A road intersection warning method and device

Abstract

The present invention provides a road intersection warning method and device. First, basic data is acquired, and the basic data includes geometric parameters and motion information of a vehicle, as well as relative position information between the vehicle and a target vehicle as a potential dangerous target at the intersection. Then, the current driving behavior of the vehicle is determined based on the geometric parameters and motion information of the vehicle, and the driving path of the vehicle and the driving path of the target vehicle corresponding to the current driving behavior of the vehicle are determined in combination with the relative position information. Further, based on the driving path of the vehicle and the driving path of the target vehicle, collision parameters between the vehicle and the target vehicle are determined, and the collision parameters include collision time and lateral displacement of the collision point. Finally, a dangerous target vehicle with the shortest collision time is selected from the target vehicles, and an intersection warning is performed according to the collision time and lateral displacement of the collision point corresponding to the dangerous target vehicle.

Description

Technical Field

[0001] The present invention relates to the field of intelligent driving technology, and more specifically, to a road intersection warning method and device. Background Art

[0002] As intelligent driving becomes more sophisticated, early warning of various traffic conditions and emergency situations is a core technology for ensuring vehicle safety. Intersections are a typical example of complex traffic environments and a key consideration for traffic safety. Therefore, early warning at intersections is a crucial tool for ensuring vehicle safety.

[0003] Currently, most studies use vehicle networking or intelligent transportation technology to achieve intersection warning and assisted passage, which requires specific equipment and traffic infrastructure requirements. It cannot be achieved unilaterally by the vehicle itself and is difficult to promote in a short period of time. Summary of the Invention

[0004] In view of this, in order to solve the above problems, the present invention provides a method and device for early warning of intersections, and the technical solutions are as follows:

[0005] A road intersection early warning method, the method comprising:

[0006] Acquiring basic data, including geometric parameters and motion information of the vehicle, and relative position information between the vehicle and a target vehicle, the target vehicle being a potentially dangerous target at the intersection;

[0007] Determining the current driving behavior of the vehicle based on the vehicle's geometric parameters and motion information, and determining the driving path of the vehicle and the target vehicle corresponding to the current driving behavior of the vehicle in combination with the relative position information;

[0008] Determining collision parameters between the host vehicle and the target vehicle based on the host vehicle's travel path and the target vehicle's travel path, the collision parameters including collision time and lateral displacement of the collision point;

[0009] A dangerous target vehicle with the shortest collision time is selected from the target vehicles, and an intersection warning is performed according to the collision time and lateral displacement of the collision point corresponding to the dangerous target vehicle.

[0010] Preferably, the geometric parameters of the vehicle include the distance d between the rear axle and the front end of the vehicle, and the motion information of the vehicle includes the vehicle speed v0, yaw angular velocity ω0, yaw angle θ, steering angle D steer , steering angle rate ω steer ;

[0011] Determining the current driving behavior of the vehicle based on the vehicle's geometric parameters and motion information includes:

[0012] Calculate the curvature radius of the vehicle

[0013] If the curvature radius ROC of the vehicle is greater than or equal to the corresponding curvature radius threshold R lowthresh , or the curvature radius ROC of the vehicle is less than the curvature radius threshold R lowthresh If the duration is less than the corresponding duration threshold latchtime, the current driving behavior of the vehicle is determined to be turning or going straight, and the driving behavior of the vehicle at the previous moment is retrieved;

[0014] If the vehicle's previous driving behavior was non-turning, calculate the vehicle's lateral displacement within a fixed time t as Δy = (v0*sinθ+d*ω0*cosθ)*t; and determine the vehicle's current driving behavior using the following formula:

[0015]

[0016] Among them, y thresh is the lateral displacement threshold;

[0017] If the vehicle's previous driving behavior was turning, the following formula is used to determine the vehicle's current driving behavior:

[0018]

[0019] Among them, changing lanes and going straight normally are both considered straight driving behaviors.

[0020] Preferably, determining the driving path of the vehicle and the driving path of the target vehicle corresponding to the current driving behavior of the vehicle in combination with the relative position information includes:

[0021] The vehicle's driving path is determined according to the following formula:

[0022]

[0023] The driving path of the target vehicle i is determined according to the following formula:

[0024]

[0025] Among them, x i is the longitudinal displacement of the target vehicle i in the vehicle coordinate system, y i is the lateral displacement of the target vehicle i in the vehicle coordinate system, a i and b i are the driving path parameters when the target vehicle i is the opposite target vehicle, c i and d i are the driving path parameters of the target vehicle i when it crosses the target vehicle, a i and b iIt is obtained by fitting multiple historical displacement points of the opposite target vehicle i using the least square method, c i and d i It is obtained by fitting multiple historical displacement points of the crossing target vehicle i using the least square method, and the multiple historical displacement points are obtained from the relative position information of the target vehicle.

[0026] Preferably, the motion information of the vehicle further includes the longitudinal acceleration a(x) of the vehicle;

[0027] The determining of the collision parameters between the host vehicle and the target vehicle based on the driving path of the host vehicle and the driving path of the target vehicle includes:

[0028] The collision point (x i ,y i ):

[0029]

[0030] Among them, x i is the longitudinal distance of the vehicle to the collision point, y i is the lateral distance of the vehicle to the collision point;

[0031] Based on the collision point (x i ,y i ), determine the collision time TTC according to the following formula i :

[0032]

[0033] Among them, β i is the center angle of the circle when the vehicle turns to reach the collision point, and v0(x) is the longitudinal component of the vehicle's speed v0.

[0034] Preferably, the geometric parameters of the vehicle also include the vehicle width w, and the basic data also include the motion information and length L of the target vehicle i. i The motion information of the target vehicle i includes the speed v of the target vehicle i i The longitudinal component of the velocity v i (x), and the longitudinal acceleration a of the target vehicle i i (x), the method further comprising:

[0035] The collision zone entry time t of target vehicle i is calculated according to the following formula: i (entry) and the time t at which the collision zone is left i (leave):

[0036]

[0037] Among them, x i (t arg et) is the longitudinal distance of target vehicle i to the collision point, x i (t arg et) can be determined based on the longitudinal displacement of the target vehicle i in the host vehicle coordinate system and the longitudinal distance of the host vehicle to the collision point;

[0038] The collision time TTC is calculated according to the following formula i Perform validity check:

[0039]

[0040] Among them, t(buff) is the reserved time interval.

