Airbag folding simulation system, method, and program

Abstract

Provided are an airbag folding simulation method and the like that reduce the difficulty of setting up an airbag folding simulation. This airbag folding simulation method comprises: a step for overlaying a three-dimensional animation with a simulation model, from the completion of deployment of an airbag to the stowed state thereof; a step for setting feature points in the simulation model that correspond to feature points set in the three-dimensional animation; a step for setting, at predetermined time intervals, how to move nodes on the simulation model so that the feature points of the simulation model approach the feature points of the three-dimensional animation; and a step for determining the set values for a numerical calculation so that the shape of the simulation model resulting from moving the nodes, as calculated by the numerical calculation, is as close as possible to the shape of the three-dimensional animation. The step for setting how to move the nodes and the step for determining the set values for the numerical calculation are repeated until the airbag is folded into the stowed state thereof.

Description

Airbag Folding Simulation System, Method, and Program

[0001] This invention relates to an airbag folding simulation system, method, and program.

[0002] In product type certification, in response to diversifying requirements, the idea of evaluating performance using simulation results instead of actual machine tests has attracted attention. For example, in the automotive industry, discussions on its application to the safety of autonomous driving and the cruising range (electricity cost) of electric vehicles have already begun. Thus, conventionally, it was common practice to consider the results of CAE and finally evaluate them with an actual machine. However, recently, there has been a tendency for the results of CAE to be directly evaluated.

[0003] Taking virtual testing in automotive collision safety as an example, the extent to which the simulation model can accurately represent real experiments is determined by the rating method of ISO18571. In the current protocol, while using the same simulation model, the validity is guaranteed by comparing the calculation results of two or more different initial conditions with the experimental results under the same conditions.

[0004] For example, an airbag is a very important device for protecting passengers during a collision. Since the simulation technology of airbags greatly affects the performance prediction accuracy, each company is striving to improve its simulation technology.

[0005] An airbag is folded and stored small so as not to interfere with the passengers as much as possible during normal driving. However, when a collision is detected, it needs to be deployed into a predetermined shape at high speed, so the folding method is very important and becomes a complex folding method.

[0006] As an example of a simulation-based complex folding method of such an airbag, the technology of Non-Patent Document 1 below can be cited.

[0007] "Case of JFOLD" (https: / / www.jsol-cae.com / product / struct / jfold / cases / )

[0008] However, the method described in Non-Patent Document 1 involves sandwiching a simulation model of the airbag between virtual plates called "tools," and then deforming these tools by performing geometric movements such as translation and rotation. This requires experienced engineers to perform the work by making inferences from work instruction drawings and observations of actual objects, resulting in a process of trial and error.

[0009] Therefore, performing highly accurate simulations presents challenges such as reliance on experienced engineers, increasing the amount of work required for rework, and errors in simulations due to discrepancies between drawings and reality. Furthermore, this method involves significant changes between steps, resulting in a large number of items that must be set at once. Consequently, it takes a long time to obtain the final result, making it difficult to speed up airbag folding simulations.

[0010] The present invention aims to solve the above problems, reduce the difficulty of setting up folding simulations, which are a bottleneck in airbag simulation technology, and provide an airbag folding simulation method, system, and program that can be implemented with little experience and is less dependent on individual expertise.

[0011] To solve the above-mentioned problems, one aspect of the airbag folding simulation system of the present invention includes: a 3D animation acquisition unit that acquires a 3D animation of folding from the end of deployment to the storage state by continuously playing back a 3D shape model of the airbag from the point when it is folded to the state when stored in a vehicle to the end of deployment, wherein multiple feature points are set at locations where the position can be relatively easily identified, such as on the folds, stitching lines, or functional parts of the airbag; a simulation model acquisition unit that acquires a simulation model of a virtual object corresponding to the airbag, created by CAE analysis software based on the finite element method; a feature point setting unit that superimposes the 3D animation and the simulation model from the end of deployment to the storage state, and sets feature points in the simulation model that correspond to the feature points set in the 3D animation; and a folding setting unit that sets how nodes on the simulation model move so that the feature points set in the simulation model are brought closer to the feature points set in the 3D animation at predetermined timings. The system includes a numerical calculation unit that performs numerical calculations to compare the shape of the simulation model obtained by moving the nodes with the shape of the 3D animation model, and determines the setting values ​​for the numerical calculation so that the shapes of the two models are as close as possible, and the processing by the folding setting unit and the processing by the numerical calculation unit are repeated until the airbag is folded to the stored state.

