whether there are pedestrians within the lane line to obtain the decision influence coefficient; wherein the first area is a range within which the distance from the lane line is less than or equal to a first distance threshold; if the target pedestrian is within a second area of ​​the pedestrian crossing area, updating the individual collision risk coefficient based on the behavioral information of the pedestrians within the first area to obtain the decision influence coefficient; wherein the second area is a range within which the distance from the lane line is greater than a first distance threshold and less than or equal to a second distance threshold, where the first distance threshold is less than the second distance threshold.

[0090] like Figure 6 As shown, the method includes the following steps:

[0091] S501: Determine a current collision risk value based on current state information of the vehicle to be predicted and the target pedestrian.

[0092] The target pedestrian is a pedestrian in the pedestrian crossing area of ​​the intersection; the pedestrian's current state information includes the relative speed, relative distance, and relative acceleration to the vehicle to be predicted; the determination of the current collision risk value can refer to Example 1.

[0093] S502: Determine the individual collision risk coefficient of the target pedestrian based on the position information and speed information of the target pedestrian.

[0094] Among them, the position information and speed information can be obtained by relying on recognition devices such as cameras and distance sensing devices such as sensors installed on the vehicle to be predicted; the individual collision risk coefficient is the possibility of a collision between the target pedestrian and the vehicle to be predicted in the pedestrian crossing area; the position of the target pedestrian when the vehicle to be predicted passes through the pedestrian crossing area is determined based on the position information and speed information of the target pedestrian. The closer the position of the target pedestrian is to the vehicle to be predicted, the higher the individual collision risk coefficient.

[0095] Specifically, Figure 7 This is a schematic diagram of the position relationship between the target pedestrian and the vehicle to be predicted when the target obstacle is the target pedestrian in the vehicle collision risk prediction method in Example 3 of the present invention. For the pedestrian crossing scene, it can be divided into 4 areas: S1, S2, S3, and S4, arranged from left to right, where area S1 is the lane line range area, S2 is the first area, S3 is the second area, and S4 is the third area recursively deduced according to the threshold. Similarly, the area can be further extended according to the range of the pedestrian crossing area and the actual prediction accuracy, and there is no limit on the number of extended areas.

[0096] For example, when the target pedestrian is in the area where the predicted vehicle 100 will pass, the faster the target pedestrian's speed is and the shorter the time required to pass, the lower the individual collision risk coefficient is. When the target pedestrian is walking away from the predicted vehicle 100, the individual collision risk coefficient can be set to 0.

[0097] The S1 area is the lane line area. The lane line area can be the width of the lane where the predicted vehicle 100 is located, a set range, or the width of the predicted vehicle 100 multiplied by a coefficient A. The coefficient A can be set according to the state of the predicted vehicle 100. The vehicle state may include: vehicle speed; for example, the width of the lane line can be 2.5 meters or the width of the predicted vehicle 100 multiplied by 1.1; if the pedestrian cannot leave the S1 area before the predicted vehicle 100 arrives, a collision will occur. If the target pedestrian speed is 0, the predicted vehicle 100 will definitely collide with the target pedestrian when passing by, and the individual collision risk coefficient can be set to 1; the S2 area is the first area, and the width of the S2 area can be set to 3m; S3 is the second area, and the width of the S3 area can be set to 3.5m; the S4 area is the third area, and the width of the S4 area can be set to 4m; in the S2, S3, and S4 areas, the faster the target pedestrian walks in the direction of the lane line, the higher the individual collision risk coefficient is, which can be adjusted according to actual tests. Figure 8 Shown is a quantitative schematic diagram for determining the individual collision risk factor based on the pedestrian's zone and speed.

[0098] S503: If the target pedestrian is within the lane line range in the pedestrian crossing area, the individual collision risk coefficient is used as the decision influence coefficient of the target pedestrian.

[0099] The decision-making influence coefficient is used to determine the likelihood that the target pedestrian will continue its current behavior based on the target pedestrian's behavior information and the behavior information of pedestrians near the target pedestrian. If the target pedestrian cannot leave the lane before the predicted vehicle 100 passes through the pedestrian crossing area, the predicted vehicle 100 will definitely collide with the target pedestrian. Therefore, the individual collision coefficient of the target pedestrian within the lane is used as the decision-making influence coefficient.