[0041] Preferably, the performing of intersection warning according to the collision time and lateral displacement of the collision point corresponding to the dangerous target vehicle includes:

[0042] When the following formula is met, the intersection warning is executed:

[0043]

[0044] Among them, TTC select is the collision time corresponding to the dangerous target vehicle, TTC threshold is the collision time threshold for early warning triggering, y0 is the lateral displacement of the vehicle, y avail is the horizontal available space for the warning trigger, and y avail =|y collision |-y brake -y threshold ,y collision is the lateral displacement of the collision point corresponding to the dangerous target vehicle, y brake Assuming the vehicle brakes to a standstill at a deceleration of 0.6g, y is the lateral movement distance. threshold The lateral displacement threshold for triggering the warning.

[0045] A road intersection warning device, comprising:

[0046] A data acquisition module is used to acquire basic data, including geometric parameters and motion information of the vehicle, and relative position information between the vehicle and a target vehicle, which is a potentially dangerous target at the intersection;

[0047] a path determination module, configured to determine the current driving behavior of the vehicle based on the vehicle's geometric parameters and motion information, and determine the driving path of the vehicle and the target vehicle corresponding to the current driving behavior of the vehicle in combination with the relative position information;

[0048] a parameter determination module, configured to determine collision parameters between the host vehicle and the target vehicle based on the host vehicle's travel path and the target vehicle's travel path, wherein the collision parameters include collision time and lateral displacement of the collision point;

[0049] The intersection warning module is used to select the dangerous target vehicle with the shortest collision time from the target vehicles, and perform intersection warning according to the collision time and lateral displacement of the collision point corresponding to the dangerous target vehicle.

[0050] Preferably, the geometric parameters of the vehicle include the distance d between the rear axle and the front end of the vehicle, and the motion information of the vehicle includes the vehicle speed v0, yaw angular velocity ω0, yaw angle θ, steering angle D steer , steering angle rate ω steer ;

[0051] The path determination module for determining the current driving behavior of the vehicle based on the vehicle's geometric parameters and motion information is specifically used to:

[0052] Calculate the curvature radius of the vehicle

[0053] If the curvature radius ROC of the vehicle is greater than or equal to the corresponding curvature radius threshold R lowthresh , or the curvature radius ROC of the vehicle is less than the curvature radius threshold R lowthresh If the duration is less than the corresponding duration threshold latchtime, the current driving behavior of the vehicle is determined to be turning or going straight, and the driving behavior of the vehicle at the previous moment is retrieved;

[0054] If the vehicle's previous driving behavior was non-turning, calculate the vehicle's lateral displacement within a fixed time t as Δy = (v0*sinθ+d*ω0*cosθ)*t; and determine the vehicle's current driving behavior using the following formula:

[0055]

[0056] Among them, y thresh is the lateral displacement threshold;

[0057] If the vehicle's previous driving behavior was turning, the following formula is used to determine the vehicle's current driving behavior:

[0058]

[0059] Among them, changing lanes and going straight normally are both considered straight driving behaviors.

[0060] Preferably, the path determination module for determining the driving path of the vehicle and the driving path of the target vehicle corresponding to the current driving behavior of the vehicle in combination with the relative position information is specifically configured to:

[0061] The vehicle's driving path is determined according to the following formula:

[0062]

[0063] The driving path of the target vehicle i is determined according to the following formula:

[0064]

[0065] Among them, x i is the longitudinal displacement of the target vehicle i in the vehicle coordinate system, y i is the lateral displacement of the target vehicle i in the vehicle coordinate system, a i and b i are the driving path parameters when the target vehicle i is the opposite target vehicle, c i and d i are the driving path parameters of the target vehicle i when it crosses the target vehicle, a i and b i It is obtained by fitting multiple historical displacement points of the opposite target vehicle i using the least square method, c i and d i It is obtained by fitting multiple historical displacement points of the crossing target vehicle i using the least square method, and the multiple historical displacement points are obtained from the relative position information of the target vehicle.

[0066] Preferably, the motion information of the vehicle further includes the longitudinal acceleration a(x) of the vehicle;

[0067] The parameter determination module for determining the collision parameters between the host vehicle and the target vehicle based on the driving path of the host vehicle and the driving path of the target vehicle is specifically configured to:

[0068] The collision point (x i ,y i ):

[0069]

[0070] Among them, x i is the longitudinal distance of the vehicle to the collision point, y i is the lateral distance of the vehicle to the collision point;

[0071] Based on the collision point (x i ,y i ), determine the collision time TTC according to the following formula i :

[0072]

[0073] Among them, β i is the center angle of the circle when the vehicle turns to reach the collision point, and v0(x) is the longitudinal component of the vehicle's speed v0.

[0074] Compared with the prior art, the present invention has the following beneficial effects:

[0075] The present invention provides a method and device for early warning at an intersection. The method first obtains basic data, which includes the geometric parameters and motion information of the vehicle, as well as the relative position information between the vehicle and a target vehicle that is a potential dangerous target at the intersection. The method then determines the current driving behavior of the vehicle based on the geometric parameters and motion information of the vehicle, and determines the driving path of the vehicle and the driving path of the target vehicle corresponding to the current driving behavior of the vehicle in combination with the relative position information. The method further determines the collision parameters between the vehicle and the target vehicle based on the driving path of the vehicle and the driving path of the target vehicle. The collision parameters include the collision time and the lateral displacement of the collision point. Finally, the dangerous target vehicle with the shortest collision time is selected from the target vehicles, and an early warning at the intersection is performed based on the collision time and the lateral displacement of the collision point corresponding to the dangerous target vehicle. Based on the present invention, an early warning at an intersection can be performed based on the real-time behavior of the vehicle. By judging the macroscopic driving behavior of the vehicle, the risk of the intersection can be determined in real time, simplifying the analysis and research of complex scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0076] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are merely embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0077] Figure 1 A flow chart of a method for an intersection warning method provided by an embodiment of the present invention;

[0078] Figure 2 A schematic diagram of a typical vehicle intersection scenario provided by an embodiment of the present invention;

[0079] Figure 3 A schematic structural diagram of a road intersection warning device provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0080] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0081] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0082] Current vehicle behavior research focuses more on predicting the vehicle's future driving trajectory, and less on judging the specific driving behavior attributes at a macro level. In fact, real-time distinction between whether the vehicle is going straight, turning, making a U-turn, or changing lanes is conducive to targeted research and analysis of the current scenario.