[0012] According to one embodiment of the simulation system, using a three-dimensional shape obtained from actual equipment measurements as a reference, the system specifies how to move the nodes of the simulation model in small steps, from the fully deployed state of the airbag to the state when it is stored in the vehicle, so that the feature points of the simulation model approach the feature points of the reference. Therefore, the changes between steps are not large, and the number of items that need to be set at once is reduced, thus lowering the difficulty of setting up the simulation model. As a result, it is possible to provide an airbag folding simulation system that can be run with little experience and is less dependent on individual expertise. In addition, the airbag folding simulation can be made faster.

[0013] Furthermore, according to another embodiment of the simulation system, the numerical calculation unit uses a plurality of setpoints with random variations, performs multiple numerical calculations for each setpoint, superimposes the shapes obtained by the calculations, compares the shape of the simulation model obtained by moving the nodes with the shape of the 3D animation, and selects from the plurality of setpoints such that the shapes of the two are as close as possible.

[0014] According to this embodiment, numerical calculations are performed multiple times for each of the multiple set values ​​with random variation, the shapes obtained from the calculations are superimposed, and the shape of the simulation model obtained by moving the nodes is compared with the shape of the 3D animation. Then, a selection is made from the multiple set values ​​so that the shapes of the two are as close as possible, thereby further improving the accuracy of the airbag folding simulation and shortening the construction period by reducing the number of reworks.

[0015] The aforementioned airbag folding simulation system can be implemented as both an airbag folding simulation method and an airbag folding simulation program.

[0016] According to the present invention, in a folding simulation that changes step by step from a fully deployed state to a folded state for storage in a vehicle, the changes between steps are not large, and the number of items that need to be set at once is reduced, thus lowering the difficulty of setting up the simulation model. As a result, it provides an airbag folding simulation system that can be run with little experience and is less dependent on individual expertise. Furthermore, the airbag folding simulation is accelerated.

[0017] This figure shows the schematic configuration of the airbag folding simulation system according to the first embodiment of the present invention. This figure shows the functional blocks of the airbag folding simulation system according to the first embodiment. This flowchart shows the flow of the process for creating a 3D shape model, setting feature points in the 3D shape model, and creating a 3D animation in the first embodiment. (A) to (D) show examples of photographs taken using photogrammetry while the airbag is unfolded little by little by hand. (A) to (C) show examples of markers attached as feature points to the 3D animation shape, and (D) to (F) show examples in which feature points corresponding to the feature points of the 3D animation shape are set in the simulation model, corresponding to (A), (B), and (C), respectively, and markers are attached to these feature points. This flowchart shows the flow of the process for acquiring the simulation model, setting feature points in the simulation model, and numerical calculation in the first embodiment. Figures (A) to (C) show three-dimensional animation shapes corresponding to (D), (E), and (G), respectively. Figures (D) to (I) show examples of moving the simulation model shape by numerical calculation by overlaying the simulation model shape onto the three-dimensional animation shape and specifying how the nodes move. Figures (A) and (B) are diagrams illustrating examples of specifying how the nodes move. Figures (A) to (C) are diagrams illustrating the folding simulation of an airbag of a comparative example. Figures (A) to (C) are diagrams illustrating the folding simulation of an airbag of a comparative example. This is a flowchart showing the flow of numerical calculation processing in the second embodiment of the present invention. Figures (A) and (B) show examples of three-dimensional animation shapes and simulation model shapes resulting from the numerical calculation processing in Figure 11. This is a diagram showing the functional blocks of the airbag folding simulation system according to the third embodiment.

