Airbag control method and device, electronic equipment and storage medium

Abstract

The invention provides an air bag restraint system control method and device, electronic equipment and a storage medium, the method comprises the steps that collision sensing data and passenger state sensing data of a vehicle are obtained, the collision sensing data comprise the collision direction and collision motion data, and the passenger state sensing data comprise the safety belt connection state; determining a collision risk value corresponding to a preset head injury criterion according to the collision direction and the collision motion data, and if the collision risk value is greater than a corresponding preset risk threshold value, determining collision strength according to the collision motion data and the passenger state sensing data, a control instruction corresponding to the collision level is generated according to a comparison result of the collision strength and a preset strength threshold value so as to perform hierarchical unfolding control on the safety air bag; according to the method and the device, the safety air bag is subjected to hierarchical unfolding control in combination with various data including the safety belt connection state, so that risk hierarchical protection is realized, and unified control of effective protection and injury minimization of a driver and passengers is realized.

Description

Technical Field

[0001] This application relates to the technical field of vehicle safety, and particularly to an airbag control method, device, electronic device, and storage medium. Background Art

[0002] In a vehicle collision accident, seat belts and airbags, as important components of the occupant restraint system, play a crucial role in protecting the safety of drivers and passengers. However, in the case of not wearing a seat belt, the body of the occupant may experience a large degree of forward movement due to the collision impact, resulting in non-ideal contact with the occupant during the high-speed deployment of the airbag. This contact may cause unnecessary impact injuries to parts such as the neck and face, reducing the protective effect of the airbag instead, and even posing a risk of secondary injury.

[0003] In the related art, the control of the airbag usually only relies on the signals collected by the collision sensor to judge the collision intensity, and decides whether to trigger the deployment action of the airbag accordingly, and it is impossible to achieve unified control of effective protection of the occupant and minimization of injuries. Summary of the Invention

[0004] This application provides an airbag control method, device, electronic device, and storage medium to solve the above technical problem that in the process of airbag control, it is impossible to achieve unified control of effective protection of the occupant and minimization of injuries.

[0005] In some embodiments of this application, this application provides an airbag control method, including: obtaining collision perception data and occupant state perception data of the vehicle, where the collision perception data includes a collision direction and collision motion data, and the occupant state perception data includes a seat belt connection state; determining a collision risk value corresponding to a preset head injury criterion according to the collision direction and the collision motion data; if the collision risk value is greater than the corresponding preset risk threshold, determining the collision intensity according to the collision motion data and the occupant state perception data, and generating a control instruction corresponding to the collision level according to the comparison result between the collision intensity and a preset intensity threshold, so as to perform hierarchical deployment control on the airbag.

[0006] In some embodiments of this application, determining the collision intensity according to the collision motion data and the occupant state perception data includes: if the seat belt connection state is not connected, determining the vehicle deceleration during the collision as the human equivalent acceleration; if the seat belt connection state is connected, determining the human equivalent acceleration according to the collision motion data, the occupant mass, and the preset energy absorption parameter of the seat belt; determining the human equivalent acceleration as the collision intensity; where the collision motion data includes the vehicle deceleration during the collision, and the occupant state perception data further includes the occupant mass.

[0007] In some embodiments of the present application, determining the human equivalent acceleration according to the collision motion data, the occupant mass, and the preset energy absorption parameters of the seat belt includes: determining the energy absorbed by the seat belt according to the actual vehicle mass, the vehicle speed before the collision, the vehicle speed after the collision, and the seat belt efficiency coefficient; determining the human equivalent acceleration according to the rated stretching distance of the seat belt, the occupant mass, and the energy absorbed by the seat belt; wherein, the collision motion data further includes the vehicle speed before the collision and the vehicle speed after the collision, the preset energy absorption parameters include the seat belt efficiency coefficient and the rated stretching distance of the seat belt, and the actual vehicle mass is obtained based on the preset vehicle mass and the occupant mass.

[0008] In some embodiments of the present application, generating a control instruction corresponding to the collision level according to the comparison result between the collision intensity and the preset intensity threshold to perform hierarchical deployment control on the airbag includes: if the collision intensity is less than or equal to the first intensity threshold, generating a control instruction corresponding to a low-intensity collision to suppress the pop-up of the airbag and generating a warning prompt message; if the collision intensity is greater than the first intensity threshold and the collision intensity is less than or equal to the second intensity threshold, generating a control instruction corresponding to a medium-intensity collision to make the inflation pressure of the airbag within a preset pressure range and make the deployment time of the airbag within a preset deployment period; if the collision intensity is greater than the second intensity threshold, generating a control instruction corresponding to a high-intensity collision to make the airbag deploy fully and directionally; wherein, the preset intensity threshold includes the first intensity threshold and the second intensity threshold.

[0009] In some embodiments of the present application, after performing hierarchical deployment control on the airbag, it further includes: determining the collision impact force corresponding to the airbag according to the collision intensity and the occupant mass in the occupant status perception data; performing injury assessment based on the collision impact force and the seat position data in the occupant status perception data to obtain a first assessment result, and the injury assessment includes at least one of chest force assessment and leg injury assessment; performing internal organ injury assessment according to the chest compression amount and the chest compression speed to obtain a second assessment result, and the chest compression amount and the chest compression speed are obtained based on the chest force assessment; adjusting the parameters of the hierarchical deployment control based on the first assessment result and the second assessment result.

