Vehicle collision detection method, system, and vehicle

Abstract

This application provides a vehicle collision detection method, system and vehicle. The vehicle collision detection method includes the steps of dividing an acquired set of obstacle information into a first obstacle map and a second obstacle map based on a preset height of an obstacle, then modeling a target vehicle with the first obstacle map and the second obstacle map, and fusing the modeling result with the first obstacle map and the second obstacle map to obtain a first collision detection model and a second collision detection model. Further, obtaining a distance relationship between an obstacle corresponding to the set of obstacle information at a preset position of the target vehicle by the first collision detection model and / or the second collision detection model, and determining whether the target vehicle will collide based on the distance relationship. This application can perform real-time and accurate collision risk analysis for different types of obstacle heights, improve the accuracy and efficiency of collision detection, and further ensure the driving safety of the vehicle and the personal safety of the people in the vehicle.

Description

[Technical Field]

[0001] This application relates to the technology of intelligent driving, and more particularly to a vehicle collision detection method, system, and vehicle. [Background technology]

[0002] With the rapid development of intelligent driving, cars equipped with autonomous driving functions are becoming increasingly common. While autonomous driving functions bring convenience to people, they can also lead to unpleasant user experiences, such as scratching rearview mirrors, driving onto curbs, or causing collisions. As a result, some users are reluctant to use autonomous driving functions, which is detrimental to the development of intelligent driving.

[0003] Conventional technology can acquire the distance between a target vehicle and an obstacle based on sensor data, and can also perform collision detection analysis and early warning. However, the data transmitted from the sensors has inaccuracies and delays, making it impossible to perform collision detection quickly. Furthermore, the sensors cannot perform appropriate collision detection for different vehicles based on the height of the obstacle, so the target vehicle may collide with an obstacle while in motion, and in serious cases, passengers and occupants may be killed or injured. [Overview of the project] [Problems that the invention aims to solve]

[0004] This invention addresses the technical problems of the prior art, where data transmitted from sensors is inaccurate and delayed, making it impossible to perform collision detection promptly, and sensors cannot perform appropriate collision detection for different vehicles based on the height of the obstacle, resulting in the target vehicle colliding with an obstacle while in motion. To solve these problems, this invention provides a method and system that can perform vehicle collision detection based on the height of an obstacle. [Means for solving the problem]

[0005] Specifically, this application provides a vehicle collision detection method that includes the following steps. S100 acquires a set of obstacle information within a preset range of the target vehicle, and divides the set of obstacle information into a first obstacle map and a second obstacle map based on the preset height of the obstacles.

[0006] In S200, the target vehicle is modeled using the first obstacle map and the second obstacle map, and the vehicle model is fused with the corresponding obstacle map to obtain a first collision detection model and a second collision detection model, respectively.

[0007] S300, wherein the first collision detection model and / or the second collision detection model obtain the distance relationship with an obstacle corresponding to a set of obstacle information at a preset position of the target vehicle, and determine whether the target vehicle will collide based on the said distance relationship.

[0008] The above technical solution divides the obstacle map based on the height of the obstacles, establishes a corresponding collision detection model, determines the distance relationship between the target vehicle and the obstacle using the collision detection model, and then obtains the collision risk. This enables real-time and accurate collision risk analysis for different types of obstacle heights, improving the accuracy and efficiency of collision detection and further guaranteeing the driving safety of the vehicle and the personal safety of those inside the vehicle.

[0009] Furthermore, in step S100, dividing the obstacle information set into a first obstacle map and a second obstacle map based on a preset obstacle height includes the following: The maximum lengths of the front and rear wheels of the target vehicle are calibrated, and based on the calibration results, a preset obstacle height is set.

[0010] The system determines whether the height of an obstacle corresponding to the obstacle information set is greater than the height of a pre-set obstacle. If it is greater, the obstacle is placed in the first obstacle map; otherwise, the obstacle is placed in the second obstacle map.

[0011] In the above technical solution, the height of pre-set obstacles is calibrated based on the lengths of the vehicle's front and rear wheels, and obstacles of different heights are placed in different maps. This makes the collision detection analysis more directional, further improving the accuracy and efficiency of collision detection.

[0012] Furthermore, step S200 includes the following: By modeling the target vehicle, a first vehicle model including the vehicle's front overhang and rear overhang is obtained, and a second vehicle model with the front overhang and rear overhang removed is obtained.

[0013] The first vehicle model and the second vehicle model are merged with the first obstacle map and the second obstacle map, respectively, to obtain the first collision detection model and the second collision detection model.

[0014] A vehicle model is obtained, and by fusing the vehicle model with a corresponding obstacle map, a suitable collision detection model is obtained for use in subsequent collision detection analysis.

[0015] In the above technical solution, by calibrating based on the maximum lengths of the front and rear wheels of the target vehicle, the model and collision detection model corresponding to each vehicle are unique, thereby making collision detection more individualized and intelligent.

[0016] Furthermore, before performing step S300, The method includes the steps of establishing a first Cartesian coordinate system with the geometric center of the target vehicle as the first origin, the width direction of the vehicle as the x-axis, and the length direction as the y-axis, and establishing a second Cartesian coordinate system with the midpoint of the rear axle of the target vehicle as the second origin, the width direction of the vehicle as the X-axis, and the length direction as the Y-axis.

