A protrusion inspection method and computer medium for a vehicle passenger compartment

Abstract

The application provides a protrusion checking method and computer medium for a vehicle passenger cabin, and the method comprises the following steps: acquiring initial data containing a reference point position and a three-dimensional geometric model to be measured; dividing all surfaces in the model into collision areas and exempt areas according to the three-dimensional geometric model to be measured and the initial data; configuring a first collision model in the collision areas, using the first collision model to determine the contactability of the surface sampling points, and screening out effective contact points that can be contacted; calculating the curvature radius of each effective contact point on the surface of the model; comparing each curvature radius with a preset threshold value, and outputting a checking result according to the comparison result. The application accurately locks the dangerous contact points through collision interference simulation, converts manual point-by-point blind measurement into automatic compliance screening, effectively eliminates subjective errors, and significantly improves the accuracy and objective consistency of protrusion checking.

Description

Technical Field

[0001] This invention relates to the field of automotive design, and more particularly to a method and computer medium for inspecting protrusions in a vehicle passenger compartment. Background Technology

[0002] In the field of automotive engineering and safety design, the inspection of protrusions inside the vehicle's passenger compartment is an indispensable core component of the passive safety system. When a vehicle experiences a sudden collision or rollover, the occupants' torsos and heads are highly susceptible to severe secondary impacts with various interior trim components due to inertia. Sharp structures or irregular edges on the interior surfaces significantly increase the risk of serious injuries such as scratches and punctures. Therefore, strictly controlling the surface geometry within the passenger compartment and ensuring that the height and rounded edges of local protrusions comply with relevant safety regulations and standards (e.g., GB1152-2009 "Passenger Car Interior Protrusions") is a mandatory safety verification benchmark during the automotive interior design phase. This step not only directly relates to the compliance of the final vehicle product but also serves as a crucial physical barrier to protect the lives of passengers.

[0003] However, current methods for inspecting protrusions inside vehicles primarily rely on manual experience. Even in a digital environment incorporating 3D geometric models, design engineers still need to manually define complex verification areas and rely on subjective identification and manual point selection to systematically check for potential hazardous protrusions. This workflow is not only inefficient but also prone to missed inspections and measurement errors due to blind spots when dealing with interior components with varied shapes and complex structures. More importantly, automotive design and development is a continuous iterative process. Once the underlying surface model undergoes even minor modifications or updates, the entire complex point-by-point inspection process needs to be manually re-executed. Existing technologies lack a mechanism that can adapt to changes in spatial boundaries and automatically simulate physical collision interference from a global perspective to locate risk characteristics, resulting in a lengthy overall development cycle and difficulty in effectively ensuring compliance and consistency.

[0004] Therefore, in order to overcome the shortcomings of traditional manual verification methods, such as strong subjectivity, low efficiency, and difficulty in coping with high-frequency design iterations, this invention provides a method for efficient, accurate, and fully automated protrusion inspection in a three-dimensional model of a vehicle's passenger compartment. Summary of the Invention

[0005] In order to overcome the above-mentioned technical defects, the purpose of this invention is to provide a method for inspecting protrusions in a vehicle passenger compartment and a computer medium.

[0006] This invention discloses a method for inspecting protrusions in a vehicle passenger compartment, comprising: Acquire initial data, including the location of the reference point; Obtain the three-dimensional geometric model to be tested. Based on the three-dimensional geometric model to be tested and the initial data, divide all surfaces in the three-dimensional geometric model to be tested into collision areas and exemption areas. A first collision model is configured within the collision area, and the first collision model is used to determine the accessibility of surface sampling points within the collision area in order to screen out effective contact points that can make contact with the first collision model. Calculate the radius of curvature of each effective contact point on the surface of the three-dimensional geometric model to be measured; The curvature radius of each effective contact point is compared with a preset threshold, and the inspection results of protrusions in the vehicle passenger compartment are output based on the comparison results.

[0007] Preferably, a three-dimensional geometric model to be tested is obtained. Based on the three-dimensional geometric model to be tested and the initial data, all surfaces in the three-dimensional geometric model to be tested are divided into collision regions and exemption regions, including: The impact zone includes the head impact zone and the knee impact zone; Obtain the height coordinates of the reference point, establish a rotation axis with the reference point as the center, and construct multiple slice planes that rotate along the rotation axis within a preset angle range; The points where each slice plane intersects with the surface of the three-dimensional geometric model to be measured are identified as collision points, and the upper and lower boundary lines of the surface of the three-dimensional geometric model to be measured are determined based on the collision points. Connect the collision points on each slice plane whose height coordinates are greater than or equal to the upper and lower boundary lines in sequence to form a closed curve, thereby enclosing and generating the head collision area. On each slice plane, the collision points with height coordinates less than the upper and lower boundaries are connected sequentially to form a closed curve, which encloses and generates the knee collision area.

[0008] Preferably, dividing all surfaces in the three-dimensional geometric model to be tested into collision regions and exemption regions further includes: Obtain the 3D boundary contour of the preset exempted component in the 3D geometric model to be tested; Based on the three-dimensional boundary contour and the preset extension offset, it extends outward to form a spatial repulsion envelope; The spatial repulsion envelope is projected along the projection direction onto the surface of the three-dimensional geometric model to be tested to obtain the preliminary exemption region. The portion of the initial exemption area whose height coordinates are less than the horizontal plane of the lowest point of the exempted component is removed to obtain the exemption area.

