Simulation design method for unmanned aerial vehicle airbag outer shape based on BIM

Abstract

This invention discloses a BIM-based simulation design method for the shape of an unmanned aerial vehicle (UAV) airbag, belonging to the field of UAV airbag design technology. The method includes: acquiring a BIM spatial model of the UAV fuselage, extracting the set of structural feature points and the fuselage envelope surface; dividing the airbag installation area on the fuselage envelope surface and obtaining the surface normal vector distribution; generating geometric constraint parameters for the airbag deployment profile based on the surface normal vector distribution and a preset flight attitude angle; constructing the initial airbag shape surface in a BIM parametric modeling environment and obtaining its assembly gap distribution with the fuselage envelope surface; adjusting the geometric constraint parameters until the assembly gap distribution meets the preset assembly tolerance, thus obtaining the target airbag shape surface. This method enables adaptive matching between the airbag shape and the fuselage surface, reducing repeated trial and error.

Description

Technical Field

[0001] This invention relates to the field of UAV airbag design technology, specifically a BIM-based simulation design method for the shape of UAV airbags. Background Technology

[0002] In the design of emergency landing or protection for UAVs, the shape of the airbag needs to fit closely to the fuselage surface to ensure effective containment or cushioning after deployment. Traditional airbag design often involves manually measuring the fuselage geometry, drawing the outline surface in general CAD software based on experience, verifying the gap status through prototype assembly, and repeatedly revising the mold until assembly requirements are met. This method relies on the designer's judgment, and the cycle from measurement to mold revision is long. Especially when the fuselage surface is a complex free-form surface, it is difficult to accurately delineate the airbag installation boundary and reasonably determine the deployment contour parameters manually, which can easily lead to large interference or gaps between the initial shape and the fuselage surface. In addition, the coordination between the airbag deployment direction and the actual fuselage surface normal vector during flight attitude changes is ignored, resulting in an airbag shape that cannot adapt to multi-attitude deployment requirements. Gap control lacks quantitative feedback, often leading to local stress concentration or burst risk.

[0003] With the application of Building Information Modeling (BIM) technology in the field of unmanned aerial vehicles (UAVs), the 3D fuselage model has achieved high accuracy. However, existing methods have not fully utilized the structural features and surface information of the fuselage in the BIM model to automatically determine the airbag installation boundary, nor have they combined the surface normal vector distribution with the flight attitude angle to generate constraint shape parameters. Furthermore, they lack closed-loop design methods based on feedback and iterative adjustment of assembly gap distribution. How to make the airbag shape adaptively match the fuselage envelope surface and quickly converge to the target shape that meets the assembly tolerance during iteration is a problem that needs to be solved. Summary of the Invention

[0004] This invention provides a BIM-based simulation design method for the shape of an airbag in a drone. It aims to utilize the fuselage structural features and surface normal vector distribution in the BIM model, combined with flight attitude angles, to automatically generate geometric constraint parameters for the airbag shape. The parameters are then adjusted by feedback from the assembly gap distribution, achieving precise adaptation between the airbag shape and the fuselage surface, thus replacing the manual trial-and-error process.

[0005] To achieve the above objectives, the present invention provides the following technical solution: The present invention provides a simulation design method for the shape of an airbag for a UAV based on BIM, comprising: acquiring a BIM spatial model of the UAV fuselage, and extracting the set of fuselage structural feature points and the fuselage envelope surface from the BIM spatial model; dividing the airbag installation area on the fuselage envelope surface according to the set of fuselage structural feature points, and obtaining the distribution of surface normal vectors within the airbag installation area; generating geometric constraint parameters for the airbag deployment profile according to the distribution of surface normal vectors and a preset UAV flight attitude angle; inputting the geometric constraint parameters into the parametric modeling environment of BIM software to construct the initial shape surface of the airbag, and obtaining the assembly gap distribution between the initial shape surface of the airbag and the fuselage envelope surface; adjusting the geometric constraint parameters according to the assembly gap distribution until the assembly gap distribution meets the preset assembly tolerance, thereby obtaining the target airbag shape surface.

[0006] This method integrates BIM model with UAV flight attitude information, precisely defining the airbag installation area on the fuselage envelope surface and extracting its surface normal vector distribution. This ensures that the geometric constraint parameters of the airbag deployment profile accurately reflect the relationship between the fuselage surface curvature changes and the incoming flow direction, significantly improving the matching degree between the airbag shape and the fuselage aerodynamic shape. By automatically generating the initial shape surface in the BIM parametric modeling environment and dynamically checking and iteratively adjusting the assembly gap, interference between the airbag and the fuselage can be effectively avoided, ensuring controllable assembly tolerances. Thus, the airbag shape surface that meets the assembly requirements can be obtained in the virtual simulation stage, eliminating the need for repeated physical trial assembly and modification, significantly shortening the airbag design cycle and reducing the cost of physical prototype manufacturing.