[0017] In the above technical solution, establishing the first and second orthogonal coordinate systems not only ensures that the positions of obstacles are reliably digitized, but also makes it easier to later determine the first and second pre-defined circular regions.

[0018] Furthermore, step S300 includes the following. S301, in the first collision detection model, obtain the initial obstacle coordinates of the obstacle in the first orthogonal coordinate system.

[0019] S302, based on the initial obstacle coordinates and the coordinates of the first origin, calculate the distance of each obstacle to the first origin.

[0020] If the distance of at least one obstacle to the first origin is less than the first preset distance, execute step S303; otherwise, execute step S305.

[0021] In the above technical solution, first, obstacles with a collision risk are screened, and after narrowing down the range, the collision risk is further detected.

[0022] S303, obtain the converted obstacle coordinates of the obstacle within the first preset circular region in the second orthogonal coordinate system, and proceed to step S304.

[0023] In the above technical solution, the converted obstacle coordinates in the second orthogonal coordinate system are used to mitigate the influence of the vehicle's yaw angle on the obstacle coordinates, thereby making the acquisition result of the distance between the obstacle and the target vehicle more accurate and further improving the accuracy of collision detection.

[0024] S304, based on the converted obstacle coordinates and the coordinates of the second origin, calculate the distance of each obstacle within the first preset circular region to the second origin.

[0025] If the distance of at least one obstacle within the first preset circular region to the second origin is less than the second preset distance, it is determined that the target vehicle will collide; otherwise, execute step S305.

[0026] The second predetermined distance is generally preferably half the width of the vehicle, the length of the vehicle's rear overhang, or the distance from the rear axle to the front overhang. This further limits the obstacles that could cause a collision, thereby determining whether the target vehicle is at risk of collision.

[0027] It is determined that the vehicle in question is S305 and therefore does not pose a collision risk.

[0028] The first predetermined circular region has the first origin as its center and the first predetermined distance as its radius.

[0029] Furthermore, step S300 further includes the following: S311, in the second collision detection model, the initial obstacle coordinates in the first Cartesian coordinate system of the obstacle are obtained.

[0030] S312, the distance of each obstacle to the first origin is calculated based on the initial obstacle coordinates and the coordinates of the first origin.

[0031] If the distance of at least one obstacle to the first origin is less than a third preset distance, step S313 is performed; otherwise, step S315 is performed.

[0032] In step S313, the transformed obstacle coordinates in the second Cartesian coordinate system of the obstacle within the second preset circular region are obtained, and the process proceeds to step S314.

[0033] S314, the distance from each obstacle within the second preset circular region to the second origin is calculated based on the coordinates of the transformed obstacle and the coordinates of the second origin.

[0034] If the distance of at least one obstacle within the second preset circular region to the second origin is less than the second preset distance, it is determined that the target vehicle will collide with it; otherwise, step S315 is executed.

[0035] The condition is S315, and it is determined that there is no collision risk for the vehicle in question.

[0036] The second predetermined circular region has the first origin as its center and the third predetermined distance as its radius.

[0037] Furthermore, step S303 or step S313 specifically includes the following: The yaw angle of the vehicle, the corresponding initial obstacle coordinates in the first Cartesian coordinate system for obstacles within a first preset circular region or a second preset circular region, and the midpoint coordinates of the vehicle's rear axle in the first Cartesian coordinate system are obtained.

[0038] Based on the initial obstacle coordinates and the midpoint coordinates of the vehicle's rear axle, a first included angle and the distance of the obstacle to the midpoint of the vehicle's rear axle are calculated. The first included angle is the angle of the line connecting the obstacle and the midpoint of the vehicle's rear axle with respect to the line segment of the obstacle perpendicular to the X-axis of the second Cartesian coordinate system.

[0039] A second included angle is calculated based on the first included angle and the yaw angle of the vehicle.

[0040] Based on the distance of the obstacle to the midpoint of the rear axle of the vehicle and the second included angle, the transformed obstacle coordinates of the obstacle are calculated.

[0041] The above technical solution avoids the influence of the vehicle's yaw angle on the converted obstacle coordinates, which helps to further improve the accuracy of subsequent collision detection.

[0042] Furthermore, if it is determined that the target vehicle will collide, step S304 or S314 further includes a step of issuing an early collision warning.

[0043] The above technical solution involves issuing early collision warning information to promptly promptly take emergency measures, thereby reducing the likelihood of an accident occurring, improving driving safety, and protecting the safety of the vehicle and its passengers.

[0044] This application also provides a vehicle collision detection system based on the same concept, the system including the following: A first acquisition module, used to acquire a set of obstacle information within a predetermined range of the target vehicle.

[0045] A division module used to divide the obstacle information set into a first obstacle map and a second obstacle map based on a predetermined obstacle height.

[0046] A second acquisition module is used to model a target vehicle using the first obstacle map and the second obstacle map, and to merge the vehicle model with the corresponding obstacle map to obtain a first collision detection model and a second collision detection model.

[0047] A decision module is used to determine whether a collision will occur, based on the distance relationship between the target vehicle and an obstacle corresponding to a set of obstacle information at a preset position of the target vehicle, using the first collision detection model and / or the second collision detection model.