[0009] Preferably, a first collision model is configured within the collision area, and the first collision model is used to determine the accessibility of surface sampling points within the collision area, so as to screen out effective contact points that can make contact with the first collision model, including: The first collision model includes a collision ball used to simulate a head; Based on the preset density parameters, obtain each surface sampling point in the collision area and extract the normal vector of each surface sampling point; Offset along the direction of the normal vector by a preset distance equal to the radius of the collision sphere to determine the center position of the collision sphere corresponding to each surface sampling point; Starting from the center of the sphere, find the other surface sampling point closest to the center of the sphere; Calculate the distance between the nearest other surface sampling point and the center of the sphere; If the distance is greater than or equal to the radius of the colliding sphere, the corresponding surface sampling point is determined to be a valid contact point.

[0010] Preferably, configuring a first collision model within the collision area and using the first collision model to determine the accessibility of surface sampling points within the collision area to screen out effective contact points capable of contacting the first collision model further includes: The first collision model also includes a triangular prism for simulating the knee. Based on preset density parameters, obtain each surface sampling point in the knee collision area and extract the normal vector of each surface sampling point. Offset along the direction of the normal vector by a value equal to the preset interference distance of the triangular prism to determine the reference center position of each surface sampling point corresponding to the triangular prism; Starting from the reference center location, find the other surface sampling points that are closest to the reference center location; Calculate the distances between the nearest other surface sampling points and the reference center location; If the distance is greater than or equal to the preset interference distance of the triangular prism, the corresponding surface sampling point is determined to be a valid contact point.

[0011] Preferably, calculating the radius of curvature of each effective contact point on the surface of the three-dimensional geometric model to be measured includes: Discretize the cross-sectional line containing the effective contact points into multiple discrete cross-sectional points; Move the first collision model along the preset reference line direction until the first collision model is tangent to the surface of the protrusion; Obtain the position of the first collision model in the tangent state, and calculate the difference in coordinate information between the highest point and the lowest point of the protrusion as the protrusion amount of the protrusion; Determine the geometric properties of the highest or lowest point. If the corresponding point is located on a curve segment, calculate the radius of curvature at that point; if the corresponding point is a corner or endpoint, determine that the radius of curvature is zero.

[0012] Preferably, when the three-dimensional geometric model to be tested includes grid-like parts, the protrusion inspection method further includes: A second collision model is configured at grille-type parts; Divide the grille-type parts into protruding parts and peripheral parts, and extract the corresponding cross-sectional lines; Discrete sampling is performed on the cross-sectional line to obtain a discrete point set. For the discrete points of the protruding parts, the nearest neighbor points are searched in the discrete point set of the surrounding parts to form discrete point pairs. Select valid discrete point pairs whose normal vector angle satisfies the preset perpendicular condition; Based on the distance between effective discrete point pairs, the minimum gap and minimum sheet thickness of the grid-type parts are calculated, and the inspection results for the grid-type parts are output in combination with the preset compliance requirements.

[0013] Preferably, when the three-dimensional geometric model to be tested includes knob-like parts, the protrusion inspection method further includes: A third collision model is configured on knob-type parts. The third collision model is a ball head model with a preset size sliding pressure head. Extract the rotation axis and the surface to be inspected from knob-type parts; The third collision model is controlled to rotate and scan along the rotation axis at preset angle intervals, and collision interference detection is performed during the scanning process; Record the interference-free limit position of the third collision model during the rotational scanning process, and extract the maximum protrusion of the surface to be inspected at the limit position and the cross-sectional area at the preset offset position, so as to output the inspection results for knob-type parts in combination with the preset compliance requirements.

[0014] Preferably, the radius of curvature of each effective contact point is compared with a preset threshold, and the inspection results of protrusions in the vehicle passenger compartment are output based on the comparison results, including: The preset thresholds include a first threshold and a second threshold, and the first threshold is less than the second threshold; If the radius of curvature is less than the first threshold, the corresponding valid contact point is determined to be unqualified. If the radius of curvature is greater than or equal to the first threshold and less than or equal to the second threshold, the corresponding effective contact point is determined to be in a dangerous state. If the radius of curvature is greater than the second threshold, the corresponding effective contact point is deemed qualified. Based on the classification of non-compliance, hazardous conditions, and compliance, the inspection results of protrusions in the vehicle's passenger compartment are output.

[0015] A second aspect of this application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements any of the foregoing methods for inspecting protrusions in a vehicle passenger compartment.