[0018] (First Embodiment) Hereinafter, an airbag folding simulation system according to the first embodiment of the present invention will be described in detail with reference to the drawings. Figure 1 is a diagram showing the schematic configuration of the airbag folding simulation system 100 according to this embodiment. As shown in Figure 1, the airbag folding simulation system 100 of this embodiment includes a central processing unit 1, a display device 2, a storage device 3, an input device 4, and an output device 5.

[0019] The central processing unit 1 is a device such as a personal computer capable of executing programs, and includes a CPU and memory. The display device 2 is a device such as a liquid crystal display capable of displaying characters and images. The storage device 3 is a device such as an HDD (Hard Disk Drive) capable of storing programs and data, and an external database server may also be used. The program of the present invention is stored in the storage device 3. The input device 4 is a device such as a keyboard that allows the user to input data or instructions. The output device 5 is a device such as a printer capable of outputting characters and images. In the airbag folding simulation system 100 of this embodiment, the output device 5 may be omitted.

[0020] Figure 2 shows a functional block that operates when the central processing unit 1 executes the program of the present invention. As shown in Figure 2, the central processing unit 1 functions as a control unit 10. The control unit 10 also functions as a 3D animation acquisition unit 12, a simulation model acquisition unit 13, a feature point setting unit 14, a folding setting unit 15, and a numerical calculation unit 16, according to the program of the present invention.

[0021] The 3D shape model is created outside of the airbag folding simulation system 100. The 3D shape model is created, for example, by first folding an actual airbag to the state it is stored in the vehicle, and then gradually unfolding the airbag by hand, while photographing it using photogrammetry or the like until the airbag is fully deployed. In addition, multiple feature points are set in this 3D shape model at locations on the airbag that are relatively easy to pinpoint as it is gradually unfolded. These feature points are set, for example, on the folds of the airbag, on the seams, or on functional parts such as vent holes. Details on setting feature points in the 3D shape model and acquiring the 3D shape model will be described later.

[0022] The 3D animation acquisition unit 12 acquires 3D animations created outside the airbag folding simulation system 100. The 3D animation is created by continuously playing the 3D shape model created as described above in reverse. In this way, a 3D animation is created showing the airbag folding from the fully deployed state to the state when it is stored in the vehicle. The 3D animation acquisition unit 12 acquires the 3D animation from an external source via a network or the like and stores it in the storage device 3.

[0023] The simulation model acquisition unit 13 acquires a simulation model of a virtual object corresponding to an airbag, which is created outside the airbag folding simulation system 100. The simulation model is created using CAE analysis software based on the finite element method. The simulation model acquisition unit 13 acquires the simulation model from an external source via a network or the like and stores it in the storage device 3.

[0024] The feature point setting unit 14 overlays the 3D animation created by the 3D animation acquisition unit 12 with the simulation model acquired by the simulation model acquisition unit 13, and sets the feature points corresponding to the feature points set in the 3D animation into the simulation model. Details of the processing of the feature point setting unit 14 will be described later.

[0025] The folding setting unit 15 sets how to move the nodes on the simulation model so that the feature points set in the simulation model by the feature point setting unit 14 are brought closer to the feature points set in the 3D animation at predetermined intervals. Details of the processing of the folding setting unit 15 will be described later.

[0026] The numerical calculation unit 16 performs numerical calculations to compare the shape of the simulation model obtained by moving the nodes on the simulation model with the shape of the 3D animation model, and determines the numerical calculation settings so that the shapes of the two models are as close as possible. Details of the processing of the numerical calculation unit 16 will be described later.

[0027] In this embodiment, the processing by the folding setting unit 15 and the processing by the numerical calculation unit 16 are repeated until the airbag is folded from a fully deployed state to the state when stored in the vehicle. Details of this process will be described later.