[0007] As a preferred technical solution of the present invention, a BIM spatial model of the UAV fuselage is obtained, and the set of structural feature points and the fuselage envelope surface are extracted from the BIM spatial model. Specifically, this includes: identifying the skeleton line structure of the UAV fuselage in the BIM spatial model, obtaining the intersection coordinates and endpoint coordinates of the skeleton line structure, and compiling them into a set of structural feature points; extracting all vertex coordinates of the triangular mesh on the outer surface of the fuselage in the BIM spatial model, and performing surface fitting on all vertex coordinates to obtain the fuselage envelope surface. By extracting the set of structural feature points from the skeleton line structure, the stress-bearing skeleton layout of the fuselage can be accurately reflected, providing a structural constraint basis for the division of the airbag installation area; based on the surface fitting of the triangular mesh vertices, the actual geometric shape of the outer surface of the fuselage can be completely reproduced, so that the subsequent surface normal vector calculation and assembly gap analysis are based on a close approximation of the actual shape.

[0008] As another preferred technical solution of the present invention, based on the set of fuselage structural feature points, an airbag installation area is divided on the fuselage envelope surface, and the surface normal vector distribution within the airbag installation area is obtained. Specifically, this includes: projecting the set of fuselage structural feature points onto the fuselage envelope surface to obtain a projected feature point set; using the outermost projected feature point in the projected feature point set as a boundary marker point, performing region growth on the fuselage envelope surface to obtain the airbag installation area; traversing each grid patch within the airbag installation area, calculating the normal vector of each grid patch, and statistically obtaining the surface normal vector distribution. Preferably, the region growth uses the closed boundary line enclosed by the boundary marker points as the seed region, and the growth condition is that the angle between the normal vectors of adjacent grid patches is less than a set angle threshold, gradually expanding outwards until reaching the closed boundary line. This method, combining structural feature point projection with region growth based on consistent normal vectors, can automatically and accurately define suitable airbag installation areas, such as concave or convex areas on the fuselage surface, avoiding subjective errors caused by manual division and ensuring the surface continuity of the installation area, which is beneficial for airbag fitting and deployment.

[0009] Further, based on the surface normal vector distribution and the preset UAV flight attitude angle, geometric constraint parameters for the airbag deployment profile are generated. Preferably, this includes: determining the vector of the airbag deployment direction based on the preset UAV flight attitude angle; calculating the angle between the principal normal vector in the surface normal vector distribution and the vector of the airbag deployment direction; marking the mesh patches corresponding to surface normal vectors with angles less than a preset angle threshold as principal deployment patches; obtaining the boundary contour line of the principal deployment patch; and measuring the maximum and minimum span values ​​of the boundary contour line as geometric constraint parameters. The maximum and minimum span values ​​of the boundary contour line can be obtained by calculating the maximum and minimum Euclidean distances between all pairs of points on the boundary contour line of the principal deployment patch, respectively. By using the flight attitude angle to determine the deployment direction and selecting the area whose normal vector is approximately consistent with the deployment direction as the main deployment surface, it can be ensured that the airbag's windward side faces the flight direction during deployment, thus optimizing the aerodynamic force state. The span parameter is directly derived from the boundary contour of the main deployment surface as a geometric constraint, so that the generation of the airbag's cross-sectional contour directly depends on the actual fuselage surface features, making the design basis more objective.

[0010] Further, the geometric constraint parameters are input into the parametric modeling environment of the BIM software to construct the initial shape surface of the airbag and obtain the assembly gap distribution between the initial shape surface of the airbag and the fuselage envelope surface. Preferably, this includes: inputting the maximum span value and the minimum span value into the parametric modeling environment of the BIM software as the major axis parameter and minor axis parameter of the airbag cross-sectional profile; generating the airbag cross-sectional profile line in the parametric modeling environment based on the major axis parameter and the minor axis parameter, and stretching the airbag cross-sectional profile line along the direction of the principal normal vector to obtain the initial shape surface of the airbag; performing a Boolean difference operation on the initial shape surface of the airbag and the fuselage envelope surface in the BIM software to extract the set of gap thickness values ​​in the non-overlapping areas as the assembly gap distribution. When stretching the airbag cross-sectional profile line along the direction of the principal normal vector, the stretching length can be dynamically determined based on the outward safety distance of the fuselage envelope surface in the direction of the principal normal vector and the maximum flight speed of the UAV. By using the major and minor axis parameters to drive the cross-sectional profile and stretch it along the principal normal direction to generate a curved surface, the shape of the airbag and the fuselage surface can maintain a coordinated transition in the normal direction, reducing airflow interference. Furthermore, by directly extracting the set of gap thickness values ​​through Boolean difference operations, the assembly gap information is fully quantified, providing accurate data support for subsequent interference correction. The dynamic determination of the stretching length is more in line with the space safety requirements for airbag deployment under combat or flight conditions.