[0016] Compared with existing technologies, the above technical solution has the following advantages: 1. This application provides a method for inspecting protrusions in a vehicle passenger compartment. In a digital environment, the method acquires the location of reference points and the three-dimensional geometric model to be tested, dividing it into collision areas and exemption areas. The collision areas are further divided into head collision areas and knee collision areas. A first collision model is then configured within each collision area to determine accessibility, thereby identifying valid contact points and calculating the radius of curvature. Finally, the inspection results are output based on a preset threshold comparison. This method utilizes discrete sampling and model interference logic to realistically simulate the possible collision contact states of the passenger's torso and head. It not only directly replaces the tedious process of manual area division and fixed-point measurement in existing technologies, avoiding errors caused by subjective experience, but also enables automated batch verification as the three-dimensional geometric model to be tested is updated, significantly improving inspection efficiency and the objectivity and consistency of results. 2. Specifically, regarding the precise definition of the inspection scope, this solution simulates the motion envelope of the occupant's relative displacement from a spatial geometric perspective. The upper and lower boundaries are determined by the height coordinates of a reference point, and head and knee collision boundaries are generated by slicing scans. This automatically adapts to various complex vehicle internal structures and accurately captures the true contact boundaries under complex curved surface shapes. Simultaneously, the solution establishes a three-dimensional spatial buffer for specific regulatory-exempt components. Through dimensionality reduction projection processing of the spatial exclusion envelope, areas unsuitable for conventional inspection are precisely stripped away and filtered using the lowest point horizontal plane of the exempted component. This mechanism avoids redundant calculations and false alarms in non-hazardous areas, ensuring the rigor of the algorithm's inspection logic and the rationality of system computing power scheduling. 3. Regarding specific curvature calculations, this solution precisely pinpoints the physical limit of force on local protrusions by combining inverse geometric deduction with dynamic simulation tangency. The algorithm uses a collision sphere and a triangular prism as collision models for the head and knee regions, respectively. It locates the center of the simulated sphere or reference center based on normal vector offset, and intelligently filters out narrow gaps that are physically inaccessible to occupants through spatial distance determination, ensuring that calculations only target exposure points with real collision risks. Subsequently, it quantifies the actual protrusion amount by extracting the coordinate difference within the tangent interval and directly intervenes in the underlying geometric topology properties, distinguishing between smooth curve segments and sharp endpoints. Finally, the solution performs a safety assessment of each protrusion based on its specific curvature radius, constructing a tiered, progressive judgment logic using multi-level thresholds. This not only directly intercepts obviously non-compliant designs but also provides clear priority quantitative guidance for subsequent model curvature optimization. 4. Addressing the unique topology of vehicle interiors, this solution introduces a customized scanning interference strategy and achieves stable system-level deployment. For densely structured, narrow-gap grille-like parts, the algorithm utilizes the orthogonality of normal vectors to accurately locate parallel opposing surfaces. It then calculates the actual physical gaps and material thicknesses through spatial distance, resolving the problem of interference misjudgments caused by dense parallel structures. For knob-like parts with operational strokes, the system virtually simulates a moving mechanism mounted on a specific axis for dynamic angle scanning, exhaustively capturing the maximum intrusion amount under any operating posture. Finally, the above methods are implemented on physical devices such as processors and computer-readable storage media, providing a standardized, high-concurrency, automated safety verification platform for the automotive R&D system. Attached Figure Description

[0017] Figure 1 A flowchart illustrating the method for inspecting protrusions in a vehicle passenger compartment provided in this application. Figure 2 A flowchart illustrating the area division in the method for inspecting protrusions in a vehicle passenger compartment provided in this application. Figure 3 A schematic diagram of the inspection process for grille-type parts in the method for inspecting protrusions in a vehicle passenger compartment provided in this application; Figure 4 This is a schematic diagram of the inspection process for knob-type parts in the method for inspecting protrusions in the passenger compartment of a vehicle provided in this application. Detailed Implementation

[0018] The advantages of the present invention will be further illustrated below with reference to the accompanying drawings and specific embodiments.

[0019] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numerals in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this disclosure. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this disclosure as detailed in the appended claims.

[0020] The terminology used in this disclosure is for the purpose of describing particular embodiments only and is not intended to be limiting of the disclosure. The singular forms “a,” “the,” and “the” as used in this disclosure and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used herein refers to and includes any and all possible combinations of one or more of the associated listed items.

[0021] It should be understood that although the terms first, second, third, etc., may be used in this disclosure to describe various information, such information should not be limited to these terms. These terms are used only to distinguish information of the same type from one another. For example, without departing from the scope of this disclosure, first information may also be referred to as second information, and similarly, second information may also be referred to as first information. Depending on the context, the word "if" as used herein may be interpreted as "when," "when," or "in response to determination."

[0022] In the description of this invention, it should be understood that the terms "longitudinal", "lateral", "up", "down", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this invention.

[0023] In the description of this invention, unless otherwise specified and limited, it should be noted that the terms "installation", "connection" and "linking" should be interpreted broadly. For example, they can refer to mechanical or electrical connections, or internal connections between two components. They can be direct connections or indirect connections through an intermediate medium. Those skilled in the art can understand the specific meaning of the above terms according to the specific circumstances.

[0024] In the following description, suffixes such as "module," "part," or "unit" used to denote elements are used only for the convenience of the description of the invention and have no specific meaning in themselves. Therefore, "module" and "part" can be used interchangeably.

[0025] Please see Figure 1 , Figure 1 This is a flowchart illustrating the method for inspecting protrusions in the passenger compartment of a vehicle, as provided in this application.

[0026] like Figure 1 As shown, this invention discloses a method for inspecting protrusions in a vehicle passenger compartment, comprising: Acquire initial data, including the location of the reference point; Obtain the three-dimensional geometric model to be tested. Based on the three-dimensional geometric model to be tested and the initial data, divide all surfaces in the three-dimensional geometric model to be tested into collision areas and exemption areas. A first collision model is configured within the collision area, and the first collision model is used to determine the accessibility of surface sampling points within the collision area in order to screen out effective contact points that can make contact with the first collision model. Calculate the radius of curvature of each effective contact point on the surface of the three-dimensional geometric model to be measured; The curvature radius of each effective contact point is compared with a preset threshold, and the inspection results of protrusions in the vehicle passenger compartment are output based on the comparison results.