[0048] Furthermore, the system further includes the following: An establishment module used to establish a first orthogonal coordinate system and a second orthogonal coordinate system, wherein the first orthogonal coordinate system has the geometric center of the target vehicle as the first origin, the width direction of the vehicle as the x-axis, and the length direction as the y-axis, and the second orthogonal coordinate system has the midpoint of the rear axle of the target vehicle as the second origin, the width direction of the vehicle as the X-axis, and the length direction as the Y-axis.

[0049] This is an early warning module used to issue collision early warning information when it is determined that the target vehicle is in danger of colliding.

[0050] Furthermore, the divided module includes the following: Calibration units are used to calibrate the maximum lengths of the front and rear wheels of the vehicle in question.

[0051] A setting unit used to set a preset obstacle height based on the calibration result of the calibration unit.

[0052] A splitting unit is used to determine whether the height of an obstacle corresponding to an obstacle information set is greater than the height of a preset obstacle. If it is greater, the obstacle is placed in a first obstacle map; if it is less, the obstacle is placed in a second obstacle map.

[0053] Furthermore, the second acquisition module includes the following: A modeling unit used to obtain a first vehicle model and a second vehicle model.

[0054] A fusion unit is used to obtain a first collision detection model and a second collision detection model by fusing the first vehicle model and the second vehicle model with the first obstacle map and the second obstacle map, respectively.

[0055] Furthermore, the decision module includes the following: An acquisition unit used in a first collision detection model or a second collision detection model to acquire the initial obstacle coordinates of an obstacle in a first orthogonal coordinate system.

[0056] A first calculation unit is used to calculate the distance of each obstacle to the first origin based on the initial obstacle coordinates and the coordinates of the first origin, and to obtain the transformed obstacle coordinates in the second Cartesian coordinate system of obstacles within a second preset circular region if the distance is less than a first preset distance or a third preset distance.

[0057] A second calculation unit is used to calculate the distance from each obstacle within the second preset circular area to the second origin based on the coordinates of the transformed obstacle and the coordinates of the second origin, and to determine that the target vehicle will collide if the distance is less than the second preset distance.

[0058] The system described above can perform accurate collision risk analysis for different types of obstacle heights, improving the accuracy of collision detection and further ensuring the safety of the vehicle while driving and the personal safety of those inside.

[0059] This invention also provides, based on the same concept, a vehicle equipped with a vehicle collision detection system, the system using the vehicle collision detection method to detect whether the target vehicle will collide with another vehicle. [Effects of the Invention]

[0060] Compared to conventional technology, the beneficial effects of this invention are as follows: This invention divides an obstacle information set into a first obstacle map and a second obstacle map based on the height of the obstacle, establishes corresponding first and second collision detection models, determines the distance relationship between the target vehicle and the obstacle using the first and / or second collision detection models, and obtains the risk of the target vehicle colliding with the obstacle. This invention can perform accurate collision risk analysis for different types of obstacle heights, improving the accuracy of collision detection and further guaranteeing the driving safety of the vehicle and the personal safety of those inside the vehicle. [Brief explanation of the drawing]

[0061] [Figure 1] This is a flowchart of the vehicle collision detection method described in this application. [Figure 2] This is a flowchart illustrating the method for performing collision detection using the first collision detection model shown in Figure 1. [Figure 3] This is a flowchart illustrating the method for performing collision detection using the second collision detection model shown in Figure 1. [Figure 4] Figure 1 is a system configuration diagram of the vehicle collision detection method. [Modes for carrying out the invention]

[0062] This invention provides a vehicle collision detection method, system, and vehicle, thereby solving the technical problems of the conventional technology, which suffers from inaccurate and inefficient collision detection due to the use of sensors to detect the distance between a vehicle and an obstacle, and the inability to perform appropriate collision detection for different vehicles based on the height of the obstacle.

[0063] This application provides a vehicle collision detection method, system, and vehicle, which are generally as follows: The system obtains a set of obstacle information within a predetermined range of the target vehicle, determines whether the height of an obstacle corresponding to the set of obstacle information is greater than the predetermined height of an obstacle, places the obstacle in the first obstacle map if it is greater, and places the obstacle in the second obstacle map if it is not. The system then models and fuses the target vehicle using the first and second obstacle maps to obtain a first collision detection model and a second collision detection model, respectively. Furthermore, the system obtains the distance relationship between the obstacle corresponding to the set of obstacle information at the geometric center of the target vehicle using the first and / or second collision detection models. If the distance of at least one obstacle to the geometric center is less than the first predetermined distance, the system obtains the distance relationship between the obstacle at the midpoint of the rear axle of the target vehicle, and determines that the target vehicle will collide if the distance of at least one obstacle to the midpoint of the rear axle of the target vehicle is less than the second predetermined distance.

[0064] The vehicle collision detection method, system, and vehicle of this application will be described in more detail below, based on specific embodiments and drawings.

[0065] (Example 1) Referring to Figure 1, the present invention provides a vehicle collision detection method comprising the following steps. S100 acquires a set of obstacle information within a preset range of the target vehicle, and divides the set of obstacle information into a first obstacle map and a second obstacle map based on the preset height of the obstacles.

[0066] Furthermore, in step S100, dividing the obstacle information set into a first obstacle map and a second obstacle map based on a preset obstacle height includes the following: The maximum lengths of the front and rear wheels of the target vehicle are calibrated, and based on the calibration results, a preset obstacle height is set.

