A vehicle driving collision risk assessment method, system, device and storage medium

Abstract

Embodiments of the present application provide a vehicle driving collision risk assessment method, system, device and storage medium, the method comprising: obtaining vehicle geometric parameters, vehicle relative position information, vehicle motion data and predicted driving path information, obtaining the host vehicle driving trajectory equation and the target vehicle driving trajectory equation under the target driving behavior according to the vehicle relative position information and the predicted driving path information, calculating the collision time parameter under the target driving behavior according to the predicted driving path information, the vehicle motion data, the host vehicle driving trajectory equation and the target vehicle driving trajectory equation, performing collision effectiveness verification according to the vehicle geometric parameters and the collision time parameter under the target driving behavior, normalizing the collision time parameter that has passed the collision effectiveness verification into a risk coefficient, and outputting a turning decision result corresponding to the risk coefficient. The present application improves the reliability of the vehicle driving collision risk assessment.

Description

Technical Field

[0001] This invention relates to the field of intelligent driving technology, and in particular to a method, system, device and storage medium for assessing vehicle collision risks. Background Technology

[0002] With the development of intelligent driving technology, decision-making and planning for driving behavior in complex traffic environments and emergency situations have become crucial technical safeguards for ensuring safe and efficient vehicle operation. Vehicle turning behavior, as a typical driving behavior decision, is a critical component of intelligent driving technology because improper turning can disrupt the normal movement of surrounding vehicles and easily lead to traffic accidents.

[0003] Currently, there is limited risk assessment available for intelligent vehicles' turning behavior decisions. Existing technologies typically assess the risk of turning behavior by calculating the relative motion between the vehicle and the target vehicle in real time. Because this real-time calculation method leaves drivers with limited reaction time in complex traffic environments or emergency situations, it still cannot prevent traffic accidents, thus reducing the reliability of vehicle turning behavior decisions. Therefore, improving the reliability of collision risk assessment has become an urgent problem to be solved. Summary of the Invention

[0004] The purpose of this invention is to provide a method, system, device, and storage medium for assessing vehicle collision risks, thereby improving the reliability of risk assessments for vehicle collisions. The specific technical solution is as follows:

[0005] A method for assessing vehicle collision risk includes:

[0006] It acquires vehicle geometric parameters, vehicle relative position information, vehicle motion data, and predicted driving path information.

[0007] Based on the relative position information of the vehicles and the predicted driving path information, the driving trajectory equations of the current vehicle and the target vehicle under the target driving behavior are obtained, wherein the target driving behavior includes at least one of turning behavior, non-turning behavior, oncoming behavior, and traversing behavior.

[0008] Based on the predicted driving path information, the vehicle motion data, the driving trajectory equation of the current vehicle, and the driving trajectory equation of the target vehicle, the collision time parameters under the target driving behavior are calculated.

[0009] The collision validity is verified based on the vehicle's geometric parameters and the collision time parameters under the target driving behavior.

[0010] The collision time parameter, which has passed the collision validity verification, is normalized into a risk coefficient, and the turning decision result corresponding to the risk coefficient is output.

[0011] Optionally, obtaining the vehicle's trajectory equation and the target vehicle's trajectory equation under the target driving behavior based on the vehicle's relative position information and the predicted driving path information includes at least one of the following four methods:

[0012] Method 1: Based on the predicted driving path information, the vehicle's turning radius R and lateral displacement y in the vehicle's coordinate system... h and longitudinal displacement x h The equation of the vehicle's trajectory under the stated turning behavior is obtained as follows:

[0013] (x h -e) 2 +(y h -f) 2 =R 2 ,

[0014] Where e and f are the parameters of the vehicle's trajectory equation under the turning behavior, and the vehicle's trajectory equation under the turning behavior represents the simulated driving trajectory of the vehicle under the turning behavior.

[0015] Based on the predicted driving path information, the vehicle's turning endpoint coordinates (y) end1 x end1 The vehicle's turning starting point coordinates (0, 0), the vehicle's turning radius R, and the vehicle's trajectory equation under the turning behavior are obtained through the formula:

[0016]

[0017] Calculate e and f.

[0018] Method 2: Based on the lateral displacement y of the vehicle in the vehicle coordinate system. h and the longitudinal displacement x h The equation of the vehicle's trajectory under the non-turning behavior is obtained as follows:

[0019] y h =gx h

[0020] Wherein, g is the parameter of the vehicle's trajectory equation under the non-turning behavior, and the vehicle's trajectory equation under the non-turning behavior represents the simulated driving trajectory of the vehicle under the non-turning behavior.

[0021] Based on the predicted driving path information, the coordinates (y) of the vehicle's non-turning endpoint are... end2 x end2The starting coordinates of the vehicle without turning (0, 0) and the trajectory equation of the vehicle under the described non-turning behavior are obtained through the formula:

[0022] y end2 =gx end2 ,

[0023] Calculate g.

[0024] Method 3: Based on the lateral displacement y of the target vehicle in the vehicle coordinate system from the relative vehicle position information. t and the longitudinal displacement x t The trajectory equation of the target vehicle under the aforementioned opposing behavior is obtained as follows:

[0025] y t =a i x t +b i ,

[0026] Among them, a i and b i The parameter is the target vehicle's trajectory equation for the oncoming target vehicle, where i is the serial number of the target vehicle identified by this vehicle. The target vehicle's trajectory equation under the oncoming behavior represents the simulated driving trajectory of the target vehicle under the oncoming behavior.

[0027] Method 4: Based on the lateral displacement y of the target vehicle in the vehicle coordinate system from the relative vehicle position information. t and the longitudinal displacement x t Obtain the trajectory equation of the target vehicle under the traversing behavior:

[0028] x t =c i y t +d i ,

[0029] Among them, c i and d i The parameters are the target vehicle's trajectory equations when it traverses the target vehicle. The target vehicle's trajectory equations under the traversing behavior characterize the simulated trajectory of the target vehicle under the traversing behavior.

[0030] Optionally, obtaining the vehicle relative position information includes: obtaining the vehicle relative position information of the target vehicle.

[0031] The acquisition of the relative position information of the target vehicle specifically includes:

[0032] Within a preset time interval, the relative vehicle position information of multiple target vehicles is continuously recorded, resulting in multiple position information groups, each corresponding one-to-one with a target vehicle. Furthermore, each relative vehicle position information included in each position information group represents the relative vehicle position information of the target vehicle corresponding to that position information group.

[0033] For each group of location information: The relative vehicle positions at all times other than the latest time are calculated according to the formula:

[0034]

[0035] Transform to the vehicle's coordinate system where the latest vehicle relative position information is located, wherein the recording time of the latest vehicle relative position information is later than the recording time of the vehicle relative position information at other times, p i The relative position information of the i-th target vehicle at other times. The relative position information of the i-th target vehicle at other times is transformed to the relative position information of the vehicle in the coordinate system of the latest time, where Δθ is the yaw angle of the vehicle from the other times to the latest time, and Δx is the yaw angle of the vehicle. h Δy represents the longitudinal displacement of the vehicle in the vehicle's coordinate system per unit time. h The displacement of the vehicle in the vehicle's coordinate system within the unit time interval is the lateral displacement of the vehicle.

[0036] Optionally, the a i The b i The c i and the d i The methods of obtaining it include:

[0037] Obtain a first number of relative position information of the target vehicle under the opposing behavior.