[0027] Specifically, the protrusion inspection method provided in this application acquires basic data of the vehicle's passenger compartment, including reference point positions. Based on these reference point positions and the three-dimensional geometric model of the passenger compartment surface to be tested, the area is divided into collision zones and exemption zones. A first collision model is then configured within the collision zone, and contactability is determined using surface sampling points. Valid contact points are then selected, and the radius of curvature is calculated. Finally, the radius of curvature is compared with a preset threshold to output the inspection result. This solution transforms the physical collision testing requirements of existing technologies into geometric spatial calculations in a digital environment. It utilizes discrete sampling and model interference logic to realistically simulate the potential collision contact states of the occupant's torso, head, and knees. This not only directly replaces the tedious process of traditional manual area division and fixed-point measurement, avoiding errors caused by subjective experience, but also enables automated batch verification as the three-dimensional geometric model to be tested is updated, significantly improving inspection efficiency and the objectivity and consistency of results.

[0028] The above is an explanation of the basic concept of this application. The specific implementation methods of each step will be explained below with reference to the accompanying drawings.

[0029] Firstly, there are no restrictions on the specific implementation method of regional division.

[0030] Please see Figure 2 , Figure 2 A flowchart illustrating the area division in the method for inspecting protrusions in a vehicle passenger compartment provided in this application. like Figure 2 As shown, in one possible implementation, a three-dimensional geometric model to be tested is obtained. Based on the three-dimensional geometric model to be tested and the initial data, all surfaces in the three-dimensional geometric model to be tested are divided into collision regions and exemption regions, including: The impact zone includes the head impact zone and the knee impact zone; Obtain the height coordinates of the reference point, establish a rotation axis with the reference point as the center, and construct multiple slice planes that rotate along the rotation axis within a preset angle range; The points where each slice plane intersects with the surface of the three-dimensional geometric model to be measured are identified as collision points, and the upper and lower boundary lines of the surface of the three-dimensional geometric model to be measured are determined based on the collision points. Connect the collision points on each slice plane whose height coordinates are greater than or equal to the upper and lower boundary lines in sequence to form a closed curve, thereby enclosing and generating the head collision area. On each slice plane, the collision points with height coordinates less than the upper and lower boundaries are connected sequentially to form a closed curve, which encloses and generates the knee collision area.

[0031] Specifically, regarding the specific division of the collision area and the exemption area, this application obtains the height coordinate (z-axis coordinate) from the coordinate information of a reference point (exemplarily, the reference point can be the H point, hip point). Areas with height coordinates less than the reference point are defined as exemption areas. A rotation axis is established with the reference point as the center, and multiple slice planes rotating along the rotation axis are constructed within a preset angle range. For example, a collision ball with a diameter of 165mm can be used to simulate a human head and rotate along multiple preset angles to obtain slice planes. The collision points on the surface of the three-dimensional geometric model that the collision ball may collide with are calculated. Based on these collision points, the upper and lower dividing lines of the three-dimensional geometric model to be tested are determined (those skilled in the art will understand that the upper and lower dividing lines here are existing technologies and will not be elaborated on here. For example, they can be formulated according to GB1152-2009 "Internal Protrusions of Passenger Cars"). For areas with height coordinates greater than or equal to the upper and lower dividing lines, a complete head collision area is generated; for areas with height coordinates less than the upper and lower dividing lines, they are divided into knee collision areas, thereby achieving a refined division of the entire upper and lower space of the dashboard.

[0032] This process simulates the motion envelope of an occupant relative to a reference point from a spatial geometry perspective. Through slicing and distance optimization algorithms, it accurately captures the true contact boundaries under complex curved surface shapes. This approach automatically adapts to various intricate vehicle interior structures, precisely mapping the abstract test area to a closed solid inspection surface in three-dimensional space. Furthermore, it accurately distinguishes the two major collision regions with distinctly different force characteristics—the head and knee—using upper and lower boundary lines, ensuring the accuracy of subsequent zone interference determination.

[0033] Furthermore, in one possible implementation, dividing all surfaces in the 3D geometric model under test into collision regions and exemption regions also includes: Obtain the 3D boundary contour of the preset exempted component in the 3D geometric model to be tested; Based on the three-dimensional boundary contour and the preset extension offset, it extends outward to form a spatial repulsion envelope; The spatial repulsion envelope is projected along the projection direction onto the surface of the three-dimensional geometric model to be tested to obtain the preliminary exemption region. The portion of the initial exemption area whose height coordinates are less than the horizontal plane of the lowest point of the exempted component is removed to obtain the exemption area.

[0034] Understandably, during a collision, certain components and their extensions are inaccessible to the impact ball. Examples include the steering wheel and its rear area, a small area at the angle between the windshield and the A-pillar, and a portion of the area behind the dashboard. Therefore, for these components, the method first extracts their three-dimensional boundary contours. Then, based on regulations or safety margins, a preset expansion offset is set (for example, 127mm for the steering wheel and 82.5mm for the pillar structure, the radius of the impact ball), thus expanding outward to construct a three-dimensional spatial exclusion envelope. Subsequently, this three-dimensional envelope is projected onto the surface of a three-dimensional geometric model, such as the main dashboard, along a specified projection direction (e.g., the positive direction parallel to the vehicle's X-axis), transforming the three-dimensional safety boundary into a two-dimensional preliminary exemption area. Finally, by comparing height coordinates, the system automatically removes invalid mesh points located below the horizontal plane of the lowest point of the exempted component and failing to meet the minimum head impact height requirement, thereby accurately locking the final exemption area and avoiding erroneous interference judgments around such safety components.

