A vehicle collision detection method, device, equipment and storage medium

Abstract

The application discloses a vehicle collision detection method, device and equipment and a storage medium, relates to the technical field of vehicle collision detection, and comprises the following steps: acquiring motion state information and size parameters of a first target and a second target; calculating the relative state of the first target and the second target based on the motion state information, wherein the relative state is used to represent the relative position relationship and the relative motion trend of the first target and the second target in a two-way local coordinate system; calculating a projection overlap parameter according to the relative state and the size parameters, wherein the projection overlap parameter is used to represent the time margin of the projection interference of the two targets within a future time window; and determining whether the first target and the second target collide according to the projection overlap parameter. The technical problem that the existing collision detection coordinate system is single, low in precision, lacks time dimension parameters and is inaccurate in risk assessment is solved.

Description

Technical Field

[0001] This application relates to the field of vehicle collision detection technology, specifically to a vehicle collision detection method, apparatus, equipment, and storage medium. Background Technology

[0002] With the rapid development of autonomous driving technology and active safety systems, vehicle collision detection, as the foundation of decision-making, planning, and control execution, directly impacts vehicle driving safety through its accuracy and real-time performance. Existing autonomous driving and active safety systems typically require precise collision detection of surrounding vehicles, obstacles, and other targets to determine the existence of collision risks and issue timely warnings or take control measures.

[0003] In existing technologies, target collision detection typically models the target as a rectangle with length, width, and heading angle. The core of collision detection lies in determining whether two directed rectangles have overlapping regions. Currently, mainstream rectangle collision detection methods mainly include geometric decomposition-based methods, separating axis theorem (SAT)-based methods, and primitive-based methods. Among them, the separating axis theorem-based method, which projects the rectangle onto candidate axes and uses the overlap of the projection intervals to determine collisions, is widely used due to its high computational efficiency.

[0004] However, existing collision detection methods still have limitations in practical applications: When performing coordinate transformations and projection calculations, existing methods often employ a single coordinate system or a processing approach with a high degree of coupling between transformation and projection. When it is necessary to simultaneously determine the relative relationship between two dynamic targets, the accuracy is limited, failing to fully reflect the complex relative motion state between the two targets. Traditional collision determination is usually based on binary overlap results, making it difficult to provide collision risk parameters with a time dimension. This results in a lack of accurate assessment of the urgency of collisions during collision risk determination, affecting the accuracy of early warning and control strategies. Summary of the Invention

[0005] This application provides a vehicle collision detection method, apparatus, device, and computer-readable storage medium, which can solve the technical problems of low accuracy due to the single coordinate system and lack of time dimension parameters in the existing collision detection technology, resulting in inaccurate risk assessment.

[0006] To achieve the above objectives, the technical solution adopted by the present invention is as follows: In a first aspect, embodiments of this application provide a vehicle collision detection method, including: Acquire the motion state information and dimensional parameters of the first and second targets; The relative states of the first target and the second target are calculated based on the motion state information. The relative states are used to characterize the relative positional relationship and relative motion trend of the first target and the second target in a two-way local coordinate system. The projection overlap parameter is calculated based on the relative state and size parameters to characterize the time margin for interference between the two projections within a future time window. The collision between the first target and the second target is determined based on the projection overlap parameters.

[0007] Based on the above technical solution, the motion state information also includes the offset information of the reference point relative to the geometric center; After calculating the relative state, it also includes: The relative position in the relative state is compensated by the offset information to obtain the compensated relative position; The projection overlap parameters are calculated based on the compensated relative position and size parameters.

[0008] Based on the above technical solution, determining whether the first target and the second target collide according to the projection overlap parameters includes: In the local coordinate system of the first target, determine whether the first target and the second target satisfy the first overlap condition in both the longitudinal and transverse directions; In the local coordinate system of the second target, determine whether the first target and the second target satisfy the second overlap condition in both the longitudinal and transverse directions; If both the first overlap condition and the second overlap condition are met, a collision is determined to have occurred; otherwise, no collision is determined to have occurred.

[0009] Based on the above technical solution, the method also includes hierarchical early warning control: If a collision is determined to have occurred, calculate the remaining time before the collision. When the remaining time before the collision is less than the first warning threshold, the first-level warning operation is executed; When the remaining time of the collision is less than the second warning threshold, a second-level braking operation is performed. The second warning threshold is less than the first warning threshold.

[0010] Based on the above technical solution, the relative position in the relative state is compensated using the offset information, including: Based on the distance from the reference point to the front end of the target and the length of the target, calculate the offset distance between the reference point and the geometric center; The position in the relative state is shifted by the offset distance along the target's heading direction to obtain the compensated relative position.

[0011] Based on the above technical solution, the calculation of the offset distance between the reference point and the geometric center includes: The absolute value of the difference between the distance from the reference point to the front end of the target and the half-length of the target is calculated as the offset distance.

[0012] Based on the above technical solution, the determination logic for the first overlap condition in the longitudinal direction is as follows: Determine whether the sum of the longitudinal projection half-length of the second target and the effective longitudinal half-length of the first target is greater than the longitudinal relative distance after compensation. Wherein, the effective longitudinal half-length of the first target is dynamically determined based on the front-back position relationship of the second target relative to the first target; If the second target is located in front of the first target, then the distance from the reference point of the first target to the front end is taken; If the second target is located behind the first target, then the total length of the first target is taken as the distance from the reference point of the first target to the front end is subtracted from the total length of the first target.

[0013] Secondly, embodiments of this application provide a vehicle collision detection device, comprising: The information acquisition module is used to acquire the motion state information and size parameters of the first target and the second target; The state calculation module is used to calculate the relative state of the first target and the second target based on the motion state information. The relative state is used to characterize the relative positional relationship and relative motion trend of the first target and the second target in a two-way local coordinate system. The parameter calculation module is used to calculate the projection overlap parameter based on the relative state and size parameters, in order to characterize the time margin for the interference of the two projections within a future time window; The collision determination module is used to determine whether the first target and the second target collide based on the projection overlap parameters.

[0014] Thirdly, embodiments of this application provide a vehicle collision detection device, which includes a processor, a memory, and a vehicle collision detection program stored in the memory and executable by the processor, wherein when the vehicle collision detection program is executed by the processor, it implements the steps of the vehicle collision detection method as described in the first aspect.