[0083] Currently, there is little research on vehicle collision analysis at intersections. The conditions at intersections are relatively complex. Vehicle behaviors at intersections include left turns, no turns, and right turns. Target vehicles at intersections include oncoming vehicles and crossing vehicles. These target vehicles may go straight or turn. It is necessary to find a unified and universal method to analyze these types of scenarios and provide early warning of intersection risks.

[0084] Most current studies use single-dimensional early warning indicators. However, collisions at intersections can occur at varying angles and locations, and single-dimensional or single-directional collision parameters are insufficient. For example, in a scenario where a turning vehicle encounters an oncoming vehicle, the longitudinal collision time is short, but the lateral displacement is significant. Therefore, it is necessary to combine both longitudinal and lateral indicators to provide early identification of vehicle intersection warnings.

[0085] To this end, the present invention provides a road intersection warning method, the method flow chart of which is as follows: Figure 1 As shown, the following steps are included:

[0086] S10, acquiring basic data, which includes geometric parameters and motion information of the vehicle, and relative position information between the vehicle and a target vehicle, where the target vehicle is a potentially dangerous target at the intersection.

[0087] In the embodiment of the present invention, basic data can be identified and obtained through the bus signal of the vehicle and the radar camera sensor. The basic data includes the geometric parameters of the vehicle (such as the distance d between the rear axle and the front end of the vehicle), the motion information of the vehicle (such as the vehicle speed v0, yaw angular velocity ω0, yaw angle θ, steering angle D steer , steering angle rate ω steer , the longitudinal acceleration a(x) of the vehicle), and the relative position information between the vehicle and the target vehicle (such as the horizontal and vertical displacement of the target vehicle in the coordinate system of the vehicle).

[0088] See also Figure 2 , Figure 2 A typical vehicle intersection scenario diagram provided by an embodiment of the present invention. Figure 2As shown, the embodiment of the present invention selects oncoming and crossing vehicles as potential danger targets (i.e., target vehicles), including oncoming vehicles in the opposite lane and vehicles approaching from the left or right of the vehicle. In real-time applications, a certain number (e.g., up to 10 each) of oncoming and crossing vehicles can be selected, and the target vehicle selection can be completed by setting a potential danger zone.

[0089] S20, determining the current driving behavior of the vehicle based on the geometric parameters and motion information of the vehicle, and determining the driving path of the vehicle and the driving path of the target vehicle corresponding to the current driving behavior of the vehicle in combination with the relative position information.

[0090] In the embodiment of the present invention, a macroscopic driving behavior judgment model of the vehicle is established to judge the straight-moving or turning behavior of the vehicle in real time, as follows:

[0091] First, the U-turn behavior of the vehicle is determined by combining the curvature radius and duration of the vehicle's turning. The embodiment of the present invention does not consider the U-turn warning of the vehicle at the intersection. The geometric parameters of the vehicle include the distance d between the rear axle and the front end of the vehicle, and the motion information of the vehicle includes the vehicle speed v0, yaw angular velocity ω0, yaw angle θ, steering angle D steer , steering angle rate ω steer The vehicle's U-turn behavior meets the following conditions:

[0092]

[0093] In this regard, the driving behavior of the vehicle can be divided into straight driving, turning and U-turn. The embodiment of the present invention can calculate the curvature radius of the vehicle. Furthermore, if the curvature radius ROC of the vehicle is greater than or equal to the corresponding curvature radius threshold R lowthresh , or the curvature radius ROC of the vehicle is less than the curvature radius threshold R lowthresh If the duration is less than the corresponding duration threshold latchtime, it can be determined that the vehicle's driving behavior is turning or going straight, so as to exclude the U-turn scenario.

[0094] Furthermore, the embodiment of the present invention can calculate the lateral displacement Δy of the vehicle within a fixed time based on the distance d between the rear axle and the front end of the vehicle, the vehicle speed v0, the yaw angular velocity ω0, and the yaw angle θ, and at the same time arbitrate the yaw angular velocity ω0, the steering angle D steer , steering angle rate ω steer Multi-dimensional information, combined with the characteristics of different macro-driving behaviors, determines whether the vehicle is turning or going straight. The straight-moving behaviors at intersections include normal straight-moving and lane-changing straight-moving. The details are as follows:

[0095] First, retrieve the vehicle's last driving behavior. If the vehicle's last driving behavior was non-turning (normal straight driving or lane-changing straight driving), the vehicle's lateral displacement within a fixed time (e.g., 20ms) can be calculated as Δy = (v0*sinθ+d*ω0*cosθ)*t. The vehicle's current driving behavior can be determined using the following formula: thresh is the lateral displacement threshold:

[0096]

[0097] If the vehicle's previous driving behavior was turning, the following formula can be used to determine the vehicle's current driving behavior, where lane change and normal straight driving are both considered straight driving behaviors:

[0098]

[0099] Furthermore, the driving path of the vehicle and the target vehicle can be obtained in real time based on the current driving behavior of the vehicle.

[0100] 1) The vehicle's driving path: The vehicle's driving path is obtained based on the vehicle's yaw angular velocity ω0, including the vehicle's turning and straight driving scenarios.