[0011] Furthermore, based on the assembly gap distribution, the geometric constraint parameters are adjusted until the assembly gap distribution meets the preset assembly tolerance, thereby obtaining the target airbag outer surface. Preferably, this includes: comparing the set of gap thickness values ​​in the assembly gap distribution with the preset assembly tolerance, and marking areas where the gap thickness values ​​exceed the preset assembly tolerance as interference regions; obtaining the coordinates of the center point of the interference region, calculating the projected area of ​​the interference region along the principal normal vector direction based on the center point coordinates, proportionally reducing the maximum span value based on the projected area to obtain an updated maximum span value; using the updated maximum span value and the minimum span value as new geometric constraint parameters to reconstruct the airbag outer surface until all gap thickness values ​​fall within the preset assembly tolerance range, thereby obtaining the target airbag outer surface. Here, when proportionally reducing the maximum span value based on the projected area, the proportionality coefficient for proportional reduction is preferably the ratio of the projected area of ​​the interference region to the total projected area of ​​the main unfolded surface. By marking the interference area and adjusting the maximum span value in a targeted manner according to the proportion of its projected area in the principal normal vector direction, closed-loop iterative optimization of the assembly gap was achieved. This can quickly eliminate local interference while ensuring the effective coverage area of ​​the airbag. The final result is a uniform and tolerant fit gap between the airbag's outer curved surface and the fuselage's envelope curved surface, which not only ensures the reliability of the airbag installation but also avoids the risk of increased aerodynamic drag caused by excessive gaps or the risk of deployment scratches caused by excessively small gaps.

[0012] The technical effects and advantages provided by the present invention in the above technical solution are as follows:

[0013] After extracting the fuselage envelope surface and dividing the airbag installation area, the distribution of surface normal vectors within this area is obtained. This, combined with preset UAV flight attitude angles, generates the geometric constraint parameters for the airbag deployment profile. The surface normal vector distribution reflects the surface orientation differences of the fuselage at various points within the installation area, while the flight attitude angles determine the airbag's deployment direction vector under specific flight conditions. By calculating the angle between the principal normal vector in the surface normal vector distribution and the airbag deployment direction vector, the principal deployment surface is selected. Based on the maximum and minimum spans of the boundary contour lines of this surface, the major and minor axis parameters of the airbag cross-sectional profile are directly derived. This method binds the initial parameters of the airbag deployment profile to the actual fuselage surface shape and flight deployment direction, avoiding the disconnect from the overall fuselage surface caused by manually estimating the profile size based on experience. This ensures that the initial shape surface has an adaptive tendency to fit the fuselage surface along the deployment direction, reducing the number of subsequent major adjustments.

[0014] Based on this, after constructing the initial airbag shape surface, the assembly gap distribution between it and the fuselage envelope surface is obtained. Interference areas are marked by comparing the set of gap thickness values ​​with preset assembly tolerances. The maximum span value is then proportionally reduced based on the ratio of the projected area of ​​the interference area along the principal normal vector direction to the total projected area of ​​the main unfolded surface, thereby updating the geometric constraint parameters and reconstructing the airbag shape surface. This feedback adjustment mechanism directly drives the correction of geometric constraint parameters using quantitative information on the gap distribution, replacing the traditional method of relying on experience to judge the amount of modification and repeated trial assembly. This ensures that each iteration has a clear adjustment direction and proportional basis. Through iteration, all gap thickness values ​​can fall within the preset assembly tolerance range. The resulting target airbag shape surface and fuselage surface have uniform and controllable gaps, maintaining the assembly relationship even during changes in flight attitude, reducing the risk of local overpressure or suspension after airbag inflation. Attached Figure Description

[0015] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0016] Figure 1 This is a flowchart of a BIM-based method for simulating the shape of an unmanned aerial vehicle (UAV) airbag.

[0017] Figure 2 This is a flowchart of the process for extracting geometric constraint parameters for drone airbag deployment;

[0018] Figure 3 This is a flowchart of the iterative optimization process for the airbag's outer curved surface based on the projected area of ​​the assembly gap;

[0019] Figure 4 It is the iterative convergence curve of the fitting error of the fuselage envelope surface;

[0020] Figure 5 This is a diagram showing the angle distribution of the normal vectors between adjacent grid patches in the airbag installation area;

[0021] Figure 6 This is a distribution diagram of the sampling point spacing on the boundary contour of the main deployment surface of the airbag;

[0022] Figure 7 It is the iterative convergence curve of the airbag shape assembly gap thickness value. Detailed Implementation