[0028] (Process for creating a 3D shape model, setting feature points in the 3D shape model, and creating a 3D animation) Next, the process for creating a 3D shape model, setting feature points in the 3D shape model, and creating a 3D animation in this embodiment will be explained with reference to Figures 3 to 5.

[0029] As described above, in this embodiment, the process of creating a three-dimensional shape model of the airbag is performed outside the airbag folding simulation system 100.

[0030] When creating a three-dimensional shape model of an airbag, first, an actual airbag is prepared in the state it is folded to when stored in a vehicle (Figure 3: Step S1). The folding method of this airbag is predetermined to suit the size and shape of the storage space, or to ensure that a desirable deployment behavior is obtained.

[0031] Next, the folded airbag is gradually unfolded by hand and photographed, for example, using photogrammetry (Figure 3: Step S2). Note that the method for creating a three-dimensional shape model of the present invention is not limited to the method using photogrammetry; for example, a method that uses an infrared sensor or laser reflection to read the distance to the airbag may also be used.

[0032] Figures 4(A) to (D) show examples of photographs taken using photogrammetry while the airbag is deployed little by little by hand. The three-dimensional shape obtained by photographing using photogrammetry is saved in, for example, STL format.

[0033] As the airbag is deployed little by little, markers are attached to characteristic points (Figure 3: Step S3). In this embodiment, these characteristic points will be called "feature points." These feature points are set so that they can be matched with the simulation model described later. Feature points are set, for example, on folds such as mountain folds and valley folds of the airbag, on or around circular stitching lines, or on or around functional parts such as vent holes.

[0034] Figures 5(A) to 5(C) show examples of markers attached as feature points 30a to 30e. It should be noted that the present invention is not limited to markers as a method for representing feature points; for example, the points may be actually written on the airbag with a marker.

[0035] Then, the airbag is slightly unfolded again (Figure 3: Step S4), and the process of taking images using photogrammetry as described above (Figure 3: Step S2), setting feature points, that is, attaching markers to the feature points (Figure 3: Step S3), and then slightly unfolding it again (Figure 3: Step S4) is repeated until the airbag is fully deployed and the deployment is complete (Figure 3: Step S5; NO).

[0036] Once the airbag is fully deployed and deployment is complete (Figure 3: Step S5; YES), a 3D animation of the airbag folding from its fully deployed state to its stored state in the vehicle is generated by sequentially playing back multiple 3D shapes captured by photogrammetry (Figure 3: Step S6).

[0037] (3D animation and simulation model acquisition process, simulation model feature point setting process, and numerical calculation process) Next, the simulation model acquisition process, the simulation model feature point setting process, and the numerical calculation process in this embodiment will be explained with reference to Figures 5 to 8.

[0038] First, the 3D animation acquisition unit 12 acquires the 3D animation created outside the airbag folding simulation system 100 as described above (Figure 6: Step S10). In this embodiment, a simulation model for a virtual object corresponding to the airbag is created outside the airbag folding simulation system 100, and this simulation model is acquired by the simulation model acquisition unit 13 (Figure 6: Step S10).

[0039] Simulation models based on the finite element method are created using CAE preprocessing software. For example, an airbag base fabric model is created using an assembly of thin membrane elements measuring 2 mm square. The simulation model may be created from CAD data used in the product development process, or from CAD drawings created based on measurements of the actual device shape.

[0040] Next, the feature point setting unit 14 overlays the three-dimensional animation from the fully deployed state to the state when the airbag is stored in the vehicle, as described above, with the simulation model of the airbag acquired by the simulation model acquisition unit 13 (Figure 6: Step S11).