[0067] The system determines whether the height of an obstacle corresponding to the obstacle information set is greater than the height of a pre-set obstacle. If it is greater, the obstacle is placed in the first obstacle map; otherwise, the obstacle is placed in the second obstacle map.

[0068] Since the chassis height and vehicle height vary depending on the vehicle, the aforementioned pre-set obstacle height is calibrated based on the actual vehicle, and therefore, the value of the pre-set obstacle height is not specifically limited here.

[0069] Furthermore, the first obstacle map described above is a map in which the vehicle body must not enter, and the second obstacle map described above is a map in which the vehicle body may enter. Here, "may be allowed to enter" means that the vehicle's front overhang or rear overhang may enter due to the presence of a limiter or low curb, etc.

[0070] Once the division of the first obstacle map and the second obstacle map is complete, step S200 can be executed.

[0071] In S200, the target vehicle is modeled using the first obstacle map and the second obstacle map, and the vehicle model is fused with the corresponding obstacle map to obtain a first collision detection model and a second collision detection model, respectively.

[0072] Furthermore, step S200 is, The process includes the step of modeling the target vehicle to obtain a first vehicle model including the vehicle's front overhang and rear overhang, and a second vehicle model in which the vehicle's front overhang and rear overhang have been removed.

[0073] Note that the vehicle model is not a three-dimensional model, but a two-dimensional planar model of the vehicle's outline around its four sides. The first vehicle model is a rounded rectangle representing the outline of the entire vehicle, including the front and rear overhangs, with the front and rear wheels each represented by four corresponding rectangles. The second vehicle model is a rectangular outline diagram with the front and rear overhangs removed, and the corners of this outline diagram represent the positions of the wheels.

[0074] The first vehicle model and the second vehicle model are merged with the first obstacle map and the second obstacle map, respectively, to obtain the first collision detection model and the second collision detection model.

[0075] The first or second collision detection model can represent the position of an obstacle in the obstacle information set relative to a corresponding vehicle model, wherein the obstacle is represented in the form of data points.

[0076] Once the acquisition of the first and second collision detection models is complete, step S300 can be executed.

[0077] S300, wherein the first collision detection model and / or the second collision detection model obtain the distance relationship with an obstacle corresponding to a set of obstacle information at a preset position of the target vehicle, and determine whether the target vehicle will collide based on the said distance relationship.

[0078] Before performing step S300, Establish a first rectangular coordinate system with the geometric center of the target vehicle as the first origin, the width direction of the vehicle as the x-axis, and the length direction as the y-axis. Also, include the step of establishing a second rectangular coordinate system with the midpoint of the rear axle of the target vehicle as the second origin, the width direction of the vehicle as the X-axis, and the length direction as the Y-axis.

[0079] Furthermore, referring to FIG. 2, step S300 includes the following. S301, in the first collision detection model, obtain the initial obstacle coordinates (x , ,

[0080] , 2 , , 0i , , i ,

[0082] , 2 , , i ,

[0083] , i , 0i , , i , ,

[0081] , y i ) of the obstacle in the first rectangular coordinate system.

[0080] S302, based on the initial obstacle coordinates (x i ) , y i ) and the coordinates (x0, y0) of the first origin, calculate the distance L 0i from each obstacle to the first origin, where L 0i = √((x i - x0) 2 + (y i - y0) 2 ).

[0081] If the distance from at least one obstacle to the first origin is less than the first preset distance, execute step S303; otherwise, execute step S305.

[0082] Note that the first preset distance is the distance from the geometric center of the target vehicle to any corner of the vehicle. Here, since the geometric center of the vehicle is selected, the distance to any corner of the vehicle from this geometric center is the same, and in practice, it can be arbitrarily selected. For example, measure the distance from the geometric center of the target vehicle to the right front corner of the vehicle, and define this distance as the first preset distance. [[ID=​When using a vehicle collision detection method in the process of automatic parking, the positional relationship of an obstacle to the target vehicle can be determined based on the distance from the vehicle's geometric center to any corner of the vehicle. This allows for screening of obstacles that pose a collision risk during the automatic parking process, for example, during forward movement, reverse movement, or turning.

[0084] S303, the transformed obstacle coordinates in the second Cartesian coordinate system of the obstacle within the first preset circular region are obtained.

[0085] Here, the first predetermined circular region has the first origin as its center and the first predetermined distance as its radius.

[0086] Furthermore, step S303 specifically means, The yaw angle θ of the vehicle, and the corresponding initial obstacle coordinates (x) of the obstacle in the first Cartesian coordinate system within the first predetermined circular region. n ,y n ), and the step of obtaining the midpoint coordinates (x,y) of the rear axle of the vehicle in the first Cartesian coordinate system.

[0087] Based on the initial obstacle coordinates and the midpoint coordinates of the vehicle's rear axle, the first included angle θ1 and the distance L from the obstacle to the midpoint of the vehicle's rear axle are calculated.

[0088] Here, θ1 = arctan(((x n -x) / (y n -y)), L=√((x n -x) 2 +(y n -y) 2 )

[0089] The first included angle θ1 is the angle between the line connecting the obstacle and the midpoint of the vehicle's rear axle and the line segment of the obstacle perpendicular to the X-axis of the second Cartesian coordinate system.