[0038] Substitute the first number of relative position information of the vehicles into the target vehicle's trajectory equation under the opposing behavior, and then, based on the least squares principle and the residual principle, simultaneously solve the target vehicle's trajectory equation under the opposing behavior after substitution:

[0039]

[0040] Calculate the a i and the b i Where j is the j-th historical time point of the target vehicle, k is the total number of historical time points, and w j Let x be the weight of the target vehicle at the j-th historical time point.ij In the second number of vehicle relative position information, the longitudinal displacement of the i-th target vehicle in the vehicle coordinate system at the j-th historical time point is y. ij The second number of vehicle relative position information is the lateral displacement of the i-th target vehicle in the vehicle coordinate system at the j-th historical time point.

[0041] And / or, obtain a second number of vehicle relative position information of the target vehicle under the crossing behavior.

[0042] Substitute the second number of vehicle relative position information into the target vehicle trajectory equation under the traversing behavior, and according to the least squares principle and the residual principle, simultaneously solve the target vehicle trajectory equation under the traversing behavior after substitution:

[0043]

[0044] Calculate the c i and the d i .

[0045] Optionally, calculating the collision time parameters under the target driving behavior based on the predicted driving path information, the driving trajectory equation of the current vehicle, and the driving trajectory equation of the target vehicle includes:

[0046] By simultaneously establishing the trajectory equations of the vehicle under the turning behavior and the trajectory equations of the target vehicle under the target driving behavior, and solving for them, the coordinates of the first collision point between the vehicle and the target vehicle under the turning behavior are obtained. Based on the coordinates of the first collision point (y... p1 x p1 ), through the formula:

[0047]

[0048] Calculate the yaw angle θ of the vehicle as it moves in an arc from the first point of impact.

[0049] Based on the vehicle's turning radius R and the vehicle speed in the vehicle motion data, using the formula:

[0050]

[0051] Calculate the angular velocity ω of the vehicle as it performs the circular motion described above.

[0052] Based on the yaw angle θ and the angular velocity ω, using the formula:

[0053]

[0054] Calculate the first collision time parameter (TTC) between the vehicle and the target vehicle under the stated turning behavior. t .

[0055] And / or, simultaneously establish the vehicle's trajectory equation under the no-turn behavior and the target vehicle's trajectory equation under the target behavior, and solve for the coordinates of the second collision point between the vehicle and the target vehicle under the no-turn behavior. Based on the second collision point coordinates (y... p2 x p2 ), through the formula:

[0056]

[0057] Calculate the second collision time parameter (TTC) between the vehicle and the target vehicle under the non-turning behavior. nt , where V0(x p2 Let a(x) be the longitudinal velocity of the vehicle in the aforementioned vehicle coordinate system. p2 Let x be the longitudinal acceleration of the vehicle in the aforementioned vehicle coordinate system. p2 In the vehicle's coordinate system, this is the longitudinal distance between the vehicle and the second collision point.

[0058] Optionally, collision validity verification is performed based on the vehicle geometry parameters and the collision time parameters under the target driving behavior, including:

[0059] Based on the vehicle's geometric parameters and the collision time parameters under the target driving behavior, the potential collision area between the vehicle and the target vehicle is determined. It is then determined whether the collision time parameters fall within the time period between entering and leaving the collision area. If so, the collision is deemed valid. The moment of entering the collision area is the moment when the front of the target vehicle reaches the first collision point or the second collision point, and the moment of leaving the collision area is the moment when the rear of the target vehicle leaves the first collision point or the second collision point.

[0060] Optionally, the method further includes:

[0061] Based on the risk coefficient, the driving assistance strategy corresponding to the risk coefficient stored in the preset database is invoked, and the driving assistance strategy is sent to the preset driving control algorithm so that the preset driving control algorithm makes a decision based on the driving assistance strategy.

[0062] A vehicle collision risk assessment system, comprising:

[0063] The data acquisition unit is used to acquire vehicle geometric parameters, vehicle relative position information, vehicle motion data, and predicted driving path information.

[0064] The trajectory acquisition unit is used to obtain the trajectory equation of the current vehicle and the trajectory equation of the target vehicle under the target driving behavior based on the relative position information of the vehicle and the predicted driving path information, wherein the target driving behavior includes at least one of turning behavior, non-turning behavior, oncoming behavior and traversing behavior.

[0065] The parameter calculation unit is used to calculate the collision time parameters under the target driving behavior based on the predicted driving path information, the vehicle motion data, the driving trajectory equation of the vehicle itself, and the driving trajectory equation of the target vehicle.

[0066] The validity verification unit is used to perform collision validity verification based on the vehicle geometric parameters and the collision time parameters under the target driving behavior.

[0067] The coefficient generation unit is used to normalize the collision time parameter after the collision validity verification into a risk coefficient, and output the turning decision result corresponding to the risk coefficient.

[0068] Optionally, the trajectory acquisition unit is configured to perform at least one of the following four methods:

[0069] Method 1: Based on the predicted driving path information, the vehicle's turning radius R and lateral displacement y in the vehicle's coordinate system are used. h and longitudinal displacement x h The equation of the vehicle's trajectory under the stated turning behavior is obtained as follows:

[0070] (x h -e) 2 +(y h -f) 2 =R 2 ,

[0071] Where e and f are the parameters of the vehicle's trajectory equation under the turning behavior, and the vehicle's trajectory equation under the turning behavior represents the simulated driving trajectory of the vehicle under the turning behavior.

[0072] Based on the predicted driving path information, the vehicle's turning endpoint coordinates (y) end1 x end1 The vehicle's turning starting point coordinates (0, 0), the vehicle's turning radius R, and the vehicle's trajectory equation under the turning behavior are obtained through the formula:

[0073]

[0074] Calculate e and f.

[0075] Method 2: Based on the lateral displacement y of the vehicle in the vehicle coordinate system. hand the longitudinal displacement x h The equation of the vehicle's trajectory under the non-turning behavior is obtained as follows:

[0076] y h =gx h

[0077] Wherein, g is the parameter of the vehicle's trajectory equation under the non-turning behavior, and the vehicle's trajectory equation under the non-turning behavior represents the simulated driving trajectory of the vehicle under the non-turning behavior.

[0078] Based on the predicted driving path information, the coordinates (y) of the vehicle's non-turning endpoint are... end2 x end2 The starting coordinates of the vehicle without turning (0, 0) and the trajectory equation of the vehicle under the described non-turning behavior are obtained through the formula:

[0079] y end2 =gx end2 ,

[0080] Calculate g.

[0081] Method 3: Based on the lateral displacement y of the target vehicle in the vehicle coordinate system from the relative vehicle position information. t and the longitudinal displacement x t The trajectory equation of the target vehicle under the aforementioned opposing behavior is obtained as follows:

[0082] y t =a i x t +b i ,

[0083] Among them, a i and b i The parameter is the target vehicle's trajectory equation for the oncoming target vehicle, where i is the serial number of the target vehicle identified by this vehicle. The target vehicle's trajectory equation under the oncoming behavior represents the simulated driving trajectory of the target vehicle under the oncoming behavior.

[0084] Method 4: Based on the lateral displacement y of the target vehicle in the vehicle coordinate system from the relative vehicle position information. t and the longitudinal displacement x t Obtain the trajectory equation of the target vehicle under the traversing behavior:

[0085] x t =c i y t +d i ,

[0086] Among them, c i and d iThe parameters are the target vehicle's trajectory equations when it traverses the target vehicle. The target vehicle's trajectory equations under the traversing behavior characterize the simulated trajectory of the target vehicle under the traversing behavior.

[0087] Optionally, the data acquisition unit is configured to: acquire the relative position information of the target vehicle.

