A Collision Detection Method, System, Device and Medium

Abstract

The present application provides a collision detection method, system, device and medium, relating to the technical field of vehicles. The method includes: determining a first influence amount of vehicle braking on the vehicle longitudinal acceleration according to the relationship curve between the braking torque and the vehicle longitudinal deceleration and the current braking torque of the vehicle; determining the shock absorber stiffness according to the mapping relationship between the suspension compression height and the shock absorber stiffness and the current suspension compression height of the vehicle; determining a second influence amount of road surface mutation on the vehicle longitudinal acceleration according to the shock absorber stiffness and the vehicle mass; determining the target acceleration in the longitudinal direction of the vehicle after correcting the braking influence and the road surface mutation influence according to the first influence amount, the second influence amount and the longitudinal acceleration of the vehicle; determining the collision event confidence according to the relationship between the target acceleration and the preset threshold corresponding to the direction of its own acceleration; and determining whether a collision occurs to the vehicle according to the collision event confidence, aiming to avoid the missed detection of collision events.

Description

Technical Field

[0001] The present application relates to the technical field of vehicles, and particularly to a collision detection method, system, device and medium. Background Art

[0002] With the continuous increase in the popularity of intelligent connected vehicles, the nature of cars is no longer simply a means of transportation. Consumers' expectations for the intelligent functions of cars are getting higher and higher, and the requirements for the intelligent services of car brands are also getting higher and higher. Correspondingly, the intelligent level of cars is also getting higher and higher. Although the increasing intelligent level of cars provides better support for avoiding traffic accidents, traffic accidents still cannot be completely eliminated.

[0003] Currently, there is an E-call system for emergency calls after a traffic accident occurs, but it is only triggered for major accidents where the airbag will pop out. When the intelligent assisted driving related functions are turned on, the system can identify limited risks and perform collision avoidance, but the system does not have an accident recognition function. For the surround view system, although collision recognition can be performed through image recognition technology, when the object being collided is an irregular object, the recognition effect is poor, and due to the limitation of the video frame rate, there is also a problem that accidents with quick contact and separation cannot be recognized. Summary of the Invention

[0004] In view of this, the present application provides a collision detection method, system, device and medium, aiming to solve or partially solve the problems existing in the background art.

[0005] The first aspect of the present application provides a collision detection method, and the method includes:

[0006] Determine a first influence amount of vehicle braking on the vehicle longitudinal acceleration according to the relationship curve between the braking torque and the vehicle longitudinal deceleration and the current braking torque of the vehicle;

[0007] Determine the shock absorber stiffness according to the mapping relationship between the suspension compression height and the shock absorber stiffness and the current suspension compression height of the vehicle;

[0008] Determine a second influence amount of road surface mutation on the vehicle longitudinal acceleration according to the shock absorber stiffness and the vehicle mass;

[0009] Determine the target acceleration in the vehicle longitudinal direction after correcting the braking influence and the road surface mutation influence according to the first influence amount, the second influence amount and the vehicle longitudinal acceleration;

[0010] Determine the collision event confidence according to the relationship between the target acceleration and the preset threshold corresponding to the direction of its own acceleration;

[0011] Determine whether the vehicle has a collision according to the collision event confidence.

[0012] The second aspect of the present application provides a collision detection system, the system includes:

[0013] A first influence quantity determination module, configured to determine a first influence quantity of vehicle braking on the vehicle longitudinal acceleration according to the relationship curve between the braking torque and the vehicle longitudinal deceleration and the current braking torque of the vehicle;

[0014] A shock absorber stiffness determination module, configured to determine the shock absorber stiffness according to the mapping relationship between the suspension compression height and the shock absorber stiffness and the current suspension compression height of the vehicle;

[0015] A second influence quantity determination module, configured to determine a second influence quantity of road surface mutation on the vehicle longitudinal acceleration according to the shock absorber stiffness and the vehicle mass;

[0016] A first acceleration determination module, configured to determine the target acceleration in the vehicle longitudinal direction after correcting the braking influence and the road surface mutation influence according to the first influence quantity, the second influence quantity and the vehicle longitudinal acceleration;

[0017] A confidence determination module, configured to determine the collision event confidence according to the relationship between the target acceleration and a preset threshold corresponding to the direction of its own acceleration;

[0018] A collision determination module, configured to determine whether a vehicle collision occurs according to the collision event confidence.

[0019] The third aspect of the present application provides an electronic device, including: a processor, a memory, and a computer program stored on the memory and running on the processor, and when the computer program is executed by the processor, it implements the steps in a collision detection method as described in the first aspect of the present application.

[0020] The fourth aspect of the present application provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, it implements the steps in a collision detection method as described in the first aspect of the present application.

[0021] The collision detection method provided by the present application has the following advantages:

[0022] A collision detection method provided by an embodiment of the present application first determines a first influence amount of vehicle braking on the vehicle longitudinal acceleration according to the relationship curve between the braking torque and the vehicle longitudinal deceleration and the current braking torque of the vehicle; determines the shock absorber stiffness according to the mapping relationship between the suspension compression height and the shock absorber stiffness and the current suspension compression height of the vehicle; determines a second influence amount of road surface mutation on the vehicle longitudinal acceleration according to the shock absorber stiffness and the vehicle mass; determines the target acceleration in the vehicle longitudinal direction after correcting the braking influence and the road surface mutation influence according to the first influence amount, the second influence amount and the vehicle longitudinal acceleration; determines the collision event confidence according to the relationship between the target acceleration and the preset threshold corresponding to the direction of its own acceleration; and determines whether a collision occurs to the vehicle according to the collision event confidence. Thus, the present application first collects the total acceleration of the vehicle through the IMU of the vehicle, and then corrects the total acceleration based on the state information of the vehicle, excluding the parts of the total acceleration of the vehicle affected by braking and road surface mutation, and finally obtains a target acceleration in the vehicle longitudinal direction. By comparing the target acceleration with a preset threshold set in advance, the confidence of the current collision event is determined. A simple way is that when the target acceleration is a positive acceleration (taking a positive value), the higher the degree that it exceeds a certain preset threshold, the higher the corresponding confidence; when the target acceleration is a negative acceleration (taking a negative value), the higher the degree that it is lower than a certain preset threshold, the higher the corresponding confidence. After the obtained confidence reaches a certain value (such as 90%), it is determined that a collision has occurred to the vehicle. Since the collision detection method provided by the present application determines whether a collision occurs to the vehicle based on the acceleration after eliminating the influence of vehicle braking and ground mutation on the vehicle acceleration, it can avoid the situation that the recognition effect of the surround view system is poor when the collision is an abnormal object, and can avoid the problem that due to the limitation of the video frame number, an accident of quick contact and separation cannot be recognized, so as to achieve the purpose of avoiding missed detection of collision events. At the same time, for the problem that it is difficult to distinguish the misoperation working conditions (such as when the user suddenly performs emergency braking on the speed bump road surface, resulting in a mutation of the acceleration) in the collision detection method based on the total acceleration of the vehicle itself, the way of the present application to determine whether a collision occurs to the vehicle based on the acceleration after eliminating the influence of vehicle braking and ground mutation on the vehicle acceleration can effectively avoid the situation of misdetecting a collision in such misoperation working conditions. At the same time, based on the detection result, the collision event is reported to the manufacturer service background in time, and accident handling guidance, rescue, maintenance service reservation, etc. are provided to the user who has an accident in time, which can greatly improve the user satisfaction and enhance the brand technology image. In addition, the collision detection method provided by the present application only needs the IMU of the vehicle to provide the total acceleration of the vehicle for subsequent algorithm correction, without relying on collision sensors, image sensors and radar sensors, and can effectively reduce the system cost. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] To more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the accompanying drawings required for the description of the embodiments of the present application. Obviously, the accompanying drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0024] Figure 1 Flowchart of a collision detection method shown in an embodiment of the present application;

[0025] Figure 2 Schematic diagram of the relationship curve between braking torque and vehicle longitudinal deceleration in a collision detection method shown in an embodiment of the present application;

[0026] Figure 3 Schematic diagram of the relationship curve between steering wheel angle and turning radius in a collision detection method shown in an embodiment of the present application;

[0027] Figure 4 Another flowchart of a collision detection method shown in an embodiment of the present application;

[0028] Figure 5 Schematic diagram of acceleration correction in the first scenario of a collision detection method shown in an embodiment of the present application;

[0029] Figure 6 Schematic diagram of acceleration correction in the second scenario of a collision detection method shown in an embodiment of the present application;

[0030] Figure 7 Schematic diagram of a collision detection system shown in an embodiment of the present application. Detailed implementation manners

[0031] The following will clearly and completely describe the technical solutions in the embodiments of the present application with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are some, but not all, of the embodiments of the present application. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present application without creative efforts fall within the protection scope of the present application.

[0032] Refer to Figure 1 , Figure 1 which is a flowchart of a collision detection method shown in an embodiment of the present application. As Figure 1 shown, the method includes:

[0033] Step S1: Determine the first influence amount of vehicle braking on vehicle longitudinal acceleration according to the relationship curve between braking torque and vehicle longitudinal deceleration and the current braking torque of the vehicle.

[0034] In this embodiment, as Figure 2 shown, the present application pre - establishes a relationship curve between the braking torque of the vehicle and the longitudinal deceleration of the vehicle. During the normal driving of the vehicle, the current braking torque of the vehicle is determined based on the relevant information collected by the vehicle sensors, and based on the relationship curve pre - established by the present application between the braking torque of the vehicle and the longitudinal deceleration of the vehicle, the longitudinal deceleration corresponding to the current braking torque of the vehicle is determined. This longitudinal deceleration of the vehicle is the first influence quantity on the longitudinal acceleration of the vehicle caused by vehicle braking.