[0035] Based on this method, this application establishes a three-dimensional spatial buffer for specific safety components or components that are permitted to be excluded by regulations. Through dimensionality reduction processing using envelope projection, the interference range of these three-dimensional components on the surrounding environment is accurately imprinted onto the three-dimensional geometric surface to be tested, forming a preliminary exemption zone. Simultaneously, validity filtering is performed using height coordinates. This step can accurately isolate areas unsuitable for routine inspection, avoiding redundant calculations and false alarms in non-hazardous areas, and ensuring the rigor of the algorithm's inspection logic and the rationality of system computing power scheduling.

[0036] After completing the above steps, a more rigorous division of the collision area and exemption area can be obtained, and then the subsequent curvature calculation stage can be started.

[0037] Those skilled in the art will understand that there is no limitation on the specific method of curvature calculation.

[0038] like Figure 1 As shown, for the head collision area, in one possible implementation, a first collision model is configured within the collision area, and the first collision model is used to determine the contactability of surface sampling points within the collision area, so as to screen out the effective contact points that can contact the first collision model, including: The first collision model includes a collision ball used to simulate a head; Based on the preset density parameters, obtain each surface sampling point in the collision area and extract the normal vector of each surface sampling point; Offset along the direction of the normal vector by a preset distance equal to the radius of the collision sphere to determine the center position of the collision sphere corresponding to each surface sampling point; Starting from the center of the sphere, find the other surface sampling point closest to the center of the sphere; Calculate the distance between the nearest other surface sampling point and the center of the sphere; If the distance is greater than or equal to the radius of the colliding sphere, the corresponding surface sampling point is determined to be a valid contact point.

[0039] Understandably, before calculating curvature, it is necessary to find effective curvature calculation points. Therefore, this application adopts the measurement principle of inverse geometric interference. The first collision model is also based on the aforementioned collision sphere with a diameter of 165mm to simulate a human head. The system first breaks down the continuous collision area surface into dense surface sampling points based on preset density parameters (for example, the chord height error of the surface mesh discretization can be set to 0.1mm, and the sampling step size to 5mm), and calculates the spatial normal vector of each surface sampling point. In order to verify whether a specific surface sampling point can be actually touched by the occupant's head, the algorithm pushes the surface sampling point outward vertically along its own normal vector by a preset distance equal to the radius of the first collision model (i.e., 82.5mm), thereby simulating the position of the sphere's center when the collision sphere just touches that point. Then, taking the sphere's center position as the starting point, it searches for other neighboring surface sampling points closest to the sphere's center in the surrounding point cloud data. If the calculated nearest distance is greater than or equal to 82.5mm, it means that inside and on the surface of this virtual sphere, apart from the target contact point, no other vehicle body digital model structure will physically interfere with the sphere. In other words, the collision ball can make contact with the point without obstruction, and the surface sampling point is determined to be a valid contact point. Conversely, if the nearest distance is less than the radius, it means that the collision ball will be blocked by the surrounding more prominent structure before it touches the point. This point is an unreachable dead angle and will be filtered out in the subsequent curvature verification.

[0040] Using the above judgment methods, the vehicle passenger compartment protrusion inspection method provided in this application can intelligently filter out narrow gaps that are physically inaccessible to passengers, ensuring that subsequent calculations only target exposure points with real collision risks.

[0041] Similarly, this application also provides a solution for the knee impact area.

[0042] In one possible implementation, a first collision model is configured within the collision region, and the first collision model is used to determine the accessibility of surface sampling points within the collision region. This process, which filters out effective contact points capable of contacting the first collision model, further includes: The first collision model also includes a triangular prism for simulating the knee. Based on preset density parameters, obtain each surface sampling point in the knee collision area and extract the normal vector of each surface sampling point. Offset along the direction of the normal vector by a value equal to the preset interference distance of the triangular prism to determine the reference center position of the triangular prism corresponding to each surface sampling point; Starting from the reference center location, find the other surface sampling points that are closest to the reference center location; Calculate the distances between the nearest other surface sampling points and the reference center location; If the distance is greater than or equal to the preset interference distance of the triangular prism, the corresponding surface sampling point is determined to be a valid contact point.

[0043] It is understandable that the trajectory and contact surface of the occupant's knee during a collision differ significantly from those of the head. Below the lower boundary line, a standard spherical model is insufficient to accurately reflect the injury characteristics. Therefore, this application uses a triangular prism as the first collision model in the knee collision area. This prism corresponds in physical parameters to the knee inspection tool required by regulations, such as a specific triangular prism shape with a width of 120mm, a vertices of 60 degrees, and rounded edges of R60mm. Since the triangular prism is not a standard sphere, the algorithm simulates its spatial repulsion envelope using a preset interference distance (i.e., the distance from the tangent point of the prism's edge to its internal reference geometric center). The system also moves the surface sampling points of the knee collision area vertically outward along the normal vector by this preset interference distance to locate the reference center position of the triangular prism and searches for the nearest point in the surrounding area. If the nearest distance is greater than or equal to this interference distance, it indicates that the triangular prism tool will not be blocked or jammed by other protruding structures under the dashboard when contacting the target point, thus confirming that this point is a valid contact point where a real impact to the knee may occur.

[0044] By introducing a triangular prism interference judgment model with a shape that conforms to the biomechanical characteristics of the knee in the lower half of the region, this solution effectively compensates for the shortcomings of a single spherical model, which is prone to missed or false judgments in the narrow lower space. This design ensures that the collision risk assessment of vulnerable parts such as the knee and lower leg strictly aligns with national occupant protection safety regulations, and greatly improves the rigor and compliance of the safety inspection of protrusions throughout the cabin.