[0088] The acquisition of the relative position information of the target vehicle specifically includes:

[0089] Within a preset time interval, the relative vehicle position information of multiple target vehicles is continuously recorded, resulting in multiple position information groups, each corresponding one-to-one with a target vehicle. Furthermore, each relative vehicle position information included in each position information group represents the relative vehicle position information of the target vehicle corresponding to that position information group.

[0090] For each group of location information: The relative vehicle positions at all times other than the latest time are calculated according to the formula:

[0091]

[0092] Transform to the vehicle's coordinate system where the latest vehicle relative position information is located, wherein the recording time of the latest vehicle relative position information is later than the recording time of the vehicle relative position information at other times, p i The relative position information of the i-th target vehicle at other times. The relative position information of the i-th target vehicle at other times is transformed to the relative position information of the vehicle in the coordinate system of the latest time, where Δθ is the yaw angle of the vehicle from the other times to the latest time, and Δx is the yaw angle of the vehicle. h Δy represents the longitudinal displacement of the vehicle in the vehicle's coordinate system per unit time. h The longitudinal displacement of the vehicle in the vehicle's coordinate system within the unit time interval is given.

[0093] Optionally, the trajectory acquisition unit further includes: a parameter calculation subunit.

[0094] Used to obtain a first number of relative position information of the target vehicle under the opposing behavior.

[0095] Substitute the first number of relative position information of the vehicles into the target vehicle's trajectory equation under the opposing behavior, and then, based on the least squares principle and the residual principle, simultaneously solve the target vehicle's trajectory equation under the opposing behavior after substitution:

[0096]

[0097] Calculate the a i and the b i Where j is the j-th historical time point of the target vehicle, k is the total number of historical time points, and w j Let x be the weight of the target vehicle at the j-th historical time point. ij In the second number of vehicle relative position information, the longitudinal displacement of the i-th target vehicle in the vehicle coordinate system at the j-th historical time point is y. ij The second number of vehicle relative position information is the lateral displacement of the i-th target vehicle in the vehicle coordinate system at the j-th historical time point.

[0098] And / or, to obtain a second number of vehicle relative position information of the target vehicle under the crossing behavior.

[0099] Substitute the second number of vehicle relative position information into the target vehicle trajectory equation under the traversing behavior, and according to the least squares principle and the residual principle, simultaneously solve the target vehicle trajectory equation under the traversing behavior after substitution:

[0100]

[0101] Calculate the c i and the d i .

[0102] Optionally, the parameter calculation unit is configured as follows:

[0103] This is used to simultaneously establish the trajectory equations of the vehicle under the turning behavior and the trajectory equations of the target vehicle under the target driving behavior, and solve for them to obtain the coordinates of the first collision point between the vehicle and the target vehicle under the turning behavior. Based on the coordinates of the first collision point (y... p1 x p1 ), through the formula:

[0104]

[0105] Calculate the yaw angle θ of the vehicle as it moves in an arc from the first point of impact.

[0106] Based on the vehicle's turning radius R and the vehicle speed in the vehicle motion data, using the formula:

[0107]

[0108] Calculate the angular velocity ω of the vehicle as it performs the circular motion described above.

[0109] Based on the yaw angle θ and the angular velocity ω, using the formula:

[0110]

[0111] Calculate the first collision time parameter (TTC) between the vehicle and the target vehicle under the stated turning behavior. t .

[0112] And / or, used to simultaneously establish the vehicle's trajectory equation under the no-turn behavior and the target vehicle's trajectory equation under the target behavior, and solve for it to obtain the coordinates of the second collision point between the vehicle and the target vehicle under the no-turn behavior. Based on the second collision point coordinates (y... p2 x p2 ), through the formula:

[0113]

[0114] Calculate the second collision time parameter (TTC) between the vehicle and the target vehicle under the non-turning behavior. nt , where v0(x p2 Let a(x) be the longitudinal velocity of the vehicle in the aforementioned vehicle coordinate system. p2 Let x be the longitudinal acceleration of the vehicle in the aforementioned vehicle coordinate system. p2 In the vehicle's coordinate system, this is the longitudinal distance between the vehicle and the second collision point.

[0115] Optionally, the validity verification unit is configured as follows:

[0116] This is used to determine the potential collision zone between the vehicle and the target vehicle based on the vehicle's geometric parameters and the collision time parameters under the target driving behavior, and to determine whether the collision time parameters are within the time period between entering the collision zone and leaving the collision zone. If so, the collision is determined to be valid. The moment of entering the collision zone is the moment when the front of the target vehicle reaches the first collision point or the second collision point, and the moment of leaving the collision zone is the moment when the rear of the target vehicle leaves the first collision point or the second collision point.

[0117] Optionally, the system further includes: a strategy invocation unit, configured to invoke a driving assistance strategy stored in a preset database corresponding to the risk coefficient according to the risk coefficient, and send the driving assistance strategy to a preset driving control algorithm so that the preset driving control algorithm makes a decision based on the driving assistance strategy.

[0118] A device for assessing vehicle collision risk includes:

[0119] processor;

[0120] Memory used to store executable instructions of the processor.

[0121] The processor is configured to execute the instructions to implement the vehicle collision risk assessment method as described in any of the preceding claims.

[0122] A computer-readable storage medium, when the instructions in the computer-readable storage medium are executed by the processor of the device, enables the device to perform the vehicle collision risk assessment method as described in any of the preceding claims.

[0123] This invention provides a vehicle collision risk assessment method, system, device, and storage medium. It utilizes existing vehicle data to predict and simulate vehicle trajectories under different target driving behaviors. Based on these trajectories, it performs hypothetical analyses of potential future collision risks. Compared to existing technologies that calculate collision occurrences by implementing relative motion relationships, this invention offers greater reliability in assessing potential vehicle collision risks while maintaining broad applicability. When validating potential collisions, this invention introduces collision time parameters and potential collision areas for comprehensive evaluation, further enhancing the reliability of collision risk assessment. Furthermore, by transferring the target vehicle's driving information to the vehicle's current coordinate system and considering historical time point weights when calculating the vehicle trajectory equation, this invention reduces reliance on external data, further ensuring the accuracy of the data used for collision validity verification. Therefore, this invention significantly improves the reliability of vehicle collision risk assessment.

[0124] Of course, any product or method implementing the present invention does not necessarily need to achieve all of the advantages described above at the same time. Attached Figure Description

[0125] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0126] Figure 1 A flowchart of a vehicle collision risk assessment method provided in an embodiment of the present invention;

[0127] Figure 2 A schematic diagram of a target driving behavior provided as an optional embodiment of the present invention;

[0128] Figure 3 A block diagram of a vehicle collision risk assessment system provided in an embodiment of the present invention;

[0129] Figure 4 A block diagram of a vehicle collision risk assessment device provided in an embodiment of the present invention. Detailed Implementation

[0130] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0131] This invention provides a method for assessing vehicle collision risk, such as... Figure 1 As shown, the method includes:

[0132] S101. Obtain vehicle geometric parameters, vehicle relative position information, vehicle motion data, and predicted driving path information.

[0133] Optionally, the aforementioned vehicle geometric parameters include, but are not limited to: the length and width of the current vehicle, the length and width of the target vehicle. The aforementioned relative vehicle position information includes, but is not limited to: the lateral and longitudinal displacements of the target vehicle in the current vehicle coordinate system. The aforementioned vehicle motion data includes, but is not limited to: the speed and acceleration of the current vehicle, the speed and acceleration of the target vehicle. The aforementioned predicted driving path information includes, but is not limited to: the turning radius of the current vehicle, the starting point of the current vehicle's turn, the ending point of the current vehicle's turn, the starting point of the current vehicle's non-turning turn, and the ending point of the current vehicle's non-turning turn. The aforementioned data can be collected through the vehicle's bus signals, in-vehicle navigation equipment, and in-vehicle sensors. The data used in this invention can be obtained through existing equipment and technologies, thus improving the universality of this invention.