[0035] Step S2: Determine the shock absorber stiffness according to the mapping relationship between the suspension compression height and the shock absorber stiffness and the current suspension compression height of the vehicle.

[0036] In this embodiment, the present application also pre - establishes a mapping relationship between the suspension compression height of the vehicle and the shock absorber stiffness. During the normal driving of the vehicle, the suspension compression height of each current suspension of the vehicle is determined based on the relevant information collected by the vehicle sensors, and based on the mapping relationship pre - established by the present application between the suspension compression height of the vehicle and the shock absorber stiffness, the shock absorber stiffness of each shock absorber corresponding to the suspension compression height of each current suspension of the vehicle is determined.

[0037] Step S3: Determine the second influence quantity on the longitudinal acceleration of the vehicle caused by road surface mutation according to the shock absorber stiffness and the vehicle mass.

[0038] In this embodiment, based on the vehicle mass and the shock absorber stiffness of each current shock absorber of the vehicle determined in step S2, an acceleration is calculated through a first algorithm. This acceleration is the second influence quantity on the longitudinal acceleration of the vehicle caused by road surface mutation. Among them, the expression of the first algorithm is: , where is the second influence quantity, k is the proportionality coefficient, is the change in suspension height of the i - th suspension, m is the vehicle mass, is the shock absorber stiffness of the i - th shock absorber.

[0039] Step S4: Determine the target longitudinal acceleration of the vehicle after correcting the braking influence and the road surface mutation influence according to the first influence quantity, the second influence quantity and the longitudinal acceleration of the vehicle.

[0040] In this embodiment, after obtaining the current first influence quantity and the second influence quantity of the vehicle through steps S1 to S3, the longitudinal acceleration of the vehicle collected by the vehicle's IMU is subtracted by this first influence quantity and the second influence quantity to obtain the target longitudinal acceleration of the vehicle after correcting the braking influence and the road surface mutation influence.

[0041] In this embodiment, another alternative implementation for determining the target acceleration in the longitudinal direction of the vehicle may be to correct the acceleration in the longitudinal direction of the vehicle collected in real time according to the vehicle state information through calculation formulas such as looking up empirical values in a table, and / or quadratic polynomials, and / or piecewise functions, and / or exponential functions, and / or trigonometric functions, etc., to obtain the target acceleration in the longitudinal direction of the vehicle after correcting the influence of braking and road surface mutations.

[0042] Step S5: Determine the confidence level of the collision event according to the relationship between the target acceleration and the preset threshold corresponding to the direction of its own acceleration.

[0043] In this embodiment, after obtaining the target acceleration in the longitudinal direction of the vehicle after eliminating the influence of braking and road surface mutations through steps S1 to S4, determine the direction of the target acceleration. For the positive target acceleration and the negative target acceleration, this application respectively sets corresponding preset thresholds. For example, the positive preset threshold is 2 and the negative one is -1.2. First, determine the direction of the target acceleration, and then compare the target acceleration with the preset threshold corresponding to the direction of its own acceleration. Based on the relationship between the target acceleration and the preset threshold, determine the confidence level of the collision event. An alternative implementation is: when the target acceleration is positive, the greater the degree to which the target acceleration exceeds the preset threshold corresponding to the direction of its own acceleration, the higher the confidence level of the collision event; when the target acceleration is negative, the greater the degree to which the target acceleration is lower than the preset threshold corresponding to the direction of its own acceleration, the higher the confidence level of the collision event. A simple implementation is to establish a mapping relationship between various degrees to which the positive target acceleration exceeds the preset threshold corresponding to the direction of its own acceleration and each value of the confidence level of the collision event, and establish a mapping relationship between various degrees to which the negative target acceleration is lower than the preset threshold corresponding to the direction of its own acceleration and each value of the confidence level of the collision event. Based on these two established mapping relationships and the determined target acceleration in the longitudinal direction of the vehicle, determine the corresponding confidence level of the collision event. Among them, the preset threshold can be set according to the actual scenario and will not be specifically limited here.

[0044] Step S6: Determine whether the vehicle has collided according to the confidence level of the collision event.

[0045] In this embodiment, based on the confidence level of the collision event obtained in step S5, determine whether the vehicle has collided currently. An alternative implementation is: preset a threshold (such as 90%), and then determine whether the obtained confidence level of the collision event is greater than or equal to the threshold. If it is greater than or equal to the threshold, determine that the vehicle has collided currently. At this time, report the collision event of the vehicle to the manufacturer's service background in a timely manner, and provide accident handling guidance, rescue, maintenance service reservation, etc. to the users who have had traffic accidents in a timely manner, so as to achieve the purpose of improving user satisfaction.

[0046] A collision detection method provided by an embodiment of the present application first determines a first influence amount of vehicle braking on the vehicle longitudinal acceleration according to the relationship curve between the braking torque and the vehicle longitudinal deceleration and the current braking torque of the vehicle; determines the shock absorber stiffness according to the mapping relationship between the suspension compression height and the shock absorber stiffness and the current suspension compression height of the vehicle; determines a second influence amount of road surface mutation on the vehicle longitudinal acceleration according to the shock absorber stiffness and the vehicle mass; determines the target acceleration in the longitudinal direction of the vehicle after correcting the braking influence and the road surface mutation influence according to the first influence amount, the second influence amount and the longitudinal acceleration of the vehicle; determines the confidence level of the collision event according to the relationship between the target acceleration and the preset threshold corresponding to the direction of its own acceleration; and determines whether the vehicle has collided according to the confidence level of the collision event. Thus, the present application first collects the total acceleration of the vehicle through the IMU of the vehicle, and then corrects the total acceleration based on the state information of the vehicle, excluding the part of the total acceleration of the vehicle affected by braking and road surface mutation, and finally obtains a target acceleration in the longitudinal direction of the vehicle. By comparing the target acceleration with a preset threshold set in advance, the confidence level of the current collision event is determined. A simple way is that when the target acceleration is a positive acceleration (taking a positive value), the higher the degree of exceeding a certain preset threshold, the higher the corresponding confidence level; when the target acceleration is a negative acceleration (taking a negative value), the higher the degree of being lower than a certain preset threshold, the higher the corresponding confidence level. After the obtained confidence level reaches a certain value (such as 90%), it is determined that the vehicle has collided. Since the collision detection method provided by the present application determines whether the vehicle has collided based on the acceleration after eliminating the influence of vehicle braking and ground mutation on the vehicle acceleration, it can avoid the situation where the recognition effect of the surround view system is poor when the collision is an irregular object, and can avoid the problem that the accident of quick contact and separation cannot be recognized due to the limitation of the video frame number, so as to achieve the purpose of avoiding missed detection of collision events. At the same time, for the problem that it is difficult to distinguish the misoperation conditions (such as sudden emergency braking by the user on the bumpy road surface causing sudden change of acceleration) in the collision detection method based on the total acceleration of the vehicle itself, the method of the present application for determining whether the vehicle has collided based on the acceleration after eliminating the influence of vehicle braking and ground mutation on the vehicle acceleration can effectively avoid the situation of misdetecting collision in such misoperation conditions. At the same time, based on the detection result, the collision event is reported to the manufacturer service background in time, and accident handling guidance, rescue, maintenance service reservation, etc. are provided to the users who have had traffic accidents in time, which can greatly improve the user satisfaction and enhance the brand technology image. In addition, the collision detection method provided by the present application only needs the IMU of the vehicle to provide the total acceleration of the vehicle for subsequent algorithm correction, and does not need to rely on collision sensors, image sensors and radar sensors, which can effectively reduce the system cost.

[0047] Combined with the above embodiments, in one implementation manner, the embodiments of the present application further provide a collision detection method. In this collision detection method, step S5 may include steps S51 to S54:

[0048] Step S51: Compare the target acceleration in the longitudinal direction of the vehicle with the preset threshold corresponding to the direction of its own acceleration.

[0049] In this embodiment, since the acceleration in the longitudinal direction of the vehicle has directions including the forward positive direction and the backward negative direction, the present application sets corresponding preset thresholds for the forward target acceleration and the backward target acceleration respectively. Compare the target acceleration in the longitudinal direction of the vehicle obtained through step S4 with the preset threshold corresponding to the direction of its own acceleration.

[0050] Step S52: When the vehicle has a target acceleration that meets the set conditions, determine the duration from the moment of occurrence when the target acceleration in the longitudinal direction of the vehicle continuously meets the set conditions. The set condition is that the absolute value of the target acceleration is greater than the absolute value of the preset threshold corresponding to the direction of its own acceleration.