[0015] Fourthly, embodiments of this application provide a computer-readable storage medium storing a vehicle collision detection program, wherein when the vehicle collision detection program is executed by a processor, it implements the steps of the vehicle collision detection method as described in the first aspect.

[0016] Compared with the prior art, the advantages of the present invention are as follows: This invention discloses a vehicle collision detection method, apparatus, device, and storage medium. By acquiring the motion state information and size parameters of a first target and a second target, and calculating the relative state of the first target and the second target in a two-way local coordinate system based on the motion state information, and particularly by calculating the projection overlap parameter, which represents the time margin of interference between the projections of the two targets within a future time window, based on the relative state and size parameters, this invention achieves a refined representation of the complex relative motion state of two targets and a quantification of collision risk in the time dimension. This method effectively reflects the relative positional relationship and relative motion trend of the first target and the second target in a two-way local coordinate system, preventing the accuracy limitations caused by a single coordinate system processing method and the lack of collision urgency assessment caused by binary overlap results. It solves the problem of inaccurate risk assessment caused by the single coordinate system and lack of time dimension parameters in existing collision detection methods, and significantly improves the accuracy of collision detection and the accuracy of early warning control strategies. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a schematic flowchart of an embodiment of a vehicle collision detection method, apparatus, device, and storage medium of the present invention. Figure 2 This is a schematic diagram of the hardware structure in an embodiment of a vehicle collision detection method, apparatus, device, and storage medium of the present invention. Detailed Implementation

[0019] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present application.

[0020] First, some of the technical terms used in this application will be explained to help those skilled in the art understand this application.

[0021] Two-way local coordinate system: refers to the local coordinate systems established for the first target and the second target respectively. By transforming target B to the coordinate system of target A and transforming target A to the coordinate system of target B, two-way relative state calculation is realized. It is the core foundation for decoupling coordinate transformation and projection calculation and reducing algorithm complexity in this scheme.

[0022] Reference point offset compensation: refers to the technical means of calculating the offset between the reference point and the geometric center in the engineering reality of intelligent driving, where the target position is usually represented as the rear axle center point rather than the geometric center of a rectangle, and then shifting the reference point position along the heading angle direction to compensate for the geometric center position when judging projection overlap; it is used to eliminate the loss of collision detection accuracy caused by reference point offset.

[0023] Projection overlap parameter: This parameter is calculated based on the relative state and size parameters and is used to characterize the time margin for interference between the projections of the two objects within a future time window. Unlike the traditional binary overlap result, this parameter provides a quantitative basis for collision risk with a time dimension.

[0024] Time Remaining for Collision (TTC): This refers to the ratio of the distance from the target to the boundary of the collision area to the relative velocity, used to measure the urgency of a collision. This scheme sets a first threshold and a second threshold based on this parameter to achieve graded early warning and control of rear collision risks.

[0025] Rear axle center point: refers to the reference point used in the vehicle kinematics model, usually located at the center of the vehicle's rear axle; it is often used as the target position output in intelligent driving perception data, and has a positional deviation from the rectangular geometric center used for collision detection in this application, and is an object that needs to be offset compensated.

[0026] Motion state information: refers to the target physical state data acquired by sensors, including position coordinates, velocity, heading angle, yaw rate, and size parameters (length, width, distance from reference point to front end); it is the data input source for this scheme to perform bidirectional coordinate system transformation and collision determination.

[0027] Vehicle collision detection needs to cover multiple targets, including surrounding vehicles and obstacles, and accurately determine collision risk based on a bidirectional projection strategy. Traditional methods rely on single coordinate system transformation or the Separated Axis Theorem (SAT), which suffers from high coupling between coordinate transformation and projection calculation, failure to consider the offset between the reference point and the geometric center, large computational redundancy, and lack of time-dimensional risk assessment. Therefore, this application proposes a target collision detection method, apparatus, device, and storage medium based on bidirectional projection.

[0028] To make the objectives, technical solutions, and advantages of this application clearer, the embodiments of this application will be described in further detail below with reference to the accompanying drawings.

[0029] In a first aspect, embodiments of this application provide a vehicle collision detection method.

[0030] In one embodiment, reference is made to Figure 1 , Figure 1 This is a flowchart illustrating the vehicle collision detection method of this application. Figure 1 As shown, a vehicle collision detection method includes: S1. Obtain the motion state information and size parameters of the first and second targets; S2. Calculate the relative state of the first target and the second target based on the motion state information. The relative state is used to characterize the relative positional relationship and relative motion trend of the first target and the second target in the two-way local coordinate system. S3. Calculate the projection overlap parameter based on the relative state and size parameters to characterize the time margin for interference between the two projections within a future time window; S4. Determine whether the first target and the second target collide based on the projection overlap parameters.

[0031] In this embodiment, by acquiring the motion state information and size parameters of the first and second targets, and calculating their relative states in a bidirectional local coordinate system based on the motion state information, particularly by calculating the projection overlap parameter representing the time margin for interference between their projections within a future time window based on the relative states and size parameters, a refined characterization of the complex relative motion states of the two targets and the quantification of collision risk in the time dimension are achieved. This method effectively reflects the relative positional relationship and relative motion trend of the first and second targets in a bidirectional local coordinate system, preventing the accuracy limitations caused by a single coordinate system processing method and the lack of collision urgency assessment due to binarized overlap results. It solves the problem of inaccurate risk assessment caused by the single coordinate system and lack of time dimension parameters in existing collision detection, significantly improving the accuracy of collision detection and the accuracy of early warning control strategies. If there is a temporary loss of sensor data, interpolation compensation can be performed using data from the previous cycle to ensure process continuity.

[0032] Furthermore, in one embodiment, the motion state information includes the target's position coordinates, velocity, heading angle, and yaw rate. Size parameters include the target's length, width, and the distance from the reference point to the target's front end. It should be noted that in intelligent driving scenarios, the target's reference point is typically defined as the rear axle center point used in the vehicle's kinematics model, rather than the geometric center of a rectangle. This information can be acquired using sensors such as millimeter-wave radar, lidar, or cameras. For example, millimeter-wave radar can be used to detect the target's velocity and distance, while cameras can be used to identify the target's type and heading angle. Multi-sensor fusion can improve the robustness of data acquisition. Obtaining the distance from the reference point to the front end provides a data foundation for subsequent offset compensation between the reference point and the geometric center.