[0010] In some embodiments of the present application, before performing hierarchical deployment control on the airbag, it further includes: inputting the current in-vehicle image in the occupant state perception data into a heatmap prediction model to obtain the current predicted heatmap and current predicted coordinates of multiple current predicted key points; determining the target limb of each driver and passenger according to the part affinity fields corresponding to multiple initial limbs, where the initial limbs are connected based on each of the current predicted coordinates; determining the limb part corresponding to the target limb of each driver and passenger based on each of the current predicted heatmaps, and determining the body forward tilt angle of each driver and passenger according to the seat position data corresponding to each driver and passenger and the limb part, where the seat position data is obtained based on the occupant state perception data; if the body forward tilt angle of a driver and passenger is greater than or equal to a preset angle threshold, activating the surface flexible coating of the airbag; where the heatmap prediction model is obtained based on historical in-vehicle images and the key point annotation information of the historical in-vehicle images.

[0011] In some embodiments of the present application, the heatmap prediction model is obtained based on historical in-vehicle images and the key point annotation information of the historical in-vehicle images, including: performing key point annotation on the historical in-vehicle image to obtain the annotation coordinates of multiple annotated key points; generating a converted heatmap for each annotated key point according to the annotation coordinates of each annotated key point; inputting the historical in-vehicle image into a network prediction model to obtain the historical predicted heatmap of multiple historical predicted key points, and determining the historical predicted coordinates of the historical predicted key points by taking the peak coordinates in the predicted heatmap; optimizing the parameters of the network prediction model according to the difference between each converted heatmap and the corresponding historical predicted heatmap, and / or the difference between each annotation coordinate and the corresponding historical predicted coordinate, to obtain the heatmap prediction model.

[0012] In some embodiments of the present application, the present application provides an airbag control device, including: a data acquisition module and a control decision module; the data acquisition module is used to acquire the collision perception data and occupant state perception data of the vehicle, the collision perception data includes the collision direction and collision motion data, and the occupant state perception data includes the seat belt connection state; the control decision module includes: a collision risk determination unit, used to determine the collision risk value corresponding to a preset head injury criterion according to the collision direction and the collision motion data; a control instruction generation unit, used to, if the collision risk value is greater than the corresponding preset risk threshold, determine the collision intensity according to the collision motion data and the occupant state perception data, and generate a control instruction corresponding to the collision level according to the comparison result between the collision intensity and the preset intensity threshold, so as to perform hierarchical deployment control on the airbag.

[0013] In some embodiments of the present application, the present application provides an electronic device, which includes: one or more processors; a storage device for storing one or more programs, and when the one or more programs are executed by the one or more processors, the electronic device implements the steps of the airbag control method as described in any one of the above.

[0014] In some embodiments of the present application, 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 of a computer, the computer executes the steps of the airbag control method as described in any one of the above.

[0015] Advantages of the embodiments of the present application: The present application provides an airbag control method, device, electronic device and storage medium. In the embodiments of the present application, the collision risk value corresponding to the preset head injury criterion is determined through the collision direction and collision motion data, the high collision risk is determined by comparing the collision risk value with the preset risk threshold, and the collision intensity is calculated based on the seat belt connection state to achieve the hierarchical deployment control of the airbag. The present application combines multiple data including the seat belt connection state to control the airbag, thereby realizing risk classification protection and achieving unified control of effectively protecting the driver and passengers and minimizing injuries.

[0016] It should be understood that the above general description and subsequent detailed description are only exemplary and explanatory, and cannot limit the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The accompanying drawings herein are incorporated into the specification and form a part of the specification, showing embodiments consistent with the present application, and are used together with the specification to explain the principles of the present application. Obviously, the drawings in the following description are only some embodiments of the present application, and those of ordinary skill in the art can obtain other drawings based on these drawings without creative efforts. In the drawings:

[0018] Figure 1 A schematic diagram of an exemplary system architecture to which the technical solution of the embodiments of the present application can be applied is shown;

[0019] Figure 2 A flowchart of an airbag control method according to an embodiment of the present application is shown;

[0020] Figure 3 A block diagram of an airbag control device according to an embodiment of the present application is shown;

[0021] Figure 4 A schematic diagram of the structure of a computer system of an electronic device suitable for implementing the embodiments of the present application is shown. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0022] The following specific examples illustrate the implementation manners of the present application. Those skilled in the art can easily understand other advantages and effects of the present application from the content disclosed in this specification. The present application can also be implemented or applied through other different specific implementation manners. Various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present application. It should be noted that, without conflict, the following embodiments and the features in the embodiments can be combined with each other.

[0023] In the following embodiments, the diagrams provided are only used to illustrate the basic concept of the present application in a schematic manner. Therefore, only the components related to the present application are shown in the diagrams, rather than being drawn according to the number, shape, and size of the components in actual implementation. The form, quantity, and proportion of each component in actual implementation can be arbitrarily changed, and the layout form of the components may also be more complex.

[0024] In the following description, a large number of details are explored to provide a more thorough explanation of the embodiments of the present application. However, it is obvious to those skilled in the art that the embodiments of the present application can be implemented without these specific details. In other embodiments, well-known structures and devices are shown in the form of block diagrams rather than in detail to avoid making the embodiments of the present application difficult to understand.

[0025] In the related art, the control of the airbag usually only depends on the signals collected by the collision sensor to judge the collision intensity, and accordingly decides whether to trigger the deployment action of the airbag, and it is impossible to achieve unified control of effectively protecting the driver and passengers and minimizing injuries. However, the present application can combine various data including the connection state of the seat belt to achieve hierarchical deployment control of the airbag, thereby realizing risk-based hierarchical protection.