[0023] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0024] See Figure 1 This invention provides a BIM-based simulation design method for the shape of an airbag for a UAV. The overall implementation process of this method is as follows: A BIM spatial model of the UAV fuselage is obtained, and the set of structural feature points and the fuselage envelope surface are extracted from the BIM spatial model. Based on the set of structural feature points, the airbag installation area is divided on the fuselage envelope surface, and the distribution of surface normal vectors within this area is obtained. Using the surface normal vector distribution and a preset UAV flight attitude angle, geometric constraint parameters for the airbag deployment profile are generated. These geometric constraint parameters are input into the parametric modeling environment of the BIM software to construct the initial airbag shape surface, and the assembly gap distribution between the initial airbag shape surface and the fuselage envelope surface is obtained. The geometric constraint parameters are iteratively adjusted based on this assembly gap distribution until the assembly gap distribution meets the preset assembly tolerance, ultimately obtaining the target airbag shape surface.

[0025] In practical implementation, a BIM spatial model of the UAV fuselage is imported. The skeleton line structure, representing the fuselage structural axis, is extracted from the component semantic information of the BIM spatial model. The skeleton line structure is composed of multiple connected line segment units. All line segment units in the skeleton line structure are traversed. When two line segment units have overlapping endpoints in space, the spatial coordinates of the overlapping position are used as the intersection coordinates of the skeleton line structure. When an endpoint of a line segment unit has no endpoints of other line segment units within a preset neighborhood, the spatial coordinates of that endpoint are used as the endpoint coordinates of the skeleton line structure. All intersection coordinates and all endpoint coordinates of the skeleton line structure are collected to form the fuselage structural feature point set.

[0026] The triangular mesh data of the fuselage outer surface is read from the BIM spatial model. Each triangular facet in the mesh is analyzed, and the coordinates of two vertices of each facet are extracted. Duplicate vertex coordinates from the three facets are then merged to obtain the coordinates of all vertices of the fuselage outer surface triangular mesh. A nonlinear least-squares surface fitting algorithm is used to fit the coordinates of all vertices of the fuselage outer surface triangular mesh into a continuous parametric surface. This fitted continuous parametric surface is then used as the fuselage envelope surface. Surface fitting is achieved by minimizing the following error function:

[0027]

[0028] Where M represents the total number of vertices of the triangular mesh on the outer surface of the fuselage; This represents the three-dimensional coordinate vector of the m-th vertex in the triangular mesh on the outer surface of the fuselage. The parameter coordinates of the m-th vertex in the parameter space represent the coordinates of all vertices of the triangular mesh on the outer surface of the fuselage. These coordinates are obtained by performing planar parameter mapping on all vertex coordinates using the initial parameterization method. Indicates The surface shape control coefficient vector b is the input parametric surface in parametric coordinates. The three-dimensional coordinate vector of a point on the surface at that location; This represents the total error of the surface fitting. It is solved using an iterative optimization algorithm. Obtain the surface shape control coefficient vector b with a minimum value, and use the solved surface shape control coefficient vector b to determine the fuselage envelope surface.

[0029] See Figure 4 In the figure, the horizontal axis represents the number of iterations in the surface fitting process, ranging from 0 to 1000 steps, and the vertical axis represents the total error F(b) of the surface fitting, ranging from 0 to 6. The broken line curve shows the trend of the total error F(b) during the actual iteration process, and the red dashed line represents the fitting trend line based on the exponential decay model. As the number of iterations increases, the total error F(b) of the surface fitting decreases rapidly. In the initial stage, the error drops rapidly from about 5.1 to about 2, with the most significant decrease occurring within 200 steps. This indicates that the nonlinear least squares surface fitting algorithm continuously optimizes the approximation effect of the continuous parametric surface of the triangular mesh vertex coordinates on the fuselage outer surface. Subsequently, the error continues to decrease slowly. After the number of iterations reaches about 800 steps, the error approaches 0 and the change becomes stable, indicating that the fitted surface has approximated the spatial distribution of the original triangular mesh data on the fuselage outer surface quite well, and the fitting accuracy meets the expected requirements. The red exponential decay trend line is highly consistent with the actual error curve, indicating that the iterative decrease of the error conforms to the exponential decay law.

[0030] In practice, each feature point in the fuselage structural feature point set is projected onto the fuselage envelope surface along the normal direction of the local surface near that feature point. During the projection process, for the k-th feature point in the fuselage structural feature point set... Searching on the fuselage envelope surface and The nearest surface point ,Will Determined as Projected feature points on the fuselage envelope surface. Traverse all feature points in the fuselage structural feature point set to obtain the projected feature point set.

[0031] In some embodiments, the projection feature points located at the outermost edge of the projection feature point set are extracted. The extraction method is as follows: calculate the convex hull or outer envelope polygon of the projection feature point set, and use the projection feature points corresponding to the vertices of the convex hull or outer envelope polygon as boundary marker points. Connect the boundary marker points sequentially according to the connectivity order on the fuselage envelope surface to form a closed boundary line.