[0033] In this embodiment, driving data is collected using multiple sensors, addressing the performance degradation of a single sensor under adverse weather or lighting conditions. Millimeter-wave radar offers strong penetration, while cameras provide rich semantic information; their fusion complements each other's strengths and weaknesses. This approach ensures the accuracy of the input data, providing a reliable foundation for subsequent trajectory modeling. The reference point is clearly defined as the rear axle center point, aligning with the actual output of the vehicle's kinematics model and avoiding model errors caused by directly using the geometric center. For vehicles with lower configurations, only millimeter-wave radar can be used, with algorithms compensating for heading angle information to suit the needs of different cost models.

[0034] Furthermore, in one embodiment, calculating the relative state includes establishing a two-way local coordinate system. First, a local coordinate system for the first target is established, with the position of the first target as the origin and the heading direction of the first target as the positive axis. Then, the second target is transformed into the coordinate system of the first target, and the relative position and relative heading angle of the second target in the local coordinate system of the first target are calculated. Similarly, using the second target as a reference, a local coordinate system for the second target is established, and the relative position and relative heading angle of the first target in the local coordinate system of the second target are calculated. Through this two-way coordinate system transformation strategy, collision detection is decomposed into two independent coordinate system transformations and two projection calculations, reducing coupling and facilitating step-by-step optimization, thus avoiding the problem of high coupling between the transformation and projection steps in traditional schemes, which makes step-by-step optimization difficult.

[0035] In this embodiment, a two-way coordinate system transformation solves the problem of high coupling between coordinate transformation and projection calculation, making step-by-step optimization difficult in traditional schemes. The introduction of a two-way local coordinate system enables adaptive recognition of the target's relative steering intention. This approach aligns with vehicle kinematics principles and significantly improves the prediction accuracy of future trajectory points. If the target is in a complex lane-changing process, the global coordinate system can be switched as an auxiliary reference to increase the model's adaptability.

[0036] Furthermore, in one embodiment, the collision determination includes judging four sets of overlap conditions. The first and second sets of conditions are respectively: in the local coordinate system of the first target, it is determined whether the first target and the second target satisfy the overlap conditions in both the longitudinal and lateral directions. The third and fourth sets of conditions are respectively: in the local coordinate system of the second target, it is determined whether the first target and the second target satisfy the overlap conditions in both the longitudinal and lateral directions. If all four sets of overlap conditions are satisfied, a collision is determined to have occurred. The entire collision detection process only requires four projection overlap judgments, without iterative or recursive calculations, significantly reducing computational complexity and meeting the real-time requirements of active safety systems.

[0037] In this embodiment, by judging four sets of overlap conditions, the computational redundancy caused by the traditional Separated Axis Theorem (SAT) requiring the checking of multiple candidate axes is resolved; accurate collision detection can be completed with only four judgments, significantly reducing the computational power consumption of the automotive embedded system. This method ensures the accuracy and reliability of the mathematical basis for collision time calculation. If the processor has extremely high computing power, the sampling point density can be increased to further pursue ultimate accuracy.

[0038] Furthermore, in one embodiment, the motion state information also includes offset information of the reference point relative to the geometric center; After calculating the relative state, it also includes: The relative position in the relative state is compensated by offset information to obtain the compensated relative position; The projection overlap parameters are calculated based on the compensated relative position and size parameters.

[0039] In this embodiment, by introducing offset information and performing compensation, the problem of decreased detection accuracy caused by the target position being represented as the rear axle center rather than the geometric center in intelligent driving is solved; in particular, the uncompensated offset in the vehicle length direction affects the accuracy of front and rear collision detection. This method ensures the stability of the subsequent step size adjustment basis. If the noise is high, weighted filtering can also be used to process the offset information to improve the real-time performance of compensation.

[0040] Furthermore, in one embodiment, compensating for the relative position in the relative state using offset information includes: considering that the target position in intelligent driving is usually represented as the rear axle center rather than the geometric center, directly using this position will lead to a decrease in collision detection accuracy, especially since the offset in the vehicle length direction is not compensated, which will affect the accuracy of front- and rear-facing collision detection. Therefore, it is necessary to calculate the offset between the reference point and the geometric center. When performing projection overlap judgment, the reference point position is translated by the offset along the heading angle direction to compensate for the geometric center position. This compensation avoids the loss of collision detection accuracy caused by the offset of the reference point and is particularly suitable for collision detection in the longitudinal direction of the vehicle.

[0041] In this embodiment, offset compensation addresses the issue of decreased detection accuracy caused by inconsistency between the target reference point and the geometric center in intelligent driving; in particular, uncompensated offsets along the vehicle's length direction affect the accuracy of front- and rear-end collision detection. This method ensures the accuracy of the projection center position and avoids loss of collision detection accuracy due to reference point offset.

[0042] Further, in one embodiment, determining whether the first target and the second target collide based on the projection overlap parameter includes: In the local coordinate system of the first target, determine whether the first target and the second target satisfy the first overlap condition in both the longitudinal and transverse directions; In the local coordinate system of the second target, determine whether the first target and the second target satisfy the second overlap condition in both the longitudinal and transverse directions; If both the first and second overlap conditions are met, a collision is determined to have occurred; otherwise, no collision is determined to have occurred.

[0043] In this embodiment, the overlapping condition judgment under the bidirectional coordinate system solves the problem of missed detection that may exist in the judgment of the unidirectional coordinate system; accurate collision detection can be completed in only four judgments, which greatly reduces the computing power consumption of the vehicle embedded system and meets the real-time requirements of the active safety system.

[0044] Furthermore, in one embodiment, the method further includes tiered early warning control: If a collision is determined to have occurred, calculate the remaining time before the collision. When the remaining time before collision is less than the first warning threshold, the first-level warning operation is executed; When the remaining time before collision is less than the second warning threshold, the second-level braking operation is performed; The second warning threshold is lower than the first warning threshold.

[0045] When the remaining time before collision is less than a first threshold, an in-vehicle warning sound and a flashing warning on the instrument panel are triggered; when the remaining time before collision is less than a second threshold, braking force is actively applied. This tiered strategy avoids false triggering caused by premature intervention while ensuring timely response in emergency situations.