[0051] In this embodiment, when the vehicle has a target acceleration in the longitudinal direction that meets the set conditions, determine that from the moment when the target acceleration that meets the set conditions appears, the subsequent target acceleration in the longitudinal direction of the vehicle continuously meets the set conditions for the duration. Among them, the set condition is that the absolute value of the target acceleration is greater than the absolute value of the preset threshold corresponding to the direction of its own acceleration, that is, when the vehicle has a target acceleration whose absolute value is greater than the absolute value of the preset threshold corresponding to the direction of its own acceleration, from the moment when the target acceleration appears, determine whether the target acceleration of the vehicle continuously meets the set conditions. If so, monitor and determine how long it lasts. For example, the vehicle determines the target acceleration in real time, compares each determined target acceleration with the preset threshold corresponding to the direction of its own acceleration. When a target acceleration whose absolute value is greater than the preset threshold corresponding to the direction of its own acceleration appears at a certain moment t0, from this moment t0, determine the relationship between the subsequent target acceleration and the preset threshold corresponding to the direction of its own acceleration. If from the moment t0 to the end of 1 second, the absolute values of all the determined target accelerations during this period are greater than the absolute value of the preset threshold corresponding to the direction of its own acceleration, and at the same time, after 1 second, a target acceleration whose absolute value is less than or equal to the absolute value of the preset threshold corresponding to the direction of its own acceleration is determined, then determine that the duration during which the target acceleration in the longitudinal direction of the vehicle continuously meets the set conditions is 1 second.

[0052] Step S53: Determine the maximum target acceleration of the vehicle during the duration according to the duration.

[0053] In this embodiment, after determining the duration for which the target acceleration in the vehicle longitudinal direction continuously satisfies the set condition through step S52, a maximum target acceleration among all the target accelerations determined during this duration is determined. Among them, when the acceleration direction of the target acceleration is positive, the maximum target acceleration among all the target accelerations refers to the one with the largest acceleration value; when the acceleration direction of the target acceleration is negative, the maximum target acceleration among all the target accelerations refers to the one with the largest absolute value of the acceleration value.

[0054] Step S54: Determine the collision event confidence level according to the relationship between the duration and the preset threshold corresponding to the maximum target acceleration and its own acceleration direction.

[0055] In this embodiment, after determining the duration for which the target acceleration in the vehicle longitudinal direction continuously satisfies the set condition and the maximum target acceleration during this duration through steps S52 and S53, based on the relationship between this duration and the preset threshold corresponding to the maximum target acceleration and its own acceleration direction, the collision event confidence level of the vehicle currently having a collision event is determined. An optional implementation method is: establish a first mapping relationship between the duration and the degree of excess of the absolute value of the maximum target acceleration exceeding the absolute value of the preset threshold corresponding to its own acceleration direction and the collision event confidence level. After determining the duration and the maximum target acceleration, first determine the degree of excess of the absolute value of the maximum target acceleration exceeding the absolute value of the preset threshold corresponding to its own acceleration direction, and then based on this degree of excess and the determined duration, by querying this first mapping relationship, determine the collision event confidence level (such as 80%) corresponding to both this degree of excess and this duration in this first mapping relationship, and determine the determined collision event confidence level (i.e., 80%) as the collision event confidence level of the vehicle currently having a collision.

[0056] Combined with the above embodiments, in one implementation manner, the embodiments of the present application further provide a collision detection method. In this collision detection method, step S6 may include: comparing the collision event confidence level with the upper and lower limits of the confidence threshold range respectively; when the collision event confidence level is greater than or equal to the upper limit of the confidence threshold range, determining that a vehicle collision event has occurred; when the collision event confidence level is less than or equal to the lower limit of the confidence threshold range, determining that no vehicle collision event has occurred; when the collision event confidence level is within the confidence threshold range, obtaining the control actions performed on the vehicle within the target duration range, where the target duration range is: a set duration range before and after the determination moment when the collision event confidence level is determined; determining whether there is a target control action among the performed control actions, where the target control action is a control action related to the collision event; correcting the collision event confidence level according to the performed target control action; and determining whether a vehicle collision has occurred according to the corrected collision event confidence level.

[0057] In this embodiment, after obtaining the collision event confidence level of the current vehicle collision event, it is further determined whether the vehicle has actually had a collision event currently. Specifically, the present application pre-sets a confidence threshold range, which can be set according to the actual application scenario and is not specifically limited here, such as (80%, 90%). First, the obtained collision event confidence level is compared with the upper and lower limits of the confidence threshold range respectively. When the collision event confidence level is greater than or equal to the upper limit of the confidence threshold range, it is determined that a vehicle collision event has occurred. At this time, the collision event of the vehicle is reported to the manufacturer service background in a timely manner, and accident handling guidelines, rescue, maintenance service reservation, etc. are provided to the user who has had a traffic accident in a timely manner, so as to achieve the purpose of improving user satisfaction. When the collision event confidence level is less than or equal to the lower limit of the confidence threshold range, it is determined that no vehicle collision event has occurred. At this time, a new round of determination of whether the vehicle has a collision event continues.

[0058] When the confidence level of the collision event is within the confidence threshold range (for example, between 80% and 90% within the confidence threshold range, and not equal to 80% and not equal to 90%), this is a suspected collision event with a high probability. In this case, the present application will correct the confidence level of the collision event in this situation. The specific correction process is as follows: Obtain the control actions performed by the user on the vehicle within the target time range. The target time range is a set time range before and after the determination time when the confidence level of the collision event is determined. The set time range can be set according to the actual application scenario and is not specifically limited here. For example, a total of 5 seconds, 2.5 seconds before the determination time and 2.5 seconds after the determination time. Then determine whether there is a target control action among all the control actions performed within the target time range. The target control action is a control action related to the collision event, that is, if a collision event actually occurs, generally a target control action will be performed on the vehicle. The target control action at least includes switching to the parking gear, pulling up the handbrake, turning on the hazard lights, opening the door, and emergency braking. If there is a target control action, the confidence level of the collision event is corrected upwards. At the same time, the relevance of different types of target control actions to the collision event is different. Therefore, the correction amplitude of the confidence level of the collision event by different types of target control actions is different. The higher the relevance, the higher the correction amplitude. Then, based on the corrected confidence level of the collision event, it is finally determined whether the vehicle has collided. The determination method is still to compare the corrected confidence level of the collision event with the upper and lower limits of the confidence threshold range. When the comparison result at this time is that the corrected confidence level of the collision event is greater than or equal to the upper limit of the confidence threshold range, it is determined that a collision event has occurred to the vehicle. If the comparison result at this time is that the corrected collision event is less than the upper limit of the confidence threshold range, it is directly determined that the vehicle has not had a collision event, and the confidence level of the collision event will not be corrected again. That is, in one round of collision event determination process of the present application, the confidence level of the collision event will only be corrected once. If the corrected confidence level of the collision event is less than the upper limit of the confidence threshold range, it is directly determined that no collision event has occurred.

[0059] Combined with the above embodiments, in one implementation manner, the embodiment of the present application further provides a collision detection method. In this collision detection method, the method further includes: during the correction process of the confidence level of the collision event after determining the duration, determining whether the vehicle appears again with a target acceleration that meets the set conditions; in the case of reappearance, exiting the correction process, and re-determining the confidence level of the collision event according to the relationship between the reappearing target acceleration and the preset threshold corresponding to its own acceleration direction; and determining whether the vehicle has collided according to the re-determined confidence level of the collision event.

[0060] In this embodiment, in order to quickly identify a collision event that has occurred, the present application provides an implementation for a special scenario. The special scenario is that in many scenarios where a collision is about to occur, users may make some incorrect control actions on the vehicle in a hurry, and thus the above implementation of a collision detection method provided by the present application identifies it as a suspected collision event and further corrects the confidence level of the collision event. In this scenario, a real collision may occur soon, and at this time, the above implementation of a collision detection method provided by the present application may still be in the process of correcting the confidence level of the collision event corresponding to the suspected collision event and cannot quickly identify this real collision. To solve this problem, the present application provides another implementation: during the process of correcting the confidence level of the collision event after determining the confidence level of the collision event based on the duration (this correction process refers to the correction process based on whether the target control action is executed in the above implementation), determine whether the vehicle appears again with a target acceleration that meets the set conditions. If it appears again, then exit the current correction process, and timely re-determine the confidence level of the collision event based on the relationship between the currently reappearing target acceleration and the preset threshold corresponding to the direction of its own acceleration, and determine whether the vehicle has a collision based on the re-determined confidence level of the collision event. Among them, the method for determining the confidence level of the collision event that timely re-determines the confidence level of the collision event based on the relationship between the currently reappearing target acceleration and the preset threshold corresponding to the direction of its own acceleration can be the same as the determination method in the implementation of step S5 above, or can be the same as the determination method in the implementation of steps S51 to S54 above. Here, the implementation for determining whether the vehicle has a collision based on the re-determined confidence level of the collision event can be the same as the implementation of step S6 above, or can be the same as another implementation of step S6 above. For example, starting from a certain moment t0, it is determined that the duration for which the target acceleration in the longitudinal direction of the vehicle continuously meets the set conditions is 1 second. After 1 second, the target acceleration of the vehicle no longer meets the set conditions. At this time, a confidence level of the collision event is calculated after this 1 second, and this confidence level of the collision event meets the situation of a suspected collision event. Therefore, the confidence level of the collision event is corrected accordingly. During the correction process, the vehicle obtains another target acceleration that meets the set conditions through real-time determination. At this time, the correction process will end, and based on the target acceleration that meets the set conditions determined during this correction process, the confidence level of the collision event is re-determined, and it is determined whether the vehicle has a collision based on the re-determined confidence level of the collision event.