[0026] Please refer to Figure 1 , Figure 1 which shows a schematic diagram of an exemplary system architecture to which the technical solution of the embodiment of the present application can be applied. As Figure 1As shown, the system architecture may include a seat belt buckle sensor 101, a collision acceleration sensor 102, a seat sensor 103, an electronic control unit 104, and a safety airbag controller 105. Among them, the seat belt connection status in the occupant state perception data can be obtained through the seat belt buckle sensor 101. The seat sensor 103 includes a weight sensor and a position sensor, so that the occupant mass and seat position data in the occupant state perception data can be obtained. The collision perception data, such as the collision direction and collision motion data, can be obtained through the collision acceleration sensor 102. The data collected by the seat belt buckle sensor 101 is transmitted to the electronic control unit 104 through Digital Input (DI). The data collected by the collision acceleration sensor 102 is transmitted to the electronic control unit 104 through the Serial Peripheral Interface (SPI). The data collected by the seat position sensor 103 is transmitted to the electronic control unit 104 through the Controller Area Network (CAN). The electronic control unit 104 generates a control instruction corresponding to the collision level based on the obtained data and transmits it to the safety airbag controller 105 through CAN, so as to realize the deployment control of the safety airbag through the safety airbag controller 105.

[0027] Please refer to Figure 2 , Figure 2 shows a schematic flow chart of a safety airbag control method according to an embodiment of the present application. As Figure 2 shown, in an exemplary embodiment, the safety airbag control method at least includes steps S210 to S230, which are introduced in detail as follows:

[0028] Step S210, obtain the collision perception data and occupant state perception data of the vehicle.

[0029] Among them, the collision perception data includes the collision direction and collision motion data, and the occupant state perception data includes the seat belt connection status.

[0030] Step S220, determine the collision risk value corresponding to the preset head injury criterion according to the collision direction and collision motion data.

[0031] Step S230, if the collision risk value is greater than the corresponding preset risk threshold, determine the collision intensity according to the collision motion data and occupant state perception data, and generate a control instruction corresponding to the collision level according to the comparison result of the collision intensity and the preset intensity threshold, so as to perform hierarchical deployment control on the safety airbag.

[0032] In some embodiments of the present application, the collision direction includes a frontal collision and a side collision.

[0033] In some embodiments of the present application, the collision motion data includes at least one of the vehicle deceleration at the time of collision, the vehicle speed before collision, the vehicle speed after collision, and the collision duration.

[0034] In some embodiments of the present application, the collision duration is obtained based on the collision start time and the collision end time.

[0035] In some embodiments of the present application, the occupant state perception data further includes the occupant mass, the front-rear position of the seat, the height position of the seat, and the current in-vehicle image.

[0036] In some embodiments of the present application, the preset head injury criterion includes a first head injury criterion corresponding to a frontal collision and a second head injury criterion corresponding to a side collision.

[0037] In some embodiments of the present application, during the collision process of the vehicle, it only includes a frontal collision, or only includes a side collision, or includes both a frontal collision and a side collision.

[0038] In some embodiments of the present application, the collision risk value includes the value corresponding to the frontal collision risk and / or the side collision risk.

[0039] In some embodiments of the present application, the first head injury criterion corresponding to the frontal collision is as follows:

[0040]

[0041] Wherein, HIC is the frontal collision risk, t2 is the collision end time, t1 is the collision start time, and a(t) is the average collision acceleration.

[0042] In some embodiments of the present application, the average collision acceleration is as follows:

[0043]

[0044] Wherein, a(t) is the average collision acceleration, v2 is the vehicle speed after collision, v1 is the vehicle speed before collision, Δt is the collision duration, t2 is the collision end time, and t1 is the collision start time.

[0045] In some embodiments of the present application, the second head injury criterion corresponding to the side collision is as follows:

[0046]

[0047] Wherein, HPC is the side collision risk, t1 is the collision start time, t2 is the collision end time, and a(t) is the average collision acceleration.

[0048] In some embodiments of the present application, in step S230, if the value of the frontal collision risk is greater than the first risk threshold, or the value of the side collision risk is greater than the second risk threshold, or the value of the frontal collision risk is greater than the first risk threshold and the value of the side collision risk is greater than the second risk threshold, the collision intensity is determined based on the collision motion data and the occupant state perception data, and a control instruction corresponding to the collision level is generated according to the comparison result between the collision intensity and the preset intensity threshold; the preset risk threshold includes the first risk threshold and / or the second risk threshold.

[0049] In some embodiments of the present application, the preset risk threshold is set based on standards such as national standards and industry standards. For example, the first risk threshold is 700 and the second risk threshold is 1000.

[0050] In some embodiments of the present application, determining the collision intensity based on the collision motion data and the occupant state perception data includes: if the seat belt connection state is not connected, determining the vehicle deceleration during the collision as the human equivalent acceleration; if the seat belt connection state is connected, determining the human equivalent acceleration based on the collision motion data, the occupant mass, and the preset energy absorption parameter of the seat belt; determining the human equivalent acceleration as the collision intensity; where the collision motion data includes the vehicle deceleration during the collision, and the occupant state perception data further includes the occupant mass.

[0051] In some embodiments of the present application, the seat belt connection state being not connected is used to indicate that the driver and passengers are not wearing seat belts. At this time, the driver and passengers will impact the vehicle interior structure due to inertia, and the human equivalent acceleration a2 received by the driver and passengers is approximately equal to the vehicle deceleration a1 during the collision: a1≈a2.