[0046] In this embodiment, a tiered warning control system addresses the issue of false alarms or delayed intervention caused by a single threshold. The first threshold provides a warning, while the second threshold executes control, forming a safe closed loop for human-machine co-driving. This approach enhances the driving experience and system reliability. If the vehicle has steering assist, steering intervention can also be applied simultaneously when the second threshold is triggered, providing multi-dimensional hazard avoidance measures.

[0047] Furthermore, in one embodiment, compensation for relative positions in relative states using offset information includes: Calculate the offset distance between the reference point and the geometric center based on the distance from the reference point to the front of the target and the length of the target; The position in the relative state is shifted by a distance along the target's heading to obtain the compensated relative position.

[0048] In this embodiment, the projection center deviation problem caused by the inconsistency between the reference point and the geometric center is solved through specific offset distance calculation and translation compensation; the module has a built-in compensation algorithm that can effectively correct position errors. This method ensures the stability of the subsequent step size adjustment basis. If the trajectory points are sparse, interpolation can also be used for smoothing to adapt to sensor data with different sampling frequencies.

[0049] Further, in one embodiment, calculating the offset distance between the reference point and the geometric center includes: Calculate the absolute value of the difference between the distance from the reference point to the front of the target and the half-length of the target, and use it as the offset distance.

[0050] In this embodiment, the offset distance is calculated using absolute values, thus solving the computational complexity problem caused by the uncertainty of the offset direction. The module provides explicit geometric calculation logic, providing a precise basis for subsequent adaptive step size. This method transforms geometric features into calculable numerical values, realizing the mapping between the mathematical model and the physical scene. If computational resources are limited, a lookup table method can also be used to approximate the offset distance, sacrificing a small amount of accuracy for improved computational speed.

[0051] Furthermore, in one embodiment, the determination logic for the first overlap condition in the longitudinal direction is as follows: Determine whether the sum of the longitudinal projection half-length of the second target and the effective longitudinal half-length of the first target is greater than the longitudinal relative distance after compensation. Among them, the effective longitudinal half-length of the first target is dynamically determined based on the front-to-back positional relationship of the second target relative to the first target; If the second target is in front of the first target, then take the distance from the reference point of the first target to the front end; If the second target is located behind the first target, then take the total length of the first target minus the distance from the reference point of the first target to its front end.

[0052] In this embodiment, by dynamically determining the effective longitudinal half-length, the problem of judgment error caused by the inability of a fixed half-length to adapt to reference point offsets is solved; the module implements differentiated judgment for forward and backward directions, and optimally allocates computing resources. This method achieves optimal allocation of computing resources. If the scenario is complex, the judgment level can be increased to achieve more refined resource control.

[0053] Furthermore, in one embodiment, the specific steps of the bidirectional coordinate system transformation and offset compensation are as follows: 101. Two-way coordinate system transformation 1011. Establish the local coordinate system of target A; Establish a local coordinate system with the position of target A as the origin, the heading direction of target A as the positive X-axis, and the direction perpendicular to the heading direction as the positive Y-axis. .

[0054] 1012. Transform target B to the coordinate system of target A; Calculate the relative position of target B in the local coordinate system of target A. and relative heading angle The calculation formula is as follows: ; in, and These are the position coordinates of target A and target B in the global coordinate system, respectively. and These are the heading angles for target A and target B, respectively.

[0055] 1013. Transform target A to the coordinate system of target B; Similarly, using target B as a reference, calculate the relative position of target A in the local coordinate system of target B. and relative heading angle The calculation formula is as follows: ; 102. Reference point offset compensation: Considering that the target position in intelligent driving usually represents the rear axle center rather than the geometric center, it is necessary to offset the projection center point.

[0056] 1021. Calculate the offset between the reference point and the geometric center. ; in: The distance from the reference point to the front end of the rectangle. For the target length, It is half the length of a rectangle. This represents the distance of the reference point from the geometric center.

[0057] 1022. Compensation for the center position of the projection When performing projection overlap judgment, the reference point position is translated along the heading angle direction. Compensation is performed to the geometric center position. For compensation of target B in the coordinate system of target A: ; in This is the offset of the reference point for target B. This represents the compensated relative position. Similarly, for the compensation of target A in the coordinate system of target B: ; in This is the offset of the reference point for target A.

[0058] 1023. Calculation of the projected half-length: Calculate the projected half-length of target B in the coordinate system of target A. The half-length of the projection of target B onto the X-axis in the coordinate system of target A is: ; In the formula, P x,B∣A L represents the half-length (i.e., the horizontal half-width) of the projection of target B onto the X-axis of the target A's body coordinate system; BW represents the total length of target B. B ψ is the total width of target B; B∣A The relative angle between the heading direction of target B and the heading direction of target A (the relative heading angle from A to B).

[0059] The half-length of the projection of target B onto the Y-axis in the coordinate system of target A is: ; In the formula, P y,B∣A L represents the half-length (vertical half-length) of the projection of target B onto the Y-axis of the target A's body coordinate system; B W represents the total length of target B. B ψ is the total width of target B; B∣A The angle between the heading direction of target B and the heading direction of target A.

[0060] 1024. Calculate the half-length of the projection of target A in the coordinate system of target B. Similarly: ; In the formula, P x,A∣B L represents the half-length of the projection of target A onto the X-axis of the target B's body coordinate system; A W represents the total length of target A. A ψ is the total width of target A; A∣B Let be the relative angle between the heading direction of target A and the heading direction of target B (the relative heading angle from B to A), and ψ B∣A They are opposites; .

[0061] In the formula, P y,A∣B L represents the half-length of the projection of target A onto the Y-axis of the target B's body coordinate system; A W represents the total length of target A. A ψ is the total width of target A; A∣B The angle between the heading direction of target A and the heading direction of target B.

[0062] In this embodiment, a two-way coordinate system transformation strategy is used to decompose collision detection into two independent coordinate system transformations and two projection calculations, reducing coupling and facilitating step-by-step optimization. Trigonometric functions are only used for coordinate transformation and projection half-length calculation, and each only needs to be calculated once, significantly reducing computational complexity.