[0090] Based on the first included angle and the yaw angle of the vehicle, the second included angle θ2 is calculated, where θ2 = θ1 - θ.

[0091] Based on the distance L from the obstacle to the midpoint of the rear axle of the vehicle and the second included angle θ2, the transformed obstacle coordinates (x m ,y m ) calculate, where x m =L×sin(θ2), y m = L × cos(θ²).

[0092] S304, the distance from each obstacle within the first preset circular region to the second origin is calculated based on the coordinates of the transformed obstacle and the coordinates of the second origin.

[0093] If the distance of at least one obstacle within the first preset circular region to the second origin is less than the second preset distance, it is determined that the target vehicle will collide with it; otherwise, step S305 is executed.

[0094] The second predetermined distance is generally preferably half the width of the vehicle, the length of the vehicle's rear overhang, or the distance from the vehicle's rear axle to the front overhang.

[0095] x m If it is less than half the width of the vehicle, or y m If it is less than the length of the vehicle's rear overhang, or y m If the distance is less than the distance from the rear axle of the vehicle to the front overhang, it is determined that the vehicle in question will collide.

[0096] Furthermore, if it is determined that the target vehicle is in danger of colliding, S304 further includes the step of issuing an early collision warning.

[0097] Here, the system may issue real-time early collision warning information via an in-vehicle display or voice notification, and at the same time, it may transmit the early warning information to the driver's mobile phone or IoT platform for connected cars to alert the driver to take emergency measures, thereby reducing the occurrence of collision accidents.

[0098] It is determined that the vehicle in question is S305 and therefore does not pose a collision risk.

[0099] (Example 2) Referring to Figure 3, the present invention also provides a modified example in which collision detection is performed by the second collision detection model for an obstacle that may be driven into, wherein step S300 further includes the following: S311, in the second collision detection model, the initial obstacle coordinates in the first Cartesian coordinate system of the obstacle are obtained.

[0100] S312, the distance of each obstacle to the first origin is calculated based on the initial obstacle coordinates and the coordinates of the first origin.

[0101] If the distance of at least one obstacle to the first origin is less than a third preset distance, step S313 is performed; otherwise, step S315 is performed.

[0102] The third predetermined distance is the distance from the geometric center of the target vehicle to either the front axle or the rear axle side of the vehicle.

[0103] S313, the transformed obstacle coordinates in the second Cartesian coordinate system of the obstacle within the second preset circular region are obtained.

[0104] Here, the second predetermined circular region has the first origin as its center and the third predetermined distance as its radius.

[0105] Furthermore, step S313 specifically includes the following: The yaw angle of the vehicle, the corresponding initial obstacle coordinates in the first Cartesian coordinate system for obstacles within a second preset circular region, and the midpoint coordinates of the vehicle's rear axle in the first Cartesian coordinate system are obtained.

[0106] Based on the initial obstacle coordinates and the midpoint coordinates of the vehicle's rear axle, the first included angle and the distance of the obstacle to the midpoint of the vehicle's rear axle are calculated.

[0107] The first included angle is the angle between the line connecting the obstacle and the midpoint of the vehicle's rear axle and the line segment of the obstacle perpendicular to the X-axis of the second Cartesian coordinate system.

[0108] A second included angle is calculated based on the first included angle and the yaw angle of the vehicle.

[0109] Based on the distance of the obstacle to the midpoint of the rear axle of the vehicle and the second included angle, the transformed obstacle coordinates of the obstacle are calculated.

[0110] S314, the distance from each obstacle within the second preset circular region to the second origin is calculated based on the coordinates of the transformed obstacle and the coordinates of the second origin.

[0111] If the distance of at least one obstacle within the second preset circular region to the second origin is less than the second preset distance, it is determined that the target vehicle will collide with it; otherwise, step S315 is executed.

[0112] Furthermore, if it is determined that the target vehicle is in danger of colliding, S314 further includes the step of issuing an early collision warning.

[0113] The condition is S315, and it is determined that there is no collision risk for the vehicle in question.

[0114] (Example 3) Referring to Figure 4, the present invention also provides a vehicle collision detection system, the system including the following: A first acquisition module, used to acquire a set of obstacle information within a predetermined range of the target vehicle.

[0115] In one possible embodiment, the obstacle information set may be acquired by a target sensor, and the target sensor may be a sensor corresponding to a laser radar and a stereo camera. The laser radar may be installed on both sides of the bottom of the vehicle, symmetrically from left to right or front to back, thereby acquiring information on obstacles that are far away, low in position, and require high detection accuracy. The stereo camera may be installed on the top of the vehicle, thereby acquiring information on obstacles that are close and high in position. By using the laser radar and stereo camera in combination, the acquired obstacle information set becomes more comprehensive and accurate, thereby enabling more accurate detection when collision detection is performed later, and preventing collision accidents from occurring due to missed obstacle detection.

[0116] A division module used to divide the obstacle information set into a first obstacle map and a second obstacle map based on a predetermined obstacle height.

[0117] The aforementioned divided module includes the following: Calibration units are used to calibrate the maximum lengths of the front and rear wheels of the vehicle in question.

[0118] A setting unit used to set a preset obstacle height based on the calibration result of the calibration unit.