[0041] Then, the feature point setting unit 14 sets the feature points corresponding to the feature points set in the 3D animation as the feature points in the simulation model (FIG. 6: step S12). In FIG. 5, the feature points corresponding to the feature points 30a to 30e set in the 3D animation shown in FIGS. 5(A) to 5(C) are set as the feature points 31a to 31e in the simulation model as shown in FIGS. 5(D) to 5(F), and an example in which markers are attached to the feature points 31a to 31e is shown.

[0042] Next, the folding setting unit 15 specifies how to move the nodes on the simulation model so that the feature points set in the simulation model by the feature point setting unit 14 approach the feature points set in the 3D animation at each predetermined timing (FIG. 6: step S13). Then, the folding setting unit 15 performs the setting on the simulation model so as to achieve such movement (FIG. 6: step S14).

[0043] FIGS. 7(D) to 7(I) are diagrams for explaining the processing by the folding setting unit 15 described above. FIG. 7(D) shows the shape of the simulation model of the airbag in a fully deployed state. Incidentally, FIG. 7(A) shows the 3D animation shape corresponding to FIG. 7(D).

[0044] FIG. 7(E) shows an example in which the 3D animation shape when a little time has elapsed from FIG. 7(D) and the shape of the simulation model of FIG. 7(D) are superimposed. Incidentally, FIG. 7(B) shows the 3D animation shape corresponding to FIG. 7(E). The folding setting unit 15 specifies how to move the nodes on the simulation model shape shown in FIG. 7(E) so that the feature points set in the simulation model shape shown in FIG. 7(E) approach the feature points set in the 3D animation shape shown in FIG. 7(E). The straight arrows shown in FIG. 7(E) indicate this way of movement.

[0045] Next, the numerical calculation unit 16 performs numerical calculations (Fig. 6: step S15), compares the shape of the simulation model resulting from moving the nodes with the shape of the 3D animation model, and determines the set value of the numerical calculation so that the shapes of both are the closest (Fig. 6: step S16). Here, if the shapes of both are separated beyond a predetermined value and the movement of the nodes as specified cannot be obtained, the numerical calculation unit 16 changes the set value of the numerical calculation, repeats the numerical calculation and the shape comparison, and determines the set value of the numerical calculation so that the shapes of both are the closest (Fig. 6: step S16).

[0046] Fig. 7(F) shows the 3D animation shape after the numerical calculation by the numerical calculation unit 16 and the simulation model shape. As in this example, in this embodiment, by specifying the way of moving the nodes of the simulation model shape so that the feature points of the simulation model shape approach the feature points of the 3D animation model, the shape of the simulation model can be made closer to the 3D animation model shape.

[0047] In this embodiment, the processing by the folding setting unit 15 and the processing by the numerical calculation unit 16 as described above are repeated until the airbag is folded into the state when stored in the vehicle (Fig. 6: step S17; NO).

[0048] Fig. 7(G) shows an example in which the 3D animation shape when a little time has passed from Fig. 7(F) and the simulation model shape of Fig. 7(F) are overlapped. Incidentally, Fig. 7(C) shows the 3D animation shape corresponding to Fig. 7(G). The folding setting unit 15 specifies the way of moving the nodes on the simulation model shape shown in Fig. 7(G) so that the feature points set on the simulation model shape shown in Fig. ৭(G) approach the feature points set on the 3D animation shape shown in Fig. 7(G). The straight arrow shown in Fig. 7(G) indicates this way of moving.

[0049] Figure 7(H) shows the 3D animation shape after numerical calculation by the numerical calculation unit 16 and the simulation model shape. As shown in this example, in this embodiment, the simulation model shape can be made to resemble the 3D animation model shape by specifying how to move the nodes of the simulation model shape so that the feature points of the simulation model shape are brought closer to the feature points of the 3D animation model. Figure 7(I) shows an example in which the 3D animation shape after a short time has passed since Figure 7(H) is superimposed with the simulation model shape of Figure 7(H). The folding setting unit 15 sets how to move the nodes on the simulation model shape shown in Figure 7(I) so that the feature points set on the simulation model shape shown in Figure 7(I) are brought closer to the feature points set on the 3D animation shape shown in Figure 7(I). The straight arrows shown in Figure 7(I) indicate this movement. The above process is repeated in the same manner until the airbag is completely folded.