[0101] Derivative of the longitudinal displacement x of the vehicle in the vehicle coordinate system for:

[0102] Derivative of the lateral displacement y of the vehicle in the vehicle coordinate system for:

[0103] Derivative of the vehicle's yaw angle θ for:

[0104] The vehicle's driving path is determined according to the following formula:

[0105]

[0106] 2) Target Vehicle i's Driving Path: Target vehicles i are classified as oncoming and crossing vehicles. These include oncoming vehicles in the opposite lane and vehicles approaching from the left or right of the target vehicle. This embodiment of the present invention primarily addresses straight-moving target vehicles and does not consider turning target vehicles. Because time and distance at intersections are relatively limited, a low-order function can be used to describe the target vehicle's driving path.

[0107]

[0108] In the above two formulas, x iis the longitudinal displacement of the target vehicle i in the vehicle coordinate system, y i is the lateral displacement of the target vehicle i in the vehicle coordinate system; a i and b i is the driving path parameter when target vehicle i is the opposite target vehicle; c i and d i is the driving path parameter of target vehicle i when it crosses the target vehicle.

[0109] For any target vehicle, the vehicle can obtain the relative position information between it and the target vehicle in real time, and continuously record the relative position information of the target vehicle at fixed time intervals, including the longitudinal displacement and lateral displacement of the target vehicle in the vehicle's coordinate system. In practical applications, the relative position information can be recorded up to a certain number of times, and the relative position information closest to the current moment is retained and stored. When the target is lost, the historical relative position information will be cleared and new relative position information will be recorded again. Each time the relative position information is recorded, a displacement point of the target vehicle is generated. This is based on the vehicle's coordinate system at that time. Using coordinate transformation, the displacement point at the previous moment is converted to the current vehicle coordinate system. In this way, the historical displacement points of a series of target vehicles are transferred to the vehicle's current coordinate system.

[0110] The displacement point p of the target vehicle i in the previous vehicle coordinate system i Transfer to the current vehicle coordinate system, the new displacement point for:

[0111]

[0112] In practical applications, in the above formula, i = 1, ..., m, ..., 10, ..., n, ..., 20, where i represents the target vehicle number, m represents the number of oncoming target vehicles, and 0 ≤ m ≤ 10 means there are at most 10 oncoming target vehicles, and 0 means there are no oncoming target vehicles, and n represents the number of crossing target vehicles, and 10 ≤ n ≤ 20 means there are at most 10 crossing target vehicles, and 10 means there are no crossing target vehicles. i is the displacement point of target vehicle i in the vehicle coordinate system at the last moment, is the displacement point p of target vehicle i i Transfer to the displacement point under the current vehicle coordinate system, Δθ is the yaw angle of the vehicle per unit time, Δx is the longitudinal displacement of the target vehicle i per unit time in the vehicle coordinate system; Δy is the lateral displacement of the target vehicle i per unit time in the vehicle coordinate system.

[0113] The target vehicle's path is predicted using historical relative position information. Given at least a certain number of sampled historical displacement points, the least squares method is used to substitute the path and calculate the residual between the sampled values. A small residual indicates the fit is complete, yielding the target vehicle's path parameters. The weights of different historical displacement points are also considered, with more recent displacement points receiving greater weights and more distant displacement points receiving smaller weights.

[0114] Driving path parameter a when target vehicle i is the oncoming target vehicle i and b i The solution satisfies the following:

[0115]

[0116] Target vehicle i is the driving path parameter c when crossing the target vehicle i and d i The solution satisfies the following:

[0117]

[0118] In practical applications, in the above two formulas, j = 0, ..., k, where j represents the historical displacement point of the target vehicle i in the jth time interval before, and 0 ≤ k ≤ 9 means that the nine most recent historical time points of the current time are recorded for multiple records, 0 is the current time point, and w j is the weight of the jth historical moment, x ij is the longitudinal displacement of the target vehicle i at the jth historical moment in the current vehicle coordinate system, y ij is the lateral displacement of the target vehicle i at the jth historical moment in the current vehicle coordinate system, a i and b i is the parameter of the driving path of the opposite target vehicle i; c i and d i is the parameter of the path that crosses the target vehicle i. The target vehicle's path is determined according to the following formula:

[0119]

[0120] S30 , determining collision parameters between the host vehicle and the target vehicle based on the driving path of the host vehicle and the driving path of the target vehicle, wherein the collision parameters include collision time and lateral displacement of the collision point.

[0121] In the embodiment of the present invention, a vehicle collision model is established, and collision parameters are calculated based on the current driving behavior of the vehicle, as follows:

[0122] Based on the vehicle's current driving behavior, the collision point between the vehicle and the target vehicle is determined. The vehicle's current driving behavior includes turning left, turning right, or going straight. Target vehicles include oncoming and crossing targets. If the vehicle's current driving behavior is determined to be turning or going straight, the coordinates of the collision point are calculated by combining the vehicle's driving path with the target vehicle's. If the vehicle is turning, the vehicle's driving path can also be used to determine the center angle of the collision point.

[0123] The motion information of the vehicle also includes the longitudinal acceleration a(x) of the vehicle, the collision point (x i ,y i )The solution satisfies the following:

[0124]

[0125] In practical applications, in the above formula, i = 1, ..., m, ..., 10, ..., n, ..., 20, where i represents the target vehicle number, m represents the number of oncoming target vehicles, and 0 ≤ m ≤ 10 indicates that there are at most 10 oncoming target vehicles, and 0 indicates that there are no oncoming target vehicles, and n represents the number of crossing target vehicles, and 10 ≤ n ≤ 20 indicates that there are at most 10 crossing target vehicles, and 10 indicates that there are no crossing target vehicles. i is the longitudinal distance of the vehicle to the collision point, y i The lateral distance of the vehicle to the collision point

[0126] Furthermore, based on the collision point (x i ,y i ), calculate the collision parameters. The collision time TTC between the vehicle and the target vehicle i i Determined based on the following formula:

[0127]

[0128] Among them, β i is the center angle of the circle when the vehicle turns to reach the collision point, and v0(x) is the longitudinal component of the vehicle's speed v0.