[0119] Since the chassis height and vehicle height vary depending on the vehicle, the aforementioned pre-set obstacle heights are calibrated based on the actual vehicle and are not limited thereto.

[0120] A splitting unit is used to determine whether the height of an obstacle corresponding to an obstacle information set is greater than the height of a preset obstacle. If it is greater, the obstacle is placed in a first obstacle map; if it is less, the obstacle is placed in a second obstacle map.

[0121] A second acquisition module is used to model a target vehicle using the first obstacle map and the second obstacle map, and to merge the vehicle model with the corresponding obstacle map to obtain a first collision detection model and a second collision detection model.

[0122] The second acquisition module includes the following: A modeling unit used to obtain a first vehicle model and a second vehicle model.

[0123] A fusion unit is used to obtain a first collision detection model and a second collision detection model by fusing the first vehicle model and the second vehicle model with the first obstacle map and the second obstacle map, respectively.

[0124] A decision module is used to determine whether a collision will occur, based on the distance relationship between the target vehicle and an obstacle corresponding to a set of obstacle information at a preset position of the target vehicle, using the first collision detection model and / or the second collision detection model.

[0125] The aforementioned system further includes the following: An establishment module used to establish a first orthogonal coordinate system and a second orthogonal coordinate system, wherein the first orthogonal coordinate system has the geometric center of the target vehicle as the first origin, the width direction of the vehicle as the x-axis, and the length direction as the y-axis, and the second orthogonal coordinate system has the midpoint of the rear axle of the target vehicle as the second origin, the width direction of the vehicle as the X-axis, and the length direction as the Y-axis.

[0126] The aforementioned decision module includes the following: An acquisition unit used in a first collision detection model or a second collision detection model to acquire the initial obstacle coordinates of an obstacle in a first orthogonal coordinate system.

[0127] A first calculation unit is used to calculate the distance of each obstacle to the first origin based on the initial obstacle coordinates and the coordinates of the first origin, and to obtain the transformed obstacle coordinates in the second Cartesian coordinate system of obstacles within a second preset circular region if the distance is less than a first preset distance or a third preset distance.

[0128] A second calculation unit is used to calculate the distance from each obstacle within the second preset circular area to the second origin based on the coordinates of the transformed obstacle and the coordinates of the second origin, and to determine that the target vehicle will collide if the distance is less than the second preset distance.

[0129] The aforementioned system further includes the following: This is an early warning module used to issue collision early warning information when it is determined that the target vehicle is in danger of colliding.

[0130] (Example 4) The present invention also provides a vehicle equipped with a vehicle collision detection system, the system using the vehicle collision detection method to detect whether the target vehicle will collide with another vehicle.

[0131] In summary, this invention provides a vehicle collision detection method, system, and vehicle, which divides an obstacle map based on the height of the obstacle, establishes a corresponding collision detection model, determines the distance relationship between the target vehicle and the obstacle using the collision detection model, and obtains the collision risk. It can perform real-time and accurate collision risk analysis for different types of obstacle heights, improving the accuracy and efficiency of collision detection, and further guaranteeing the driving safety of the vehicle and the personal safety of those inside the vehicle.

[0132] Here, exemplary embodiments are described with reference to the drawings, but it should be understood that these embodiments are merely illustrative and are not intended to limit the scope of this application. A person skilled in the art may make various changes and modifications thereto without departing from the scope and spirit of this application. All such changes and modifications shall be included within the scope of this application as asserted in the claims.

[0133] Those skilled in the art will recognize that each exemplary unit and algorithmic step described in the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design limitations relating to the technical solution. Those skilled in the art may implement the functions described using different methods for each specific application, but such implementations should not be considered beyond the scope of this application.

[0134] It should be understood that the devices and methods disclosed in some embodiments provided herein may be implemented in other ways. For example, the embodiments of the devices described above are merely illustrative, and the division of the units is merely a logical functional division and may be divided in other ways in actual implementation, for example, multiple units or components may be combined or integrated into another device, or some features may be omitted or not implemented at all.

[0135] Embodiments relating to each component of the present application may be implemented in hardware, or in software modules running on one or more processors, or in combination thereof. Those skilled in the art will understand that, in practice, some or all of the functions of some of the modules based on the embodiments of the present application may be implemented using a microprocessor or a digital signal processor (DSP). The present application may be implemented as an apparatus program (e.g., a computer program, a computer program product) used to perform some or all of the methods described herein. Such a program implementing the present application may be stored on a computer-readable medium, or may comprise one or more signals. Such signals may be obtained by downloading from a website on the Internet, or may be provided in carrier signals, or may be provided in any other form.

[0136] In this specification, relational terms such as "First," "Second," etc., are merely used to distinguish one entity or operation from another, and do not necessarily stipulate or suggest that such a relationship or order actually exists between these entities or operations. Furthermore, the term "includes" or any other variation is intended to cover cases of non-exclusive inclusion, thereby meaning that a process, method, article, or device containing a set of elements includes not only those elements but also other elements not explicitly listed, or elements that are unique to such process, method, article, or device. Unless further limited, an element limited by the phrase "includes one..." does not preclude the existence of other identical elements in a process, method, article, or device containing such element.

[0137] Although the description in this application is based on the specific embodiments described above, it is obvious to those skilled in the art that various substitutions, modifications, and changes may be made based on the above content. Therefore, all such substitutions, improvements, and changes shall be included in the intent and scope of the claims.