[0100] S504. If the target pedestrian is within the first area of ​​the pedestrian crossing area, the individual collision risk coefficient is updated based on whether there are pedestrians within the lane line range to obtain a decision influence coefficient; wherein the first area is a range where the distance from the lane line is less than or equal to the first distance threshold.

[0101] The first threshold is greater than the width of the lane line. For example, the first threshold can be 3 meters. When the target pedestrian is in the first area and there are other pedestrians within the lane line, it will affect the behavior of the target pedestrian, so the decision influence coefficient of the target pedestrian needs to be updated. If the target pedestrian in the first area sees other pedestrians in the lane line and passing through the pedestrian crossing area, the possibility of the target pedestrian following increases, and the decision influence coefficient of the target pedestrian will increase. The decision influence coefficient needs to be updated to the decision influence coefficient of the first pedestrian.

[0102] Specifically, such as Figure 7 As shown, area S1 is the lane line range, area S2 is the first area, and the presence of pedestrians in area S1 serves as a crossing demonstration for target pedestrians in area S2, which will inspire the target pedestrians in area S2 to take crossing actions.

[0103] For example, the decision influence coefficient of the target pedestrian in the S2 area can be obtained according to the following formula:

[0104] In the above formula, R S2,man is the decision influence coefficient of the target pedestrian in the S2 area under the influence of the pedestrians in the S1 area, R′ S2,man is the individual collision risk coefficient of the target pedestrian. When there is a pedestrian in the S1 area, determine R S2,man The coefficient 1.2 can be modified according to actual conditions. The specific value of the coefficient is not limited here, and the value is greater than 1.

[0105] S505. If the target pedestrian is in the second area of ​​the pedestrian crossing area, the individual collision risk coefficient is updated based on the behavior information of the pedestrian in the first area to obtain a decision influence coefficient; wherein the behavior information includes at least position information and speed information.

[0106] The second area is a range where the distance from the lane line is greater than a first distance threshold and less than or equal to a second distance threshold, and the first distance threshold is less than the second distance threshold.

[0107] Among them, the target pedestrian in the second area will be affected by the pedestrians in the first area. When the pedestrians in the first area move toward the pedestrian crossing area, it will affect the behavior of the target pedestrian in the second area, increasing the possibility of the target pedestrian following through the pedestrian crossing area, and the decision influence coefficient of the target pedestrian in the second area is updated to the second pedestrian decision influence coefficient; when the pedestrians in the first area are in a stationary state waiting for the predicted vehicle 100 to pass, the possibility of the target pedestrian waiting for the predicted vehicle 100 to pass is reduced, and the decision influence coefficient of the target pedestrian in the second area is updated to the third pedestrian decision influence coefficient.

[0108] Specifically, such as Figure 7 As shown, area S1 represents the lane markings, area S2 represents the first area, and area S3 represents the second area. When pedestrians are present in all three areas, their behavior is influenced by the actions of those around them. For example, pedestrian 1 is in area S2, pedestrian 2 is in area S3, and pedestrian 3 is in area S4. Pedestrian 1 perceives a collision risk in area S2 and stops crossing the road. Pedestrian 2, while unaware of the risk, sees pedestrian 1 standing still. Therefore, when pedestrian 2 arrives in area S2, they will likely adopt the same strategy. Similarly, pedestrian 3 will be influenced by pedestrian 2's actions. If pedestrian 3 notices pedestrian 2 slowing down, they will also adopt a similar strategy.