[0129] On this basis, the embodiment of the present invention can further analyze the collision time TTC i Perform validity check. The collision point (x i ,y i ) lacks the time dimension, so when considering the time from the vehicle to the collision point (x i ,y i ), it is also necessary to consider whether the target vehicle i is in the potential collision area of ​​the collision point. Considering the vehicle shape, the time to enter the collision area (i.e., the collision area entry time t i(entry)) is the time when the front of target vehicle i reaches the collision point (x i ,y i ), the time to leave the collision area (i.e., the time to leave the collision area t i (leave)) is the time when the rear end of the target vehicle leaves the collision point, so the collision time TTC i Only occurs when the collision zone is entered at time t i (entry) and the time t at which the collision zone is left i To this end, the present invention performs effective collision verification on the collision time between the host vehicle and the target vehicle i.

[0130] The geometric parameters of the vehicle also include the vehicle width w, and the basic data also includes the motion information and length L of the target vehicle i i , the motion information of target vehicle i includes the speed v of target vehicle i i The longitudinal component of the velocity v i (x), and the longitudinal acceleration a of the target vehicle i i (x).

[0131] The collision zone entry time t of target vehicle i is calculated according to the following formula: i (entry) and the time t at which the collision zone is left i (leave):

[0132]

[0133] In the above formula, i = 1, L, m, L, 10, L, n, L 20, where i represents the target vehicle number, m represents the number of oncoming target vehicles, and 0 ≤ m ≤ 10 indicates that there are at most 10 oncoming target vehicles, and 0 indicates that there are no oncoming target vehicles, and n represents the number of crossing target vehicles, and 10 ≤ n ≤ 20 indicates that there are at most 10 crossing target vehicles, and 10 indicates that there are no crossing target vehicles. i (t arg et) is the longitudinal distance of target vehicle i to the collision point, x i (t arg et) can be determined based on the longitudinal displacement of the target vehicle i in the own vehicle coordinate system and the longitudinal distance of the own vehicle to the collision point.

[0134] The collision time TTC is calculated according to the following formula i Perform validity check:

[0135]

[0136] Wherein, t(buff) is the reserved time interval.

[0137] S40: Select a dangerous target vehicle with the shortest collision time from the target vehicles, and issue an intersection warning based on the collision time and lateral displacement of the collision point corresponding to the dangerous target vehicle.

[0138] In this embodiment of the present invention, the collision times between target vehicles are compared, and the target vehicle with the shortest collision time is selected as the dangerous target vehicle. The corresponding collision time and lateral displacement of the collision point are also obtained. Furthermore, the dangerous target vehicles can be distinguished and screened for oncoming and crossing target vehicles, and both oncoming and crossing scenarios can be displayed simultaneously, supporting the instrument panel to display dangerous target vehicles in both directions.

[0139] The dangerous oncoming target vehicle i satisfies the following conditions:

[0140]

[0141] The dangerous crossing target vehicle i satisfies the following conditions:

[0142]

[0143] Dangerous target vehicles meet the following requirements:

[0144]

[0145] In actual applications, in the above three formulas, i = 1,…,m,…,10,…,n,…,20, where i represents the number of the target vehicle, m is the number of oncoming target vehicles, and 0≤m≤10 indicates that there are at most 10 oncoming target vehicles, and 0 indicates that there are no oncoming target vehicles, and n is the number of crossing target vehicles, and 10≤n≤20 indicates that there are at most 10 crossing target vehicles, and 10 indicates that there are no crossing target vehicles.

[0146] Furthermore, appropriate multi-dimensional thresholds are selected for analysis, while considering the collision time and lateral displacement of the collision point to output warning information.

[0147] Assuming that the vehicle brakes to a stop at a deceleration of 0.6g, the estimated braking distance is obtained. Based on the vehicle's yaw angular velocity ω0, the lateral and longitudinal movement distances during the turning process can be obtained.

[0148] When the following formula is met, the intersection warning is executed:

[0149]

[0150] Among them, TTC select is the collision time corresponding to the dangerous target vehicle, TTC threshold is the collision time threshold for early warning triggering, y0 is the lateral displacement of the vehicle, y avail is the horizontal available space for the warning trigger, and yavail =|y collision |-y brake -y threshold ,y collision is the lateral displacement of the collision point corresponding to the dangerous target vehicle, y brake Assuming the vehicle brakes to a standstill at a deceleration of 0.6g, y is the lateral movement distance. threshold The lateral displacement threshold for triggering the warning.

[0151] It should be noted that the estimated stopping distance, assuming the vehicle brakes to a standstill at 0.6g, may be a curve. Generally speaking, based on the vehicle's yaw rate ω0, the lateral movement distance during the turn can be obtained. If the vehicle brakes to a curve (i.e., a turn), the curvature radius can be determined based on the vehicle's yaw rate ω0 and vehicle speed v0. The horizontal movement distance can be obtained based on the geometric projection.

[0152] It should also be noted that y0, y collision , and y brake The displacement parameters are based on the earth coordinate system, and the coordinate system transformation is used to transform these displacement parameters from the vehicle coordinate system to the earth coordinate system. It should also be noted that all thresholds in the embodiment of the present invention are pre-set.