[0061] Combined with the above embodiments, in one implementation manner, the embodiments of the present application further provide a collision detection method. In this collision detection method, the method further includes: determining the current turning radius of the vehicle according to the relationship curve between the steering wheel angle and the turning radius and the current steering wheel angle of the vehicle; determining a third influence amount of the vehicle steering on the lateral acceleration of the vehicle according to the turning radius and the current speed of the vehicle; determining the target acceleration in the lateral direction of the vehicle after correcting the steering influence according to the third influence amount and the lateral acceleration of the vehicle; determining a fourth influence amount of the vehicle steering on the yaw acceleration of the vehicle according to the third influence amount and the current speed of the vehicle; determining the target acceleration of the vehicle yaw rotation after correcting the steering influence according to the fourth influence amount and the yaw acceleration of the vehicle; determining the confidence level of the collision event according to the relationship between the target acceleration and the preset threshold corresponding to the direction of its own acceleration, including: determining the first collision event confidence level according to the relationship between the target acceleration in the lateral direction of the vehicle and the lateral setting threshold corresponding to the direction of its own acceleration; determining the second collision event confidence level according to the relationship between the target acceleration in the longitudinal direction of the vehicle and the longitudinal setting threshold corresponding to the direction of its own acceleration; determining the third collision event confidence level according to the relationship between the combined target acceleration of the vehicle and the combined setting threshold corresponding to the direction of its own acceleration, where the combined target acceleration is the acceleration synthesized by the target acceleration in the lateral direction of the vehicle and the target acceleration in the longitudinal direction of the vehicle; determining the fourth collision event confidence level according to the relationship between the target acceleration of the vehicle yaw rotation and the yaw setting threshold corresponding to the direction of its own acceleration; determining the maximum confidence level among the first collision event confidence level, the second collision event confidence level, the third collision event confidence level, and the fourth collision event confidence level as the final collision event confidence level of the vehicle.

[0062] In this embodiment, as Figure 3 shown, the present application pre - establishes a relationship curve between the steering wheel angle and the turning radius. During the normal driving of the vehicle, the current steering wheel angle of the vehicle is determined by collecting relevant information through the vehicle sensors, and based on the relationship curve between the steering wheel angle and the turning radius of the vehicle pre - established by the present application, the current turning radius of the vehicle corresponding to the current steering wheel angle of the vehicle is determined. According to this turning radius and the current speed of the vehicle, an acceleration is calculated through a second algorithm, and this acceleration is determined as the third influence amount of the vehicle steering on the lateral acceleration of the vehicle. Among them, the expression of this second algorithm is: , where, is the third influencing quantity, R is the turning radius, and V is the current speed of the vehicle. Subtract the current lateral acceleration of the vehicle obtained by the vehicle IMU from this third influencing quantity to obtain the target acceleration in the lateral direction of the vehicle after correcting the influence of vehicle steering. At the same time, divide the current third influencing quantity of the vehicle by the current speed of the vehicle to obtain the fourth influencing quantity of vehicle steering on the yaw acceleration of the vehicle, and then subtract this fourth influencing quantity from the current yaw acceleration of the vehicle obtained by the vehicle IMU to obtain the target acceleration of the vehicle's yaw rotation after correcting the influence of vehicle steering. Here, two target accelerations under new acceleration types will be obtained.

[0063] In this embodiment, when a certain implementation manner of this application obtains the target accelerations under the above two new acceleration types, another implementation manner of step S5 is as follows: For the target acceleration in the vehicle lateral acceleration type, this application will determine a first collision event confidence level according to the relationship between the target acceleration in the vehicle lateral direction and the lateral setting threshold corresponding to its own acceleration direction. Its determination method is similar to the implementation manners of steps S51 to S54 above. Two lateral setting thresholds corresponding to the acceleration direction will be set for the target acceleration in the vehicle lateral direction respectively, and then the target acceleration in the vehicle lateral direction will be compared with the lateral preset threshold corresponding to its own acceleration direction; when there is a target acceleration in the vehicle lateral direction that meets the set conditions, determine the duration from the occurrence moment when the target acceleration in the vehicle lateral direction continuously meets the set conditions. The set condition is that the absolute value of the target acceleration in the vehicle lateral direction is greater than the absolute value of the lateral preset threshold corresponding to its own acceleration direction; according to this duration, determine the maximum target acceleration in the vehicle lateral direction during this duration; according to this duration and the relationship between the maximum target acceleration and the lateral preset threshold corresponding to its own acceleration direction, determine a first collision event confidence level.

[0064] Similarly, for the target acceleration under the vehicle longitudinal acceleration type, this application will determine a second collision event confidence level according to the relationship between the target acceleration in the vehicle longitudinal direction and the longitudinal set threshold corresponding to the direction of its own acceleration. The determination method is the same as the implementation manners of the above steps S51 to S54. Two longitudinal set thresholds corresponding to the acceleration direction will be set for the target acceleration in the vehicle longitudinal direction, and then the target acceleration in the vehicle longitudinal direction will be compared with the longitudinal preset threshold corresponding to the direction of its own acceleration. When a target acceleration that meets the set conditions appears in the vehicle longitudinal direction, determine the duration from the moment of appearance during which the target acceleration in the vehicle longitudinal direction continuously meets the set conditions. The set condition is that the absolute value of the target acceleration in the vehicle longitudinal direction is greater than the absolute value of the longitudinal preset threshold corresponding to the direction of its own acceleration. According to this duration, determine the maximum target acceleration in the vehicle longitudinal direction during this duration. According to this duration and the relationship between the maximum target acceleration and the longitudinal preset threshold corresponding to the direction of its own acceleration, determine a second collision event confidence level.

[0065] This application first calculates the target acceleration in the vehicle lateral direction and the target acceleration in the vehicle longitudinal direction through the square root method to obtain a combined target acceleration. Similarly, this application will determine a third collision event confidence level according to the relationship between the combined target acceleration of the vehicle and the combined set threshold corresponding to the direction of its own acceleration. The determination method is similar to the implementation manners of the above steps S51 to S54. Among them, the direction of the combined target acceleration is the same as the target acceleration in the vehicle longitudinal direction at the same moment. Two combined set thresholds corresponding to the acceleration direction will be set for the combined target acceleration of the vehicle, and then the combined target acceleration will be compared with the combined preset threshold corresponding to the direction of its own acceleration. When a combined target acceleration that meets the set conditions appears in the vehicle, determine the duration from the moment of appearance during which the combined target acceleration continuously meets the set conditions. The set condition is that the absolute value of the combined target acceleration is greater than the absolute value of the combined preset threshold corresponding to the direction of its own acceleration. According to this duration, determine the maximum combined target acceleration in the vehicle longitudinal direction during this duration. According to this duration and the relationship between the maximum combined target acceleration and the combined preset threshold corresponding to the direction of its own acceleration, determine a third collision event confidence level.

[0066] Similarly, for the target acceleration in the vehicle yaw acceleration type, the present application will determine a fourth collision event confidence level according to the relationship between the target acceleration of the vehicle yaw rotation and the yaw setting threshold corresponding to the direction of its own acceleration. The determination method is similar to the implementation manners of the above steps S51 to S54. Among them, the direction of the target acceleration of the vehicle yaw rotation is the same as the target acceleration in the vehicle lateral direction at the same moment. The present application will set two longitudinal setting thresholds corresponding to the acceleration direction for the target acceleration in the vehicle longitudinal direction respectively, and then compare the target acceleration of the vehicle yaw rotation with the yaw preset threshold corresponding to the direction of its own acceleration; in the case where the target acceleration of the vehicle yaw rotation that meets the set conditions appears, determine the duration from the occurrence moment when the target acceleration of the vehicle yaw rotation continuously meets the set conditions. The set condition is that the absolute value of the target acceleration of the vehicle yaw rotation is greater than the absolute value of the yaw preset threshold corresponding to the direction of its own acceleration; according to this duration, determine the maximum target acceleration of the vehicle yaw rotation within this duration; according to this duration and the relationship between the maximum target acceleration and the yaw preset threshold corresponding to the direction of its own acceleration, determine a fourth collision event confidence level.

[0067] In this embodiment, compare the obtained first collision event confidence level, second collision event confidence level, third collision event confidence level and fourth collision event confidence level, and select the largest one of them as the final collision event confidence level of the vehicle, and then use this final collision event confidence level to participate in step S6 and the optional implementation manners of step S6 to determine whether the vehicle has a collision.

[0068] Combined with the above embodiments, in one implementation manner, the embodiment of the present application also provides a collision detection method. In this collision detection method, the method further includes: during the correction process of the collision event confidence level after determining the duration, whether a target acceleration that meets the set conditions appears again under at least one acceleration type of the vehicle; in the case of reappearance, exit the correction process, and according to the relationship between the reappeared target acceleration and the set threshold under the acceleration type to which the vehicle itself belongs corresponding to the direction of its own acceleration, re-determine the collision event confidence level under the acceleration type to which the reappeared target acceleration belongs; in the case of obtaining multiple collision event confidence levels after re-determination, determine the largest confidence level among the re-determined multiple collision event confidence levels as the final new collision event confidence level of the vehicle; according to the final new collision event confidence level of the vehicle, determine whether the vehicle has a collision.