[0109] In a feasible embodiment, based on the behavior information of pedestrians located in the first area, the individual collision risk coefficient is updated to obtain the decision influence coefficient, including:

[0110] determining a candidate individual collision risk coefficient for each candidate pedestrian based on position information and speed information of each candidate pedestrian located within the first area;

[0111] determining a regional average collision risk coefficient for the first region based on the candidate individual collision risk coefficients;

[0112] If the regional average collision risk coefficient is zero and there is a pedestrian in the first area, determining the decision influence coefficient of the target pedestrian based on the product of the individual collision risk coefficient of the target pedestrian and a preset first weakening coefficient; wherein the preset first weakening coefficient is less than 1;

[0113] If the average collision risk coefficient of the area is zero and there is no pedestrian in the first area, the individual collision risk coefficient of the target pedestrian is used as the decision influence coefficient of the target pedestrian;

[0114] If the regional average collision risk coefficient is not equal to zero, the decision influence coefficient of the target pedestrian is determined according to the product of the individual collision risk coefficient of the target pedestrian and the regional average collision risk coefficient.

[0115] Specifically, the decision influence coefficient of the target pedestrian in the second area is determined according to the following formula:

[0116]

[0117] In the above formula, R S(i-1) For S i-1 Regional average collision risk factor of the region, R′ Si,man For S i The individual collision risk coefficient of the target pedestrian in the area, R Si,man For S i The decision-making influence coefficient of the target pedestrian in the area, 0.4 is the preset first weakening coefficient, and its specific value can be adjusted according to actual conditions and is not limited here.

[0118] The candidate pedestrians may be some or all of the pedestrians in the first area. The average collision risk coefficient of the first area is the average of the collision risk coefficients of the candidate pedestrians in the area. For example, the following formula may be used to calculate the average collision risk coefficient of each candidate pedestrian in the area:

[0119]

[0120] Among them, R 群 is the average collision risk coefficient of each candidate pedestrian in the area, n is the number of candidate pedestrians, R i is the individual collision risk coefficient of each candidate pedestrian in the area; when the average collision risk coefficient of the first area is zero, it means that the pedestrians in the current area will not collide when the predicted vehicle 100 travels according to the current state.

[0121] S506 : Predicting the final collision risk between the target pedestrian and the vehicle to be predicted based on the current collision risk value and the decision influence coefficient.

[0122] The risk assessment of the vehicle and a single pedestrian is quantified as:

[0123] R 系统，(Si,man) =R 本车，(Si,man) *R Si,man ;

[0124] Among them, R 本车，(Si,man) is the current collision risk value between the vehicle 100 to be predicted and the target pedestrian in the Si area, R 本车，(Si,man) It can be R 本车，目标 In the formula, the target is calculated when it is a pedestrian;

[0125] R 系统，(Si,man) is the final collision risk value between the vehicle 100 to be predicted and the target pedestrian in the Si area;

[0126] R Si,man is the decision-making influence coefficient of the target pedestrian in the Si area.

[0127] The risk assessment of vehicles and people is quantified as:

[0128] R 系统，群 =R 本车，群 *R 群 ;

[0129] Among them, R 本车，群 is the current collision risk value between the vehicle 100 and the crowd to be predicted, R 本车，群 It can be R 本车，目标 In the formula, it is calculated when the target is a crowd;

[0130] R 系统，群 is the final collision risk value between the vehicle 100 and the crowd to be predicted;

[0131] R 群 is the decision-making influence coefficient of the crowd.

[0132] The technical solution of the embodiment of the present application quantifies the pedestrian risk value by distinguishing the area where pedestrians are located at the intersection, judges the behavior of the target pedestrian and the influence of nearby pedestrians on the target pedestrian, and judges the decision-making influence coefficient of the target pedestrian. It solves the problems of unnecessary braking of the predicted vehicle when pedestrians are at the intersection under the traditional method and the inability to accurately judge the behavior of pedestrians when multiple pedestrians are in different areas of the intersection, thereby achieving the effect of eliminating unnecessary risk items and effectively and early identifying risks.

[0133] Example 4

[0134] Figure 9 This is a flowchart of a method for predicting vehicle collision risk when the target obstacle is a stop line, according to a fourth embodiment of the present invention. This embodiment, based on the above embodiment, features a target stop line as the target obstacle, and the decision information includes target traffic light information associated with the target stop line. Accordingly, determining the decision influence coefficient of the target obstacle based on the decision information of the target obstacle includes: determining the decision influence coefficient of the target stop line based on the target traffic light information, based on a predetermined mapping relationship between traffic light information and the decision influence coefficient of the stop line.