[0052] In some embodiments of the present application, determining the human equivalent acceleration based on the collision motion data, the occupant mass, and the preset energy absorption parameter of the seat belt includes: determining the energy absorbed by the seat belt according to the actual vehicle mass, the vehicle speed before the collision, the vehicle speed after the collision, and the seat belt efficiency coefficient; determining the human equivalent acceleration according to the rated stretching distance of the seat belt, the occupant mass, and the energy absorbed by the seat belt; where the collision motion data further includes the vehicle speed before the collision and the vehicle speed after the collision, the preset energy absorption parameter includes the seat belt efficiency coefficient and the rated stretching distance of the seat belt, and the actual vehicle mass is obtained based on the preset vehicle mass and the occupant mass.

[0053] In some embodiments of the present application, the seat belt connection state being connected is used to indicate that the driver and passengers are wearing seat belts. The seat belt absorbs energy through stretching deformation, and the human equivalent acceleration is reduced. At this time, the seat belt efficiency coefficient η is introduced. The energy absorbed by the seat belt is proportional to the seat belt efficiency coefficient η, and the energy absorbed by the seat belt is determined by the kinetic energy theorem as follows:

[0054]

[0055] Among them, E belt is the energy absorbed by the seat belt, η is the seat belt efficiency coefficient, M is the actual mass of the whole vehicle, v1 is the vehicle speed before collision, and v2 is the vehicle speed after collision.

[0056] In some embodiments of the present application, the equivalent human acceleration is determined through the energy absorbed by the seat belt, Newton's second law, and the definition of work, as follows:

[0057]

[0058] Among them, a3 is the equivalent human acceleration when the seat belt connection state is connected, m p is the occupant mass, d1 is the rated stretching distance of the seat belt, η is the seat belt efficiency coefficient, M is the actual mass of the whole vehicle, v1 is the vehicle speed before collision, and v2 is the vehicle speed after collision.

[0059] In some embodiments of the present application, control instructions corresponding to the collision level are generated according to the comparison result between the collision intensity and the preset intensity threshold to perform hierarchical deployment control on the airbag, including: if the collision intensity is less than or equal to the first intensity threshold, control instructions corresponding to a low-intensity collision are generated to inhibit the airbag from popping up and generate a warning prompt message; if the collision intensity is greater than the first intensity threshold and less than or equal to the second intensity threshold, control instructions corresponding to a medium-intensity collision are generated to make the inflation pressure of the airbag within the preset pressure range and make the deployment time of the airbag within the preset deployment period; if the collision intensity is greater than the second intensity threshold, control instructions corresponding to a high-intensity collision are generated to make the airbag deploy with full pressure and directionally; where the preset intensity threshold includes the first intensity threshold and the second intensity threshold.

[0060] In some embodiments of the present application, the warning prompt message is used to implement seat warning and sound and light alarm.

[0061] In some embodiments of the present application, the preset pressure range includes 50% to 70% of the full-pressure inflation, that is, the inflation pressure is reduced by 30% to 50% on the basis of the full-pressure inflation. The preset deployment period includes a delay in deployment of 0.1 second to 0.3 second.

[0062] In some embodiments of the present application, the directional deployment includes guiding the airbag to expand directionally towards the chest of the driver and passengers rather than the head through a multi-chamber structure.

[0063] In some embodiments of the present application, before performing hierarchical deployment control on the airbag, it further includes: inputting the current in-vehicle image in the occupant status perception data into a heatmap prediction model to obtain the current predicted heatmap and current predicted coordinates of multiple current predicted key points; determining the target limb of each driver and passenger according to the part affinity fields corresponding to multiple initial limbs, where the initial limbs are connected based on the respective current predicted coordinates; determining the limb part corresponding to the target limb of each driver and passenger based on the respective current predicted heatmaps, and determining the body forward tilt angle of each driver and passenger according to the seat position data and limb part corresponding to each driver and passenger, where the seat position data is obtained based on the occupant status perception data; if the body forward tilt angle of a driver and passenger is greater than or equal to a preset angle threshold, activating the surface flexible coating of the airbag; wherein, the heatmap prediction model is obtained based on historical in-vehicle images and key point annotation information of historical in-vehicle images

[0064] In some embodiments of the present application, the current predicted coordinates of a current predicted key point are obtained based on the peak coordinates of the corresponding current predicted heatmap.

[0065] In some embodiments of the present application, an initial limb is formed by connecting two current predicted key points.

[0066] In some embodiments of the present application, part affinity fields (PAFs) are used for multi-person pose estimation (such as OpenPose), and initial limbs are formed by connecting current predicted key points. The vector field when a current predicted key point belongs to the i-th initial limb is as follows to characterize the direction of the current predicted key point:

[0067]

[0068] where, (x, y) are the current predicted coordinates of the current predicted key point, L i (x, y) is the vector field when the current predicted key point belongs to the i-th initial limb, V i is the direction vector of the i-th initial limb.

[0069] In some embodiments of the present application, V i is obtained based on the current predicted coordinates corresponding to the starting point and the current predicted coordinates corresponding to the ending point in the i-th initial limb. For example, V i can characterize the vector from the left wrist to the elbow.

[0070] In some embodiments of the present application, the determination of the part affinity field of an initial limb is as follows:

[0071]

[0072] where, E is the part affinity field of an initial limb, pj is the current predicted coordinate corresponding to the end point of an initial limb, p i is the current predicted coordinate corresponding to the starting point of an initial limb, p(t) is a sampling point between the starting point and the end point of an initial limb, L i (p(t)) is the vector field of the sampling point p(t), ||p j -p i || is the linear interpolation path between the starting point and the end point of an initial limb.