[0134] Optionally, in one alternative embodiment of the present invention, the aforementioned vehicle coordinate system can be a coordinate system with the vehicle as the origin, the longitudinal axis of the vehicle as the vertical axis, and the lateral axis of the vehicle as the horizontal axis. By transforming the position of the target vehicle at different historical moments during its movement to the same coordinate system, the accuracy of the calculation results is ensured while improving processing efficiency.

[0135] S102. Based on the relative position information of the vehicles and the predicted driving path information, obtain the driving trajectory equation of the current vehicle and the driving trajectory equation of the target vehicle under the target driving behavior, wherein the target driving behavior includes at least one of turning behavior, non-turning behavior, oncoming behavior and traversing behavior.

[0136] Optionally, in one embodiment of the present invention, both the vehicle's trajectory equation and the target vehicle's trajectory equation represent the predicted trajectory of the vehicle. The present invention solves for the intersection point of the vehicle's trajectory equations under different target driving behaviors by simultaneously solving the equations. This intersection point is the predicted collision point between the vehicle and the target vehicle. Based on this predicted collision point, the potential collision between the vehicle and the target vehicle is predicted. This allows the present invention, compared to existing real-time monitoring methods, to predict and assess the risk of vehicle turning behavior over a future period, thereby improving the reliability of risk assessment for vehicle turning behavior decisions.

[0137] Optionally, in another alternative embodiment of the invention, such as Figure 2 As shown, the objects of the aforementioned turning and non-turning behaviors are this vehicle 201, while the objects of the aforementioned oncoming and crossing behaviors are the target vehicle 202. The aforementioned non-turning behavior is the target driving behavior of this vehicle going straight.

[0138] S103. Based on the predicted driving path information, vehicle motion data, the driving trajectory equation of this vehicle and the driving trajectory equation of the target vehicle, calculate the collision time parameters under the target driving behavior.

[0139] Optionally, the aforementioned Time to Collision (TTC) parameter reflects the risk of a collision between the vehicle and the target vehicle over time; as the time decreases, the collision risk increases. By introducing the collision time parameter, the collision risk is digitized, ensuring the accuracy of the data used for vehicle collision risk assessment while improving computational efficiency and further shortening reaction time.

[0140] S104. Based on the vehicle's geometric parameters and the collision time parameters under the target driving behavior, perform collision validity verification.

[0141] Optionally, in one embodiment of the present invention, the above-mentioned collision validity verification may involve calculating the potential collision area between the vehicle and the target vehicle based on the vehicle's geometric parameters, and determining whether the collision time parameter falls within the time period between the target vehicle entering and leaving the potential collision area. When the collision time parameter falls within the time period of the potential collision area, it is determined that there is a risk of collision, i.e., the collision is valid. The potential collision area includes, but is not limited to: the predicted collision point being located in the front area of ​​the target vehicle and / or the predicted collision point being located in the rear area of ​​the target vehicle.

[0142] This invention verifies the validity of potential collisions based on collision time parameters of various target driving behaviors. By adding data types for assessing collision risk, the accuracy of the invention's prediction results for possible collisions is improved.

[0143] S105. Normalize the collision time parameters that have passed the collision validity verification into risk coefficients, and output the turning decision results corresponding to the risk coefficients.

[0144] Optionally, in one alternative embodiment of the present invention, the above-mentioned normalization of the collision time parameter after collision validity verification into a risk coefficient refers to converting the collision time parameter after collision validity verification into a visualized risk coefficient, using time as a reference factor, and using the risk coefficient to measure the risk level of a possible collision. By normalizing the collision time parameter into a risk coefficient, the present invention allows the prediction results of possible future collisions to be intuitively displayed to the driver, facilitating the driver's risk assessment. Simultaneously, the output risk coefficient can be directly applied to other driving assistance algorithms, shortening the overall reaction time and improving the reliability and efficiency of invoking vehicle turning decisions.

[0145] This invention utilizes existing vehicle data to predict and simulate vehicle trajectories under different target driving behaviors. Based on these trajectories, it performs hypothetical analyses of potential future collision risks. Compared to existing technologies that calculate collision occurrences by implementing relative motion relationships, this invention offers greater applicability and improved reliability in assessing potential vehicle collision risks. When validating potential collisions, this invention incorporates collision time parameters and potential collision areas for comprehensive evaluation, further enhancing the reliability of collision risk assessment. Furthermore, by transferring the target vehicle's driving information to the current vehicle's coordinate system and considering historical time point weights when calculating the vehicle's trajectory equation, this invention reduces reliance on external data, further ensuring the accuracy of the data used for collision validity verification. Therefore, this invention significantly improves the reliability of risk assessment for vehicle collisions.

[0146] Optionally, based on the vehicle's relative position information and the predicted driving path information, the trajectory equations of the current vehicle and the target vehicle under the target driving behavior are obtained, including at least one of the following four methods:

[0147] Method 1: Based on the vehicle's turning radius R and lateral displacement y in the vehicle's coordinate system from the predicted driving path information. h and longitudinal displacement x h Obtain the equation of the vehicle's trajectory under turning behavior:

[0148] (x h -e) 2 +(y h -f) 2 =R 2 ,

[0149] Where e and f are the parameters of the vehicle's trajectory equation under turning behavior, and the vehicle's trajectory equation under turning behavior represents the simulated driving trajectory of the vehicle under turning behavior;

[0150] Based on the predicted driving path information, the vehicle's turning endpoint coordinates (y end1 x end1 The vehicle's turning starting coordinates (0, 0), turning radius R, and trajectory equation under turning behavior are given by the formula:

[0151]

[0152] Calculate e and f.

[0153] Method 2: Based on the lateral displacement y of the vehicle in its own coordinate system h and longitudinal displacement x h The equation of the vehicle's trajectory without turning is obtained as follows:

[0154] y h =gx h

[0155] Where g is the parameter of the vehicle's trajectory equation under non-turning behavior, and the vehicle's trajectory equation under non-turning behavior represents the simulated driving trajectory of the vehicle under non-turning behavior;

[0156] Based on the predicted driving path information, the coordinates of the vehicle's non-turning endpoint (y) are... end2 x end2 The starting coordinates of the vehicle without turning (0, 0) and the equation of the vehicle's trajectory under non-turning behavior are obtained through the formula:

[0157] y end2 =gx end2 ,

[0158] Calculate g.

[0159] Method 3: Based on the lateral displacement y of the target vehicle in its own coordinate system from the relative vehicle position information. t and longitudinal displacement x t The trajectory equation of the target vehicle under opposing behavior is obtained as follows:

[0160] y t =a i x t +b i ,

[0161] Among them, a i and b iThe parameters of the target vehicle's trajectory equation under the oncoming target vehicle are: i is the sequence number of the target vehicle when it is identified by this vehicle. The target vehicle's trajectory equation under the oncoming behavior represents the simulated trajectory of the target vehicle under the oncoming behavior.

[0162] Method 4: Based on the lateral displacement y of the target vehicle in its own coordinate system from the vehicle relative position information. t and longitudinal displacement x t Obtain the trajectory equation of the target vehicle under the traversing behavior:

[0163] x t =c i y t +d i ,

[0164] Among them, c i and d i The parameters are the target vehicle's trajectory equations for crossing the target vehicle. The target vehicle's trajectory equations under the crossing behavior characterize the simulated driving trajectory of the target vehicle under the crossing behavior.