[0135] S701: Determine a current collision risk value based on current state information of the vehicle to be predicted and the target stop line.

[0136] Because the target stop line can be considered a target obstacle for the predicted vehicle as the target traffic light changes, for example, the predicted vehicle needs to stop before the target stop line when the target traffic light turns red. The current state of the target traffic light information may include the traffic light color, remaining time, and the distance and speed of the predicted vehicle from the target stop line. The current collision risk value can be determined by referring to Example 1.

[0137] S702 : Based on a predetermined mapping relationship between traffic light information and stop line decision influence coefficients, determine a target stop line decision influence coefficient according to target traffic light information.

[0138] Figure 10 This is a schematic diagram of the relationship between the target stop line and the position of the vehicle to be predicted when the target obstacle is the target stop line in the method for predicting vehicle collision risk in the third embodiment of the present invention.

[0139] When traffic light information changes, the behavior of pedestrians near the intersection also changes, thus affecting the pedestrian's decision-making influence coefficient. For example, when the traffic light in the lane where the predicted vehicle 100 is located is green, the likelihood of pedestrians passing through the intersection is relatively low, so the decision-making influence coefficient is set to the seventh decision-making influence coefficient. When the traffic light in the lane where the predicted vehicle 100 is located is yellow, the likelihood of pedestrians passing through the intersection increases, so the decision-making influence coefficient is set to the eighth decision-making influence coefficient. When the traffic light in the lane where the predicted vehicle 100 is located is red, the stop line becomes a risk obstacle. Although the collision risk for stationary pedestrians is zero at this time, pedestrians may pass through the pedestrian crossing area because the red light is on, meeting the traffic conditions. Therefore, the decision-making influence coefficient is set to the ninth decision-making influence coefficient. The ninth decision-making influence coefficient is greater than the eighth decision-making influence coefficient, and the eighth decision-making influence coefficient is greater than the seventh decision-making influence coefficient.

[0140] For example, the decision influence coefficient R of the target stop line can be determined according to the following formula: 线 , when the traffic light is green, the decision-making influence coefficient of the target stop line is 0; when the traffic light is yellow, the decision-making influence coefficient of the target stop line is 0.9; when the traffic light is red, the decision-making influence coefficient of the target stop line is 1;

[0141]

[0142] S703 : Predicting the final collision risk between the target stop line and the vehicle to be predicted based on the current collision risk value and the decision influence coefficient.

[0143] The risk assessment of the vehicle to be predicted 100 and the target stop line is quantified as:

[0144] R 系统，线 =R本车，线 *R 线 ;

[0145] Among them, R 本车，线 is the current collision risk value between the vehicle 100 to be predicted and the target stop line;

[0146] R 线 is the decision influence coefficient of the target stop line;

[0147] R 系统，线 is the final collision risk between the vehicle 100 and the target stop line to be predicted.

[0148] The technical solution of the embodiment of the present application solves the defect that traditional methods cannot predict changes in pedestrian behavior when traffic light information changes, by obtaining traffic light information and changing the decision influence coefficient according to the traffic light information, and achieves the effect of predicting pedestrian behavior in advance according to different traffic light information.

[0149] Example 5

[0150] Figure 11 FIG. 1 is a schematic diagram of the structure of the vehicle collision risk prediction device in the fifth embodiment of the present invention. This embodiment is applicable to the case of vehicle collision risk prediction. Figure 11 As shown, the device includes:

[0151] The current collision risk value determination module 801 is configured to determine the current collision risk value based on the current state information of the vehicle to be predicted and the target obstacle; wherein the obstacle includes at least one of the following: a vehicle, a pedestrian, and a stop line;

[0152] A decision influence coefficient determination module 802 is configured to determine a decision influence coefficient of a target obstacle based on the decision information of the target obstacle;

[0153] The final collision risk prediction module 803 is used to predict the final collision risk between the target obstacle and the vehicle to be predicted based on the current collision risk value and the decision influence coefficient.