[0073] In some embodiments of the present application, the determination of the sampling point is as follows:

[0074] p(t) = p i + t(p j - p i ) Equation (8)

[0075] wherein, p(t) is the sampling point between the starting point and the end point of an initial limb, p i is the current predicted coordinate corresponding to the starting point of an initial limb, p j is the current predicted coordinate corresponding to the end point of an initial limb, t ∈ [0, 1].

[0076] In some embodiments of the present application, when the part affinity field E is close to 1, it is determined that p j and p i belong to the limbs of the same occupant, that is, the initial limb composed of p j and p i is the target limb of an occupant.

[0077] In some embodiments of the present application, the body forward tilt angle is used to characterize the forward tilt angle of the occupant relative to the seat back position. Among them, the seat back position is obtained based on the seat position data.

[0078] In some embodiments of the present application, the impact force is reduced by activating the surface flexible coating of the airbag.

[0079] In some embodiments of the present application, the heat map prediction model is obtained based on the historical in-vehicle images and the key point annotation information of the historical in-vehicle images, including: performing key point annotation on the historical in-vehicle images to obtain the annotation coordinates of multiple annotated key points; generating the conversion heat map of each annotated key point according to the annotation coordinates of each annotated key point; inputting the historical in-vehicle images into the network prediction model to obtain the historical prediction heat maps of multiple historical predicted key points, and determining the historical prediction coordinates of the historical predicted key points as the peak coordinates in the prediction heat maps; optimizing the parameters of the network prediction model according to the gap between each conversion heat map and the corresponding historical prediction heat map, and / or the gap between each annotation coordinate and the corresponding historical prediction coordinate, to obtain the heat map prediction model.

[0080] In some embodiments of the present application, the generation of a transformed heatmap for annotating key points is as follows:

[0081]

[0082] where H k (x, y) is the value at the coordinate (x, y) in the transformed heatmap corresponding to the k-th annotated key point, μ x,k is the annotated abscissa of the k-th annotated key point, μ y,k is the annotated ordinate of the k-th annotated key point, and σ is the Gaussian kernel radius.

[0083] In some embodiments of the present application, the Gaussian kernel radius is usually taken as 10% of the pixel distance.

[0084] In some embodiments of the present application, the determination of the historical prediction coordinates is as follows:

[0085]

[0086] where is the historical prediction abscissa of the k-th historical prediction key point, is the historical prediction ordinate of the k-th historical prediction key point, is the value at the coordinate (x, y) in the historical prediction heatmap corresponding to the k-th historical prediction key point.

[0087] In some embodiments of the present application, the gap between each transformed heatmap and the corresponding predicted heatmap is determined by a preset loss function; the gap between each annotated coordinate and the corresponding predicted coordinate is determined by the correct key point ratio and / or key point similarity.

[0088] In some embodiments of the present application, the preset loss function is as follows:

[0089]

[0090] where is the preset loss function, H k (x, y) is the value at the coordinate (x, y) in the transformed heatmap corresponding to the k-th annotated key point, is the value at the coordinate (x, y) in the historical prediction heatmap corresponding to the k-th historical prediction key point, and K is the total number of key points.

[0091] In some embodiments of the present application, if the coordinate distance between a historical predicted coordinate and the corresponding annotated coordinate is less than or equal to a preset distance threshold, the historical predicted key point corresponding to the predicted coordinate is determined as the correct key point; wherein, the preset distance threshold is obtained based on a preset error threshold and the height and width of a preset bounding box.

[0092] In some embodiments of the present application, the preset distance threshold = α·max(h, w), where α is the preset error threshold, and (h, w) is the height and width of the preset bounding box.

[0093] In some embodiments of the present application, the preset bounding box may be a human body bounding box, α = 0.2, that is, the coordinate distance between a predicted coordinate and the corresponding annotated coordinate is less than or equal to 20% of the human body bounding size.

[0094] In some embodiments of the present application, the determination of the correct key point ratio is as follows:

[0095]

[0096] where PCK@α is the correct key point ratio.

[0097] In some embodiments of the present application, the determination of the key point similarity is as follows:

[0098]

[0099] where OKS is the key point similarity, d k is the Euclidean distance between the k-th historical predicted coordinate and the corresponding annotated coordinate, s is the scale parameter of the preset bounding box, K k is the preset relaxation factor corresponding to the type of the k-th historical predicted key point, V k is the key point visibility label corresponding to the k-th historical predicted key point, and δ is an indicator function used to filter the historical predicted key points participating in the calculation.

[0100] In some embodiments of the present application, in formula (14), only the similarity contributions of the annotated historical predicted key points are accumulated in the numerator, and the number of all annotated historical predicted key points is counted in the denominator to normalize the final score.

[0101] In some embodiments of the present application, the scale parameter of the human body detection box is as follows:

[0102]

[0103] where s is the scale parameter of the human body detection box, w is the width of the preset bounding box, and h is the height of the preset bounding box.

[0104] In some embodiments of the present application, K kUsed to adjust the error tolerance of different historical prediction key points, reflecting the importance or positioning difficulty of different joints or parts in human pose estimation (which is a manually annotated value by experience). For example, K 眼 = 0.025, K 髋 = 0.1, then for the same Euclidean distance d k = 10 pixels, the deduction for the eyes is more severe than that for the hips.

[0105] In some embodiments of the present application, the screening of historical prediction key points is as follows

[0106]

[0107] Among them, δ is an indicator function used to screen the historical prediction key points participating in the calculation, and V k is the key point visibility label.

[0108] In some embodiments of the present application, if the annotated key point corresponding to the historical prediction key point is annotated as visible, or invisible but exists, then δ(V k > 0) = 1.

