Airbag device

Abstract

To prevent an inflated airbag from failing to exhibit expected restraining performance and from adversely affecting an occupant.SOLUTION: A control device 54 of an airbag device 10 determines whether a collision of a vehicle has occurred or not, based on an output value from a collision sensor 50 which detects the collision of the vehicle; determines whether or not a head part of an occupant is positioned within a predetermined range corresponding to a loop space of an airbag 32 including a pair of front and rear chambers, left and right, in which a loop space is formed between an airbag main body deployed on a vehicle front side of the occupant during inflation thereof and the airbag main body extending in a vehicle longitudinal direction during the inflation, based on an output value from an indoor camera 52 which photographs inside of the photographic range including the head part of the occupant seated on a seat 12 and based on a determination model which is learned in advance from learning data including data about a case where the head part of the occupant is not positioned within the predetermined range; and inflates the airbag 32 when it is determined that the collision of the vehicle has occurred and that the head part of the occupant is positioned within the predetermined range.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present disclosure relates to an airbag device. [Background technology]

[0002] Patent Document 1 describes an airbag device that includes a torso support part that deploys from behind the occupant's head to the front of the occupant, and a pair of head support parts that deploy on both the left and right sides of the occupant's head and are connected to the torso support part. In this airbag device, in the deployed state, a clearance part is formed that passes through in the vertical direction between the pair of head support parts to avoid the occupant's head.

[0003] Patent Document 2 also describes an airbag device having a rear inflatable portion that deploys on the back side of the seat, a pair of side inflatable portions that extend forward from both the left and right sides of the rear inflatable portion, and a pair of front inflatable portions that extend from the side inflatable portions to the center and are connected to each other at the center to cover the front of the occupant. In this airbag device, gas flow paths are formed in the order of the rear inflatable portion, the side inflatable portion, and the front inflatable portion. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] JP 2019-018593 A [Patent Document 2] JP 2019-218013 A Summary of the Invention [Problem to be solved by the invention]

[0005] In the case of the airbags disclosed in Patent Documents 1 and 2, which include an airbag body that is deployed in front of the vehicle occupant when deployed, and a pair of left and right front and rear chambers that extend in the vehicle longitudinal direction when deployed and form a loop space between the airbag body and the airbag body, there are the following problems. That is, if the head of the occupant is not in a normal position, for example, if the occupant is sleeping and leaning against the window, or if the occupant is in a forward-leaning position, the head of the occupant may not enter the loop space when the airbag is deployed, or may enter only partially. In this case, there is a concern that the deployed airbag may not be able to exert the desired restraint performance, or may have a negative effect on the occupant.

[0006] The present disclosure has been made in consideration of the above-mentioned circumstances, and has an object to provide an airbag device that can prevent a deployed airbag from failing to exhibit the desired restraint performance or from adversely affecting an occupant. [Means for solving the problem]

[0007] The airbag device of the first aspect includes a control unit that determines whether or not a vehicle collision has occurred based on an output value from a collision sensor that detects a vehicle collision, and determines whether or not the head of an occupant seated in a seat is located within a predetermined range corresponding to the loop space of an airbag that includes an airbag body that is deployed in front of the occupant when deployed and a pair of left and right front and rear chambers that extend in the fore-and-aft direction of the vehicle when deployed and form a loop space between the airbag body, based on an output value from a camera that captures an image within a capturing range including the head of the occupant, and deploys the airbag when it determines that a vehicle collision has occurred and that the occupant's head is located within the predetermined range.

[0008] In the first aspect, the presence or absence of a vehicle collision is determined based on an output value from a collision sensor. Also, based on an output value from a camera that captures an image of a range including the head of the occupant, it is determined whether or not the head of the occupant is located within a predetermined range corresponding to the loop space of the airbag. Then, when it is determined that there is a vehicle collision and that the head of the occupant is located within the predetermined range, the airbag is deployed. As a result, when it is determined that the head of the occupant is not located within the predetermined range, the airbag is not deployed, so that it is possible to prevent the deployed airbag from failing to exert the desired restraint performance and from adversely affecting the occupant.

[0009] In a second aspect, in the first aspect, the control unit determines whether or not the occupant's head is located within the specified range based on a determination model that has been pre-trained from learning data that includes, in addition to the output value from the camera, data for when the occupant's head is not located within the specified range.

[0010] According to the second aspect, it is possible to accurately determine whether or not the occupant's head is located within a predetermined range corresponding to the loop space of the airbag, without complicating the configuration by, for example, making the camera a stereo camera.