[0154] In this technical solution, optionally, the current state information includes at least the relative distance, relative speed, and relative acceleration between the vehicle to be predicted and the target obstacle;

[0155] Accordingly, the current collision risk value determination module 801 includes:

[0156] Determine the collision time based on the relative distance, relative speed and relative acceleration between the vehicle to be predicted and the target obstacle;

[0157] The current collision risk value is determined based on the collision time.

[0158] In this technical solution, optionally, the target obstacle is a target vehicle, and the decision information includes taillight information; wherein the taillight information includes at least one of the following: brake light information and turn signal information;

[0159] Accordingly, the decision influence coefficient determination module 802 includes:

[0160] a lane information determination unit, configured to obtain lane information of a target vehicle;

[0161] The decision influence coefficient determination unit is used to determine the decision influence coefficient of the target vehicle based on the taillight information and lane information.

[0162] In this technical solution, optionally, the lane information includes the same lane and adjacent lanes as the vehicle to be predicted;

[0163] Accordingly, the decision influence coefficient determination unit is specifically used to:

[0164] If the lane information indicates that the target vehicle is in the same lane as the target vehicle, the decision influence coefficient of the target vehicle is determined based on the brake light information and the deceleration state of the target vehicle;

[0165] If the lane information indicates that the target vehicle is in an adjacent lane, the decision influence coefficient of the target vehicle is determined based on the turn signal information and the motion trajectory of the target vehicle.

[0166] In this technical solution, optionally, the target obstacle is a target pedestrian, and the decision information includes behavior information of pedestrians in the same pedestrian crossing area as the target pedestrian; wherein the behavior information includes at least position information and speed information;

[0167] Accordingly, the decision influence coefficient determination module 802 includes:

[0168] An individual collision risk coefficient determination unit, configured to determine an individual collision risk coefficient of a target pedestrian based on the position information and speed information of the target pedestrian;

[0169] a first decision-making influence coefficient determining unit, configured to use the individual collision risk coefficient as the decision-making influence coefficient of the target pedestrian if the target pedestrian is within the lane line range in the pedestrian crossing area;

[0170] a second decision-influence coefficient determination unit configured to, if the target pedestrian is within a first region of the pedestrian crossing area, update the individual collision risk coefficient based on whether there are pedestrians within the lane line to obtain the decision-influence coefficient; wherein the first region is a range within which the distance from the lane line is less than or equal to a first distance threshold;

[0171] The third determination unit for the decision influence coefficient is used to update the individual collision risk coefficient based on the behavioral information of the pedestrians in the first area if the target pedestrian is in the second area of ​​the pedestrian crossing area, thereby obtaining the decision influence coefficient; wherein the second area is a range in which the distance from the lane line is greater than the first distance threshold and less than or equal to the second distance threshold, and the first distance threshold is less than the second distance threshold.

[0172] In this technical solution, optionally, the third determination unit of the decision influence coefficient is specifically used to:

[0173] determining a candidate individual collision risk coefficient for each candidate pedestrian based on position information and speed information of each candidate pedestrian located within the first area;

[0174] determining a regional average collision risk coefficient for the first region based on the candidate individual collision risk coefficients;

[0175] If the regional average collision risk coefficient is zero and there is a pedestrian in the first area, determining the decision influence coefficient of the target pedestrian based on the product of the individual collision risk coefficient of the target pedestrian and a preset first weakening coefficient; wherein the preset first weakening coefficient is less than 1;

[0176] If the average collision risk coefficient of the area is zero and there is no pedestrian in the first area, the individual collision risk coefficient of the target pedestrian is used as the decision influence coefficient of the target pedestrian;

[0177] If the regional average collision risk coefficient is not equal to zero, the decision influence coefficient of the target pedestrian is determined according to the product of the individual collision risk coefficient of the target pedestrian and the regional average collision risk coefficient.

[0178] In this technical solution, optionally, the target obstacle is a target stop line, and the decision information includes target traffic light information associated with the target stop line;

[0179] Accordingly, the decision influence coefficient determination module 802 is specifically configured to:

[0180] Based on a predetermined mapping relationship between traffic light information and a stop line decision influence coefficient, the decision influence coefficient of the target stop line is determined according to the target traffic light information.