[0069] In this embodiment, when the maximum confidence level among multiple confidence levels is determined as the confidence level of the vehicle's final collision event and participates in the determination of whether the vehicle has a collision, the collision detection method provided by this application further includes: during the process of correcting the maximum collision event confidence level after obtaining the maximum collision event confidence level based on the continuous duration (the correction process refers to the correction process based on whether to execute the target control action in the above-mentioned embodiment), determine whether at least one of all target accelerations under various acceleration types of the vehicle (including the target acceleration in the lateral direction of the vehicle, the target acceleration in the longitudinal direction of the vehicle, the combined target acceleration, and the target acceleration of the vehicle's yaw rotation) appears again and meets the set conditions (for example, it is assumed that a combined target acceleration and a target acceleration of the vehicle's yaw rotation appear again and both meet the corresponding set conditions). If it appears again, then exit the current correction process, and timely re-determine the corresponding collision event confidence level based on the relationship between the currently reappearing target acceleration that meets the corresponding set conditions and the preset threshold corresponding to the direction of its own acceleration. It should be understood that if there are multiple reappearing target accelerations that meet the set conditions among all target accelerations under various acceleration types of the vehicle (such as including both a combined target acceleration and a target acceleration of the vehicle's yaw rotation), then correspondingly, timely re-determine the corresponding collision event confidence level based on the relationship between the currently reappearing target accelerations that meet the set conditions (that is, the combined target acceleration and the target acceleration of the vehicle's yaw rotation that appear again simultaneously) and the preset thresholds corresponding to the directions of their own accelerations (that is, the combined preset threshold corresponding to the direction of its own acceleration of the combined target acceleration, and the yaw preset threshold corresponding to the direction of its own acceleration of the target acceleration of the vehicle's yaw rotation), and there will also be multiple corresponding collision event confidence levels (that is, including the third collision event confidence level corresponding to the combined target acceleration and the fourth collision event confidence level corresponding to the target acceleration of the vehicle's yaw rotation). This application will determine the maximum confidence level among the re-determined multiple collision event confidence levels as the vehicle's final new collision event confidence level, and determine whether the vehicle has a collision based on the re-determined vehicle's final new collision event confidence level. Among them, the method of timely re-determining the corresponding collision event confidence level based on the relationship between the currently reappearing target acceleration that meets the set conditions and the preset threshold corresponding to the direction of its own acceleration is the same as the method of determining the first collision event confidence level, the second collision event confidence level, the third collision event confidence level, and the fourth collision event confidence level, and is determined based on the acceleration type corresponding to the currently reappearing target acceleration that meets the set conditions. Here, the implementation method of determining whether the vehicle has a collision based on the re-determined vehicle's final new collision event confidence level can be the same as the implementation method of step S6 above, or can be the same as another implementation method of step S6 above.

[0070] Combined with the above embodiments, in one implementation, the embodiment of the present application further provides a collision detection method. In this collision detection method, the method further includes: predefining that the positive value of the lateral acceleration is the acceleration received by the vehicle to the right, predefining that the positive value of the longitudinal acceleration is the acceleration received by the vehicle forward, and predefining that the positive value of the resultant acceleration is the acceleration received by the vehicle from the rear side; in the case of determining that a vehicle collision event occurs, according to the acceleration type corresponding to the final collision event confidence of the vehicle and the positive or negative value of the corresponding target acceleration, determining the collision type of the collision event; when the corresponding acceleration type is the lateral acceleration and the value of the target acceleration in the vehicle's lateral direction is positive, determining that the collision type is the left collision type; when the corresponding acceleration type is the lateral acceleration and the value of the target acceleration in the vehicle's lateral direction is negative, determining that the collision type is the right collision type; when the corresponding acceleration type is the longitudinal acceleration and the value of the target acceleration in the vehicle's longitudinal direction is positive, determining that the collision type is the rear-end collision type; when the corresponding acceleration type is the longitudinal acceleration and the value of the target acceleration in the vehicle's longitudinal direction is negative, determining that the collision type is the frontal collision type; when the corresponding acceleration type is the resultant acceleration and the value of the resultant target acceleration is positive, determining that the collision type is the rear-side collision type; when the corresponding acceleration type is the resultant acceleration and the value of the resultant target acceleration is negative, determining that the collision type is the front-side collision type; when the corresponding acceleration type is the yaw acceleration, determining that the collision type is the complex collision type.

[0071] In this embodiment, in order to determine the collision type of the collision when it is determined that a vehicle collision occurs, a collision detection method provided by the present application further includes: predefining that the positive value of the vehicle's lateral acceleration is the acceleration received by the vehicle to the right, predefining that the positive value of the vehicle's longitudinal acceleration is the acceleration received by the vehicle forward, and predefining that the positive value of the resultant acceleration is the acceleration received by the vehicle forward. Then, in the case of determining that a vehicle collision occurs, based on the acceleration type corresponding to the final collision event confidence of the vehicle (such as the lateral acceleration) and the positive or negative value of the target acceleration corresponding to the final collision event confidence of the vehicle (i.e., the target acceleration in the vehicle's lateral direction), determining the collision type of the collision event. For example, if the final collision event confidence of the vehicle obtained is determined based on the target acceleration in the vehicle's lateral direction corresponding to the lateral acceleration type, in the case of determining that a vehicle collision occurs, based on this lateral acceleration type and the positive or negative value of the target acceleration in the vehicle's lateral direction, determining the collision type of the collision event.

[0072] When it is determined that a collision has occurred, and the type of acceleration based on which the confidence level of the vehicle's final collision event is determined is lateral acceleration and the value of the target acceleration in the lateral direction of the vehicle is positive, determine that the collision type is a left-side collision type; when it is determined that a collision has occurred, and the type of acceleration based on which the confidence level of the vehicle's final collision event is determined is lateral acceleration and the value of the target acceleration in the lateral direction of the vehicle is negative, determine that the collision type is a right-side collision type; when it is determined that a collision has occurred, and the type of acceleration based on which the confidence level of the vehicle's final collision event is determined is longitudinal acceleration and the value of the target acceleration in the longitudinal direction of the vehicle is positive, determine that the collision type is a rear-end collision type; when it is determined that a collision has occurred, and the type of acceleration based on which the confidence level of the vehicle's final collision event is determined is longitudinal acceleration and the value of the target acceleration in the longitudinal direction of the vehicle is negative, determine that the collision type is a frontal collision type; when it is determined that a collision has occurred, and the type of acceleration based on which the confidence level of the vehicle's final collision event is determined is combined acceleration and the value of the combined target acceleration is positive, determine that the collision type is a rear-side collision type; when it is determined that a collision has occurred, and the type of acceleration based on which the confidence level of the vehicle's final collision event is determined is combined acceleration and the value of the combined target acceleration is negative, determine that the collision type is a front-side collision type; when it is determined that a collision has occurred, and the type of acceleration based on which the confidence level of the vehicle's final collision event is determined is yaw acceleration, determine that the collision type is a complex collision type.

[0073] Combined with the above embodiments, in one implementation, the embodiment of the present application further provides a collision detection method. In this collision detection method, the method further includes: determining and saving the identity information of the in-vehicle user to the local according to the collected image data of the in-vehicle user; when it is determined that the vehicle has a current collision and the collision type that has occurred is obtained, continuously sending a collision event notification for a preset duration to trigger the in-vehicle image acquisition device to collect in-vehicle and out-of-vehicle video data; synchronously saving the collected in-vehicle and out-of-vehicle video data and the flag bit of the collision type to the local and the cloud, and saving the identity information of the in-vehicle user to the cloud.

[0074] In this embodiment, after the user gets in the vehicle and starts the vehicle, image data of all users in the vehicle is collected. Based on the collected image data, the identity information of all users in the vehicle is determined, and the identity information of all current users in the vehicle and the personal health information of all users are stored from the persistent storage database to a specific storage location. The specific storage location stores the identity information of all current users in the vehicle and the personal health information of all users, and the information stored in the specific storage location will be cleared after the user finishes using the vehicle. And when a user gets in the vehicle and starts the vehicle next time, a new round of determination and storage of the identity information of all current users in the vehicle and the personal health information of all users in the specific storage location is performed. Among them, the user identity information corresponding to the image data, and the personal health information of the user corresponding to the user identity information are actively recorded and stored in the persistent storage database by the main user who owns the vehicle. The personal health information corresponding to the user can facilitate rescue personnel to take timely and correct rescue measures in the event of a collision. In the case of determining that the vehicle has collided and determining the type of the collision, in order to prevent the vehicle from not receiving the collision event notification due to the sending of other event notifications, resulting in the vehicle being unable to collect corresponding countermeasures in time. At this time, this application will continuously send collision event notifications for a preset duration to ensure as much as possible that the in-vehicle and out-of-vehicle video data collection by the in-vehicle image acquisition device is triggered, and the operation of synchronously saving the collected in-vehicle and out-of-vehicle video data and the flag bit of the collision type locally and to the cloud is triggered, and the identity information and personal health information of all current users in the vehicle stored in the specific storage location are triggered to be uploaded to the cloud. In this way, accident handling guidance, rescue, maintenance service reservation, etc. are provided to users who have had a traffic accident in a timely manner, thereby improving user satisfaction.

[0075] Combined with the above embodiments, in one implementation, the embodiment of the present application also provides a collision detection method. In this collision detection method, the method further includes: collecting various state information of the vehicle, where the various state information at least includes braking torque, suspension compression height, steering wheel angle, and speed; determining whether the collected various state information meets preset conditions, where the preset conditions at least include state information collected during the vehicle power-on process, state information collected during the system restart process, the signal value of the state information is invalid, and the state information is intermittently lost; in the case where the state information meets the preset conditions, replacing the collected value of the state information with a preset default value corresponding to the type of the state information; performing corresponding formula conversion on the state information that needs formula conversion in the collection; performing time step filtering processing on the state information that belongs to high-frequency signals in the collection according to the scheduling period.