[0045] Once the effective contact point is obtained, the radius of curvature can be calculated.

[0046] In one possible implementation, the radius of curvature of each effective contact point on the surface of the three-dimensional geometric model to be measured is calculated, including: Discretize the cross-sectional line containing the effective contact points into multiple discrete cross-sectional points; Move the first collision model along the preset reference line direction until the first collision model is tangent to the surface of the protrusion; Obtain the position of the first collision model in the tangent state, and calculate the difference in coordinate information between the highest point and the lowest point of the protrusion as the protrusion amount of the protrusion; Determine the geometric properties of the highest or lowest point. If the corresponding point is located on a curve segment, calculate the radius of curvature at that point; if the corresponding point is a corner or endpoint, determine that the radius of curvature is zero.

[0047] In the specific curvature calculation stage, for the selected exposure risk points, this method also uses the positioning action of the first collision model (such as a collision ball for the head, or a triangular prism for the knee) to simultaneously calculate the relative height difference and sharpness of the local surface. The method first slices along a specific direction, cutting the complex curved surface containing effective contact points into two-dimensional cross-sectional lines, and further refines it into multiple discrete cross-sectional points, simplifying the three-dimensional calculation into a planar geometric determination.

[0048] Subsequently, based on the first collision model, the model is simulated to slide or approach along a preset reference line until the solid surface of the model (such as the spherical surface of a collision ball or the side surface of a triangular prism) and the two side surfaces of the local structure containing the effective contact point (i.e., the "protrusion" referred to in this application) reach a physical limit state of close tangency. Within the tangent-locked geometric envelope, the system automatically searches for and captures the highest point (peak) and the lowest point (reference surface or root of part) on the protrusion. By calculating the difference in coordinate information between the two, the protrusion amount of the protrusion is accurately calculated, thereby assessing the absolute degree of its intrusion into the passenger compartment.

[0049] Simultaneously, to verify the sharpness of the point, the algorithm checks the topological geometry of the highest or lowest point in situ: if the location is on a smoothly transitioning curve segment, the algorithm uses calculus to calculate the radius of its approximate circle, which serves as the corresponding radius of curvature; if the location is itself a bend (corner point) or an overhanging edge (endpoint) of a component, since it physically appears as a sharp protrusion that cannot form a rounded corner, the algorithm will no longer perform conventional curvature fitting, but will directly force the radius of curvature to be zero, thus using it as a punitive conclusion to classify it as a high-risk unqualified feature. This processing method can accurately extract the relative size of local features and directly negate sharp edges that may cause cutting damage, faithfully restoring the true assessment logic of damage intensity. The above is a complete explanation of the analytical method for the first collision model provided in this application. Those skilled in the art will understand that the protrusion inspection method provided in this application may also include more computational models to perform more targeted inspections of different types of protrusions.

[0050] Please see Figure 3 , Figure 3 This is a schematic diagram of the inspection process for grille-type parts in the method for inspecting protrusions in a vehicle passenger compartment provided in this application.

[0051] For example, such as Figure 3As shown, in one possible implementation, when the 3D geometric model to be tested includes grid-like parts, the protrusion inspection method further includes: A second collision model is configured at grille-type parts; Divide the grille-type parts into protruding parts and peripheral parts, and extract the corresponding cross-sectional lines; Discrete sampling is performed on the cross-sectional line to obtain a discrete point set. For the discrete points of the protruding parts, the nearest neighbor points are searched in the discrete point set of the surrounding parts to form discrete point pairs. Select valid discrete point pairs whose normal vector angle satisfies the preset perpendicular condition; Based on the distance between effective discrete point pairs, the minimum gap and minimum sheet thickness of the grid-type parts are calculated, and the inspection results for the grid-type parts are output in combination with the preset compliance requirements.

[0052] For grille-like parts such as air conditioning vents and grille covers, which feature high-density parallel blades or dense gaps, conventional omnidirectional surface rolling collision ball detection is prone to interference errors due to dense blind zones caused by their unique spatial morphology. Therefore, this application introduces a specialized second collision model for directional verification. In actual execution, the second collision model performs specific geometric measurements of the characteristic step length and thickness for the special gaps of the grille. The system first identifies the grille body as a protruding part, marks the frame containing the grille or adjacent grilles as peripheral parts, and extracts the corresponding cross-sectional lines in the key load-bearing directions and performs discrete sampling. To calculate the true opening of each blade gap, the algorithm performs a nearest neighbor search on each discrete point on the protruding part, establishing spatial discrete point pairs within the discrete point set of the peripheral parts. Since the opposite sides of the grille gaps are usually parallel or quasi-parallel, the system calculates the angle between the normal vectors of the discrete point pairs and filters out effective discrete point pairs with an angle close to 90 degrees (i.e., satisfying the preset perpendicularity condition, meaning the normals of the two points are aligned or orthogonal). Ultimately, by directly extracting the true spatial distance between these effective discrete point pairs, the minimum gap of grille-like components and the minimum blade thickness can be accurately calculated. This method successfully transforms the intricate grille structure into local relative distance quantification, ensuring that the dimensional inspection results of densely packed, multi-slit components are rigorously aligned with regulatory indicators. This targeted processing strategy solves the technical challenge of interference-induced misjudgments in the overall inspection of dense parallel structures, enabling the accurate measurement of the safety of multi-slit components such as vents.

[0053] Please see Figure 4 , Figure 4 This is a schematic diagram of the inspection process for knob-type parts in the method for inspecting protrusions in the passenger compartment of a vehicle provided in this application.