[0165] Optionally, obtaining vehicle relative position information includes: obtaining the target vehicle's relative position information.

[0166] Obtain the relative position information of the target vehicle, specifically including:

[0167] Within a preset time interval, the relative position information of multiple target vehicles is continuously recorded, and multiple position information groups are obtained. Each position information group corresponds to a target vehicle. Furthermore, the relative position information of each vehicle included in each position information group is the relative position information of the target vehicle corresponding to that position information group.

[0168] For each location information group: Take the vehicle relative position information from this location information group, excluding the latest time's relative position information, and calculate the vehicle relative position information from other times using the formula:

[0169]

[0170] Transform to the vehicle's coordinate system at the latest relative position information, where the recording time of the latest relative position information is later than the recording time of other relative position information, p i For the relative position information of the i-th target vehicle at other times, This transforms the relative position information of the i-th target vehicle at other times into the relative position information of the target vehicle in the latest vehicle coordinate system, where Δθ is the yaw angle of the target vehicle from other times to the latest time, and Δx is the yaw angle of the target vehicle. h Δy represents the longitudinal displacement of the vehicle in its own coordinate system per unit time. hIt represents the longitudinal displacement of the vehicle in its own coordinate system within a unit of time.

[0171] Optionally, in one alternative embodiment of the present invention, the number of vehicle relative position information entries for each target vehicle recorded can be limited. For example, starting from the latest time, 10 vehicle relative position information entries for the target vehicle are recorded in reverse chronological order. Starting from the 11th vehicle relative position information entry, previously recorded vehicle relative position information entries are deleted, thereby reducing storage resources occupied by useless information. Alternatively, a corresponding number can be set for the multiple target vehicles recorded according to their target driving behavior. For example, 10 entries are collected for oncoming target vehicles, and 10 entries are collected for traversing target vehicles. When a target vehicle leaves the detection range of the present invention, the vehicle relative position information corresponding to that target vehicle is deleted. The present invention does not impose excessive restrictions on the specific numerical value of the above-mentioned number.

[0172] Optional, a i b i c i and d i The methods of obtaining it include:

[0173] Obtain the first number of relative positions of the target vehicle under oncoming traffic.

[0174] Substitute the relative position information of the first number of vehicles into the target vehicle's trajectory equation under the oncoming behavior, and then, based on the least squares principle and the residual principle, simultaneously solve the target vehicle's trajectory equation under the oncoming behavior after substitution:

[0175]

[0176] Calculate a i and b i Where j is the j-th historical time point of the target vehicle, k is the total number of historical time points, and w j Let x be the weight of the target car at the j-th historical moment. ij In the second set of relative vehicle position information, y represents the longitudinal displacement of the i-th target vehicle in its own coordinate system at the j-th historical time point. ij In the second number of relative position information of vehicles, the lateral displacement of the i-th target vehicle in its own coordinate system at the j-th historical time point.

[0177] And / or, obtain the second number of vehicle relative position information of the target vehicle under the traversing behavior.

[0178] Substitute the relative position information of the second number of vehicles into the trajectory equation of the target vehicle under the traversing behavior, and then, based on the least squares principle and the residual principle, simultaneously solve the trajectory equation of the target vehicle under the traversing behavior after substitution:

[0179]

[0180] Calculate c i and d i .

[0181] Optionally, based on the predicted driving path information, the vehicle's trajectory equation, and the target vehicle's trajectory equation, collision time parameters under the target driving behavior are calculated, including:

[0182] By simultaneously solving the equations of the vehicle's trajectory under turning behavior and the target vehicle's trajectory under target behavior, the coordinates of the first collision point between the vehicle and the target vehicle under turning behavior are obtained. Based on the coordinates of the first collision point (y... p1 x p1 ), through the formula:

[0183]

[0184] Calculate the yaw angle θ of the vehicle as it moves in a circular arc from the first point of impact.

[0185] Based on the vehicle's turning radius R and the vehicle speed from the vehicle motion data, using the formula:

[0186]

[0187] Calculate the angular velocity ω of the vehicle as it moves in a circular arc.

[0188] Based on the yaw angle θ and angular velocity ω, using the formula:

[0189]

[0190] Calculate the first collision time (TTC) parameter between the vehicle and the target vehicle during cornering. t .

[0191] And / or, simultaneously solve the trajectory equations of the vehicle under the non-turning behavior and the trajectory equations of the target vehicle under the target driving behavior to obtain the coordinates of the second collision point between the vehicle and the target vehicle under the non-turning behavior.

[0192] Based on the collision point coordinates (y) of this vehicle and the target vehicle under non-turning behavior. p2 x p2 ), through the formula:

[0193]

[0194] Calculate the second collision time parameter (TTC) between the vehicle and the target vehicle under non-turning behavior. nt , where v0(x p2 Let a(x) be the longitudinal velocity of the vehicle in the vehicle's coordinate system.p2 Let x be the longitudinal acceleration of the vehicle in the vehicle's coordinate system. p2 This represents the longitudinal distance between the vehicle and the second point of collision in the vehicle's coordinate system.

[0195] Optionally, collision validity verification can be performed based on vehicle geometry parameters and collision time parameters under the target driving behavior, including:

[0196] Based on the vehicle's geometric parameters and the collision time parameters under the target driving behavior, the potential collision zone between the vehicle and the target vehicle is determined. It is then determined whether the collision time parameters fall within the time period between entering and leaving the collision zone. If so, the collision is deemed valid. The moment of entering the collision zone is the moment when the front of the target vehicle reaches the first or second collision point, and the moment of leaving the collision zone is the moment when the rear of the target vehicle leaves the first or second collision point.

[0197] Optionally, in one alternative embodiment of the present invention, the time t when the target vehicle i enters the collision area is... i (entry) can be obtained through the formula:

[0198]

[0199] Solve for m p (target) is the ordinate x of the first collision point mentioned above. p1 Or the ordinate x of the second collision point mentioned above p2 v i (m) represents the longitudinal velocity component of the target vehicle i, a i (m) represents the longitudinal acceleration of the target vehicle i, and d represents the width of the vehicle.

[0200] Optionally, in another alternative embodiment of the present invention, the time t when the target vehicle i leaves the collision area is... i (leave) can be achieved through the formula:

[0201]

[0202] Solve for L i Let i be the length of the target vehicle i.

[0203] Optionally, in another alternative embodiment of the present invention, under the above-described target driving behavior, the collision time parameter TTC between the vehicle and the target vehicle i is... i Collision validity can be verified using the arbitration formula:

[0204]

[0205] Verification is performed. Here, t(buff) is a preset time interval, and this preset value is the collision time parameter TTC when the conditions in the second line of the arbitration formula are met. i The value of represents the value that indicates that the collision validity check will not fail, that is, the vehicle will not collide with the target vehicle i.

[0206] Optionally, in another optional embodiment of the present invention, the risk coefficient O between the vehicle and the target vehicle i under different target driving behaviors is... i It can be done through the formula:

[0207]

[0208] Processing is performed. Specifically, when the collision time parameter TTC has passed validity verification... i A value greater than 0 and less than 1 indicates that a collision between this vehicle and target vehicle i is likely or the risk of collision is extremely high. When TTC... i When greater than or equal to 1 and less than 10, according to the formula Calculate the collision risk level when TTC i When the value is greater than or equal to 10, it is determined that the vehicle and the target vehicle i will not collide or the risk of collision is extremely low.

[0209] It should be noted that the above arbitration formula can be used to verify the collision validity under different target driving behaviors. The calculation parameters such as formulas and coordinates involved can be calculated using the formulas provided in the above method embodiments of the present invention. The present invention will not elaborate further here.