Claims

1. Step S100, comprising: acquiring a set of obstacle information within a preset range of the target vehicle; and dividing the set of obstacle information into a first obstacle map and a second obstacle map based on the preset height of the obstacles; Step S200 is the step of modeling the target vehicle using the first obstacle map and the second obstacle map, and fusing the vehicle model with the corresponding obstacle map to obtain a first collision detection model and a second collision detection model, respectively. Step S300 includes obtaining a distance relationship between the target vehicle and an obstacle corresponding to a set of obstacle information at a preset position of the target vehicle using the first collision detection model and / or the second collision detection model, and determining whether the target vehicle will collide based on the said distance relationship. In step S100, the obstacle information set is divided into a first obstacle map and a second obstacle map based on the predetermined height of the obstacles. Each step involves calibrating the maximum dimensions of the front and rear wheels of the target vehicle, and then setting a predetermined obstacle height based on the calibration results. The process includes the steps of determining whether the height of an obstacle corresponding to the obstacle information set is greater than the height of a preset obstacle, and if it is greater, placing the obstacle in the first obstacle map, and otherwise placing the obstacle in the second obstacle map. A vehicle collision detection method characterized by the following features.

2. The aforementioned step S200 is, The steps include: obtaining a first vehicle model including the front overhang and rear overhang of the vehicle, and a second vehicle model in which the front overhang and rear overhang of the vehicle are removed, by modeling the target vehicle; The vehicle collision detection method according to claim 1, characterized by comprising the step of fusing the first vehicle model and the second vehicle model with the first obstacle map and the second obstacle map, respectively, to obtain a first collision detection model and a second collision detection model.

3. Before executing step S300, The vehicle collision detection method according to claim 2, characterized by including the step of establishing a first orthogonal coordinate system in which the geometric center of the target vehicle is the first origin, the width direction of the vehicle is the x-axis, and the length direction is the y-axis.

4. Before performing step S300, further, The vehicle collision detection method according to claim 3, characterized by including the step of establishing a second orthogonal coordinate system in which the midpoint of the rear axle of the target vehicle is the second origin, the width direction of the vehicle is the X-axis, and the length direction is the Y-axis.

5. The above step S300 is, Step S301, comprising the step of obtaining the initial obstacle coordinates in a first Cartesian coordinate system of the obstacle in the first collision detection model, Step S302, the distance of each obstacle to the first origin is calculated based on the initial obstacle coordinates and the coordinates of the first origin. If the distance of at least one obstacle to the first origin is less than a first preset distance, step S303 is performed; otherwise, step S305 is performed. Step S303 is the step of obtaining the transformed obstacle coordinates in the second Cartesian coordinate system of the obstacle within the first preset circular region, and proceeding to step S304, Step S304, based on the coordinates of the transformed obstacle and the coordinates of the second origin, calculate the distance of each obstacle within the first preset circular region to the second origin, If the distance of at least one obstacle within the first preset circular region to the second origin is less than the second preset distance, it is determined that the target vehicle will collide with it; otherwise, step S305 is executed. Step S305 includes the step of determining that there is no collision risk for the target vehicle, The vehicle collision detection method according to claim 4, characterized in that the first preset circular region has the first origin as its center and the first preset distance as its radius.

6. The above step S300 is, Step S311, the second collision detection model, includes the step of obtaining the initial obstacle coordinates in the first Cartesian coordinate system of the obstacle, Step S312, the distance of each obstacle to the first origin is calculated based on the initial obstacle coordinates and the coordinates of the first origin. If the distance of at least one obstacle to the first origin is less than a third preset distance, step S313 is performed; otherwise, step S315 is performed. Step S313 is the step of obtaining the transformed obstacle coordinates in the second Cartesian coordinate system of the obstacle within the second preset circular region, and proceeding to step S314, Step S314, based on the coordinates of the transformed obstacle and the coordinates of the second origin, calculate the distance of each obstacle within the second preset circular region to the second origin, If the distance of at least one obstacle within the second preset circular region to the second origin is less than the second preset distance, it is determined that the target vehicle will collide with it; otherwise, step S315 is executed. Step S315 includes the step of determining that there is no collision risk for the target vehicle, The vehicle collision detection method according to claim 4, characterized in that the second preset circular region has the first origin as its center and the third preset distance as its radius.

7. Specifically, step S303 or step S313 is: Steps include obtaining the yaw angle of the vehicle, the corresponding initial obstacle coordinates in the first Cartesian coordinate system of an obstacle within a first preset circular region or a second preset circular region, and the midpoint coordinates of the vehicle's rear axle in the first Cartesian coordinate system. A step of calculating a first included angle and the distance of the obstacle to the midpoint of the vehicle's rear axle based on the initial obstacle coordinates and the midpoint coordinates of the vehicle's rear axle, wherein the first included angle is the included angle of the line connecting the obstacle and the midpoint of the vehicle's rear axle with respect to the line segment of the obstacle perpendicular to the X-axis of the second Cartesian coordinate system, A step of calculating a second included angle based on the first included angle and the yaw angle of the vehicle, A vehicle collision detection method according to claim 5 or 6, characterized by comprising the step of calculating the transformed obstacle coordinates of the obstacle based on the distance of the obstacle to the midpoint of the rear axle of the vehicle and a second included angle.