[0076] In this embodiment, to ensure that the relevant status information collected by the vehicle sensors can be used to correctly correct the acceleration of the vehicle, this application will preprocess various types of collected status information. Specifically: determine whether the collected various types of status information meet the preset conditions, and the preset conditions at least include: determining that the preset conditions are met when the status information is the status information collected during the vehicle power-on process; determining that the preset conditions are met when the status information is the status information collected during the system restart process; determining that the preset conditions are met when the signal value of the status information is invalid; determining that the preset conditions are met when the status information is intermittently lost. A preset default value is preset for each type of status information. When the status information meets the preset conditions, the collected value of the status information is replaced with the preset default value corresponding to the type to which the status information belongs. For the status information that needs to be formula-converted in the collection, the corresponding formula conversion is performed, and at the same time, according to the scheduling period, time-step filtering processing is performed on the status information belonging to the high-frequency signal in the collection. Then, all the relevant preprocessed status information is used to correct the acceleration of the vehicle. Among them, various types of status information at least include braking torque, suspension compression height, steering wheel angle, and speed, etc.

[0077] In this embodiment, Figure 4 Another flowchart of a collision detection method shown in an embodiment of this application is as Figure 4As shown in the figure, the entire implementation process of a collision detection method provided by this application can be as follows: First, obtain the relevant state information of the vehicle. Based on the obtained relevant state information, correct the acceleration of the current vehicle to obtain the target acceleration in the longitudinal direction of the vehicle after correcting the influence of vehicle braking and road surface mutations. Then, determine whether the target acceleration of the vehicle meets the set conditions. If the target acceleration of the vehicle does not meet the set conditions, it is determined that no collision event has occurred. At this time, directly end the execution of the subsequent steps and start a new round of determination of whether a collision has occurred. If the target acceleration of the vehicle meets the set conditions, further determine the duration during which the target acceleration of the vehicle meets the set conditions. Then, based on the relationship between this duration and the relationship between the maximum target acceleration of the vehicle and the set threshold within this duration, determine the confidence level of the collision event. Then, further determine the relationship between the confidence level of this collision event and the upper and lower limits of the confidence level threshold range. When it is determined that the confidence level of this collision event is greater than or equal to the upper limit of the confidence level threshold range, it is determined that a collision event has occurred to the vehicle. At this time, end the determination of whether a collision has occurred, and then report the subsequent collision event and relevant information to the cloud platform. When it is determined that the confidence level of this collision event is less than or equal to the lower limit of the confidence level threshold range, it is determined that no collision has occurred to the vehicle. At this time, directly end the execution of the subsequent steps and start a new round of determination of whether a collision has occurred. When it is determined that the confidence level of this collision event is within the confidence level threshold range (the values within this range do not include the upper and lower limits of the confidence level threshold range), further determine whether a target control action has been performed on the vehicle within the target duration. If no target control action has been performed on the vehicle within the target duration, no correction of the confidence level of the collision event is performed. At this time, it is determined that no collision has occurred, and the execution of the subsequent steps will be ended and a new round of determination of whether a collision has occurred will be started. If a target control action has been performed on the vehicle within the target duration, then correct the confidence level of the collision event to obtain the corrected confidence level of the collision event. At this time, determine whether the corrected confidence level of the collision event is greater than or equal to the upper limit of the confidence level threshold range. If the corrected confidence level of the collision event is greater than or equal to the upper limit of the confidence level threshold range, it is determined that a collision has occurred, and then report the subsequent collision event and relevant information to the cloud platform. If the corrected confidence level of the collision event is less than the upper limit of the confidence level threshold range, it is determined that no collision has occurred, and the execution of the subsequent steps will be directly ended and a new round of determination of whether a collision has occurred will be started.

[0078] In this embodiment, Figure 5 is a schematic diagram of acceleration correction in the first scenario of a collision detection method shown in an embodiment of this application. As Figure 5 shown, Figure 5Shown is the first scenario where a driver collides during a misoperation. Assume this scenario is that the vehicle changes lanes to an adjacent lane of its own lane and collides with a vehicle in front in that adjacent lane (the vehicle itself rear-ends the vehicle in front). At the same time, during this process, the user performs a braking control on the vehicle. At this time, the vehicle itself has a forward acceleration, and this collision generates a backward acceleration on the vehicle itself. At this time, the total acceleration of the vehicle is Figure 5 the solid line A1 in Figure 5 . At this time, the absolute value of the total acceleration of the vehicle itself will not exceed a preset threshold. If no correction is made, correspondingly, it will not be recognized that the vehicle itself has collided, resulting in this collision being missed. By using a collision detection method provided in this application to correct the total acceleration of the vehicle itself to eliminate the influence of vehicle braking and the influence of ground mutation on vehicle acceleration, the remaining is like

[0079] the dotted line B1 in Figure 6 such a target acceleration. At this time, the absolute value of the target acceleration B1 of the vehicle itself will exceed the absolute value of the preset threshold corresponding to the direction of its own acceleration, and the exceeding amplitude is relatively large. Correspondingly, this collision will be accurately detected. Figure 6 Shown in Figure 6 is a schematic diagram of acceleration correction in the second scenario of a collision detection method shown in an embodiment of this application. As Figure 6 shown, Figure 6 A2 in it represents the total acceleration of the vehicle, and B2 represents the target acceleration of the vehicle after correction. If A2 is not corrected, the absolute value of the total acceleration of the vehicle will exceed the absolute value of the preset threshold corresponding to the direction of its own acceleration, and the exceeding amplitude is relatively large. Therefore, it will be misjudged as a collision event. But actually, it is caused by the user suddenly braking emergently when the vehicle passes through a bumpy road surface, which is a misoperation condition and should not be triggered. Therefore, by using a collision detection method provided in this application to correct the total acceleration of the vehicle itself to eliminate the influence of braking and the influence of ground mutation on vehicle acceleration, the remaining is like

[0080] In this embodiment, a collision detection method provided by the present application does not rely on a collision acceleration sensor or a collision pressure sensor. Instead, through the IMU and vehicle state information, it uses confidence evaluation to detect collision events, featuring high detection accuracy and low misjudgment probability. It has very good safety redundancy and can almost avoid all misoperation events of airbag ignition strategies. At the same time, it is not used for the ignition control of the occupant restraint system, but is mainly used for the manufacturer to master the clues of collision events and for services such as customer care, rescue, repair appointment, and risk analysis after an accident.

[0081] Based on the same inventive concept, an embodiment of the present application provides a collision detection system, as Figure 7 shown. The collision detection system 700 includes:

[0082] A first influence quantity determination module 701, configured to determine a first influence quantity of vehicle braking on the vehicle longitudinal acceleration according to the relationship curve between the braking torque and the vehicle longitudinal deceleration and the current braking torque of the vehicle;

[0083] A shock absorber stiffness determination module 702, configured to determine the shock absorber stiffness according to the mapping relationship between the suspension compression height and the shock absorber stiffness and the current suspension compression height of the vehicle;

[0084] A second influence quantity determination module 703, configured to determine a second influence quantity of road surface mutation on the vehicle longitudinal acceleration according to the shock absorber stiffness and the vehicle mass;

[0085] A first acceleration determination module 704, configured to determine the target acceleration in the vehicle longitudinal direction after correcting the braking influence and the road surface mutation influence according to the first influence quantity, the second influence quantity, and the vehicle longitudinal acceleration;

[0086] A confidence determination module 705, configured to determine the collision event confidence according to the relationship between the target acceleration and a preset threshold corresponding to its own acceleration direction;

[0087] A collision determination module 706, configured to determine whether the vehicle has collided according to the collision event confidence.

[0088] Optionally, the confidence determination module 705 includes:

[0089] A first comparison module, configured to compare the target acceleration in the vehicle longitudinal direction with a preset threshold corresponding to its own acceleration direction;

[0090] A duration determination module, configured to determine, when a vehicle has a target acceleration that meets a set condition, the duration during which the longitudinal target acceleration of the vehicle continuously meets the set condition starting from the occurrence moment, where the set condition is that the absolute value of the target acceleration is greater than the absolute value of a preset threshold corresponding to its own acceleration direction;

[0091] A maximum target acceleration determination module, configured to determine the maximum target acceleration of the vehicle during the duration according to the duration;

[0092] A collision event confidence determination sub-module, configured to determine the collision event confidence according to the duration and the relationship between the maximum target acceleration and the preset threshold corresponding to its own acceleration direction.

[0093] Optionally, the collision determination module 706 includes:

[0094] A second comparison module, configured to compare the collision event confidence with the upper and lower limits of the confidence threshold range respectively;

[0095] A first collision determination module, configured to determine that a vehicle collision event occurs when the collision event confidence is greater than or equal to the upper limit of the confidence threshold range;

[0096] A second collision determination module, configured to determine that no vehicle collision event occurs when the collision event confidence is less than or equal to the lower limit of the confidence threshold range;

[0097] A control action acquisition module, configured to acquire the control actions performed on the vehicle within a target duration range when the collision event confidence is within the confidence threshold range, where the target duration range is a set duration range before and after the determination moment when the collision event confidence is determined;

[0098] A target control action determination module, configured to determine whether there is a target control action among the performed control actions, where the target control action is a control action related to the collision event;

[0099] A confidence correction module, configured to correct the collision event confidence according to the performed target control action;

[0100] A collision determination sub-module, configured to determine whether a vehicle collision occurs according to the corrected collision event confidence.

[0101] Optionally, the collision detection system 700 further includes:

[0102] A target acceleration determination module, configured to determine whether the vehicle has a target acceleration that meets the set condition again during the correction process of the collision event confidence after the duration;

[0103] A collision event confidence determination module, configured to, in the case of re-occurrence, exit the correction process, and re-determine the collision event confidence according to the relationship between the target acceleration that re-occurs and the preset threshold corresponding to the direction of its own acceleration;

[0104] A collision determination module 706, configured to determine whether a vehicle collision has occurred according to the re-determined collision event confidence.