[0109] In some embodiments of the present application, after performing hierarchical deployment control on the airbag, it further includes: determining the collision impact force corresponding to the airbag according to the collision intensity and the occupant mass in the occupant state perception data; performing injury assessment based on the collision impact force and the seat position data in the occupant state perception data to obtain a first assessment result, and the injury assessment includes at least one of chest force assessment and leg injury assessment; performing internal organ injury assessment according to the chest compression amount and the chest compression speed to obtain a second assessment result, and the chest compression amount and the chest compression speed are obtained based on the chest force assessment; adjusting the parameters of the hierarchical deployment control based on the first assessment result and the second assessment result.

[0110] In some embodiments of the present application, the determination of the collision impact force is as follows:

[0111]

[0112] Among them, F is the collision impact force, m p is the occupant mass, a2 is the human equivalent acceleration when the seat belt connection state is not connected, and a3 is the human equivalent acceleration when the seat belt connection state is connected.

[0113] In some embodiments of the present application, the chest force is equal to the collision impact force. Compare the chest force with the corresponding preset force threshold to obtain the first comparison result of the chest force assessment.

[0114] In some embodiments of the present application, the leg impact loss probability corresponding to the leg injury assessment is as follows:

[0115] TBM = F·d2 Equation (17)

[0116] Wherein, TBM is the probability of leg impact loss, F is the impact force of the collision, and d2 is the length of the force arm, which represents the distance from the tibia force application point to the joint and is obtained based on the seat position data.

[0117] In some embodiments of the present application, the probability of leg impact loss is compared with the corresponding preset impact loss probability to obtain a second comparison result for leg injury assessment.

[0118] In some embodiments of the present application, the first assessment result includes the first comparison result of chest force assessment and / or the second comparison result of leg injury assessment.

[0119] In some embodiments of the present application, the chest compression amount can be determined by Hooke's law and the chest force as follows:

[0120] F = k F ·δ F Equation (18)

[0121] Wherein, F is the chest force, and k F is the chest stiffness coefficient, and δ F is the chest compression amount.

[0122] In some embodiments of the present application, the chest deformation amount is the maximum compression displacement generated by the chest due to external force during the collision. The relationship between the chest compression amount and the chest deformation amount is as follows:

[0123]

[0124] Wherein, δ F is the chest compression amount.

[0125] In some embodiments of the present application, the above-mentioned chest stiffness coefficient, chest deformation amount, and chest original thickness can be obtained by using the test data of the test dummy. The degree of damage to the human chest is characterized by the chest compression amount. The minor injury threshold, moderate injury threshold, and fatal injury threshold corresponding to the chest compression amount, as well as the viscous criterion threshold corresponding to the chest compression speed, can be determined by using the test data of the test dummy.

[0126] In some embodiments of the present application, the chest compression speed can be determined based on the test data of the test dummy, the chest force, and the chest deformation amount.

[0127] In some embodiments of the present application, the internal organ injury assessment is as follows:

[0128] Table 1 Internal Organ Injury Assessment Table

[0129] Collision direction Minor damage threshold Moderate damage threshold Fatal damage threshold Viscosity criterion threshold Frontal collision ≤20％ 20％～35％ >35％ ≤1.0m / s Side collision ≤15％ 15％～25％ >25％ ≤0.8m / s

[0130] In some embodiments of the present application, for the internal organ damage assessment form shown in Table 1, in a side collision, due to the narrow impact area, attention should be focused on the injuries to the head, chest and pelvis. Therefore, for a side collision, the risk of side collision is determined by the second head injury criterion. Moreover, due to the lack of muscle and scapula protection, the lateral rib structure is more vulnerable, resulting in a lower injury threshold for side collisions. In addition, since the lateral organs (such as the liver and spleen) are more sensitive to rapid compression, the viscosity criterion threshold for side collisions is more stringent.

[0131] In some embodiments of the present application, frontal collisions mainly focus on the injury risks to the head, chest and lower limbs.

[0132] In some embodiments of the present application, the parameters for adjusting the hierarchical deployment control include at least one of adjusting the preset intensity threshold, the preset angle threshold, the inflation pressure of the airbag, the deployment time and the deployment direction.

[0133] The present application binds the collision intensity to the seat belt connection state to avoid the protection failure of the "all or nothing" type airbag. Moreover, by identifying data such as the seat belt connection state and the collision intensity, the deployment speed, inflation pressure and coverage range of the airbag are dynamically adjusted to achieve risk-based hierarchical protection. In addition, in the scenario of not wearing a seat belt, the secondary injury rate of the airbag can be reduced by at least 40%. In high-intensity collisions, the necessary protection ability is maintained, and the chest injury increases by no more than 15% compared with the traditional system.

[0134] Please refer to Figure 3 , Figure 3 which shows a block diagram of an airbag control device according to an embodiment of the present application. This device can be applied to Figure 1 the implementation environment shown, and is specifically configured in the electronic control unit 103. This device can also be applicable to other exemplary implementation environments and is specifically configured in other devices. The embodiment does not limit the implementation environment applicable to this device.

[0135] As Figure 3 shown, an airbag control device 300 according to an embodiment of the present application includes: a data acquisition module 301 and a control decision module 302.