[0181] The vehicle collision risk prediction device provided in the embodiment of the present invention can execute the vehicle collision risk prediction method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of executing the vehicle collision risk prediction method.

[0182] Example 6

[0183] Figure 12This is a structural diagram of an electronic device provided in Example 6 of the present invention. Figure 12 A block diagram of an exemplary electronic device 12 suitable for implementing embodiments of the present invention is shown. Figure 12 The electronic device 12 shown is only an example and should not limit the functionality and scope of use of the embodiments of the present invention.

[0184] like Figure 12 As shown, electronic device 12 is implemented as a general-purpose computing device. Components of electronic device 12 may include, but are not limited to, one or more processors or processing units 16, system memory 28, and a bus 18 connecting various system components (including system memory 28 and processing unit 16).

[0185] Bus 18 represents one or more of several types of bus structures, including a storage device bus or storage device controller, a peripheral bus, an accelerated graphics port, a processor, or a local bus using any of a variety of bus architectures. Examples of these architectures include, but are not limited to, an Industry Standard Architecture (ISA) bus, a Micro Channel Architecture (MAC) bus, an Enhanced ISA bus, a Video Electronics Standards Association (VESA) local bus, and a Peripheral Component Interconnect (PCI) bus.

[0186] The electronic device 12 typically includes a variety of computer system readable media. These media can be any available media that can be accessed by the electronic device 12, including volatile and non-volatile media, removable and non-removable media.

[0187] The system storage 28 may include computer system readable media in the form of volatile storage devices, such as random access memory (RAM) 30 or cache memory 32. The electronic device 12 may further include other removable or non-removable, volatile or non-volatile computer system storage media. By way of example only, the storage system 34 may be configured to read and write non-removable, non-volatile magnetic media ( Figure 12 Not shown, often called a "hard drive"). Although Figure 12 Although not shown, a magnetic disk drive for reading and writing to a removable non-volatile magnetic disk (e.g., a "floppy disk"), and an optical disk drive for reading and writing to a removable non-volatile optical disk (e.g., a CD-ROM, DVD-ROM, or other optical media) may be provided. In these cases, each drive may be connected to bus 18 via one or more data media interfaces. Storage device 28 may include at least one program product having a set (e.g., at least one) of program modules configured to perform the functions of various embodiments of the present invention.

[0188] A program / utility 40 having a set (at least one) of program modules 42 may be stored, for example, in storage device 28. Such program modules 42 include, but are not limited to, an operating system, one or more application programs, other program modules, and program data, each of which, or some combination thereof, may include an implementation of a network environment. Program modules 42 generally implement the functions and / or methods of the embodiments described herein.

[0189] The electronic device 12 may also communicate with one or more external devices 14 (e.g., a keyboard, a pointing device, a display 24, etc.), one or more devices that enable a user to interact with the device 12, and / or any device that enables the device 12 to communicate with one or more other computing devices (e.g., a network card, a modem, etc.). Such communication may be performed through an input / output (I / O) interface 22. Furthermore, the electronic device 12 may also communicate with one or more networks (e.g., a local area network (LAN), a wide area network (WAN), and / or a public network, such as the Internet) through a network adapter 20. Figure 12 As shown, the network adapter 20 communicates with other modules of the electronic device 12 via the bus 18. Figure 12 Not shown, other hardware and / or software modules may be used in conjunction with the electronic device 12, including but not limited to microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0190] The processing unit 16 executes various functional applications and data processing by running programs stored in the system storage device 28, such as implementing the vehicle collision risk prediction method provided by the embodiment of the present invention, including:

[0191] Determining a current collision risk value based on current state information of the vehicle to be predicted and the target obstacle; wherein the obstacle includes at least one of the following: a vehicle, a pedestrian, and a stop line;

[0192] Determine the decision influence coefficient of the target obstacle according to the decision information of the target obstacle;

[0193] The final collision risk between the target obstacle and the vehicle to be predicted is predicted based on the current collision risk value and the decision influence coefficient.