[0050] Then, if the airbag is completely folded (Figure 6: Step S17; YES), the numerical calculation unit 16 outputs a simulation model of the folded airbag (Figure 6: Step S18).

[0051] Furthermore, the folding setting unit 15 can specify how to move the nodes on the simulation model shape, as shown in Figure 8(A). For example, it can specify that a spring constant be set to apply force to the nodes, or it can specify that the triangle formed by connecting the nodes be pushed or pulled as a surface to force movement, as shown in Figure 8(A).

[0052] As described above, according to this embodiment, using the three-dimensional shape obtained from actual measurements as a reference, the movement of the nodes of the simulation model is specified in fine steps to bring the feature points of the simulation model closer to the reference feature points, from the fully deployed state of the airbag to the state when it is stored in the vehicle. Therefore, the changes between steps are not large, and the number of items that need to be set at once is reduced, thus lowering the difficulty of setting up the simulation model. As a result, it is possible to provide an airbag folding simulation system that can be run even with little experience and is less dependent on individual expertise. In addition, it is possible to speed up the airbag folding simulation.

[0053] (Comparative Example) Next, a conventional comparative example, which is compared with the first embodiment, will be described. Figures 9 and 10 are diagrams illustrating a method for simulating the folding of an airbag in a comparative example.

[0054] In the comparative example, as shown in Figures 9 and 10, the airbag simulation model is sandwiched between virtual plates called "tools," and the tools are deformed by performing geometric movements such as translation and rotation, and the folding is represented by repeating this process.

[0055] These geometric movements are performed by dragging and dropping with the mouse or by setting numerical coordinates.

[0056] Such work is performed by experienced engineers based on inferences from work instruction drawings and observations of actual objects, resulting in a process of trial and error. Therefore, there are challenges such as the reliance on experienced engineers to perform highly accurate simulations, and the increased workload due to rework. Furthermore, there is the problem of simulation errors due to differences between drawings and reality. In addition, this method has the challenge of large changes between steps and a large number of items that must be set at once. As a result, it takes a lot of time to obtain the final result, making it difficult to speed up airbag folding simulations.

[0057] However, according to the first embodiment described above, the changes between steps are not as large as in the conventional comparative example, and the number of items that need to be set at once is reduced, thus lowering the difficulty of setting up the simulation model. As a result, it is possible to provide an airbag folding simulation system that can be run even with little experience and is less dependent on individual expertise. Furthermore, it is possible to speed up the airbag folding simulation.

[0058] (Second Embodiment) Next, a second embodiment of the present invention will be described with reference to Figures 11 and 12. In the first embodiment, the folding setting unit 15 determines setting values ​​so that the nodes move as specified, and the numerical calculation unit 16 performs numerical calculations. The shape of the simulation model obtained from the calculation is then compared with the shape of the 3D animation model, and if the two shapes are separated by more than a predetermined value, the numerical calculation unit 16 changes the setting values ​​and repeats the numerical calculation and shape comparison.

[0059] In contrast, in this embodiment, multiple setting values ​​with random variation are used as setting values ​​for numerical calculation (Figure 11: Steps S20-1 to S20-n). Then, the numerical calculation unit 16 performs multiple numerical calculations for each setting value (Figure 11: Steps S21-1 to S21-n).

[0060] As a result, as shown in Figure 12(A), multiple simulation shapes corresponding to multiple setting values ​​are obtained for a reference 3D animation shape.

[0061] Next, the numerical calculation unit 16 superimposes multiple simulation shapes (Figure 11: Step S22) and, for each node, selects the simulation shape closest to the reference 3D animation shape and performs fitting (Figure 11: Step S23). In other words, for each node, it selects from multiple set values ​​so that the simulation shape closest to the reference 3D animation shape is obtained. As a result, as shown in Figure 12(B), the simulation shape closest to the reference 3D animation shape can be obtained at high speed. Thus, according to this embodiment, the airbag folding simulation can be made even faster.