[0153] The intersection warning method provided by the embodiments of the present invention can provide intersection warnings based on real-time vehicle behavior. By assessing the macroscopic driving behavior of the vehicles, it predicts the paths of the host vehicle and the target vehicle. It then calculates the collision parameters of the host vehicle and the target vehicle to screen out the most dangerous target, selects appropriate thresholds, and issues warning information. Compared with existing technologies, the present invention has the following advantages and beneficial effects:

[0154] 1) A new intersection warning method based on real-time vehicle behavior is proposed. This method determines macroscopic vehicle behavior and calculates the current risk based on specific real-time driving behaviors, simplifying the analysis and research of complex scenarios. 2) By combining the study of different macroscopic driving behaviors at intersections, characteristic parameters and relevant thresholds are extracted, analyzing the real-time vehicle operating state and determining specific vehicle driving behaviors at intersections, which has strong practical value. 3) Vehicle paths are predicted based on historical vehicle trajectory information using a least squares method. Through coordinate transformation, different paths are transferred to the current vehicle coordinate system, and the collision point is calculated using geometric methods, which is simpler, more efficient, and more robust. 4) A collision model is established based on specific macroscopic driving behaviors. Different vehicle path equations and collision parameter solutions are used for different driving behaviors, enabling more accurate real-time analysis of intersection collisions and providing insights for intersection risk warning. 5) The most dangerous target vehicles are screened using collision time parameters, and multi-dimensional threshold analysis is used to generate hazard warning information, which has important guiding significance for intelligent vehicle warning.

[0155] Based on the intersection warning method provided in the above embodiment, the embodiment of the present invention provides a device for executing the above intersection warning method. The structural diagram of the device is shown in FIG. Figure 3 Shown, including:

[0156] The data acquisition module 10 is used to acquire basic data, including geometric parameters and motion information of the vehicle, as well as relative position information between the vehicle and a target vehicle, which is a potential dangerous target at the intersection;

[0157] A path determination module 20 is configured to determine the current driving behavior of the vehicle based on the vehicle's geometric parameters and motion information, and to determine the driving path of the vehicle and the target vehicle corresponding to the current driving behavior of the vehicle based on the relative position information;

[0158] A parameter determination module 30 is used to determine the collision parameters between the host vehicle and the target vehicle based on the driving path of the host vehicle and the driving path of the target vehicle, wherein the collision parameters include the collision time and the lateral displacement of the collision point;

[0159] The intersection warning module 40 is used to select a dangerous target vehicle with the shortest collision time from the target vehicles, and perform an intersection warning based on the collision time and lateral displacement of the collision point corresponding to the dangerous target vehicle.

[0160] Optionally, the vehicle's geometric parameters include the distance d between the vehicle's rear axle and front end, and the vehicle's motion information includes the vehicle's speed v0, yaw angular velocity ω0, yaw angle θ, and steering angle D. steer , steering angle rate ω steer ;

[0161] The path determination module 20 is used to determine the current driving behavior of the vehicle based on the vehicle's geometric parameters and motion information, specifically for:

[0162] Calculate the curvature radius of the vehicle

[0163] If the curvature radius ROC of the vehicle is greater than or equal to the corresponding curvature radius threshold R lowthresh , or the curvature radius ROC of the vehicle is less than the curvature radius threshold R lowthresh If the duration is less than the corresponding duration threshold latchtime, the current driving behavior of the vehicle is determined to be turning or going straight, and the driving behavior of the vehicle at the previous moment is retrieved;

[0164] If the vehicle's previous driving behavior was non-turning, calculate the vehicle's lateral displacement within a fixed time t as Δy = (v0*sinθ+d*ω0*cosθ)*t; and determine the vehicle's current driving behavior using the following formula:

[0165]

[0166] Among them, y thresh is the lateral displacement threshold;

[0167] If the vehicle's previous driving behavior was turning, the following formula is used to determine the vehicle's current driving behavior:

[0168]

[0169] Among them, changing lanes and going straight normally are both considered straight driving behaviors.

[0170] Optionally, the path determination module 20 for determining the driving path of the vehicle and the driving path of the target vehicle corresponding to the current driving behavior of the vehicle in combination with the relative position information is specifically configured to:

[0171] The vehicle's driving path is determined according to the following formula:

[0172]

[0173] The target vehicle's driving path is determined according to the following formula:

[0174]

[0175] Among them, x i is the longitudinal displacement of the target vehicle i in the vehicle coordinate system, y i is the lateral displacement of the target vehicle i in the vehicle coordinate system, a i and b i They are the driving path parameters when target vehicle i is the opposite target vehicle, c i and d i They are the driving path parameters when target vehicle i crosses the target vehicle, a i and b i It is obtained by fitting multiple historical displacement points of the opposite target vehicle i using the least square method, c i and d i It is obtained by fitting multiple historical displacement points of the crossing target vehicle i using the least squares method. The multiple historical displacement points are obtained from the relative position information of the target vehicle.

[0176] Optionally, the motion information of the vehicle also includes the longitudinal acceleration a(x) of the vehicle;

[0177] The parameter determination module 30 is used to determine the collision parameters between the host vehicle and the target vehicle based on the driving path of the host vehicle and the driving path of the target vehicle, specifically for:

[0178] The collision point (x i,y i ):

[0179]

[0180] Among them, x i is the longitudinal distance of the vehicle to the collision point, y i is the lateral distance of the vehicle to the collision point;

[0181] Based on the collision point (x i ,y i ), determine the collision time TTC according to the following formula i :

[0182]

[0183] Among them, β i is the center angle of the circle when the vehicle turns to reach the collision point, and v0(x) is the longitudinal component of the vehicle's speed v0.

[0184] Optionally, the vehicle's geometric parameters also include the vehicle's width w, and the basic data also includes the target vehicle i's motion information and length L i , the motion information of target vehicle i includes the speed v of target vehicle i i The longitudinal component of the velocity v i (x), and the longitudinal acceleration a of the target vehicle i i (x), parameter determination module 30, is further configured to:

[0185] The collision zone entry time t of target vehicle i is calculated according to the following formula: i (entry) and the time t at which the collision zone is left i (leave):

[0186]

[0187] Among them, x i (t arg et) is the longitudinal distance of target vehicle i to the collision point, x i (t arg et) can be determined based on the longitudinal displacement of the target vehicle i in the host vehicle coordinate system and the longitudinal distance of the host vehicle to the collision point;

[0188] The collision time TTC is calculated according to the following formula i Perform validity check:

[0189]

[0190] Wherein, t(buff) is the reserved time interval.