[0194] Example 7

[0195] The seventh embodiment of the present invention further provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the method for predicting vehicle collision risk provided in the embodiment of the present invention is implemented, including:

[0196] Determining a current collision risk value based on current state information of the vehicle to be predicted and the target obstacle; wherein the obstacle includes at least one of the following: a vehicle, a pedestrian, and a stop line;

[0197] Determine the decision influence coefficient of the target obstacle according to the decision information of the target obstacle;

[0198] The final collision risk between the target obstacle and the vehicle to be predicted is predicted based on the current collision risk value and the decision influence coefficient.

[0199] The computer storage medium of the embodiment of the present invention can adopt any combination of one or more computer-readable media. Computer-readable media can be computer-readable signal media or computer-readable storage media. Computer-readable storage media can be, for example, but not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices or components, or any combination thereof. More specific examples (non-exhaustive list) of computer-readable storage media include: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In this document, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by an instruction execution system, device or device or used in combination with it.

[0200] A computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, which carries computer-readable program code. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium that can transmit, propagate, or transport a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0201] Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

[0202] The computer program code for performing the operations of the present invention can be written in one or more programming languages, or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a stand-alone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computer (e.g., through the Internet using an Internet service provider).

[0203] Note that the above are only preferred embodiments of the present invention and the technical principles employed. Those skilled in the art will understand that the present invention is not limited to the specific embodiments described herein, and that various obvious changes, readjustments, and substitutions can be made by those skilled in the art without departing from the scope of protection of the present invention. Therefore, although the present invention has been described in detail through the above embodiments, the present invention is not limited to the above embodiments and may include many other equivalent embodiments without departing from the concept of the present invention. The scope of the present invention is determined by the scope of the appended claims.

Claims

1. A method for predicting vehicle collision risk, characterized in that: include: Determining a current collision risk value based on current state information of the vehicle to be predicted and the target obstacle; wherein the obstacle includes at least one of the following: a vehicle, a pedestrian, and a stop line; determining a decision influence coefficient of the target obstacle according to the decision information of the target obstacle; Predicting a final collision risk between the target obstacle and the vehicle to be predicted based on the current collision risk value and the decision influence coefficient; Wherein, the target obstacle is a target pedestrian, and the decision information includes behavior information of pedestrians in the same pedestrian crossing area as the target pedestrian; wherein the behavior information includes at least position information and speed information; Accordingly, determining the decision influence coefficient of the target obstacle according to the decision information of the target obstacle includes: determining an individual collision risk coefficient of the target pedestrian based on the position information and speed information of the target pedestrian; If the target pedestrian is within the lane line range in the pedestrian crossing area, the individual collision risk coefficient is used as the decision influence coefficient of the target pedestrian; If the target pedestrian is within a first area of ​​the pedestrian crossing area, the individual collision risk coefficient is updated based on whether there is a pedestrian within the lane line range to obtain a decision influence coefficient; wherein the first area is a range where the distance from the lane line is less than or equal to a first distance threshold; If the target pedestrian is in the second area of ​​the pedestrian crossing area, the individual collision risk coefficient is updated based on the behavior information of the pedestrian in the first area to obtain a decision influence coefficient; wherein, the second area is a range where the distance between the target pedestrian and the lane line is greater than the first distance threshold and less than or equal to the second distance threshold, and the first distance threshold is less than the second distance threshold.

2. The method according to claim 1, characterized in that The current state information includes at least the relative distance, relative speed, and relative acceleration between the vehicle to be predicted and the target obstacle; Accordingly, the current collision risk value is determined based on the current state information of the vehicle to be predicted and the target obstacle, including: Determining a collision time based on a relative distance, relative speed, and relative acceleration between the vehicle to be predicted and the target obstacle; A current collision risk value is determined according to the collision time.

3. The method according to claim 1, characterized in that The target obstacle is a target vehicle, and the decision information includes taillight information; wherein the taillight information includes at least one of the following: brake light information and turn signal information; Accordingly, determining the decision influence coefficient of the target obstacle according to the decision information of the target obstacle includes: Obtaining lane information of the target vehicle; A decision influence coefficient of the target vehicle is determined according to the taillight information and the lane information.