[0105] Optionally, the collision detection system 700 further includes:

[0106] A turning radius determination module, configured to determine the current turning radius of the vehicle according to the relationship curve between the steering wheel angle and the turning radius and the current steering wheel angle of the vehicle;

[0107] A third influence amount determination module, configured to determine a third influence amount of the vehicle steering on the lateral acceleration of the vehicle according to the turning radius and the current speed of the vehicle;

[0108] A target acceleration determination module for the lateral direction of the vehicle, configured to determine the target acceleration for the lateral direction of the vehicle after correcting the steering influence according to the third influence amount and the lateral acceleration of the vehicle;

[0109] A fourth influence amount determination module, configured to determine a fourth influence amount of the vehicle steering on the yaw acceleration of the vehicle according to the third influence amount and the current speed of the vehicle;

[0110] A target acceleration determination module for the yaw rotation of the vehicle, configured to determine the target acceleration for the yaw rotation of the vehicle after correcting the steering influence according to the fourth influence amount and the yaw acceleration of the vehicle;

[0111] The confidence determination module 705 is configured to determine a first collision event confidence according to the relationship between the target acceleration for the lateral direction of the vehicle and the lateral preset threshold corresponding to the direction of its own acceleration; and is configured to determine a second collision event confidence according to the relationship between the target acceleration for the longitudinal direction of the vehicle and the longitudinal preset threshold corresponding to the direction of its own acceleration; and is configured to determine a third collision event confidence according to the relationship between the combined target acceleration of the vehicle and the combined preset threshold corresponding to the direction of its own acceleration, where the combined target acceleration is the acceleration synthesized by the target acceleration for the lateral direction of the vehicle and the target acceleration for the longitudinal direction of the vehicle; and is configured to determine a fourth collision event confidence according to the relationship between the target acceleration for the yaw rotation of the vehicle and the yaw preset threshold corresponding to the direction of its own acceleration; and is configured to determine the maximum confidence among the first collision event confidence, the second collision event confidence, the third collision event confidence, and the fourth collision event confidence as the final collision event confidence of the vehicle.

[0112] Optionally, the collision detection system 700 further includes:

[0113] A target acceleration determination module, configured to determine whether a target acceleration that meets a set condition appears again under at least one acceleration type of the vehicle during the correction process of the confidence level of the collision event after the duration;

[0114] A confidence level determination module 705, configured to, in the case of reappearance, exit the correction process, and re-determine the confidence level of the collision event under the acceleration type to which the reappearing target acceleration belongs according to the relationship between the reappearing target acceleration and the set threshold under the acceleration type to which the vehicle itself belongs corresponding to the direction of its own acceleration; and configured to, in the case of obtaining multiple confidence levels of the collision event after re-determination, determine the maximum confidence level among the multiple re-determined confidence levels of the collision event as the final new confidence level of the collision event of the vehicle;

[0115] A collision determination module 706, configured to determine whether the vehicle has a collision according to the final new confidence level of the collision event of the vehicle.

[0116] Optionally, the collision detection system 700 further includes:

[0117] A predefined module, configured to predefine that a positive value of the lateral acceleration means the vehicle is subjected to an acceleration to the right, a positive value of the longitudinal acceleration means the vehicle is subjected to an acceleration forward, and a positive value of the combined acceleration means the vehicle is subjected to an acceleration from the rear side;

[0118] A collision type determination module, configured to, in the case of determining that the vehicle has a collision event, determine the collision type of the collision event according to the acceleration type corresponding to the final confidence level of the collision event of the vehicle and the positive or negative value of the corresponding target acceleration; and configured to, when the corresponding acceleration type is the lateral acceleration and the value of the target acceleration in the vehicle lateral direction is positive, determine the collision type as the left collision type; and configured to, when the corresponding acceleration type is the lateral acceleration and the value of the target acceleration in the vehicle lateral direction is negative, determine the collision type as the right collision type; and configured to, when the corresponding acceleration type is the longitudinal acceleration and the value of the target acceleration in the vehicle longitudinal direction is positive, determine the collision type as the rear-end collision type; and configured to, when the corresponding acceleration type is the longitudinal acceleration and the value of the target acceleration in the vehicle longitudinal direction is negative, determine the collision type as the head-on collision type; and configured to, when the corresponding acceleration type is the combined acceleration and the value of the combined target acceleration is positive, determine the collision type as the rear-side collision type; and configured to, when the corresponding acceleration type is the combined acceleration and the value of the combined target acceleration is negative, determine the collision type as the front-side collision type; and configured to, when the corresponding acceleration type is the yaw acceleration, determine the collision type as the complex collision type.

[0119] Optionally, the collision detection system 700 further includes:

[0120] An identity information determination module, configured to determine and save the identity information of the in-vehicle user to the local according to the acquired image data of the in-vehicle user;

[0121] An event trigger module, configured to continuously send a collision event notification for a preset duration to trigger the in-vehicle image acquisition device to acquire video data inside and outside the vehicle when it is determined that the vehicle has a current collision and the type of collision that has occurred;

[0122] A storage module, configured to synchronously save the acquired video data inside and outside the vehicle and the flag bit of the collision type to the local and the cloud, and save the identity information of the in-vehicle user to the cloud.

[0123] Optionally, the collision detection system 700 further includes:

[0124] A status information acquisition module, configured to acquire various status information of the vehicle, where the various status information at least includes braking torque, suspension compression height, steering wheel angle, and speed;

[0125] A preprocessing module, configured to determine whether the acquired various status information meets preset conditions, where the preset conditions at least include status information acquired during the vehicle power-on process, status information acquired during the system restart process, the signal value of the status information is invalid, and the status information is intermittently lost; and, in the case where the status information meets the preset conditions, replacing the acquired value of the status information with a preset default value corresponding to the type of the status information; and, performing corresponding formula conversion on the acquired status information that requires formula conversion; and, performing time step filtering processing on the acquired status information belonging to high-frequency signals according to the scheduling period.

[0126] Based on the same inventive concept, an embodiment of the present application provides an electronic device, including: a processor, a memory, and a computer program stored on the memory and running on the processor, where when the computer program is executed by the processor, the steps in a collision detection method as described in the first aspect of the present application are implemented.

[0127] Based on the same inventive concept, an embodiment of the present application provides a computer-readable storage medium, where a computer program is stored on the computer-readable storage medium, and when the computer program is executed by a processor, the steps in a collision detection method as described in the first aspect of the present application are implemented.

[0128] For the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and for the related parts, refer to the partial description of the method embodiment.

[0129] It should be noted that, for the method embodiments, for the sake of simple description, they are all expressed as a series of action combinations. However, those skilled in the art should understand that the embodiments of the present application are not limited by the described action sequences, because according to the embodiments of the present application, certain steps can be performed in other sequences or simultaneously. Secondly, those skilled in the art should also understand that the embodiments described in the specification are all preferred embodiments, and the actions involved are not necessarily essential to the embodiments of the present application.

[0130] Each embodiment in this specification is described in a progressive manner. The key point of each embodiment is to illustrate the differences from other embodiments. For the same or similar parts among the embodiments, reference can be made to each other.

[0131] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the embodiments of the present application can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the embodiments of the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk memory, CD-ROM, optical memory, etc.) containing computer-usable program code.

[0132] The embodiments of the present application are described with reference to the flowcharts and / or block diagrams of methods, terminal devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each flow and / or block in the flowchart and / or block diagram, as well as the combination of flows and / or blocks in the flowchart and / or block diagram, can be realized by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing terminal devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing terminal devices generate a device for realizing the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0133] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing terminal device to work in a specific manner, so that the instructions stored in the computer-readable memory generate a manufactured product including an instruction device, and the instruction device realizes the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0134] These computer program instructions can also be loaded onto a computer or other programmable data processing terminal device, so that a series of operation steps are executed on the computer or other programmable terminal device to generate a computer-implemented process. Thus, the instructions executed on the computer or other programmable terminal device provide steps for implementing the functions specified in one process or multiple processes and / or blocks. Figure 1 one process or multiple processes and / or blocks Figure 1 steps for implementing the functions specified in one block or multiple blocks.

[0135] Although the preferred embodiments of the embodiments of the present application have been described, those skilled in the art can make additional changes and modifications once they learn the basic creative concepts. Therefore, the appended claims are intended to be construed as including the preferred embodiments and all changes and modifications falling within the scope of the embodiments of the present application.

[0136] Finally, it should also be noted that in this text, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or terminal device comprising a series of elements not only includes those elements, but also includes other elements not expressly listed, or also includes elements inherent to such process, method, article or terminal device. Without further limitation, an element defined by the statement "comprising one..." does not exclude the existence of additional identical elements in the process, method, article or terminal device comprising the said element.

[0137] The above provides a detailed introduction to a collision detection method, system, device and medium provided by the present application. Specific examples are used in this text to elaborate on the principle and implementation manner of the present application. The description of the above embodiments is only used to help understand the method and its core idea of the present application; at the same time, for those of ordinary skill in the art, according to the idea of the present application, there will be changes in the specific implementation manner and application scope. In summary, the content of this specification should not be construed as a limitation to the present application.