[0191] Optionally, the intersection warning module 40 for providing an intersection warning based on the collision time and lateral displacement of the collision point corresponding to the dangerous target vehicle is specifically configured to:

[0192] When the following formula is met, the intersection warning is executed:

[0193]

[0194] Among them, TTC select is the collision time corresponding to the dangerous target vehicle, TTC threshold is the collision time threshold for early warning triggering, y0 is the lateral displacement of the vehicle, y avail is the horizontal available space for the warning trigger, and y avail =|y collision |-y brake -y threshold ,y collision is the lateral displacement of the collision point corresponding to the dangerous target vehicle, y brake Assuming the vehicle brakes to a standstill at a deceleration of 0.6g, y is the lateral movement distance. threshold The lateral displacement threshold for triggering the warning.

[0195] It should be noted that the detailed functions of each functional module in the embodiment of the present invention can be found in the corresponding disclosed part of the above-mentioned intersection warning method embodiment, and will not be repeated here.

[0196] The above is a detailed introduction to the intersection warning method and device provided by the present invention. Specific examples are used in this article to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core ideas. At the same time, for those skilled in the art, according to the ideas of the present invention, there will be changes in the specific implementation methods and application scopes. In summary, the content of this specification should not be understood as limiting the present invention.

[0197] It should be noted that the various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. Reference can be made to the common and similar parts between the various embodiments. For the devices disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the description is relatively simple, and the relevant parts can be referred to the method description.

[0198] It should also be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion, such that the process, method, article, or apparatus comprising a series of elements inherent to the elements, or also including elements inherent to these processes, methods, articles, or apparatus. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of additional identical elements in the process, method, article, or apparatus comprising the element.

[0199] The above description of the disclosed embodiments is intended to enable one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be readily apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention is not limited to the embodiments shown herein but is intended to conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A road intersection warning method, characterized in that: The method comprises: Acquiring basic data, including geometric parameters and motion information of the vehicle, and relative position information between the vehicle and a target vehicle, the target vehicle being a potentially dangerous target at the intersection. The geometric parameters of the vehicle include the distance between the rear axle and the front end of the vehicle, and the motion information of the vehicle includes the vehicle's speed, yaw rate, yaw angle, steering angle, and steering angle rate. Calculating the lateral displacement of the vehicle within a fixed time based on the distance between the rear axle and the front end of the vehicle, the vehicle speed, the yaw rate, and the yaw angle, while arbitrating the yaw rate, steering angle, and steering rate of the vehicle, and judging the current driving behavior of the vehicle based on the characteristics of different driving behaviors, and determining the driving path of the vehicle and the driving path of the target vehicle corresponding to the current driving behavior of the vehicle based on the relative position information; Determining collision parameters between the host vehicle and the target vehicle based on the host vehicle's travel path and the target vehicle's travel path, the collision parameters including collision time and lateral displacement of the collision point; A dangerous target vehicle with the shortest collision time is selected from the target vehicles, and an intersection warning is performed according to the collision time and lateral displacement of the collision point corresponding to the dangerous target vehicle.

2. The method according to claim 1, characterized in that The calculation of the lateral displacement of the vehicle within a fixed time based on the distance between the rear axle and the front end of the vehicle, the vehicle speed, the yaw rate, and the yaw angle, while arbitrating the yaw rate, the steering angle, and the steering angle rate of the vehicle, and combining the characteristics of different driving behaviors to determine the current driving behavior of the vehicle, includes: Calculate the curvature radius of the vehicle Among them, v0 is the vehicle speed, ω0 is the yaw rate of the vehicle; If the curvature radius ROC of the vehicle is greater than or equal to the corresponding curvature radius threshold R lowthersh , or the curvature radius ROC of the vehicle is less than the curvature radius threshold R lowthresh If the duration is less than the corresponding duration threshold latchtime, the current driving behavior of the vehicle is determined to be turning or going straight, and the driving behavior of the vehicle at the previous moment is retrieved; If the vehicle's previous driving behavior was non-turning, calculate the vehicle's lateral displacement within a fixed time t as Δy = (v0*sinθ+d*ω0*cosθ)*t, where d is the distance between the vehicle's rear axle and front end, and θ is the vehicle's yaw angle. The current driving behavior of the vehicle is determined according to the following formula: Among them, y thresh is the lateral displacement threshold, D steer is the steering angle of the vehicle, ω steer is the steering angle rate of the vehicle; If the vehicle's previous driving behavior was turning, the following formula is used to determine the vehicle's current driving behavior: Among them, changing lanes and going straight normally are both considered straight driving behaviors.

3. The method according to claim 2, characterized in that Determining the driving path of the vehicle and the driving path of the target vehicle corresponding to the current driving behavior of the vehicle in combination with the relative position information includes: The vehicle's driving path is determined according to the following formula: The driving path of the target vehicle i is determined according to the following formula: Among them, x i is the longitudinal displacement of the target vehicle i in the vehicle coordinate system, y i is the lateral displacement of the target vehicle i in the vehicle coordinate system, a i and b i are the driving path parameters when the target vehicle i is the opposite target vehicle, c i and d i are the driving path parameters of the target vehicle i when it crosses the target vehicle, a i and b i It is obtained by fitting multiple historical displacement points of the opposite target vehicle i using the least square method, c i and d i It is obtained by fitting multiple historical displacement points of the crossing target vehicle i using the least square method, and the multiple historical displacement points are obtained from the relative position information of the target vehicle.

4. The method according to claim 3, characterized in that The vehicle's motion information also includes the vehicle's longitudinal acceleration a(x); The determining of the collision parameters between the host vehicle and the target vehicle based on the driving path of the host vehicle and the driving path of the target vehicle includes: The collision point (x i ,y i ): Among them, x i is the longitudinal distance of the vehicle to the collision point, y i is the lateral distance of the vehicle to the collision point; Based on the collision point (x i ,y i ), determine the collision time TTC according to the following formula i : Among them, β i is the center angle of the circle when the vehicle turns to reach the collision point, and v0(x) is the longitudinal component of the vehicle's speed v0.