4. The method according to claim 3, characterized in that The lane information includes the same lane as the vehicle to be predicted and the adjacent lanes; Accordingly, determining the decision influence coefficient of the target vehicle according to the taillight information and the lane information includes: If the lane information indicates that the target vehicle is in the same lane as the target vehicle, determining the decision influence coefficient of the target vehicle according to the brake light information and the deceleration state of the target vehicle; If the lane information indicates that the target vehicle is in an adjacent lane to the target vehicle, the decision influence coefficient of the target vehicle is determined according to the turn signal information and the motion trajectory of the target vehicle.

5. The method according to claim 1, wherein Based on the behavior information of pedestrians located in the first area, the individual collision risk coefficient is updated to obtain a decision influence coefficient, including: determining a candidate individual collision risk coefficient for each candidate pedestrian based on position information and speed information of each candidate pedestrian located within the first area; determining a regional average collision risk coefficient for a first region based on the candidate individual collision risk coefficients; If the regional average collision risk coefficient is zero and there is a pedestrian in the first area, determining the decision influence coefficient of the target pedestrian based on the product of the individual collision risk coefficient of the target pedestrian and a preset first weakening coefficient; wherein the preset first weakening coefficient is less than 1; If the average collision risk coefficient of the area is zero and there are no pedestrians in the first area, the individual collision risk coefficient of the target pedestrian is used as the decision influence coefficient of the target pedestrian; If the regional average collision risk coefficient is not equal to zero, the decision influence coefficient of the target pedestrian is determined according to the product of the individual collision risk coefficient of the target pedestrian and the regional average collision risk coefficient.

6. The method according to claim 1, characterized in that The target obstacle is a target stop line, and the decision information includes target traffic light information associated with the target stop line; Accordingly, determining the decision influence coefficient of the target obstacle according to the decision information of the target obstacle includes: Based on a predetermined mapping relationship between traffic light information and a stop line decision influence coefficient, the decision influence coefficient of the target stop line is determined according to the target traffic light information.

7. A vehicle collision risk prediction device, characterized in that: include: A current collision risk value determination module is used to determine a current collision risk value based on current state information of the vehicle to be predicted and the target obstacle; wherein the obstacle includes at least one of the following: a vehicle, a pedestrian, and a stop line; A decision influence coefficient determination module, configured to determine the decision influence coefficient of the target obstacle based on the decision information of the target obstacle; a final collision risk prediction module, configured to predict a final collision risk between the target obstacle and the vehicle to be predicted based on the current collision risk value and the decision influence coefficient; The target obstacle is a target pedestrian, and the decision information includes behavior information of pedestrians in the same pedestrian crossing area as the target pedestrian; wherein the behavior information includes at least position information and speed information; Accordingly, the decision-making influence coefficient determination module includes: An individual collision risk coefficient determination unit, configured to determine an individual collision risk coefficient of a target pedestrian based on the position information and speed information of the target pedestrian; a first decision-making influence coefficient determining unit, configured to use the individual collision risk coefficient as the decision-making influence coefficient of the target pedestrian if the target pedestrian is within the lane line range in the pedestrian crossing area; a second decision-influence coefficient determination unit configured to, if the target pedestrian is within a first region of the pedestrian crossing area, update the individual collision risk coefficient based on whether there are pedestrians within the lane line to obtain the decision-influence coefficient; wherein the first region is a range within which the distance from the lane line is less than or equal to a first distance threshold; The third determination unit for the decision influence coefficient is used to update the individual collision risk coefficient based on the behavioral information of the pedestrians in the first area if the target pedestrian is in the second area of ​​the pedestrian crossing area, thereby obtaining the decision influence coefficient; wherein the second area is a range in which the distance from the lane line is greater than the first distance threshold and less than or equal to the second distance threshold, and the first distance threshold is less than the second distance threshold.

8. An electronic device, characterized in that: include: one or more processors; a storage device for storing one or more programs, When the one or more programs are executed by the one or more processors, the one or more processors implement the method for predicting vehicle collision risk as described in any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method for predicting vehicle collision risk as claimed in any one of claims 1 to 6 is implemented.

Images