[0063] Furthermore, in one embodiment, the specific process of collision risk determination and control output is as follows: 201. Collision risk assessment includes: 2011. Overlap determination in the coordinate system of target A: X-axis overlap determination: Calculate the half-length of the projection interval of target A: If target B is located in front of target A ( ≥0), the length of the first half of target A is If target B is located behind target A ( <0), the length of the second half of target A is ; recorded as The X-axis overlap condition is then: >| |; Y-axis overlap determination: The half-length of the projection of target A in the Y-axis direction is... The Y-axis overlap condition is then: .

[0064] 2012. Overlap determination in the target B coordinate system.

[0065] Similarly, by determining the overlap condition of the X-axis and Y-axis directions in the target B coordinate system, two more sets of inequalities are obtained.

[0066] 2013. Comprehensive collision determination: If all four sets of overlap conditions mentioned above are met, then target A and target B are determined to have collided; otherwise, no collision is determined to have occurred. The determination logic can be expressed by the following formula: ; in, This indicates that the X-axis direction in the coordinate system of target A satisfies the overlap condition. This indicates that the overlap condition is satisfied in the Y-axis direction under the coordinate system of target A, and so on. Symbol This represents a logical AND operation, meaning all conditions must be true.

[0067] 202. Warning and Control Output. If a collision is determined to have occurred, the collision time TTC (Time To Collision) is further calculated: ; Where D is the distance from the target to the boundary of the collision area. This refers to relative velocity.

[0068] This invention achieves a tiered response to rear-collision risk by setting a first threshold and a second threshold. When the time to collision (TTC) is less than the first threshold, an in-vehicle warning sound and a flashing warning on the instrument panel are triggered; when the TTC is less than the second threshold, braking force is actively applied. This tiered strategy avoids false triggering caused by premature intervention while ensuring timely response in emergency situations.

[0069] In this embodiment, a tiered response to rear collision risk is achieved by setting a first threshold and a second threshold. This tiered strategy avoids premature intervention leading to false triggers while ensuring timely response in emergencies, thus improving the driving experience and system reliability. The vehicle collision detection device enables the systematic implementation of vehicle collision risk calculation. This system solves the problem of lacking hardware support for method implementation, and its modular design clarifies the responsibilities of each functional unit, ensuring the streamlined execution of data collection, modeling, extraction, and control. If the system is deployed in the cloud, the modules can also operate in a distributed manner, as long as the communication latency meets real-time requirements.

[0070] Secondly, embodiments of this application provide a vehicle collision detection device, comprising: The information acquisition module is used to acquire the motion state information and size parameters of the first target and the second target; The state calculation module is used to calculate the relative state of the first target and the second target based on the motion state information. The relative state is used to characterize the relative positional relationship and relative motion trend of the first target and the second target in the bidirectional local coordinate system. The parameter calculation module is used to calculate the projection overlap parameter based on the relative state and size parameters, which is used to characterize the time margin for the interference of the two projections within a future time window; The collision determination module is used to determine whether the first target and the second target collide based on the projection overlap parameters.

[0071] In this embodiment, the systematic implementation of vehicle collision risk calculation is achieved through the collaborative work of the information acquisition module, state calculation module, parameter calculation module, and collision determination module. This system solves the problem of lacking hardware support for method implementation, and its modular design clarifies the responsibilities of each functional unit, ensuring the streamlined execution of data collection, modeling, extraction, and control. If the system is deployed in the cloud, the modules can also operate in a distributed manner, as long as the communication latency meets real-time requirements.

[0072] Furthermore, in one embodiment, the information acquisition module is specifically used to: acquire the target's position coordinates, velocity, heading angle, and yaw rate; acquire the target's length, width, and the distance from the reference point to the front end of the target; the reference point is defined as the rear axle center point used in the vehicle kinematics model; and acquire information through sensors such as millimeter-wave radar, lidar, or cameras.

[0073] In this embodiment, multiple sensors are configured through the information acquisition module, solving the problem of performance degradation of a single sensor under adverse weather or lighting conditions. The module supports sensor switching and fusion, ensuring the accuracy of the input data. This approach provides a reliable foundation for subsequent trajectory modeling. If the vehicle configuration is low, the module can also be configured to receive only millimeter-wave radar data and compensate for the heading angle information through internal algorithms to adapt to the needs of different cost models.

[0074] Furthermore, in one embodiment, the state calculation module is specifically used to: establish a local coordinate system of the first target, with the position of the first target as the origin and the heading direction of the first target as the positive axis; transform the second target to the coordinate system of the first target, and calculate the relative position and relative heading angle of the second target in the local coordinate system of the first target; establish a local coordinate system of the second target with the second target as the reference, and calculate the relative position and relative heading angle of the first target in the local coordinate system of the second target.

[0075] In this embodiment, the built-in two-way coordinate system transformation in the state calculation module solves the trajectory distortion problem caused by the assumption of straight-line movement in traditional methods. The module can adaptively identify the target's steering intention, which conforms to the principles of vehicle kinematics. This approach significantly improves the prediction accuracy of future trajectory points. If the target is in a complex lane-changing process, the module can also switch to other motion models as alternatives to increase the model's adaptability.

[0076] Furthermore, in one embodiment, the collision determination module is specifically used to: determine whether the first target and the second target satisfy the overlap condition in both the longitudinal and lateral directions in the local coordinate system of the first target; determine whether the first target and the second target satisfy the overlap condition in both the longitudinal and lateral directions in the local coordinate system of the second target; if all four sets of overlap conditions are satisfied, then a collision is determined to have occurred.

[0077] In this embodiment, a collision determination module performs four sets of overlap condition judgments, solving the problem of curvature calculation difficulties caused by the discontinuity of discrete trajectory points; the module's built-in judgment algorithm can effectively filter out high-frequency noise. This method ensures the stability of the subsequent step size adjustment basis. If the trajectory points are sparse, the module can also use spline interpolation for smoothing to adapt to sensor data with different sampling frequencies.

[0078] Furthermore, in one embodiment, the state calculation module is also used to: obtain offset information of the reference point relative to the geometric center; after calculating the relative state, compensate the relative position in the relative state through the offset information to obtain the compensated relative position; and calculate the projection overlap parameter based on the compensated relative position and the size parameter.