[0032] In practice, during region growing, the set of mesh patches traversed by the closed boundary line is used as the seed region. The seed region is determined as follows: for each mesh patch in the triangular mesh representation of the fuselage envelope surface, if the distance between any vertex of the mesh patch and the closed boundary line is less than a preset tolerance, and the mesh patch is located within the region enclosed by the closed boundary line, then that mesh patch is included in the seed region. The preset tolerance is set to half the average side length of the fuselage envelope surface mesh.

[0033] During the region growth process, an angle threshold is set. , The value is 15 degrees. Set the angle threshold. The 15-degree angle is chosen because when the angle between the normal vectors of two adjacent mesh patches exceeds 15 degrees, it indicates a significant curvature change in the fuselage envelope surface in adjacent regions. The airbag installation area should not cross such curvature abrupt boundary to ensure the airbag's fit and adaptability to the fuselage surface. Starting with each mesh patch in the seed region, examine adjacent mesh patches sharing an edge with the seed mesh patch, and calculate the angle β between the normal vector of the seed mesh patch and the normal vector of the adjacent mesh patch. The formula for calculating angle β is:

[0034]

[0035] in, Represents the unit normal vector of the seed mesh patch; Represents the unit normal vector of adjacent mesh patches; Represents the magnitude of the unit normal vector of the seed mesh patch; This represents the magnitude of the unit normal vector of adjacent mesh patches; This represents the inverse cosine function. When the included angle β is less than a set angle threshold... When adjacent grid patches are located within the area enclosed by the closed boundary line, the adjacent grid patches are marked as new seed grid patches and added to the growth region. The above expansion process is repeated until all grid patches within the closed boundary line that meet the growth conditions are marked, forming a continuous region composed of marked grid patches. This continuous region is the airbag installation area.

[0036] In practice, for each mesh surface within the airbag installation area, the coordinates of its three vertices are obtained. The normal vector of the mesh surface is calculated using the right-hand rule and then normalized. All mesh surfaces within the airbag installation area are traversed, and the unit normal vector of each mesh surface is collected, resulting in a surface normal vector distribution. This surface normal vector distribution is a set containing the unit normal vectors of all mesh surfaces within the airbag installation area.

[0037] See Figure 5 The figure shows the distribution of the angle β between the normal vectors of adjacent grid panels within the airbag installation area. The horizontal axis represents the degree of the angle β between the normal vectors of adjacent grid panels, ranging from 0 to 45 degrees, and the vertical axis represents the number of grid panels within the corresponding angle range. The bar chart uses blue bars to represent the frequency distribution of the normal vector angle. The figure also uses red dashed lines to indicate the set angle threshold. This threshold is the critical included angle when dividing the airbag installation area in this embodiment.

[0038] As can be seen from the figure, the included angle of the normal vectors of most adjacent grid patches is concentrated between 0 and 15 degrees, and the number of grid patches reaches its peak in the included angle range of about 10 degrees. This indicates that the curvature variation between most grid patches in the airbag installation area is small, the surface is relatively smooth, and the included angle is less than the threshold. These conditions are conducive to maintaining a tight fit between the airbag and the fuselage's envelope surface. To the right of the 15-degree threshold line, there are gradually decreasing grid patches with a maximum included angle of approximately 40 degrees, indicating that there are significant curvature changes in some areas. The airbag installation area should not cross these curvature abrupt boundary boundaries to ensure the continuity and adaptability of the airbag's shape when it deploys.

[0039] In specific implementation, please refer to Figure 2 The system obtains the preset UAV flight attitude angles, which include three angular components: pitch, roll, and yaw. Based on the pitch, roll, and yaw angles, a rotation matrix is ​​constructed. The reference airbag deployment direction vector in the UAV's body coordinate system is multiplied by the rotation matrix to obtain the airbag deployment direction vector of the UAV in the preset flight attitude. The reference airbag deployment direction vector is defined in the UAV's body coordinate system as a unit vector perpendicular to the UAV's symmetrical plane and pointing outwards.

[0040] In some embodiments, the principal normal vector is extracted from the surface normal vector distribution. The extraction process is as follows: the arithmetic mean of the unit normal vectors of all grid patches in the surface normal vector distribution is taken to obtain the average normal vector. The average normal vector is then normalized, and the normalized average normal vector is taken as the principal normal vector in the surface normal vector distribution. The angle between the principal normal vector in the surface normal vector distribution and the airbag deployment direction vector is calculated. Each grid patch unit normal vector in the surface normal vector distribution is traversed, and the angle between each grid patch unit normal vector and the airbag deployment direction vector is calculated. The angle between each grid patch unit normal vector and the airbag deployment direction vector is compared with a preset angle threshold, which is 30 degrees. The preset angle threshold of 30 degrees is set as follows: when the angle between the grid patch unit normal vector and the airbag deployment direction vector is less than 30 degrees, it indicates that the fuselage surface area corresponding to the grid patch is basically perpendicular to the airbag deployment direction. When the airbag deploys in this area, it can obtain positive support along the normal direction. Such grid patches are marked as principal deployment patches. Mark all meshes with an angle less than 30 degrees that correspond to the unit normal vector as the main unfolded mesh.