5. The method according to claim 4, characterized in that The geometric parameters of the vehicle also include the vehicle width w, and the basic data also includes the motion information and length L of the target vehicle i. i The motion information of the target vehicle i includes the speed v of the target vehicle i i The longitudinal component of the velocity v i (x), and the longitudinal acceleration a of the target vehicle i i (x), the method further comprising: The collision zone entry time t of target vehicle i is calculated according to the following formula: i (entry) and the time t at which the collision zone is left i (leave): Among them, x i (target) is the longitudinal distance of target vehicle i to the collision point, x i (target) can be determined based on the longitudinal displacement of the target vehicle i in the host vehicle coordinate system and the longitudinal distance of the host vehicle to the collision point; The collision time TTC is calculated according to the following formula i Perform validity check: Wherein, t(buff) is the reserved time interval.

6. The method according to claim 4, characterized in that The intersection warning is performed according to the collision time and lateral displacement of the collision point corresponding to the dangerous target vehicle, including: When the following formula is met, the intersection warning is executed: Among them, TTC select is the collision time corresponding to the dangerous target vehicle, TTC threshold is the collision time threshold for early warning triggering, y0 is the lateral displacement of the vehicle, y avail is the horizontal available space for the warning trigger, and y avail =|y collision |-y brake -y threshold ,y collision is the lateral displacement of the collision point corresponding to the dangerous target vehicle, y brake Assuming the vehicle brakes to a standstill at a deceleration of 0.6g, y is the lateral movement distance. threshold The lateral displacement threshold for triggering the warning.

7. A road intersection warning device, characterized in that: The device comprises: a data acquisition module, configured to acquire basic data, including geometric parameters and motion information of the vehicle, and relative position information between the vehicle and a target vehicle, the target vehicle being a potentially dangerous target at the intersection; the geometric parameters of the vehicle including the distance between the rear axle and the front end of the vehicle; and the motion information of the vehicle including the vehicle's speed, yaw rate, yaw angle, steering angle, and steering angle rate; a path determination module for calculating the lateral displacement of the vehicle within a fixed time based on the distance between the rear axle and the front end of the vehicle, the vehicle speed, the yaw rate, and the yaw angle; arbitrating the yaw rate, the steering angle, and the steering rate of the vehicle; and determining the current driving behavior of the vehicle based on the characteristics of different driving behaviors. The module then determines the driving path of the vehicle corresponding to the current driving behavior of the vehicle and the driving path of the target vehicle based on the relative position information; a parameter determination module, configured to determine collision parameters between the host vehicle and the target vehicle based on the host vehicle's travel path and the target vehicle's travel path, wherein the collision parameters include collision time and lateral displacement of the collision point; The intersection warning module is used to select the dangerous target vehicle with the shortest collision time from the target vehicles, and perform intersection warning according to the collision time and lateral displacement of the collision point corresponding to the dangerous target vehicle.

8. The device according to claim 7, characterized in that The path determination module is used to calculate the lateral displacement of the vehicle within a fixed time based on the distance between the rear axle and the front end of the vehicle, the vehicle speed, the yaw rate, and the yaw angle, and to arbitrate the yaw rate, steering angle, and steering angle rate of the vehicle, and to judge the current driving behavior of the vehicle based on the characteristics of different driving behaviors. Specifically, it is used to: Calculate the curvature radius of the vehicle Among them, v0 is the vehicle speed, ω0 is the yaw rate of the vehicle; If the curvature radius ROC of the vehicle is greater than or equal to the corresponding curvature radius threshold R lowthtesh , or the curvature radius ROC of the vehicle is less than the curvature radius threshold R lowthresh If the duration is less than the corresponding duration threshold latchtime, the current driving behavior of the vehicle is determined to be turning or going straight, and the driving behavior of the vehicle at the previous moment is retrieved; If the vehicle's previous driving behavior was non-turning, calculate the vehicle's lateral displacement within a fixed time t as Δy = (v0*sinθ+d*ω0*cosθ)*t, where d is the distance between the vehicle's rear axle and front end, and θ is the vehicle's yaw angle. The current driving behavior of the vehicle is determined according to the following formula: Among them, y thresh is the lateral displacement threshold, D steer is the steering angle of the vehicle, ω steer is the steering angle rate of the vehicle; If the vehicle's previous driving behavior was turning, the following formula is used to determine the vehicle's current driving behavior: Among them, changing lanes and going straight normally are both considered straight driving behaviors.

9. The device according to claim 8, characterized in that The path determination module is configured to determine the driving path of the vehicle and the driving path of the target vehicle corresponding to the current driving behavior of the vehicle in combination with the relative position information, specifically configured to: The vehicle's driving path is determined according to the following formula: The driving path of the target vehicle i is determined according to the following formula: Among them, x i is the longitudinal displacement of the target vehicle i in the vehicle coordinate system, y i is the lateral displacement of the target vehicle i in the vehicle coordinate system, a i and b i are the driving path parameters when the target vehicle i is the opposite target vehicle, c i and d i are the driving path parameters of the target vehicle i when it crosses the target vehicle, a i and b i It is obtained by fitting multiple historical displacement points of the opposite target vehicle i using the least square method, c i and d i It is obtained by fitting multiple historical displacement points of the crossing target vehicle i using the least square method, and the multiple historical displacement points are obtained from the relative position information of the target vehicle.

10. The device according to claim 9, characterized in that The vehicle's motion information also includes the vehicle's longitudinal acceleration a(x); The parameter determination module for determining the collision parameters between the host vehicle and the target vehicle based on the driving path of the host vehicle and the driving path of the target vehicle is specifically configured to: The collision point (x i ,y i ): Among them, x i is the longitudinal distance of the vehicle to the collision point, y i is the lateral distance of the vehicle to the collision point; Based on the collision point (x i ,y i ), determine the collision time TTC according to the following formula i : Among them, β i is the center angle of the circle when the vehicle turns to reach the collision point, and v0(x) is the longitudinal component of the vehicle's speed v0.

Images