[0054] like Figure 4As shown, in another possible implementation, when the 3D geometric model to be tested includes knob-like parts, the protrusion inspection method also includes: A third collision model is configured on knob-type parts. The third collision model is a ball head model with a preset size sliding pressure head. Extract the rotation axis and the surface to be inspected from knob-type parts; The third collision model is controlled to rotate and scan along the rotation axis at preset angle intervals, and collision interference detection is performed during the scanning process; Record the interference-free limit position of the third collision model during the rotational scanning process, and extract the maximum protrusion of the surface to be inspected at the limit position and the cross-sectional area at the preset offset position, so as to output the inspection results for knob-type parts in combination with the preset compliance requirements.

[0055] For knob-like parts on interior trim panels, such as volume knobs and air conditioning control switches, which have rotational or pressing travel, these are isolated and multi-positional movable features. Conventional static unidirectional projection cannot cover all aspects of their safety risks. Therefore, this application introduces a dedicated third collision model. The third collision model physically corresponds precisely to the tool-based inspection device required by regulations. Specifically, it consists of a 165mm diameter collision ball and a 50mm wide sliding pressure head with a specific travel range, mounted at the center of the collision ball. The system first extracts the central axis of rotation of the knob-like part and the exposed surface to be inspected. Then, it controls the asymmetrical third collision model to be positioned along the axis of rotation and drives it to perform a full-cycle rotational sweep scan around the axis of rotation at preset angular intervals (exemplarily, every 10 degrees). During the scan, the algorithm performs real-time three-dimensional geometric collision interference detection through the collision ball, dynamically finding the rotational orientation at which the combined tool can most deeply approach the interior reference surface. The system ultimately records the interference-free limit position of the third collision model during the rotational scan, where it does not forcibly disrupt or interfere with surrounding components. Under this limit posture, the maximum protrusion of the surface under inspection relative to the reference surface is calculated. Simultaneously, the algorithm also slices and calculates the cross-sectional area of ​​the knob-like part at a preset offset position at a specific vertical depth from the peak (exemplarily, a planar position offset downwards by 2.5mm or 6.5mm), thus fully meeting the dynamic safety control requirements for the intrusion volume and abrupt change in cross-sectional intensity of knob-like components.

[0056] Therefore, the third calculation model uses a virtual, movable scanning mechanism mounted on a specific axis to perform a full-envelope dynamic angular scan around the part, exhaustively searching for the maximum intrusion volume it might cause in any working posture. This solves the problem that isolated movable structures are difficult to statically evaluate using a single reference plane, ensuring a comprehensive capture of the safety performance of such multi-posture interior components.

[0057] Finally, there are no restrictions on the specific method for outputting the inspection results.

[0058] For example, in one possible implementation, the radius of curvature of each effective contact point is compared with a preset threshold, and the inspection results of protrusions in the vehicle passenger compartment are output based on the comparison results, including: The preset thresholds include a first threshold and a second threshold, and the first threshold is less than the second threshold; If the radius of curvature is less than the first threshold, the corresponding valid contact point is determined to be unqualified. If the radius of curvature is greater than or equal to the first threshold and less than or equal to the second threshold, the corresponding effective contact point is determined to be in a dangerous state. If the radius of curvature is greater than the second threshold, the corresponding effective contact point is deemed qualified. Based on the classification of non-compliance, hazardous conditions, and compliance, the inspection results of protrusions in the vehicle's passenger compartment are output.

[0059] Specifically, to directly transform the underlying spatial geometry calculations into quantitative decisions guiding engineering modifications, this method also establishes a progressive multi-level threshold control mechanism. Specifically, according to relevant national standards, the first threshold can be set to 2.5mm to ensure compliance. For situations with higher requirements, a second threshold can be flexibly set (for example, 5mm) to further raise the standard and facilitate risk verification by designers. For instance, when the algorithm calculates that the radius of curvature of a certain effective contact point is strictly less than the first threshold of 2.5mm, it means that the rounded corner is too sharp, posing a very high risk of directly injuring occupants in a collision. The system judges it as unqualified and marks it prominently in red on the surface cloud map. When the radius of curvature is between 2.5mm and 5mm (i.e., greater than or equal to the first threshold and less than or equal to the second threshold), it indicates that although the rounded corner is barely within the line, it still has potential safety hazards. The system judges it as dangerous and displays it in yellow on the cloud map. Only when the radius of curvature is greater than the second threshold of 5mm is it judged as fully compliant and displayed in green. This multi-level interval-based data classification output can automatically create a health profile of complex vehicle cabin surfaces, helping mathematical model developers accurately locate geometric areas where rounded corners need to be adjusted, and enabling targeted deployment of design corrections.

[0060] The above is a complete description of the method for inspecting protrusions in a vehicle passenger compartment provided in this application. It should be noted that the specific method for selecting the aforementioned special components (such as the grilles, knobs, etc.) is also not limited. In one possible implementation, the special components are automatically selected by a targeted-trained artificial intelligence model. In another possible implementation, the special components can also be manually pre-selected by the designer and submitted to the system; this application also makes no restrictions on this.

[0061] Another aspect of this application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements any of the foregoing methods for inspecting protrusions in a vehicle passenger compartment.

[0062] Understandably, relying on physical devices such as processors and computer-readable storage media to run the above methods provides a standardized, high-concurrency automated security verification platform for the automotive R&D system, which is conducive to platform-based, large-scale, and standardized deployment.