[0041] In practice, the main unfolded facets located at the edges of the main unfolded facet set are extracted. The extraction method is as follows: traverse each main unfolded facet in the set, check adjacent mesh faces that share an edge with the main unfolded facet. When at least one adjacent mesh facet does not belong to the main unfolded facet set, mark the shared edge between the main unfolded facet and this adjacent mesh facet that does not belong to the main unfolded facet set as a boundary edge. Connect all boundary edges to form the boundary contour line of the main unfolded facet set.

[0042] In practice, the boundary contour line is discretely sampled to obtain a set of sampling points on the boundary contour line. The number of sampling points in the set is set to 200. The Euclidean distance between any two sampling points in the set is calculated to obtain a set of distance values. The maximum distance is selected from the set of distance values ​​as the maximum span value, and the minimum distance is selected from the set of distance values ​​as the minimum span value. The maximum and minimum span values ​​are obtained using the following distance calculation formula:

[0043]

[0044] in, Let represent the Euclidean distance between the p-th and q-th sampling points in the sampling point set; p represents the sampling point number, ranging from 1 to 200; q represents the sampling point number, ranging from 1 to 200, and... ; This represents the coordinates of the p-th sampling point in three-dimensional space; This represents the coordinates of the q-th sampling point in three-dimensional space. The maximum and minimum span values ​​are used as geometric constraint parameters.

[0045] See Figure 6 This figure is a frequency distribution histogram of the Euclidean distances between sampling points calculated from the discrete sampling point set of the boundary contour line of the main unfolded patch set. The horizontal axis represents the Euclidean distance between sampling points. The units are millimeters, ranging from approximately 30 mm to 520 mm; the vertical axis represents the frequency of sampling point pairs within the corresponding distance range. The green bar chart reflects the distribution characteristics of the distance between sampling points, showing an overall right-skewed distribution. The distances are concentrated in the range of approximately 50 mm to 250 mm, with the frequency peak at approximately 80 mm, indicating that the distances between sampling points are mostly clustered in a short to medium span range; the number of sampling point pairs with distances exceeding 350 mm decreases significantly, indicating that the maximum span of the main unfolded area boundary contour line is limited.

[0046] The maximum span value, represented by the orange dotted line in the figure, is approximately 520 mm, indicating the maximum Euclidean distance between any two sampling points in the sampling point set, thus determining the geometric constraint parameters of the major axis of the airbag cross-sectional profile. The minimum span value, represented by the purple dotted line, is approximately 40 mm, indicating the minimum distance between sampling points in the sampling point set, thus determining the geometric constraint parameters of the minor axis of the airbag cross-sectional profile. The determination of the maximum and minimum span values ​​is based on the extreme values ​​of the distance distribution, ensuring that the major and minor axes of the ellipse of the airbag cross-sectional profile fully cover the boundary range of the main deployment surface, satisfying the geometric constraint requirements of the airbag design.

[0047] In specific implementation, please refer to Figure 3 The process involves obtaining a set of gap thickness values ​​from the assembly gap distribution and comparing each gap thickness value in the set with a preset assembly tolerance. The preset assembly tolerance is a closed interval with a lower limit of 0 mm and an upper limit of 5 mm. The upper limit of the preset assembly tolerance is set to 5 mm because, according to the assembly process specifications for the UAV fuselage and airbag, the allowable gap between the airbag's outer curved surface and the fuselage's envelope curved surface must not exceed 5 mm. A gap exceeding 5 mm will cause airflow disturbance between the airbag and the fuselage during flight, affecting the aerodynamic shape. A gap less than 5 mm ensures that the airbag, after inflation, fits snugly against the fuselage's envelope curved surface within the assembly tolerance range. When a gap thickness value in the set exceeds 5 mm, the non-overlapping surface patch corresponding to the sampling point of that gap thickness value is marked as an interference region. The process iterates through all gap thickness values ​​in the set, merging all corresponding surface patches exceeding the preset assembly tolerance to form a set of interference regions.

[0048] In some embodiments, the vertex coordinates of each interference region in the interference region set are obtained, the arithmetic mean of all vertex coordinates is calculated, and the spatial point determined by the arithmetic mean is used as the center point coordinates of the interference region.