[0062] (Third Embodiment) Next, a third embodiment of the present invention will be described with reference to Figure 13. In the first embodiment, a three-dimensional model was created outside the airbag folding simulation system 100, and a three-dimensional animation created based on this three-dimensional model was acquired by the three-dimensional animation acquisition unit 12 in the airbag folding simulation system 100.

[0063] In contrast, in this embodiment, instead of the 3D animation acquisition unit 11, a 3D formation model creation unit 17 is used to create a 3D formation model within the airbag folding simulation system 100. The method for creating the 3D formation model is the same as that described in the first embodiment. Furthermore, a 3D animation is created using the 3D animation creation unit 18. The method for creating the 3D animation is the same as that described in the first embodiment.

[0064] In this embodiment as well, the changes between steps are not large, and the number of items that need to be set at once is reduced, thus lowering the difficulty of setting up the simulation model. As a result, it is possible to provide an airbag folding simulation system that can be run even with little experience and is less dependent on individual expertise. Furthermore, it is possible to speed up the airbag folding simulation.

[0065] (Variations) As described above, the creation of a three-dimensional shape model may be done by photographing the airbag using photogrammetry, or, as described above, by using methods such as reading the distance of the airbag, which is the subject, using infrared sensors or laser reflection.

[0066] The distinctive features can be handwritten on the airbag base fabric with a marker, or by attaching stickers. Alternatively, the distinctive features can be represented by attaching colored ID stickers or QR codes to the airbag base fabric. Furthermore, the distinctive features can be automatically set by automatically recognizing color differences such as folds and seams from photographs obtained using photogrammetry. The distinctive features can also be added during the photogrammetry process, or created using image processing after capturing an image of the 3D shape model. When creating the 3D shape model, an actual airbag can be used, or a pattern printed from CAD or fabric traced from the pattern can be used.

[0067] The airbag folding simulation program described above can be provided in a form stored on a computer-readable recording medium and installed on a computer. The recording medium is, for example, a non-transitory recording medium, such as an optical recording medium like a CD-ROM, but can include any known form of recording medium such as a semiconductor recording medium or a magnetic recording medium. It is also possible to provide the program via a communication network and install it on a computer.

[0068] The airbag folding simulation method, airbag folding simulation system, and airbag folding simulation program according to embodiments of the present invention have been described above. However, the present invention is not limited thereto, and various modifications are possible without departing from the spirit of the invention.

[0069] 1 Central Processing Unit 2 Display Device 3 Storage Device 4 Input Device 5 Output Device 10 Control Unit 12 3D Animation Acquisition Unit 13 Simulation Model Acquisition Unit 14 Feature Point Setting Unit 15 Folding Setting Unit 16 Numerical Calculation Unit 17 3D Shape Model Creation Unit 100 Airbag Folding Simulation System

Claims

1. A 3D animation acquisition unit acquires a 3D animation of the airbag folding from the end of deployment to the end of storage by continuously playing in reverse the 3D shape model of the airbag from the point when it is folded to the state when stored in the vehicle, wherein multiple feature points are set on locations where the position can be relatively easily identified, such as on the folds, stitching lines, or functional parts of the airbag; a simulation model acquisition unit acquires a simulation model of a virtual object corresponding to the airbag, created by CAE analysis software based on the finite element method; a feature point setting unit superimposes the 3D animation and the simulation model from the end of deployment to the end of storage, and sets feature points in the simulation model that correspond to the feature points set in the 3D animation; and a folding setting unit sets how to move nodes on the simulation model so that the feature points set in the simulation model are brought closer to the feature points set in the 3D animation at predetermined timings. An airbag folding simulation system comprising: a numerical calculation unit that performs numerical calculations to compare the shape of the simulation model obtained by moving the nodes with the shape of the three-dimensional animation model, and determines the setting values ​​for the numerical calculation so that the shapes of the two are as close as possible; and repeating the processing by the folding setting unit and the processing by the numerical calculation unit until the airbag is folded to the stored state.