[0079] In this embodiment, offset information is introduced and compensated through a state calculation module, which solves the problem of decreased detection accuracy caused by the target position being represented as the rear axle center rather than the geometric center in intelligent driving. The module has a built-in compensation algorithm that can effectively correct position errors. This method ensures the stability of the subsequent step size adjustment basis. If the noise is high, the module can also use weighted filtering to process the offset information to improve the real-time performance of the compensation.

[0080] Furthermore, in one embodiment, the state calculation module is also used to: calculate the offset between the reference point and the geometric center; and when performing projection overlap judgment, shift the reference point position by the offset along the heading angle direction to compensate to the geometric center position.

[0081] In this embodiment, offset compensation is performed by the state calculation module, which solves the problem of decreased detection accuracy caused by the target position in intelligent driving being usually represented as the rear axle center rather than the geometric center; the module has a built-in compensation algorithm that can effectively correct position errors.

[0082] Furthermore, in one embodiment, the collision determination module is also used to: determine whether the first target and the second target satisfy the first overlap condition in both the longitudinal and transverse directions in the local coordinate system of the first target; determine whether the first target and the second target satisfy the second overlap condition in both the longitudinal and transverse directions in the local coordinate system of the second target; if the first overlap condition and the second overlap condition are satisfied at the same time, then a collision is determined to have occurred; otherwise, a collision is determined not to have occurred.

[0083] In this embodiment, the collision determination module performs overlap condition judgment in a two-way coordinate system, solving the potential missed detection problem in a one-way coordinate system. The module implements refined step size at sharp curves and coarser step size at gentle curves, optimally allocating computing resources. This method achieves optimal allocation of computing resources. If the scene is complex, the module can also increase the number of step size levels to achieve more refined resource control.

[0084] Furthermore, in one embodiment, the system also includes a warning control module, used for: calculating the remaining time of collision if a collision is determined to have occurred; executing a first-level warning operation when the remaining time of collision is less than a first warning threshold; and executing a second-level braking operation when the remaining time of collision is less than a second warning threshold; wherein the second warning threshold is less than the first warning threshold. In this embodiment, a graded response strategy is established through the warning control module, solving the problem of either false alarms or delayed intervention caused by a single threshold; the module forms a safety closed loop for human-machine co-driving, improving the driving experience and system reliability. If the vehicle has a steering assist function, the module can also apply steering intervention simultaneously when the second threshold is triggered to provide multi-dimensional risk avoidance measures.

[0085] Furthermore, in one embodiment, the state calculation module is also used to: calculate the offset distance between the reference point and the geometric center based on the distance from the reference point to the front end of the target and the length of the target, where the target is a collective term for the first target and the second target; and shift the position in the relative state along the heading direction of the target by the offset distance to obtain the compensated relative position.

[0086] In this embodiment, the state calculation module performs specific offset distance calculations and translation compensation, solving the projection center deviation problem caused by the inconsistency between the reference point and the geometric center. The module has a built-in compensation algorithm that can effectively correct position errors. This method ensures the stability of the subsequent step size adjustment basis. If the trajectory points are sparse, the module can also use interpolation to smooth the data, adapting to sensor data with different sampling frequencies.

[0087] Furthermore, in one embodiment, the state calculation module is also used to: calculate the absolute value of the difference between the distance from the reference point to the front end of the target and the half length of the target, as the offset distance.

[0088] In this embodiment, the state calculation module calculates the offset distance using absolute values, thus solving the computational complexity problem caused by the uncertainty of the offset direction. The module provides explicit geometric calculation logic, providing a precise basis for subsequent adaptive step size. This method transforms geometric features into calculable values, realizing the mapping between the mathematical model and the physical scene. If computational resources are limited, the module can also use a lookup table method to approximate the offset distance, sacrificing a small amount of accuracy for improved computational speed.

[0089] Furthermore, in one embodiment, the collision determination module is also used to: determine whether the sum of the longitudinal projection half-length of the second target and the effective longitudinal half-length of the first target is greater than the longitudinal relative distance after compensation; wherein, the effective longitudinal half-length of the first target is dynamically determined according to the front-rear position relationship of the second target relative to the first target; if the second target is in front of the first target, the distance from the reference point of the first target to the front end is taken; if the second target is behind the first target, the total length of the first target is taken minus the distance from the reference point of the first target to the front end is taken.

[0090] In this embodiment, the effective longitudinal half-length is dynamically determined by the collision determination module, solving the problem of determination error caused by the inability of a fixed half-length to adapt to the offset of the reference point; the module implements differentiated determination for forward and backward directions, and optimally allocates computing resources. This method achieves optimal allocation of computing resources. If the scene is complex, the module can also add determination levels to achieve more refined resource control.

[0091] Furthermore, in one embodiment, the state calculation module is also used to perform specific steps of bidirectional coordinate system transformation and offset compensation: 101. Two-way coordinate system transformation 1011. Establish the local coordinate system of target A; Establish a local coordinate system with the position of target A as the origin, the heading direction of target A as the positive X-axis, and the direction perpendicular to the heading direction as the positive Y-axis. .

[0092] 1012. Transform target B to the coordinate system of target A; Calculate the relative position of target B in the local coordinate system of target A. and relative heading angle The calculation formula is as follows: ; in, and These are the position coordinates of target A and target B in the global coordinate system, respectively. and These are the heading angles for target A and target B, respectively.

[0093] 1013. Transform target A to the coordinate system of target B; Similarly, using target B as a reference, calculate the relative position of target A in the local coordinate system of target B. and relative heading angle The calculation formula is as follows: ; 102. Reference point offset compensation: Considering that the target position in intelligent driving usually represents the rear axle center rather than the geometric center, it is necessary to offset the projection center point.

[0094] 1021. Calculate the offset between the reference point and the geometric center. ; in: The distance from the reference point to the front end of the rectangle. For the target length, It is half the length of a rectangle. This represents the distance of the reference point from the geometric center.

[0095] 1022. Compensation for the center position of the projection When performing projection overlap judgment, the reference point position is translated along the heading angle direction. Compensation is performed to the geometric center position. For compensation of target B in the coordinate system of target A: ; in This is the offset of the reference point for target B. This represents the compensated relative position. Similarly, for the compensation of target A in the coordinate system of target B: ; in This is the offset of the reference point for target A.