[0011] In a third aspect, in the second aspect, the data when the occupant's head is not located within the specified range includes data when the occupant is lying down and leaning against the window, and data when the occupant is in a leaning forward position.

[0012] According to the third aspect, in a judgment based on a judgment model previously learned from learning data, when an occupant is lying down and leaning against the window or when an occupant is leaning forward, it can be accurately determined that the occupant's head is not located within a specified range.

[0013] A fourth aspect is the second aspect, wherein the learning data includes first data in which the output value of the camera when the occupant's head is not located within the specified range is associated with position information representing the position of the occupant's head, and second data in which the output value of the camera when the occupant's head is located within the specified range is associated with position information representing the position of the occupant's head.

[0014] According to the fourth aspect, since the learning data includes the first data and the second data, in a judgment based on a judgment model previously learned from the learning data, position information indicating the position of the occupant's head can be obtained from the output value from the camera. Then, it can be accurately judged from the position information whether the occupant's head is located within a predetermined range. Effect of the Invention

[0015] The present disclosure has an effect of suppressing a situation in which a deployed airbag is unable to exert the desired restraint performance and has a detrimental effect on an occupant. [Brief description of the drawings]

[0016] [Figure 1] 1 is a schematic block diagram of an airbag device according to an embodiment. [Diagram 2] FIG. 1 is a perspective view showing a vehicle airbag. [Diagram 3] 11 is an image diagram showing the arrangement of interior cameras and a specific example of the position of an occupant's head when it is determined that the occupant's head is not located within a predetermined range. FIG. [Figure 4] 11A and 11B are conceptual diagrams showing an example of the positional relationship between an occupant and an airbag when the occupant is in a normal posture and when the head is shifted to the side. [Diagram 5] 11A and 11B are conceptual diagrams showing an example of the positional relationship between an occupant and an airbag when the occupant is in a normal posture and when the head is shifted forward. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0017] An example of an embodiment of the present disclosure will be described in detail below with reference to the drawings. As shown in Fig. 1, an airbag device 10 according to this embodiment is provided for each seat of a vehicle, and includes a collision sensor 50, an interior camera 52, a control device 54, an inflator 44, and an airbag 32. Hereinafter, the configuration of the airbag 32 and its surroundings will be described with reference to Fig. 2.

[0018] The seat 12 shown in Fig. 2 is a front seat or a rear seat of a vehicle. The seat 12 has a seat cushion 14, a seat back 16 erected above the rear end of the seat cushion 14, and a headrest 18 connected to the upper end of the seat back 16. An occupant D seated on the seat cushion 14 of the seat 12 is restrained in the seat 12 by a seat belt 22 provided in a seat belt device 20. The seat belt device 20 is a three-point seat belt device, and is a so-called seat-attached seat belt device in which a retractor, an anchor, and a buckle (reference numbers omitted) (not shown) are provided in the seat 12.

[0019] The airbag 32 is normally stored in a folded state together with a pair of inflators 44 in a module case 46. The module case 46 is formed in a hollow rectangular parallelepiped shape. The module case 46 is disposed behind the upper part of the seat 12 (more specifically, behind the upper end of the seat back 16 and the headrest 18) and is fixed to the upper end of the seat back 16 or to the vehicle body (not shown).

[0020] The airbag 32 receives gas supply from a pair of inflators 44 and inflates (deploys and inflates) from the rear side to the front side above the seat 12. The airbag 32 includes a pair of left and right front and rear chambers 34 and an airbag body 40. When deployed, the airbag body 40 is deployed in front of the vehicle occupant D. Furthermore, when deployed, the pair of left and right front and rear chambers 34 extend in the front-rear direction of the vehicle, and a loop space 36 is formed between the airbag body 40 and the pair of left and right front and rear chambers 34.

[0021] More specifically, the front-rear chamber 34 has a pair of left and right front-rear extending portions 34A that inflate and deploy forward through both the left and right sides of the head H of the occupant D when the occupant D is in a standard posture, and a connecting portion 34B that connects the front ends of the pair of front-rear extending portions 34A in the left-right direction. The airbag body 40 inflates and deploys toward the occupant D (rearward) behind the connecting portion 34B, later than the front-rear chamber 34.