[0210] Optionally, the above methods also include:

[0211] Based on the risk coefficient, the system calls the driving assistance strategy corresponding to the risk coefficient stored in the preset database and sends the driving assistance strategy to the preset driving control algorithm so that the preset driving control algorithm can make a decision based on the driving assistance strategy.

[0212] This invention utilizes existing vehicle data to predict and simulate vehicle trajectories under different target driving behaviors. Based on these trajectories, it performs hypothetical analyses of potential future collision risks. Compared to existing technologies that calculate collision occurrences by implementing relative motion relationships, this invention offers greater applicability and improved reliability in assessing potential vehicle collision risks. When validating potential collisions, this invention incorporates collision time parameters and potential collision areas for comprehensive evaluation, further enhancing the reliability of collision risk assessment. Furthermore, by transferring the target vehicle's driving information to the current vehicle's coordinate system and considering historical time point weights when calculating the vehicle's trajectory equation, this invention reduces reliance on external data, further ensuring the accuracy of the data used for collision validity verification. Therefore, this invention significantly improves the reliability of risk assessment for vehicle collisions.

[0213] Corresponding to the above-described vehicle collision risk assessment method embodiments, the present invention also provides a vehicle collision risk assessment system, such as... Figure 3 As shown, the system includes:

[0214] The data acquisition unit 301 is used to acquire vehicle geometric parameters, vehicle relative position information, vehicle motion data, and predicted driving path information.

[0215] The trajectory acquisition unit 302 is used to obtain the trajectory equation of the vehicle and the trajectory equation of the target vehicle under the target driving behavior based on the relative position information of the vehicle and the predicted driving path information. The target driving behavior includes at least one of turning behavior, non-turning behavior, oncoming behavior and traversing behavior.

[0216] The parameter calculation unit 303 is used to calculate the collision time parameters under the target driving behavior based on the predicted driving path information, vehicle motion data, the driving trajectory equation of the vehicle itself and the driving trajectory equation of the target vehicle. The collision time parameters are negatively correlated with the collision risk.

[0217] The validity verification unit 304 is used to perform collision validity verification based on vehicle geometric parameters and collision time parameters under target driving behavior.

[0218] The coefficient generation unit 305 is used to normalize the collision time parameters that have been verified for collision validity into risk coefficients and output the turning decision results corresponding to the risk coefficients.

[0219] Optionally, the trajectory acquisition unit 302 is configured to perform at least one of the following four methods:

[0220] Method 1: Based on the vehicle's turning radius R and lateral displacement y in the vehicle's coordinate system from the predicted driving path information. h and longitudinal displacement x h Obtain the equation of the vehicle's trajectory under turning behavior:

[0221] (x h -e) 2 +(y h -f) 2 =R 2 ,

[0222] Where e and f are the parameters of the vehicle's trajectory equation under turning behavior, and the vehicle's trajectory equation under turning behavior represents the simulated driving trajectory of the vehicle under turning behavior.

[0223] Based on the predicted driving path information, the vehicle's turning endpoint coordinates (y end1 x end1 The vehicle's turning starting coordinates (0, 0), turning radius R, and trajectory equation under turning behavior are given by the formula:

[0224]

[0225] Calculate e and f.

[0226] Method 2: Based on the lateral displacement y of the vehicle in its own coordinate system h and longitudinal displacement x h The equation of the vehicle's trajectory without turning is obtained as follows:

[0227] y h =gx h

[0228] Where g is the parameter of the vehicle's trajectory equation under non-turning behavior, and the vehicle's trajectory equation under non-turning behavior represents the simulated driving trajectory of the vehicle under non-turning behavior.

[0229] Based on the predicted driving path information, the coordinates of the vehicle's non-turning endpoint (y) are... end2 x end2 The starting coordinates of the vehicle without turning (0, 0) and the equation of the vehicle's trajectory under non-turning behavior are obtained through the formula:

[0230] y end2 =gx end2 ,

[0231] Calculate g.

[0232] Method 3: Based on the lateral displacement y of the target vehicle in its own coordinate system from the relative vehicle position information. t and longitudinal displacement x tThe trajectory equation of the target vehicle under opposing behavior is obtained as follows:

[0233] y t =a i x t +b i ,

[0234] Among them, a i and b i The parameters of the target vehicle's trajectory equation under the oncoming target vehicle are: i is the sequence number of the target vehicle when it is identified by this vehicle. The target vehicle's trajectory equation under the oncoming behavior represents the simulated trajectory of the target vehicle under the oncoming behavior.

[0235] Method 4: Based on the lateral displacement y of the target vehicle in its own coordinate system from the vehicle relative position information. t and longitudinal displacement x t Obtain the trajectory equation of the target vehicle under the traversing behavior:

[0236] x t =c i y t +d i ,

[0237] Among them, c i and d i The parameters are the target vehicle's trajectory equations for crossing the target vehicle. The target vehicle's trajectory equations under the crossing behavior characterize the simulated driving trajectory of the target vehicle under the crossing behavior.

[0238] Optionally, the data acquisition unit 301 is configured to acquire the relative position information of the target vehicle.

[0239] Obtain the relative position information of the target vehicle, specifically including:

[0240] Within a preset time interval, the relative position information of multiple target vehicles is continuously recorded, resulting in multiple position information groups, each corresponding to a target vehicle. Furthermore, the relative position information of each vehicle included in each position information group is the relative position information of the target vehicle corresponding to that position information group.

[0241] For each location information group: Take the vehicle relative position information from this location information group, excluding the latest time's relative position information, and calculate the vehicle relative position information from other times using the formula:

[0242]

[0243] Transform to the vehicle's coordinate system at the latest relative position information, where the recording time of the latest relative position information is later than the recording time of other relative position information, p iFor the relative position information of the i-th target vehicle at other times, This transforms the relative position information of the i-th target vehicle at other times into the relative position information of the target vehicle in the latest vehicle coordinate system, where Δθ is the yaw angle of the target vehicle from other times to the latest time, and Δx is the yaw angle of the target vehicle. h Δy represents the longitudinal displacement of the vehicle in its own coordinate system per unit time. h It represents the longitudinal displacement of the vehicle in its own coordinate system within a unit of time.

[0244] Optionally, the trajectory acquisition unit 302 further includes: a parameter calculation subunit.

[0245] Used to obtain the first number of vehicle relative position information of the target vehicle under oncoming behavior.

[0246] Substitute the relative position information of the first number of vehicles into the target vehicle's trajectory equation under the oncoming behavior, and then, based on the least squares principle and the residual principle, simultaneously solve the target vehicle's trajectory equation under the oncoming behavior after substitution:

[0247]

[0248] Calculate a i and b i Where j is the j-th historical time point of the target vehicle, k is the total number of historical time points, and w j Let x be the weight of the target car at the j-th historical moment. ij In the second set of relative vehicle position information, the longitudinal displacement of the i-th target vehicle in its own coordinate system at the j-th historical time point is y. ij In the second number of relative position information of vehicles, the lateral displacement of the i-th target vehicle in its own coordinate system at the j-th historical time point.

[0249] And / or, used to obtain a second number of vehicle relative position information of the target vehicle under the traversing behavior.

[0250] Substitute the relative position information of the second number of vehicles into the trajectory equation of the target vehicle under the traversing behavior, and then, based on the least squares principle and the residual principle, simultaneously solve the trajectory equation of the target vehicle under the traversing behavior after substitution:

[0251]

[0252] Calculate c i and d i .