Claims

1. A collision detection method, characterized in that, The method includes: Determining a first influence amount of vehicle braking on the vehicle longitudinal acceleration according to a relationship curve between a braking torque and a vehicle longitudinal deceleration and the current braking torque of the vehicle; Determining a shock absorber stiffness according to a mapping relationship between a suspension compression height and a shock absorber stiffness and the current suspension compression height of the vehicle; Determining a second influence amount of a road surface mutation on the vehicle longitudinal acceleration according to the shock absorber stiffness and the vehicle mass; Determining a target acceleration in the vehicle longitudinal direction after correcting for the braking influence and the road surface mutation influence according to the first influence amount, the second influence amount, and the vehicle longitudinal acceleration; Determining a collision event confidence level according to a relationship between the target acceleration and a preset threshold corresponding to the direction of its own acceleration; Determining whether a vehicle collision has occurred according to the collision event confidence level; 2. The collision detection method according to claim 1, wherein Determining a collision event confidence level according to a relationship between the target acceleration and a preset threshold corresponding to the direction of its own acceleration includes: Comparing the target acceleration in the vehicle longitudinal direction with a preset threshold corresponding to the direction of its own acceleration; When the vehicle has a target acceleration that satisfies a set condition, determining a duration during which the target acceleration in the vehicle longitudinal direction continuously satisfies the set condition starting from the occurrence moment, where the set condition is that the absolute value of the target acceleration is greater than the absolute value of the preset threshold corresponding to the direction of its own acceleration; Determining the maximum target acceleration of the vehicle during the duration according to the duration; Determining a collision event confidence level according to the duration and a relationship between the maximum target acceleration and a preset threshold corresponding to the direction of its own acceleration; 3. The collision detection method according to claim 2, wherein Determining whether a vehicle collision has occurred according to the collision event confidence level includes: Comparing the collision event confidence level with the upper and lower limits of a confidence threshold range respectively; When the collision event confidence level is greater than or equal to the upper limit of the confidence threshold range, determining that a vehicle collision event has occurred; When the collision event confidence level is less than or equal to the lower limit of the confidence threshold range, determining that no vehicle collision event has occurred; When the collision event confidence level is within the confidence threshold range, obtaining control actions performed on the vehicle within a target duration range, where the target duration range is a set duration range before and after the determination moment when the collision event confidence level is determined; Determining whether there is a target control action among the performed control actions, where the target control action is a control action related to a collision event; Correcting the collision event confidence level according to the performed target control action; Determining whether a vehicle collision has occurred according to the corrected collision event confidence level; 4. A collision detection method according to claim 3, characterized in that, The method further includes: Determining whether the vehicle has a target acceleration that satisfies the set condition again during the correction process of the collision event confidence level after determining the duration; In the case of a recurrence, exiting the correction process and re-determining a collision event confidence level according to a relationship between the recurrence target acceleration and a preset threshold corresponding to the direction of its own acceleration; Determining whether a vehicle collision has occurred according to the re-determined collision event confidence level; 5. A collision detection method according to claim 3, characterized in that, The method further includes: Determine the current turning radius of the vehicle according to the relationship curve between the steering wheel angle and the turning radius and the current steering wheel angle of the vehicle; Determine the third influence quantity of the vehicle steering on the lateral acceleration of the vehicle according to the turning radius and the current speed of the vehicle; Determine the target acceleration in the lateral direction of the vehicle after correcting the steering influence according to the third influence quantity and the lateral acceleration of the vehicle; Determine the fourth influence quantity of the vehicle steering on the yaw acceleration of the vehicle according to the third influence quantity and the current speed of the vehicle; Determine the target acceleration of the vehicle yaw rotation after correcting the steering influence according to the fourth influence quantity and the yaw acceleration of the vehicle; The determining the confidence level of the collision event according to the relationship between the target acceleration and the preset threshold corresponding to the direction of its own acceleration includes: Determine the first collision event confidence level according to the relationship between the target acceleration in the lateral direction of the vehicle and the lateral preset threshold corresponding to the direction of its own acceleration; Determine the second collision event confidence level according to the relationship between the target acceleration in the longitudinal direction of the vehicle and the longitudinal preset threshold corresponding to the direction of its own acceleration; Determine the third collision event confidence level according to the relationship between the combined target acceleration of the vehicle and the combined preset threshold corresponding to the direction of its own acceleration, where the combined target acceleration is the acceleration synthesized by the target acceleration in the lateral direction of the vehicle and the target acceleration in the longitudinal direction of the vehicle; Determine the fourth collision event confidence level according to the relationship between the target acceleration of the vehicle yaw rotation and the yaw preset threshold corresponding to the direction of its own acceleration; Determine the maximum confidence level among the first collision event confidence level, the second collision event confidence level, the third collision event confidence level, and the fourth collision event confidence level as the final collision event confidence level of the vehicle.

6. A collision detection method according to claim 5, wherein The method further includes: Determine whether a target acceleration that meets the set conditions appears again under at least one acceleration type of the vehicle during the correction process of the collision event confidence level after determining the duration; In the case of reappearance, exit the correction process, and re-determine the collision event confidence level under the acceleration type to which the reappearing target acceleration belongs according to the relationship between the reappearing target acceleration and the set threshold under the acceleration type to which its own acceleration belongs corresponding to the direction of its own acceleration; In the case of obtaining multiple collision event confidence levels after re-determination, determine the maximum confidence level among the re-determined multiple collision event confidence levels as the final new collision event confidence level of the vehicle; Determine whether the vehicle has a collision according to the final new collision event confidence level of the vehicle.

7. A collision detection method according to claim 5, characterized in that, The method further includes: Pre-define the positive value of the lateral acceleration as the vehicle is subjected to an acceleration to the right, the positive value of the longitudinal acceleration as the vehicle is subjected to an acceleration forward, and the positive value of the combined acceleration as the vehicle is subjected to an acceleration from the rear side; In the case of determining that the vehicle has a collision event, determine the collision type of the collision event according to the acceleration type corresponding to the final collision event confidence level of the vehicle and the positive or negative value of the corresponding target acceleration; When the corresponding acceleration type is the lateral acceleration and the value of the corresponding target acceleration in the lateral direction of the vehicle is positive, determine the collision type as the left collision type; When the corresponding acceleration type is lateral acceleration and the value of the target acceleration in the corresponding vehicle lateral direction is negative, determine that the collision type is the right-side collision type; When the corresponding acceleration type is longitudinal acceleration and the value of the target acceleration in the corresponding vehicle longitudinal direction is positive, determine that the collision type is the rear-end collision type; When the corresponding acceleration type is longitudinal acceleration and the value of the target acceleration in the corresponding vehicle longitudinal direction is negative, determine that the collision type is the head-on collision type; When the corresponding acceleration type is combined acceleration and the value of the corresponding combined target acceleration is positive, determine that the collision type is the rear-side collision type; When the corresponding acceleration type is combined acceleration and the value of the corresponding combined target acceleration is negative, determine that the collision type is the front-side collision type; When the corresponding acceleration type is yaw acceleration, determine that the collision type is the complex collision type.

8. A collision detection method according to claim 1, characterized in that, The method further includes: Determine and save the identity information of the in-vehicle user to the local according to the collected image data of the in-vehicle user; In the case of determining that the vehicle has a current collision and obtaining the collision type that has occurred, continuously send a collision event notification for a preset duration to trigger the in-vehicle image acquisition device to collect the in-vehicle and out-of-vehicle video data; Synchronously save the collected in-vehicle and out-of-vehicle video data and the flag bit of the collision type to the local and the cloud, and save the identity information of the in-vehicle user to the cloud.

9. A collision detection method according to claim 1, wherein The method further includes: Collect various state information of the vehicle, and the various state information at least includes braking torque, suspension compression height, steering wheel angle and speed; Determine whether the collected various state information meets the preset conditions, and the preset conditions at least include the state information collected during the vehicle power-on process, the state information collected during the system restart process, the signal value of the state information is invalid, and the state information is intermittently lost; In the case where the state information meets the preset conditions, replace the collected value of the state information with the preset default value corresponding to the type of the state information; Perform corresponding formula conversion on the state information that needs formula conversion collected; According to the scheduling period, perform time-step filtering processing on the state information belonging to the high-frequency signal collected.

10. A collision detection system, characterized in that, The system includes: The first influence quantity determination module is used to determine the first influence quantity of the vehicle braking on the vehicle longitudinal acceleration according to the relationship curve between the braking torque and the vehicle longitudinal deceleration and the current braking torque of the vehicle; The shock absorber stiffness determination module is used to determine the shock absorber stiffness according to the mapping relationship between the suspension compression height and the shock absorber stiffness and the current suspension compression height of the vehicle; The second influence quantity determination module is used to determine the second influence quantity of the road surface mutation on the vehicle longitudinal acceleration according to the shock absorber stiffness and the vehicle mass; The first acceleration determination module is used to determine the target acceleration in the vehicle longitudinal direction after correcting the braking influence and the road surface mutation influence according to the first influence quantity, the second influence quantity and the longitudinal acceleration of the vehicle; The confidence determination module is used to determine the collision event confidence according to the relationship between the target acceleration and the preset threshold corresponding to its own acceleration direction; A collision determination module, configured to determine whether a vehicle has collided according to the confidence level of the collision event.

11. An electronic device, characterized in that, It includes: A processor, a memory, and a computer program stored on the memory and running on the processor. When the computer program is executed by the processor, the steps in a collision detection method according to any one of claims 1 to 9 are implemented.

12. A computer-readable storage medium, characterized in that, A computer program is stored on the computer-readable storage medium. When the computer program is executed by a processor, the steps in a collision detection method according to any one of claims 1 to 9 are implemented.

Images