[0096] 1023. Calculation of half-length projection: Calculate the half-length projection of target B in the coordinate system of target A. The half-length of the projection of target B onto the X-axis in the coordinate system of target A is: ; In the formula, P x,B∣A L represents the half-length (i.e., the horizontal half-width) of the projection of target B onto the X-axis of the target A's body coordinate system; B W represents the total length of target B. B ψ is the total width of target B; B∣A The relative angle between the heading direction of target B and the heading direction of target A (the relative heading angle from A to B).

[0097] The half-length of the projection of target B onto the Y-axis in the coordinate system of target A is: ; In the formula, P y,B∣A L represents the half-length (vertical half-length) of the projection of target B onto the Y-axis of the target A's body coordinate system; B W represents the total length of target B. B ψ is the total width of target B; B∣A The angle between the heading direction of target B and the heading direction of target A.

[0098] 1024. Calculate the half-length of the projection of target A in the coordinate system of target B. Similarly: ; In the formula, P x,A∣B L represents the half-length of the projection of target A onto the X-axis of the target B's body coordinate system; A W represents the total length of target A. A ψ is the total width of target A; A∣B Let be the relative angle between the heading direction of target A and the heading direction of target B (the relative heading angle from B to A), and ψ B∣A They are opposites; .

[0099] In the formula, P y,A∣B L represents the half-length of the projection of target A onto the Y-axis of the target B's body coordinate system; A W represents the total length of target A. A ψ is the total width of target A; A∣B The angle between the heading direction of target A and the heading direction of target B.

[0100] In this embodiment, detailed modeling and coordinate transformation are performed by the state calculation module, which solves the problem of high coupling between coordinate transformation and projection calculation; the module provides the relative state in a two-way local coordinate system, which provides an accurate basis for subsequent collision determination.

[0101] Furthermore, in one embodiment, the collision determination module and the early warning control module are also used to perform the specific process of collision risk determination and control output: 201. Collision risk assessment includes: 2011. Overlap determination in the coordinate system of target A: X-axis overlap determination: Calculate the half-length of the projection interval of target A: If target B is located in front of target A ( ≥0), the length of the first half of target A is If target B is located behind target A ( <0), the length of the second half of target A is ; recorded as The X-axis overlap condition is then: >| |; Y-axis overlap determination: The half-length of the projection of target A in the Y-axis direction is... The Y-axis overlap condition is then: .

[0102] 2012. Overlap determination in the target B coordinate system.

[0103] Similarly, by determining the overlap condition of the X-axis and Y-axis directions in the target B coordinate system, two more sets of inequalities are obtained.

[0104] 2013. Comprehensive collision determination: If all four sets of overlap conditions mentioned above are met, then target A and target B are determined to have collided; otherwise, no collision is determined to have occurred. The determination logic can be expressed by the following formula: ; in, This indicates that the X-axis direction in the coordinate system of target A satisfies the overlap condition. This indicates that the overlap condition is satisfied in the Y-axis direction under the coordinate system of target A, and so on. Symbol This represents a logical AND operation, meaning all conditions must be true.

[0105] 202. Warning and Control Output. If a collision is determined to have occurred, the collision time TTC (Time To Collision) is further calculated: ; Where D is the distance from the target to the boundary of the collision area. This refers to relative velocity.

[0106] This invention achieves a tiered response to rear-collision risk by setting a first threshold and a second threshold. When the time to collision (TTC) is less than the first threshold, an in-vehicle warning sound and a flashing warning on the instrument panel are triggered; when the TTC is less than the second threshold, braking force is actively applied. This tiered strategy avoids false triggering caused by premature intervention while ensuring timely response in emergency situations.

[0107] In this embodiment, the collision determination module and the early warning control module perform accurate collision risk determination, which solves the problem of false alarms caused by the collision time being too small due to the straight distance estimation; the module reduces the collision risk estimation error by calculating the accurate projection overlap parameters.

[0108] This application also provides a hardware architecture for a vehicle collision detection device. The system includes a perception layer, a computation layer, and an execution layer. The perception layer includes sensors such as millimeter-wave radar and lidar for acquiring target data; the computation layer includes an onboard domain controller or embedded processor for running the aforementioned collision detection algorithm and calculating relative states and projection overlap parameters; the execution layer includes a braking system and an alarm unit for executing graded warning control based on the collision determination results. The layers communicate with each other via a controller area network or Ethernet. This architecture supports the deployment of the algorithm in an onboard embedded system, with short single-frame computation time, meeting the real-time requirements of active safety systems. This method is independent of specific sensor types and can be applied to target data output from different perception sources, demonstrating strong algorithm versatility.

[0109] The functions of each module in the above-mentioned vehicle collision detection device correspond to the steps in the above-mentioned vehicle collision detection method embodiment, and their functions and implementation processes will not be described in detail here.

[0110] Thirdly, embodiments of this application provide a vehicle collision detection device, which can be a personal computer (PC), laptop computer, server, or other device with data processing capabilities.

[0111] Reference Figure 2 , Figure 2 This is a schematic diagram of the hardware structure of a vehicle collision detection device involved in an embodiment of this application. In this embodiment, the vehicle collision detection device may include a processor, a memory, a communication interface, and a communication bus.

[0112] The communication bus can be of any type and is used to interconnect the processor, memory, and communication interface.

[0113] The communication interface includes input / output (I / O) interfaces, physical interfaces, and logical interfaces used for interconnecting components within the vehicle collision detection equipment, as well as interfaces used for interconnecting the vehicle collision detection equipment with other devices (such as other computing devices or user equipment). Physical interfaces can be Ethernet interfaces, fiber optic interfaces, ATM interfaces, etc.; user equipment can be displays, keyboards, etc.

[0114] Memory can be various types of storage media, such as random access memory (RAM), read-only memory (ROM), non-volatile RAM (NVRAM), flash memory, optical storage, hard disk, programmable ROM (PROM), erasable PROM (EPROM), electrically erasable PROM (EEPROM), etc.

[0115] The processor can be a general-purpose processor, which can call the vehicle collision detection program stored in the memory and execute the vehicle collision detection method provided in the embodiments of this application. For example, the general-purpose processor can be a central processing unit (CPU). The method executed when the vehicle collision detection program is called can be referred to in the various embodiments of the vehicle collision detection method of this application, and will not be repeated here.