[0136] Among them, the data acquisition module 301 is used to acquire the collision perception data and the occupant state perception data of the vehicle. The collision perception data includes the collision direction and the collision motion data, and the occupant state perception data includes the seat belt connection state;

[0137] The control decision module 302 includes:

[0138] A collision risk determination unit, configured to determine a collision risk value corresponding to a preset head injury criterion according to a collision direction and collision motion data;

[0139] A control instruction generation unit, configured to, if the collision risk value is greater than a corresponding preset risk threshold, determine a collision intensity according to the collision motion data and occupant state perception data, and generate a control instruction of a corresponding collision level according to a comparison result between the collision intensity and a preset intensity threshold, so as to perform hierarchical deployment control on an airbag.

[0140] The airbag control device provided in the above embodiment and the airbag control method provided in the above embodiment belong to the same concept. The specific manners in which each module and unit perform operations have been described in detail in the method embodiment, and will not be repeated here. In practical applications, the airbag control device provided in the above embodiment may, according to needs, allocate the above functions to different functional modules, that is, divide the internal structure of the device into different functional modules to complete all or part of the functions described above, and this is not limited here either.

[0141] An embodiment of the present application further provides an electronic device, including: one or more processors; a storage device, configured to store one or more programs, and when the one or more programs are executed by the one or more processors, enable the electronic device to implement the airbag control method provided in each of the above embodiments.

[0142] Please refer to Figure 4 , Figure 4 which shows a schematic structural diagram of a computer system of an electronic device suitable for implementing an embodiment of the present application. Figure 4 The computer system 400 of the electronic device shown is only an example, and should not bring any limitation to the functions and usage scope of the embodiments of the present application.

[0143] As Figure 4 shown, the computer system 400 includes a central processing unit 401, which can perform various appropriate actions and processes according to a program stored in a read-only memory 402 or a program loaded from a storage section 408 into a random access memory 403, such as executing the method described in the above embodiment. In the random access memory 403, various programs and data required for system operation are also stored. The central processing unit 401, the read-only memory 402, and the random access memory 403 are connected to each other through a bus 404. An input / output interface 405 is also connected to the bus 404.

[0144] The following components are connected to the input / output interface 405: an input section 406 including a keyboard, a mouse, etc.; an output section 407 including such as a cathode ray tube (CRT), a liquid crystal display (LCD), etc. and a speaker, etc.; a storage section 408 including a hard disk, etc.; and a communication section 409 including a network interface card such as a LAN (Local Area Network) card, a modem, etc. The communication section 409 performs communication processing via a network such as the Internet. A drive 410 is also connected to the input / output interface 405 as needed. A removable medium 411, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 410 as needed so that a computer program read therefrom is installed into the storage section 408 as needed.

[0145] Specifically, according to an embodiment of the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present application includes a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a computer program for performing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication section 409, and / or installed from the removable medium 411. When the computer program is executed by a central processing unit (CPU) 401, various functions defined in the system of the present application are executed.

[0146] The computer-readable medium as shown in the embodiments of the present application may be a computer-readable signal medium, a computer-readable storage medium, or any combination of the two. A computer-readable storage medium may be, for example, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the above. More specific examples of the computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM), a flash memory, an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present application, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable computer program. Such a propagated data signal may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The computer program contained on the computer-readable medium may be transmitted by any appropriate medium, including but not limited to: wireless, wired, etc., or any suitable combination of the above.

[0147] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architectures, functions, and operations of systems, methods, and computer program products according to various embodiments of the present application. Among them, each block in the flowchart or block diagram may represent a module, a program segment, or a part of code, and the above module, program segment, or part of code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than marked in the accompanying drawings. For example, two consecutive blocks shown may actually be executed substantially in parallel, and they may sometimes be executed in the reverse order, depending on the functions involved. It should also be noted that each block in the block diagram or flowchart, and the combination of blocks in the block diagram or flowchart, may be implemented by a dedicated hardware-based system for performing the specified functions or operations, or may be implemented by a combination of dedicated hardware and computer instructions.

[0148] The units involved in the embodiments described in this application can be implemented in software or in hardware, and the described units can also be provided in a processor. Among them, the names of these units do not, in some cases, constitute a limitation on the units themselves. Therefore, the technical solution according to the embodiments of this application can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, including several instructions to enable a computing device (which can be a personal computer, a server, a touch terminal, or a network device, etc.) to execute the method according to the embodiments of this application.

[0149] On the other hand, this application also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor of the computer, the computer is enabled to execute the airbag control method provided in each of the above embodiments. The computer-readable storage medium can be included in the electronic device described in the above embodiments, or can exist alone without being assembled into the electronic device.

[0150] In the above embodiments, unless otherwise specified, when using serial numbers such as "first" and "second" to describe a common object, it only represents different instances of the same object, rather than indicating that the object to be described must be in a given order, whether in terms of time, space, sorting, or any other way.

[0151] The above embodiments only exemplarily illustrate the principles and effects of this application, rather than being used to limit this application. Any person familiar with this technology can modify or change the above embodiments without departing from the spirit and scope of this application. Therefore, all equivalent modifications or changes completed by those with ordinary knowledge in the technical field without departing from the spirit and technical idea disclosed in this application should still be covered by the claims of this application.

Claims

1. An airbag control method, characterized in that, The method includes: Obtaining collision perception data and occupant status perception data of the vehicle, where the collision perception data includes collision direction and collision motion data, and the occupant status perception data includes seat belt connection status; Determining a collision risk value corresponding to a preset head injury criterion according to the collision direction and the collision motion data; If the collision risk value is greater than a corresponding preset risk threshold, determining a collision intensity according to the collision motion data and the occupant status perception data, and generating a control command corresponding to the collision level according to the comparison result between the collision intensity and a preset intensity threshold to perform hierarchical deployment control on the airbag.