2. The airbag folding simulation system according to claim 1, wherein the numerical calculation unit uses a plurality of set values ​​with random variation, performs multiple numerical calculations for each set value, superimposes the shapes obtained by the calculations, compares the shape of the simulation model obtained by moving the nodes with the shape of the three-dimensional animation, and selects from the plurality of set values ​​such that the shapes of the two are closest.

3. A step of acquiring a 3D animation in which a 3D shape model of the airbag from the point when it is folded to the state when stored in the vehicle until the end of deployment is played in reverse order, thereby acquiring a 3D animation in which the airbag is folded from the end of deployment back to the state when stored, and in which multiple feature points are set at locations where it is relatively easy to pinpoint the position, such as on the folds, seams, or functional parts of the airbag; a step of acquiring a simulation model of a virtual object corresponding to the airbag, created by CAE analysis software based on the finite element method; a step of overlaying the 3D animation and the simulation model from the end of deployment back to the state when stored; a step of setting feature points in the simulation model that correspond to the feature points set in the 3D animation; a step of setting how to move nodes on the simulation model at predetermined timings so that the feature points set in the simulation model are brought closer to the feature points set in the 3D animation; a step of performing numerical calculations to compare the shape of the simulation model as a result of moving the nodes with the shape of the 3D animation, and determining the setting values ​​of the numerical calculations so that the shapes of the two are as close as possible. A method for simulating the folding of an airbag, comprising the steps of setting how to move the nodes on the simulation model and determining the set values ​​for the numerical calculation, and repeating these steps until the airbag is folded to the stored state.

4. The method for simulating the folding of an airbag according to claim 3, wherein the step of determining the setting values ​​for the numerical calculation is to prepare a plurality of setting values ​​with random variation as the setting values, perform the numerical calculation multiple times for each setting value, superimpose the shapes obtained by the calculation, compare the shape of the simulation model obtained by moving the nodes with the shape of the three-dimensional animation, and select from the plurality of setting values ​​such that the shapes of the two are closest.

5. An airbag folding simulation program that causes a computer to perform an airbag folding simulation, the program comprising: a 3D animation acquisition unit that continuously reverses the playback of a 3D shape model of the airbag from the point when it is folded to the state when stored in the vehicle to the end of deployment, thereby acquiring a 3D animation of the airbag folding from the end of deployment to the state when stored, wherein multiple feature points are set at locations that are relatively easy to identify, such as on the folds, seams, or functional parts of the airbag; a simulation model acquisition unit that acquires a simulation model of a virtual object corresponding to the airbag, created by CAE analysis software based on the finite element method; a feature point setting unit that superimposes the 3D animation and the simulation model from the end of deployment to the state when stored, and sets feature points in the simulation model that correspond to the feature points set in the 3D animation; a folding setting unit that sets how nodes on the simulation model move so that the feature points set in the simulation model are brought closer to the feature points set in the 3D animation at predetermined timings; and An airbag folding simulation program that functions as a numerical calculation unit that performs numerical calculations, compares the shape of the simulation model obtained by moving the nodes with the shape of the simulation model, and determines the setting values ​​for the numerical calculation so that the shapes of the two are as close as possible, and repeats the processing by the folding setting unit and the processing by the numerical calculation unit until the airbag is folded to the stored state.

6. The airbag folding simulation program according to claim 5, wherein the numerical calculation unit uses a plurality of set values ​​with random variation, performs multiple numerical calculations for each set value, superimposes the shapes obtained by the calculations, compares the shape of the simulation model obtained by moving the nodes with the shape of the three-dimensional animation, and selects from the plurality of set values ​​such that the shapes of the two are closest.

Images