[0049] In practice, the projected area of ​​the interference region along the principal normal vector direction is calculated based on the coordinates of the center point of the interference region. The calculation method is as follows: using the principal normal vector direction as the projection direction and a plane perpendicular to the principal normal vector as the projection plane, all vertices of the interference region are projected onto the projection plane along the principal normal vector direction, resulting in a set of projected points of the interference region. A convex hull operation is performed on this set of projected points to obtain the projected convex polygon of the interference region. The area of ​​this projected convex polygon is calculated, and this area is used as the projected area of ​​the interference region along the principal normal vector direction. The projected areas of the main unfolded surface set along the principal normal vector direction are obtained in the same way, and the total projected area of ​​the main unfolded surface set is calculated.

[0050] In practical implementation, the scaling factor for proportional reduction is calculated. This scaling factor is equal to the ratio of the projected area of ​​the interference region along the principal normal vector direction to the total projected area of ​​the main unfolded surface. The formula for calculating proportional reduction is:

[0051]

[0052] in, R represents the updated maximum span value. max This represents the maximum span value before shrinking; This represents the projected area of ​​the interference region along the direction of the principal normal vector; This indicates the total projected area of ​​the main unfolded surface; This represents the ratio of the projected area of ​​the interference region to the total projected area of ​​the main unfolded surface. The updated maximum span value... The minimum span value is used as a new geometric constraint parameter.

[0053] In practice, the updated maximum and minimum span values ​​are re-input into the parametric modeling environment of the BIM software. Following the same process as constructing the initial airbag shape surface, an updated airbag shape surface is generated. The assembly gap distribution between the updated airbag shape surface and the fuselage envelope surface is recalculated, resulting in an updated set of gap thickness values. All gap thickness values ​​in the updated set are compared with the preset assembly tolerance to determine if all gap thickness values ​​fall within the preset assembly tolerance range. If any gap thickness value still exceeds the preset assembly tolerance, the interference area is marked again, the projected area is calculated, the maximum span value is updated, and the above process is repeated iteratively until all gap thickness values ​​fall within the preset assembly tolerance range. When all gap thickness values ​​fall within the preset assembly tolerance range, the airbag shape surface generated in the current iteration step is taken as the target airbag shape surface.

[0054] See Figure 7 In the figure, the horizontal axis represents the iterative steps in the BIM-based UAV airbag shape simulation design method, with values ​​ranging from 1 to 10. The vertical axis represents the assembly gap thickness value corresponding to each iterative step, in millimeters. The blue scatter dots represent the distribution of gap thickness values ​​at sampling points in each iterative step, and the dark red solid line and dots represent the median gap thickness value of each iterative step. The orange dashed line and parallel orange dotted line represent the upper limit of the preset assembly tolerance (5 mm) and the lower limit (0 mm), respectively.

[0055] As shown in the figure, the gap thickness values ​​are widely distributed in the initial iteration step (step 1), with a large number of gap thickness values ​​exceeding the preset assembly tolerance upper limit of 5 mm. The median gap thickness value is approximately 5.8 mm, significantly higher than the tolerance upper limit. With the increase of the number of iterations, the median gap thickness value shows a significant decreasing trend. From step 3 to step 6, the median gap thickness value gradually decreases from approximately 4.3 mm to approximately 3.0 mm. At the same time, the scatter distribution gradually converges, and the number of gap thickness points exceeding 5 mm gradually decreases. In the subsequent iteration steps (steps 7 to 10), the median gap thickness value further decreases to close to 1.2 mm. The gap thickness values ​​of all sampling points fall within the preset assembly tolerance closed interval [0, 5] mm, indicating that the assembly gap between the airbag outer surface and the fuselage envelope surface meets the design requirements.

[0056] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application.

Claims

1. A BIM-based simulation design method for the shape of an unmanned aerial vehicle (UAV) airbag, characterized in that, include: Obtain the BIM spatial model of the UAV fuselage, and extract the fuselage structural feature point set and fuselage envelope surface from the BIM spatial model; Based on the set of feature points of the fuselage structure, the airbag installation area is divided on the envelope surface of the fuselage, and the distribution of the surface normal vector within the airbag installation area is obtained. Based on the surface normal vector distribution and the preset UAV flight attitude angle, the geometric constraint parameters of the airbag deployment profile are generated; The geometric constraint parameters are input into the parametric modeling environment of the BIM software to construct the initial shape surface of the airbag and obtain the assembly gap distribution between the initial shape surface of the airbag and the fuselage envelope surface. Based on the assembly gap distribution, the geometric constraint parameters are adjusted until the assembly gap distribution meets the preset assembly tolerance, thereby obtaining the target airbag outer surface.

2. The simulation design method for the shape of an unmanned aerial vehicle (UAV) airbag based on BIM according to claim 1, characterized in that, Obtain the BIM spatial model of the UAV fuselage, and extract the fuselage structural feature point set and fuselage envelope surface from the BIM spatial model, including: In the BIM spatial model, the skeleton line structure of the UAV fuselage is identified, and the intersection coordinates and endpoint coordinates of the skeleton line structure are obtained and aggregated into a set of fuselage structural feature points. Based on the triangular mesh of the fuselage outer surface in the BIM spatial model, the coordinates of all vertices of the triangular mesh are extracted, and surface fitting is performed on the coordinates of all vertices to obtain the fuselage envelope surface.