[0253] Optionally, the parameter calculation unit 303 is configured as follows:

[0254] This is used to simultaneously establish the trajectory equations of the current vehicle under turning behavior and the trajectory equation of the target vehicle under target behavior, and then solve for the coordinates of the first collision point between the current vehicle and the target vehicle under turning behavior. Based on the coordinates of the first collision point (y... p1 x p1 ), through the formula:

[0255]

[0256] Calculate the yaw angle θ of the vehicle as it moves in a circular arc from the first point of impact.

[0257] Based on the vehicle's turning radius R and the vehicle speed from the vehicle motion data, using the formula:

[0258]

[0259] Calculate the angular velocity ω of the vehicle as it moves in a circular arc.

[0260] Based on the yaw angle θ and angular velocity ω, using the formula:

[0261]

[0262] Calculate the first collision time (TTC) parameter between the vehicle and the target vehicle during cornering. t .

[0263] And / or, used to simultaneously establish the trajectory equations of the vehicle under non-turning behavior and the trajectory equations of the target vehicle under target behavior, and solve for them to obtain the coordinates of the second collision point between the vehicle and the target vehicle under non-turning behavior. Based on the coordinates of the second collision point (y... p2 x p2 ), through the formula:

[0264]

[0265] Calculate the second collision time parameter (TTC) between the vehicle and the target vehicle under non-turning behavior. nt , where v0(x p2 Let a(x) be the longitudinal velocity of the vehicle in the vehicle's coordinate system. p2 Let x be the longitudinal acceleration of the vehicle in the vehicle's coordinate system. p2 This represents the longitudinal distance between the vehicle and the second point of collision in the vehicle's coordinate system.

[0266] Optionally, the validity verification unit 304 is configured as follows:

[0267] This is used to determine the potential collision zone between the vehicle and the target vehicle based on the vehicle's geometric parameters and the collision time parameters under the target driving behavior. It determines whether the collision time parameters are within the time period between entering and leaving the collision zone. If so, the collision is determined to be valid. The moment of entering the collision zone is the moment when the front of the target vehicle reaches the first or second collision point, and the moment of leaving the collision zone is the moment when the rear of the target vehicle leaves the first or second collision point.

[0268] Optionally, the system also includes: a strategy invocation unit, used to invoke a driving assistance strategy stored in a preset database corresponding to the risk coefficient according to the risk coefficient, and send the driving assistance strategy to a preset driving control algorithm so that the preset driving control algorithm makes a decision based on the driving assistance strategy.

[0269] This invention provides a device for assessing vehicle collision risk, such as... Figure 4 As shown, the device includes:

[0270] Processor 401;

[0271] Memory 402 is used to store executable instructions of processor 401.

[0272] The processor 401 is configured to execute instructions to implement a vehicle collision risk assessment method as described above.

[0273] This invention provides a computer-readable storage medium that, when the instructions in the computer-readable storage medium are executed by the processor 401 of the device, enables the device to perform the vehicle collision risk assessment method as described above.

[0274] Memory may include non-persistent memory in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, like read-only memory (ROM) or flash RAM, and memory includes at least one memory chip. Memory is an example of computer-readable media.

[0275] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0276] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0277] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes the element.

[0278] The various embodiments in this specification are described in a related manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the system embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0279] The above are merely embodiments of this application and are not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

Claims

1. A method for assessing vehicle collision risk, characterized in that, include: Acquire vehicle geometric parameters, vehicle relative position information, vehicle motion data, and predicted driving path information; Based on the vehicle relative position information and the predicted driving path information, the trajectory equations of the current vehicle and the target vehicle under the target driving behavior are obtained. The target driving behavior includes at least one of turning behavior, non-turning behavior, oncoming behavior, and traversing behavior. The method of obtaining the trajectory equations of the current vehicle and the target vehicle under the target driving behavior includes at least one of the following four methods: Method 1: Obtain the trajectory equation of the current vehicle under the turning behavior based on the turning radius, lateral displacement, and longitudinal displacement of the current vehicle in the current vehicle coordinate system from the predicted driving path information; Method 2: Obtain the trajectory equation of the current vehicle under the non-turning behavior based on the lateral displacement and longitudinal displacement of the current vehicle in the current vehicle coordinate system; Method 3: Obtain the trajectory equation of the target vehicle under the oncoming behavior based on the lateral displacement and longitudinal displacement of the target vehicle in the current vehicle coordinate system from the vehicle relative position information; Method 4: Obtain the trajectory equation of the target vehicle under the traversing behavior based on the lateral displacement and longitudinal displacement of the target vehicle in the current vehicle coordinate system from the vehicle relative position information. Based on the predicted driving path information, the vehicle motion data, the vehicle's own driving trajectory equation, and the target vehicle's driving trajectory equation, the collision time parameters under the target driving behavior are calculated. Based on the vehicle's geometric parameters and the collision time parameters under the target driving behavior, a collision validity check is performed to determine the potential collision area between the vehicle and the target vehicle. It is then determined whether the collision time parameters fall within the time period between entering and leaving the collision area. If so, the collision is deemed valid. Here, the first collision point is the collision point between the vehicle and the target vehicle under the turning behavior, and the second collision point is the collision point between the vehicle and the target vehicle under the non-turning behavior. The moment of entering the collision area is the moment when the front of the target vehicle reaches the first collision point or the second collision point, and the moment of leaving the collision area is the moment when the rear of the target vehicle leaves the first collision point or the second collision point. The collision time parameter, which has passed the collision validity verification, is normalized into a risk coefficient, and the turning decision result corresponding to the risk coefficient is output. The predicted driving path information includes the pre-planned turning radius of the vehicle, the starting point of the vehicle's turn, the ending point of the vehicle's turn, the starting point of the vehicle's non-turn, and the ending point of the vehicle's non-turn.

2. The method according to claim 1, characterized in that, The step of obtaining the vehicle's trajectory equation and the target vehicle's trajectory equation under the target driving behavior based on the vehicle's relative position information and the predicted driving path information includes at least one of the following four methods: Method 1: Based on the vehicle's turning radius in the predicted driving path information. Lateral displacement of the vehicle in its own coordinate system and longitudinal displacement The equation of the vehicle's trajectory under the stated turning behavior is obtained as follows: ， in, These are the parameters of the vehicle's trajectory equation under the turning behavior, and the vehicle's trajectory equation under the turning behavior represents the simulated trajectory of the vehicle under the turning behavior. Based on the predicted driving path information, the coordinates of the vehicle's turning endpoint ( , The starting coordinates of the vehicle's turn (0, 0), and the turning radius of the vehicle. The equation of the vehicle's trajectory under the described turning behavior is derived from the formula: Calculate the ; Method 2: Based on the lateral displacement of the vehicle in the vehicle's coordinate system. and longitudinal displacement The equation of the vehicle's trajectory under the non-turning behavior is obtained as follows: in, These are the parameters of the vehicle's trajectory equation under the non-turning behavior, and the vehicle's trajectory equation under the non-turning behavior represents the simulated trajectory of the vehicle under the non-turning behavior. Based on the predicted driving path information, the coordinates of the vehicle's non-turning endpoint ( The starting coordinates of the vehicle without turning (0, 0) and the equation of the vehicle's trajectory under the described non-turning behavior are obtained through the formula: ， Calculate the ; Method 3: Based on the lateral displacement of the target vehicle in the vehicle coordinate system from the relative vehicle position information. and the longitudinal displacement The trajectory equation of the target vehicle under the aforementioned opposing behavior is obtained as follows: ， in, and These are the parameters of the target vehicle's trajectory equation under the aforementioned opposing behavior. The target vehicle is identified by this vehicle as its serial number, and the target vehicle's trajectory equation under the opposing behavior represents the simulated trajectory of the target vehicle under the opposing behavior. Method 4: Based on the lateral displacement of the target vehicle in the vehicle coordinate system from the relative vehicle position information. and the longitudinal displacement Obtain the trajectory equation of the target vehicle under the traversing behavior: ， in, and The parameters are the target vehicle's trajectory equation under the traversing behavior, which represents the simulated trajectory of the target vehicle under the traversing behavior.