2. The airbag control method according to claim 1, wherein, Determining the collision intensity according to the collision motion data and the occupant status perception data includes: If the seat belt connection status is unconnected, determining the vehicle deceleration during the collision as the human equivalent acceleration; If the seat belt connection status is connected, determining the human equivalent acceleration according to the collision motion data, the occupant mass, and a preset energy absorption parameter of the seat belt; Determining the human equivalent acceleration as the collision intensity; Wherein, the collision motion data includes the vehicle deceleration during the collision, and the occupant status perception data further includes the occupant mass.

3. The airbag control method according to claim 2, characterized in that, Determining the human equivalent acceleration according to the collision motion data, the occupant mass, and a preset energy absorption parameter of the seat belt includes: Determining the energy absorbed by the seat belt according to the actual mass of the whole vehicle, the vehicle speed before the collision, the vehicle speed after the collision, and the seat belt efficiency coefficient; Determining the human equivalent acceleration according to the rated stretching distance of the seat belt, the occupant mass, and the energy absorbed by the seat belt; Wherein, the collision motion data further includes the vehicle speed before the collision and the vehicle speed after the collision, the preset energy absorption parameter includes the seat belt efficiency coefficient and the rated stretching distance of the seat belt, and the actual mass of the whole vehicle is obtained based on a preset vehicle mass and the occupant mass.

4. The airbag control method according to claim 1, characterized in that Generating a control command corresponding to the collision level according to the comparison result between the collision intensity and a preset intensity threshold to perform hierarchical deployment control on the airbag includes: If the collision intensity is less than or equal to a first intensity threshold, generating a control command corresponding to a low-intensity collision to suppress the airbag from popping out and generating a warning prompt message; If the collision intensity is greater than the first intensity threshold and less than or equal to a second intensity threshold, generating a control command corresponding to a medium-intensity collision to make the inflation pressure of the airbag within a preset pressure range and make the deployment time of the airbag within a preset deployment period; If the collision intensity is greater than the second intensity threshold, generating a control command corresponding to a high-intensity collision to make the airbag deploy fully and directionally; Wherein, the preset intensity threshold includes the first intensity threshold and the second intensity threshold.

5. The airbag control method according to claim 4, wherein After performing hierarchical deployment control on the airbag, it further includes: Determining the collision impact force corresponding to the airbag according to the collision intensity and the occupant mass in the occupant status perception data; Based on the impact force and the seat position data in the occupant status perception data, a damage assessment is performed to obtain a first assessment result, and the damage assessment includes at least one of chest force assessment and leg injury assessment; Based on the chest compression amount and the chest compression speed, an internal organ damage assessment is performed to obtain a second assessment result, and the chest compression amount and the chest compression speed are obtained based on the chest force assessment; Based on the first assessment result and the second assessment result, the parameters for controlling the grading are adjusted.

6. The airbag control method according to any one of claims 1-5, characterized in that, Before performing the hierarchical deployment control on the airbag, it further includes: Inputting the current in-vehicle image in the occupant status perception data into the heat map prediction model to obtain the current predicted heat map and current predicted coordinates of multiple current predicted key points; Determining the target limb of each occupant according to the part affinity fields corresponding to multiple initial limbs, and the initial limbs are connected based on each of the current predicted coordinates; Based on each of the current predicted heat maps, determining the limb part corresponding to the target limb of each occupant, and determining the body forward tilt angle of each occupant according to the seat position data corresponding to each occupant and the limb part, and the seat position data is obtained based on the occupant status perception data; If the body forward tilt angle of an occupant is greater than or equal to a preset angle threshold, activate the surface flexible coating of the airbag; Wherein, the heat map prediction model is obtained based on historical in-vehicle images and the key point annotation information of the historical in-vehicle images.

7. The airbag control method according to claim 6, characterized in that, The heat map prediction model is obtained based on historical in-vehicle images and the key point annotation information of the historical in-vehicle images, including: Performing key point annotation on the historical in-vehicle image to obtain the annotation coordinates of multiple annotated key points; Generating a converted heat map of each annotated key point according to the annotation coordinates of each annotated key point; Inputting the historical in-vehicle image into the network prediction model to obtain the historical predicted heat map of multiple historical predicted key points, and determining the historical predicted coordinates of the historical predicted key points as the peak coordinates in the predicted heat map; Optimizing the parameters of the network prediction model according to the difference between each converted heat map and the corresponding historical predicted heat map, and / or the difference between each annotation coordinate and the corresponding historical predicted coordinate, to obtain the heat map prediction model.

8. An airbag control device, characterized in that, The device includes: a data acquisition module and a control decision module; The data acquisition module is used to acquire the collision perception data and the occupant status perception data of the vehicle, the collision perception data includes the collision direction and the collision motion data, and the occupant status perception data includes the seat belt connection status; The control decision module includes: A collision risk determination unit, configured to determine a collision risk value corresponding to a preset head injury criterion according to the collision direction and the collision motion data; A control instruction generation unit, configured to, if the collision risk value is greater than the corresponding preset risk threshold, determine the collision intensity according to the collision motion data and the occupant status perception data, and generate a control instruction corresponding to the collision level according to the comparison result between the collision intensity and the preset intensity threshold, so as to perform hierarchical deployment control on the airbag.

9. An electronic device, characterized in that, The electronic device includes: One or more processors; A storage device for storing one or more programs, which, when executed by the one or more processors, cause the electronic device to implement the airbag control method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, A computer program is stored thereon, which, when executed by a processor of a computer, causes the computer to execute the airbag control method according to any one of claims 1 to 7.

Images