[0022] The front and rear chambers 34 are formed in a long bag shape, and the airbag body 40 is also formed in a bag shape. Both longitudinal ends of the front and rear chambers 34 are formed as a pair of left and right inflator housing sections that house a pair of inflators 44. The longitudinal center of the front and rear chambers 34 is formed as a connecting section 34B, and the sections between the connecting section 34B and the pair of inflator housing sections are formed as front and rear extending sections 34A. The pair of inflators 44 are combustion-type or cold gas-type cylinder-type inflators that generate gas when activated. The activation of these inflators 44 is controlled by a control device 54.

[0023] On the other hand, the collision sensor 50 shown in FIG. 1 is composed of an acceleration sensor that detects acceleration in the longitudinal direction of the vehicle, and detects a vehicle collision when a state in which acceleration equal to or greater than a predetermined value, which corresponds to the vehicle colliding with an object, is detected for a predetermined period of time or longer.

[0024] The interior camera 52 captures an image within a capture range including the head H of the occupant D seated in the seat 12. More specifically, the interior camera 52 includes a monocular interior camera 52F (see FIG. 3(A)) for capturing an image of the occupant D seated in the front seat, and a monocular interior camera 52R (see FIG. 3(B)) for capturing an image of the occupant D seated in the rear seat.

[0025] As shown in Fig. 3(A), the interior camera 52F is attached to the front end of the roof of the vehicle and in the center in the vehicle width direction. The imaging range including the angle of view of the interior camera 52F is adjusted so that the head H of the occupant D seated in the driver's seat is in the imaging range even when the head H of the occupant D seated in the driver's seat is in the position 70A, 70B, 70C, etc. in Fig. 3(C) or when the head H of the occupant D seated in the passenger seat is in the position 70D, 70E, 70F, etc. in Fig. 3(C).

[0026] Positions 70A and 70C indicate the position of the head H of the occupant D when the occupant D sitting in the driver's seat is lying down with the head H tilted to the left or right (e.g. leaning against the window). Position 70B indicates the position of the head H of the occupant D when the occupant D sitting in the driver's seat is leaning forward.

[0027] Positions 70D and 70F indicate the position of the head H of the passenger D when the passenger D is lying down and tilted to the left or right (e.g., leaning against the window). Position 70E indicates the position of the head H of the passenger D when the passenger D is leaning forward.

[0028] 3B, the interior camera 52R is attached to the center of the vehicle's roof in the vehicle front-rear direction and in the vehicle width direction. As with the interior camera 52F, the imaging range of the interior camera 52R, including the angle of view, is adjusted so that the head H of the occupant D is included in the imaging range even when the occupant D sitting in the rear seat is lying down with his / her head H tilted to the left or right, or when the occupant D is leaning forward.

[0029] The control device 54 includes a collision determination unit 56, an occupant attitude determination unit 58, and a comprehensive determination unit 60. The collision determination unit 56 determines whether or not a vehicle collision has occurred based on the output value from the collision sensor 50, and outputs the determination result to the comprehensive determination unit 60. The occupant attitude determination unit 58 includes a storage unit in which a trained determination model is stored in advance. The occupant attitude determination unit 58 determines whether or not the head H of the occupant D is located within a predetermined range corresponding to the loop space 36 of the airbag 32, based on the output value from the interior camera 52 and the trained determination model, and outputs the determination result to the comprehensive determination unit 60.

[0030] The above-mentioned trained judgment model is generated using learning data including data when the head H of the occupant D is not located within a predetermined range corresponding to the loop space 36 of the airbag 32. More specifically, the learning data includes first data and second data. The first data is data in which the output value of the interior camera 52 when the head H of the occupant D is not located within the predetermined range is associated with position information (an example of teacher data) representing the position of the head H of the occupant D. The second data is data in which the output value of the interior camera 52 when the head H of the occupant D is located within the predetermined range is associated with position information (an example of teacher data) representing the position of the head H of the occupant D. The first data includes data in which the occupant D is lying down and leaning against the window, and in which the occupant D is leaning forward. The above-mentioned position information is, for example, information such as "300 mm to the left of the center (position in the standard position)".

[0031] In this embodiment, a model for outputting the position of the head H of the occupant D in a situation represented by an output value of the interior camera 52 is learned based on the above learning data, thereby generating a learned judgment model in advance. In this embodiment, a neural network as shown in Fig. 1 can be used as an example of the model, and deep learning can be used as an example of the learning algorithm.