3. The method according to claim 2, characterized in that, Obtaining the relative position information of the vehicle includes: Obtain the relative position information of the target vehicle. The acquisition of the relative position information of the target vehicle specifically includes: Within a preset time interval, the relative vehicle position information of multiple target vehicles is continuously recorded, and multiple position information groups are obtained. Each position information group corresponds to a target vehicle. Furthermore, each relative vehicle position information included in each position information group is the relative vehicle position information of the target vehicle corresponding to that position information group. For each group of location information: The relative vehicle positions at all times other than the latest time are calculated according to the formula: ， The coordinates are then transformed to the vehicle's own coordinate system at the latest time, wherein the recording time of the latest vehicle relative position information is later than the recording time of the vehicle relative position information at other times. For the first The relative position information of the target vehicle at other times. For the first The relative position information of the target vehicle at other times is converted to the relative position information of the vehicle in the vehicle's coordinate system at the latest time. This refers to the yaw angle of the vehicle from the other times to the latest time. It represents the longitudinal displacement of the vehicle in the vehicle's coordinate system per unit time. The displacement of the vehicle in the vehicle's coordinate system within the unit time interval is the lateral displacement of the vehicle.

4. The method according to claim 2, characterized in that, The The above The above and stated The methods of obtaining it include: Obtain a first number of relative position information of the target vehicle under the oncoming behavior; Substitute the first number of relative position information of the vehicles into the target vehicle's trajectory equation under the opposing behavior, and then, based on the least squares principle and the residual principle, simultaneously solve the target vehicle's trajectory equation under the opposing behavior after substitution: ， Calculate the and stated ,in, For the target vehicle A historical moment, The total number of historical time points. For the target vehicle The weight of each historical moment In the relative position information of the vehicles, the i-th target vehicle is in the i-th position. The longitudinal displacement in the vehicle coordinate system at each historical moment. In the relative position information of the vehicles, the i-th target vehicle is in the i-th position. The lateral displacement of the vehicle in the coordinate system at each historical moment; And / or, obtain a second number of vehicle relative position information of the target vehicle under the crossing behavior; Substitute the second number of vehicle relative position information into the target vehicle trajectory equation under the traversing behavior, and according to the least squares principle and the residual principle, simultaneously solve the target vehicle trajectory equation under the traversing behavior after substitution: ， Calculate the and stated .

5. The method according to claim 2, characterized in that, The step of calculating the collision time parameters under the target driving behavior based on the predicted driving path information, the driving trajectory equation of the current vehicle, and the driving trajectory equation of the target vehicle includes: By simultaneously establishing the trajectory equations of the vehicle under the turning behavior and the trajectory equations of the target vehicle under the target driving behavior, and solving for them, the coordinates of the first collision point between the vehicle and the target vehicle under the turning behavior are obtained; based on the coordinates of the first collision point ( , ), through the formula: ， Calculate the yaw angle of the vehicle as it moves in an arc from the first point of impact. ; Based on the vehicle's turning radius The vehicle speed in the vehicle motion data is obtained through the formula: ， Calculate the angular velocity of the vehicle during the circular motion described above. ; According to the yaw angle and the angular velocity Through the formula: ， Calculate the first collision time parameters between the vehicle and the target vehicle under the stated turning behavior. ; And / or, simultaneously establish the vehicle's trajectory equation under the no-turn behavior and the target vehicle's trajectory equation under the target behavior, and solve for the coordinates of the second collision point between the vehicle and the target vehicle under the no-turn behavior; based on the coordinates of the second collision point ( , ), through the formula: , Calculate the second collision time parameters between the vehicle and the target vehicle under the non-turning behavior. ,in, Let be the longitudinal speed of the vehicle in the aforementioned vehicle coordinate system. Let X be the longitudinal acceleration of the vehicle in the aforementioned vehicle coordinate system. In the vehicle's coordinate system, this is the longitudinal distance between the vehicle and the second collision point.

6. The method according to claim 1, characterized in that, The method further includes: Based on the risk coefficient, the driving assistance strategy corresponding to the risk coefficient stored in the preset database is invoked, and the driving assistance strategy is sent to the preset driving control algorithm so that the preset driving control algorithm makes a decision based on the driving assistance strategy.

7. A vehicle collision risk assessment system, characterized in that, include: The data acquisition unit is used to acquire vehicle geometric parameters, vehicle relative position information, vehicle motion data, and predicted driving path information; The trajectory acquisition unit is used to obtain the trajectory equations of the current vehicle and the target vehicle under the target driving behavior based on the vehicle relative position information and the predicted driving path information. The target driving behavior includes at least one of turning behavior, non-turning behavior, oncoming behavior, and traversing behavior. The method of obtaining the trajectory equations of the current vehicle and the target vehicle under the target driving behavior includes at least one of the following four methods: Method 1: Obtain the trajectory equation of the current vehicle under the turning behavior based on the turning radius, lateral displacement, and longitudinal displacement of the current vehicle in the current vehicle coordinate system from the predicted driving path information; Method 2: Obtain the trajectory equation of the current vehicle under the non-turning behavior based on the lateral displacement and longitudinal displacement of the current vehicle in the current vehicle coordinate system; Method 3: Obtain the trajectory equation of the target vehicle under the oncoming behavior based on the lateral displacement and longitudinal displacement of the target vehicle in the current vehicle coordinate system from the vehicle relative position information; Method 4: Obtain the trajectory equation of the target vehicle under the traversing behavior based on the lateral displacement and longitudinal displacement of the target vehicle in the current vehicle coordinate system from the vehicle relative position information. The parameter calculation unit is used to calculate the collision time parameters under the target driving behavior based on the predicted driving path information, the vehicle motion data, the driving trajectory equation of the vehicle itself, and the driving trajectory equation of the target vehicle. The validity verification unit is used to perform collision validity verification based on the vehicle geometric parameters and the collision time parameters under the target driving behavior, determine the potential collision area where the vehicle and the target vehicle collide, and determine whether the collision time parameters are within the time period between entering the collision area and leaving the collision area. If so, the collision is determined to be valid. The first collision point is the collision point between the vehicle and the target vehicle under the turning behavior, and the second collision point is the collision point between the vehicle and the target vehicle under the non-turning behavior. The moment of entering the collision area is the moment when the front of the target vehicle reaches the first collision point or the second collision point, and the moment of leaving the collision area is the moment when the rear of the target vehicle leaves the first collision point or the second collision point. The coefficient generation unit is used to normalize the collision time parameter after the collision validity verification into a risk coefficient, and output the turning decision result corresponding to the risk coefficient. The predicted driving path information includes the pre-planned turning radius of the vehicle, the turning start point of the vehicle, the turning end point of the vehicle, the non-turning start point of the vehicle, and the non-turning end point of the vehicle.

8. A device for assessing vehicle collision risk, characterized in that, include: processor; Memory used to store the processor's executable instructions; The processor is configured to execute the instructions to implement the vehicle collision risk assessment method as described in any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that, When the instructions in the computer-readable storage medium are executed by the processor of the vehicle collision risk assessment device, the device is enabled to perform the vehicle collision risk assessment method as described in any one of claims 1 to 6.

Images