3. The simulation design method for the shape of an unmanned aerial vehicle (UAV) airbag based on BIM according to claim 1, characterized in that, Based on the set of feature points of the fuselage structure, the airbag installation area is divided on the envelope surface of the fuselage, and the distribution of the surface normal vector within the airbag installation area is obtained, including: The set of fuselage structural feature points is projected onto the fuselage envelope surface to obtain the projected feature point set; Using the outermost projection feature point in the set of projection feature points as boundary marker points, region growth is performed on the fuselage envelope surface to obtain the airbag installation area; Traverse each grid patch within the airbag installation area, calculate the normal vector of each grid patch, and statistically obtain the surface normal vector distribution.

4. The simulation design method for the shape of a UAV airbag based on BIM according to claim 1, characterized in that, Based on the surface normal vector distribution and the preset UAV flight attitude angle, the geometric constraint parameters for the airbag deployment profile are generated, including: The vector of the airbag deployment direction is determined based on the preset drone flight attitude angle; Calculate the angle between the principal normal vector in the surface normal vector distribution and the vector of the airbag deployment direction, and mark the mesh surface corresponding to the surface normal vector with an angle less than a preset angle threshold as the principal deployment surface. Obtain the boundary contour line of the main unfolded surface, and measure the maximum and minimum span values ​​of the boundary contour line as geometric constraint parameters.

5. The simulation design method for the shape of a UAV airbag based on BIM according to claim 4, characterized in that, The geometric constraint parameters are input into the parametric modeling environment of the BIM software to construct the initial shape surface of the airbag, and the assembly gap distribution between the initial shape surface of the airbag and the fuselage envelope surface is obtained, including: The maximum and minimum span values ​​are input into the parametric modeling environment of the BIM software as the major and minor axis parameters of the airbag cross-sectional profile. Based on the major axis parameters and minor axis parameters, an airbag cross-sectional contour line is generated in the parametric modeling environment, and the airbag cross-sectional contour line is stretched along the direction of the principal normal vector to obtain the initial shape surface of the airbag. In the BIM software, the initial shape surface of the airbag and the envelope surface of the fuselage are subjected to Boolean difference operation to extract the set of gap thickness values ​​of the non-overlapping areas as the assembly gap distribution.

6. The simulation design method for the shape of a UAV airbag based on BIM according to claim 5, characterized in that, Based on the assembly gap distribution, the geometric constraint parameters are adjusted until the assembly gap distribution meets the preset assembly tolerance, thereby obtaining the target airbag outer surface, including: The set of gap thickness values ​​in the assembly gap distribution is compared with the preset assembly tolerance, and the area where the gap thickness value exceeds the preset assembly tolerance is marked as the interference area; Obtain the coordinates of the center point of the interference region, calculate the projected area of ​​the interference region in the direction of the principal normal vector based on the center point coordinates, and proportionally reduce the maximum span value based on the projected area to obtain the updated maximum span value. The updated maximum span value and the minimum span value are used as new geometric constraint parameters to reconstruct the airbag shape surface until all gap thickness values ​​fall within the preset assembly tolerance range, thus obtaining the target airbag shape surface.

7. The simulation design method for the shape of a UAV airbag based on BIM according to claim 3, characterized in that, Using the outermost projection feature point in the projection feature point set as the boundary marker point, when performing region growth on the fuselage envelope surface, the closed boundary line enclosed by the boundary marker point is used as the seed region, and the growth condition is that the angle between the normal vectors of adjacent mesh patches is less than a set angle threshold. The region gradually expands outward until it reaches the closed boundary line.

8. The simulation design method for the shape of a UAV airbag based on BIM according to claim 4, characterized in that, The maximum and minimum span values ​​of the boundary contour line are obtained by calculating the maximum and minimum Euclidean distances between all pairs of points on the boundary contour line of the main unfolded patch, respectively.

9. The simulation design method for the shape of an unmanned aerial vehicle (UAV) airbag based on BIM according to claim 5, characterized in that, When stretching the airbag cross-sectional profile along the principal normal vector direction, the stretching length is dynamically determined based on the outward expansion safety distance of the fuselage envelope surface in the principal normal vector direction and the maximum flight speed of the UAV.

10. The BIM-based simulation design method for the shape of an unmanned aerial vehicle (UAV) airbag according to claim 6, characterized in that, When the maximum span value is reduced proportionally according to the projected area, the proportionality coefficient of the proportional reduction is the ratio of the projected area of ​​the interference region to the total projected area of ​​the main unfolded surface.

Images