[0032] The occupant posture determination unit 58 acquires a determination result of the position of the head H of the occupant D by performing signal processing of a multilayer neural network based on the output value from the indoor camera 52 and the learned determination model every time an output value from the indoor camera 52 is input. In this embodiment, the determination result is obtained as data in a format such as, for example, an 80% probability that the position of the head H of the occupant D is "400 mm to the left of the center," a 10% probability that the position is "300 mm to the left of the center," a 10% probability that the position is "200 mm to the left of the center," and a 0% probability that the position is "100 mm to the left of the center." In the above example, the position of the head H of the occupant D is determined to be "400 mm to the left of the center," and "400 mm to the left" is compared with a predetermined value to determine whether or not the head H of the occupant D is located within a predetermined range.

[0033] When the collision determination unit 56 determines that a vehicle collision has occurred and the occupant posture determination unit 58 determines that the head H of the occupant D is located within a predetermined range, the comprehensive determination unit 60 activates the inflator 44 to deploy the airbag 32. The control device 54 including the collision determination unit 56, the occupant posture determination unit 58, and the comprehensive determination unit 60 described above is an example of a control unit in the present disclosure.

[0034] Next, the operation of this embodiment will be described. As shown in Figures 4(A) and 5(A), when the occupant D is seated in the seat 12 in a standard position, or when the seat back 16 is reclined and the occupant D is seated in a comfortable position, the head H of the occupant D is located within a predetermined range corresponding to the loop space 36. Therefore, if the vehicle collides with an object in this state and the airbag 32 is deployed, as shown by the notation "in normal position", the head H of the occupant D will be located within the loop space 36 of the deployed airbag 32, and the head H of the occupant D can be appropriately restrained.

[0035] However, as shown in Fig. 4(B), when the occupant D seated in the seat 12 is lying down and leaning against the window, the head H of the occupant D is positioned to the side outside the predetermined range corresponding to the loop space 36. Therefore, if the vehicle collides with an object in this state and the airbag 32 is deployed, there is a risk that the head H of the occupant D will fall out of the loop space 36 of the deployed airbag 32, or the front and rear chambers 34 of the airbag 32 will directly hit the back of the head of the occupant D, as shown by the notation "when the head is shifted to the side."

[0036] 5(B), when the occupant D seated in the seat 12 is in a forward-leaning position, the head H of the occupant D is located forward and outside the predetermined range corresponding to the loop space 36. For this reason, if the vehicle collides with an object in this state and the airbag 32 is deployed, as shown by the notation "when the head is shifted forward," the head H of the occupant D may come into contact with the airbag body 40, and by restraining only the head H of the occupant D, the backward bending of the neck may become large, or the head H of the occupant D may come out of the loop space 36 of the deployed airbag 32.

[0037] In contrast, in the airbag device 10 according to the present embodiment, when the occupant posture determination unit 58 determines that the head H of the occupant D is not located within a predetermined range corresponding to the loop space 36, the airbag 32 of the corresponding seat 12 is not deployed even if the vehicle collides with an object in this state. As a result, in the seat 12 in which the seated occupant D is lying down and leaning against the window as shown in FIG. 4B, the head H of the occupant D is prevented from coming out of the loop space 36 of the deployed airbag 32, or the front and rear chambers 34 of the airbag 32 are prevented from directly hitting the back of the head of the occupant D. In addition, in the seat 12 in which the seated occupant D is leaning forward as shown in FIG. 5B, the head H of the occupant D comes into contact with the airbag main body 40, and only the head H of the occupant D is restrained, so that the backward bending of the neck becomes large, or the head H of the occupant D is prevented from coming out of the loop space 36 of the deployed airbag 32.

[0038] In this embodiment, the control device 54 of the airbag device 10 determines whether or not a vehicle collision has occurred based on an output value from the collision sensor 50 that detects a vehicle collision, and determines whether or not the head H of the occupant D seated in the seat 12 is located within a predetermined range corresponding to the loop space 36 of the airbag 32, which includes the airbag body 40 that is deployed in front of the occupant when deployed and the pair of left and right front and rear chambers 34 that extend in the front-rear direction of the vehicle when deployed and form the loop space 36 between the airbag body 40, based on an output value from the interior camera 52 that captures an image of a range including the head H of the occupant D seated in the seat 12, and deploys the airbag 32 when it determines that a vehicle collision has occurred and that the head H is located within the predetermined range. This prevents the airbag 32 from being deployed when it is determined that the head H of the occupant D is not located within the predetermined range, thereby preventing the deployed airbag 32 from failing to exert the desired restraint performance and from adversely affecting the occupant D.