8. The vehicle collision detection method according to claim 7, characterized in that when it is determined that the target vehicle is going to collide, step S304 or step S314 further includes a step of issuing early collision warning information.

9. Step S100, comprising: acquiring a set of obstacle information within a preset range of the target vehicle, and dividing the set of obstacle information into a first obstacle map and a second obstacle map based on the preset height of the obstacles; Step S200 is the step of modeling the target vehicle using the first obstacle map and the second obstacle map, and fusing the vehicle model with the corresponding obstacle map to obtain a first collision detection model and a second collision detection model, respectively. Step S300 includes obtaining a distance relationship between the target vehicle and an obstacle corresponding to a set of obstacle information at a preset position of the target vehicle using the first collision detection model and / or the second collision detection model, and determining whether the target vehicle will collide based on the said distance relationship. Before performing step S300, the first step is to establish a first orthogonal coordinate system in which the geometric center of the target vehicle is the first origin, the width direction of the vehicle is the x-axis, and the length direction is the y-axis. The process includes the step of establishing a second Cartesian coordinate system with the midpoint of the rear axle of the target vehicle as the second origin, the width direction of the vehicle as the X-axis, and the length direction as the Y-axis, The above step S300 is, Step S301, comprising the step of obtaining the initial obstacle coordinates in a first Cartesian coordinate system of the obstacle in the first collision detection model, Step S302, the distance of each obstacle to the first origin is calculated based on the initial obstacle coordinates and the coordinates of the first origin. If the distance of at least one obstacle to the first origin is less than a first preset distance, step S303 is performed; otherwise, step S305 is performed. Step S303 is the step of obtaining the transformed obstacle coordinates in the second Cartesian coordinate system of the obstacle within the first preset circular region, and proceeding to step S304, Step S304, based on the coordinates of the transformed obstacle and the coordinates of the second origin, calculate the distance of each obstacle within the first preset circular region to the second origin, If the distance of at least one obstacle within the first preset circular region to the second origin is less than the second preset distance, it is determined that the target vehicle will collide with it; otherwise, step S305 is executed. Step S305 includes the step of determining that there is no collision risk for the target vehicle, A vehicle collision detection method characterized in that the first preset circular region has the first origin as its center and the first preset distance as its radius.

10. A first acquisition module used to acquire a set of obstacle information within a pre-set range of the target vehicle, A division module used to divide the obstacle information set into a first obstacle map and a second obstacle map based on pre-set obstacle heights, A second acquisition module is used to model the target vehicle using the first obstacle map and the second obstacle map, and to merge the vehicle model with the corresponding obstacle map to obtain a first collision detection model and a second collision detection model. A vehicle collision detection system comprising a first collision detection model and / or a second collision detection model, which obtains a distance relationship with an obstacle corresponding to a set of obstacle information at a preset position of the target vehicle, and a determination module used to determine whether the target vehicle will collide based on said distance relationship, The aforementioned system An establishment module used to establish a first orthogonal coordinate system and a second orthogonal coordinate system, wherein the first orthogonal coordinate system has the geometric center of the target vehicle as the first origin, the width direction of the vehicle as the x-axis, and the length direction as the y-axis, and the second orthogonal coordinate system has the midpoint of the rear axle of the target vehicle as the second origin, the width direction of the vehicle as the X-axis, and the length direction as the Y-axis, A vehicle collision detection system further comprising an early warning module used to issue early collision warning information when it is determined that a target vehicle is in danger of colliding.

11. The aforementioned split module Each of these consists of a calibration unit used to calibrate the maximum dimensions of the front and rear wheels of the vehicle in question, A setting unit used to set a preset obstacle height based on the calibration result of the calibration unit, The vehicle collision detection system according to claim 10, characterized in that it includes a splitting unit used to determine whether the height of an obstacle corresponding to an obstacle information set is greater than the height of a preset obstacle, and if it is greater, to place the obstacle in a first obstacle map, and if it is less than that, to place the obstacle in a second obstacle map.

12. The second acquisition module described above, A modeling unit used to obtain a first vehicle model and a second vehicle model, The vehicle collision detection system according to claim 11, characterized in that it includes a fusion unit used to fuse the first vehicle model and the second vehicle model with the first obstacle map and the second obstacle map, respectively, in order to obtain a first collision detection model and a second collision detection model.

13. The aforementioned decision module An acquisition unit used to acquire the initial obstacle coordinates of an obstacle in a first orthogonal coordinate system in a first collision detection model or a second collision detection model, A first calculation unit used to calculate the distance of each obstacle to the first origin based on the initial obstacle coordinates and the coordinates of the first origin, and to obtain the transformed obstacle coordinates in the second Cartesian coordinate system of obstacles within a second preset circular region if the distance is smaller than a first preset distance or a third preset distance, The vehicle collision detection system according to claim 12, further comprising: a second calculation unit used to calculate the distance from each obstacle within the second preset circular region to the second origin based on the transformed obstacle coordinates and the coordinates of the second origin, and to determine that a collision will occur if the distance is less than the second preset distance.

14. A vehicle characterized by being equipped with a vehicle collision detection system that detects whether a target vehicle will collide using the vehicle collision detection method described in any one of claims 1 to 6, or any one of claims 8 and 9.

Images