[0116] Those skilled in the art will understand that Figure 2 The hardware structure shown does not constitute a limitation of this application and may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0117] Fourthly, embodiments of this application also provide a computer-readable storage medium.

[0118] The present application provides a computer-readable storage medium storing a vehicle collision detection program, wherein when the vehicle collision detection program is executed by a processor, it implements the steps of the vehicle collision detection method described above.

[0119] The method implemented when the vehicle collision detection procedure is executed can be referred to in various embodiments of the vehicle collision detection method of this application, and will not be repeated here.

[0120] It should be noted that the sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0121] The terms "comprising" and "having," and any variations thereof, in the specification, claims, and accompanying drawings of this application are intended to cover non-exclusive inclusion. For example, a process, method, apparatus, product, or device that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to such processes, methods, products, or devices. The terms "first," "second," and "third," etc., are used to distinguish different objects, etc., and do not indicate a sequence, nor do they limit "first," "second," and "third" to different types.

[0122] In the description of the embodiments of this application, terms such as "exemplary," "for example," or "for instance" are used to indicate examples, illustrations, or explanations. Any embodiment or design described as "exemplary," "for example," or "for instance" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of terms such as "exemplary," "for example," or "for instance" is intended to present the relevant concepts in a concrete manner.

[0123] In the description of the embodiments of this application, unless otherwise stated, " / " means "or". For example, A / B can mean A or B. The "and / or" in the text is merely a description of the relationship between related objects, indicating that there can be three relationships. For example, A and / or B can mean: A exists alone, A and B exist simultaneously, and B exists alone. In addition, in the description of the embodiments of this application, "multiple" means two or more.

[0124] In some processes described in the embodiments of this application, multiple operations or steps are included in a specific order. However, it should be understood that these operations or steps may not be executed in the order they appear in the embodiments of this application, or they may be executed in parallel. The sequence number of the operation is only used to distinguish different operations, and the sequence number itself does not represent any execution order. In addition, these processes may include more or fewer operations, and these operations or steps may be executed sequentially or in parallel, and these operations or steps may be combined.

[0125] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) as described above, and includes several instructions to cause a terminal device to execute the methods described in the various embodiments of this application.

[0126] The above are merely preferred embodiments of this application and do not limit the patent scope of this application. Any equivalent structural or procedural transformations made using the content of this application's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of this application.

Claims

1. A vehicle collision detection method, characterized in that, include: Acquire the motion state information and dimensional parameters of the first and second targets; The relative states of the first target and the second target are calculated based on the motion state information. The relative states are used to characterize the relative positional relationship and relative motion trend of the first target and the second target in a two-way local coordinate system. The projection overlap parameter is calculated based on the relative state and size parameters to characterize the time margin for interference between the two projections within a future time window. The collision between the first target and the second target is determined based on the projection overlap parameters.

2. The vehicle collision detection method as described in claim 1, characterized in that, The motion state information also includes the offset information of the reference point relative to the geometric center; After calculating the relative state, it also includes: The relative position in the relative state is compensated by the offset information to obtain the compensated relative position; The projection overlap parameters are calculated based on the compensated relative position and size parameters.

3. The vehicle collision detection method as described in claim 1, characterized in that, Determining whether the first target and the second target collide based on the projection overlap parameters includes: In the local coordinate system of the first target, determine whether the first target and the second target satisfy the first overlap condition in both the longitudinal and transverse directions; In the local coordinate system of the second target, determine whether the first target and the second target satisfy the second overlap condition in both the longitudinal and transverse directions; If both the first overlap condition and the second overlap condition are met, a collision is determined to have occurred; otherwise, no collision is determined to have occurred.

4. The vehicle collision detection method as described in claim 1, characterized in that, The method also includes tiered early warning control: If a collision is determined to have occurred, calculate the remaining time before the collision. When the remaining time before the collision is less than the first warning threshold, the first-level warning operation is executed; When the remaining time of the collision is less than the second warning threshold, a second-level braking operation is performed. The second warning threshold is less than the first warning threshold.

5. The vehicle collision detection method as described in claim 2, characterized in that, The offset information is used to compensate for the relative position in the relative state, including: Based on the distance from the reference point to the front end of the target and the length of the target, calculate the offset distance between the reference point and the geometric center; The position in the relative state is shifted by the offset distance along the target's heading direction to obtain the compensated relative position.

6. The vehicle collision detection method as described in claim 5, characterized in that, The calculation of the offset distance between the reference point and the geometric center includes: The absolute value of the difference between the distance from the reference point to the front end of the target and the half-length of the target is calculated as the offset distance.

7. The vehicle collision detection method as described in claim 3, characterized in that, The logic for determining the first overlap condition in the vertical direction is as follows: Determine whether the sum of the longitudinal projection half-length of the second target and the effective longitudinal half-length of the first target is greater than the longitudinal relative distance after compensation. Wherein, the effective longitudinal half-length of the first target is dynamically determined based on the front-back position relationship of the second target relative to the first target; If the second target is located in front of the first target, then the distance from the reference point of the first target to the front end is taken; If the second target is located behind the first target, then the total length of the first target is taken as the distance from the reference point of the first target to the front end is subtracted from the total length of the first target.

8. A vehicle collision detection device, characterized in that, include: The information acquisition module is used to acquire the motion state information and size parameters of the first target and the second target; The state calculation module is used to calculate the relative state of the first target and the second target based on the motion state information. The relative state is used to characterize the relative positional relationship and relative motion trend of the first target and the second target in a two-way local coordinate system. The parameter calculation module is used to calculate the projection overlap parameter based on the relative state and size parameters, in order to characterize the time margin for the interference of the two projections within a future time window; The collision determination module is used to determine whether the first target and the second target collide based on the projection overlap parameters.

9. A vehicle collision detection device, characterized in that, The vehicle collision detection device includes a processor, a memory, and a vehicle collision detection program stored in the memory and executable by the processor, wherein when the vehicle collision detection program is executed by the processor, it implements the steps of the vehicle collision detection method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a vehicle collision detection program, wherein when the vehicle collision detection program is executed by a processor, it implements the steps of the vehicle collision detection method as described in any one of claims 1 to 7.

Images