[0039] Furthermore, in this embodiment, the control device 54 judges whether or not the head H is located within the predetermined range based on a judgment model previously learned from learning data including data when the head H of the occupant D is not located within the predetermined range, in addition to the output value from the interior camera 52. This makes it possible to accurately judge whether or not the head H of the occupant D is located within the predetermined range corresponding to the loop space 36 of the airbag 32, without complicating the configuration.

[0040] In this embodiment, the data when the head H of the occupant D is not located within the predetermined range includes data when the occupant D is lying down and leaning against the window, and data when the occupant D is leaning forward. This makes it possible to accurately determine that the head H of the occupant D is not located within the predetermined range when the occupant D is lying down and leaning against the window or when the occupant D is leaning forward, in a determination based on a determination model learned in advance from learning data.

[0041] Furthermore, in this embodiment, the learning data includes first data in which the output value of the indoor camera 52 when the head H of the occupant D is not located within a predetermined range is associated with position information indicative of the position of the head H, and second data in which the output value of the indoor camera 52 when the head H of the occupant D is located within the predetermined range is associated with position information indicative of the position of the head H. As a result, in a determination based on a determination model previously learned from the learning data, position information indicative of the position of the head H of the occupant D can be obtained from the output value from the indoor camera 52, and it can be accurately determined from the position information whether or not the head H of the occupant D is located within the predetermined range.

[0042] In the above embodiment, a trained judgment model that outputs the position of the head H of the occupant D from the output value of the indoor camera 52 is generated by using, as the learning data, data in which the output value of the indoor camera 52 is associated with position information indicating the position of the head H of the occupant D. However, the present disclosure is not limited to this. For example, as the learning data, data in which the output value of the indoor camera 52 is associated with identification information indicating whether the head H of the occupant D is located within a predetermined range may be used. By using this learning data, a trained judgment model that outputs whether the head H of the occupant D is located within the predetermined range is generated from the output value of the indoor camera 52. In this case, based on the generated judgment model, it is possible to directly determine whether the head H of the occupant D is located within the predetermined range from the output value of the indoor camera 52.

[0043] In the above embodiment, the neural network model as an example of a model is trained by deep learning, but the present invention is not limited to this. For example, a model other than the neural network model may be trained by a learning method other than deep learning.

[0044] In the above embodiment, whether or not the head H of the occupant D is located within the predetermined range is determined based on the output value from the interior camera 52 and a determination model that is trained in advance from learning data including data when the head H of the occupant D is not located within the predetermined range corresponding to the loop space 36. However, the present disclosure is not limited to this. For example, the interior camera 52 may be a stereo camera to obtain three-dimensional coordinates of the head H of the occupant D, and whether or not the head H of the occupant D is within the predetermined range may be determined based on the obtained three-dimensional coordinates. [Explanation of symbols]

[0045] 10 Airbag device 32 Airbag 34 Front and rear chambers 36 Loop Space 40 Airbag body 50 Collision Sensor 52 Indoor Camera 54 Control device (control unit) 56 Collision determination section 58 Occupant attitude determination unit 60 General Judging Section

Claims

1. an airbag device including: a control unit that determines whether or not a vehicle collision has occurred based on an output value from a collision sensor that detects a vehicle collision; and that determines whether or not the head of an occupant seated in a seat is located within a predetermined range corresponding to a loop space of an airbag that includes an airbag body that is deployed in front of the occupant when deployed, and a pair of left and right front and rear chambers that extend in the fore-and-aft direction of the vehicle when deployed and form a loop space between the airbag body and the airbag body, based on output values ​​from a camera that captures an image within a shooting range including the head of the occupant; and that deploys the airbag when it determines that a vehicle collision has occurred and that the occupant's head is located within the predetermined range.

2. 2. The airbag device according to claim 1, wherein the control unit determines whether or not the occupant's head is located within the specified range based on a determination model that has been pre-trained from learning data that includes, in addition to the output value from the camera, data for when the occupant's head is not located within the specified range.

3. 3. The airbag device according to claim 2, wherein the data when the occupant's head is not located within the specified range includes data when the occupant is lying down and leaning against the window, and data when the occupant is leaning forward.

4. 3. The airbag device according to claim 2, wherein the learning data includes first data in which the output value of the camera when the occupant's head is not located within the specified range is associated with position information representing the position of the occupant's head, and second data in which the output value of the camera when the occupant's head is located within the specified range is associated with position information representing the position of the occupant's head.

Images