[0063] It should be noted that the embodiments of the present invention have better implementability and are not intended to limit the present invention in any way. Any person skilled in the art may use the above-disclosed technical content to change or modify it into equivalent effective embodiments. However, any modifications or equivalent changes and modifications made to the above embodiments based on the technical essence of the present invention without departing from the content of the technical solution of the present invention shall still fall within the scope of the technical solution of the present invention.

Claims

1. A method for inspecting protrusions in a vehicle passenger compartment, characterized in that, The method for inspecting protrusions includes: Acquire initial data, including the location of the reference point; Obtain the three-dimensional geometric model to be tested, and divide all surfaces in the three-dimensional geometric model to be tested into collision regions and exemption regions based on the three-dimensional geometric model to be tested and the initial data. A first collision model is configured within the collision area, and the first collision model is used to determine the accessibility of surface sampling points within the collision area in order to screen out effective contact points that can contact the first collision model. Calculate the radius of curvature of each effective contact point on the surface of the three-dimensional geometric model to be measured; The radius of curvature of each effective contact point is compared with a preset threshold, and the inspection result of the protrusions in the vehicle passenger compartment is output based on the comparison result.

2. The method for inspecting protrusions in a vehicle passenger compartment as described in claim 1, characterized in that, The process involves acquiring the three-dimensional geometric model to be tested, and based on the three-dimensional geometric model to be tested and the initial data, dividing all surfaces in the three-dimensional geometric model to be tested into collision regions and exemption regions. include: The collision area includes a head collision area and a knee collision area; Obtain the height coordinates of the reference point position, establish a rotation axis with the reference point position as the center, and construct multiple slice planes rotating along the rotation axis within a preset angle range; The points where each slice plane intersects with the surface of the three-dimensional geometric model to be tested are identified as collision points, and the upper and lower boundary lines of the surface of the three-dimensional geometric model to be tested are determined based on the collision points. The collision points on each slice plane whose height coordinates are greater than or equal to the upper and lower dividing lines are connected sequentially to form a closed curve, thereby enclosing and generating the head collision area. On each of the slice planes, the collision points with height coordinates less than the upper and lower dividing lines are sequentially connected to form a closed curve, thereby enclosing and generating the knee collision area.

3. The method for inspecting protrusions in a vehicle passenger compartment as described in claim 2, characterized in that, The step of dividing all surfaces in the three-dimensional geometric model under test into collision regions and exemption regions also includes: Obtain the three-dimensional boundary contour of the preset exempted component in the three-dimensional geometric model to be tested; Based on the three-dimensional boundary contour and the preset expansion offset, a spatial repulsion envelope is formed by expanding outward. The spatial repulsion envelope is projected along the projection direction onto the surface of the three-dimensional geometric model to be tested to obtain a preliminary exemption region. The portion of the preliminary exemption region whose height coordinates are less than the horizontal plane of the lowest point of the exempted component is removed to obtain the exemption region.

4. The method for inspecting protrusions in a vehicle passenger compartment as described in claim 3, characterized in that, The step of configuring a first collision model within the collision area and using the first collision model to determine the accessibility of surface sampling points within the collision area, in order to filter out effective contact points that can make contact with the first collision model, includes: The first collision model includes a collision ball for simulating a head; Based on preset density parameters, obtain each surface sampling point in the collision area, and extract the normal vector of each surface sampling point; Offset along the direction of the normal vector by a preset distance equal to the radius of the collision sphere to determine the center position of the collision sphere corresponding to each of the surface sampling points; Starting from the center of the sphere, find the other surface sampling point closest to the center of the sphere; Calculate the distance between the nearest other surface sampling point and the center of the sphere; If the distance is greater than or equal to the radius of the colliding sphere, then the corresponding surface sampling point is determined to be the effective contact point.

5. The method for inspecting protrusions in a vehicle passenger compartment as described in claim 3, characterized in that, The first collision model is configured within the collision area, and the first collision model is used to determine the accessibility of surface sampling points within the collision area, so as to screen out effective contact points that can make contact with the first collision model. include: The first collision model also includes a triangular prism for simulating the knee. Based on preset density parameters, obtain each surface sampling point in the knee collision area, and extract the normal vector of each surface sampling point; Along the direction of the normal vector, offset by a value equal to the preset interference distance of the triangular prism, to determine the reference center position of each of the surface sampling points corresponding to the triangular prism; Starting from the reference center position, find other surface sampling points that are closest to the reference center position; Calculate the distance between the nearest other surface sampling point and the reference center location; If the distance is greater than or equal to the preset interference distance of the triangular prism, then the corresponding surface sampling point is determined to be the effective contact point.

6. The method for inspecting protrusions in a vehicle passenger compartment as described in claim 4 or 5, characterized in that, The calculation of the radius of curvature of each of the effective contact points on the surface of the three-dimensional geometric model to be measured includes: The cross-sectional line containing the effective contact point is discretized into multiple discrete cross-sectional points; Move the first collision model along the preset reference line direction until the first collision model is tangent to the surface of the protrusion; Obtain the position of the first collision model in the tangent state, and calculate the difference in coordinate information between the highest point and the lowest point of the protrusion as the protrusion amount of the protrusion; Determine the geometric properties of the highest or lowest point. If the corresponding point is located on a curve segment, calculate the radius of curvature at that point. If the corresponding point is a corner or endpoint, determine that the radius of curvature is zero.

7. The method for inspecting protrusions in a vehicle passenger compartment as described in claim 1, characterized in that, When the three-dimensional geometric model to be tested includes grid-like parts, the protrusion inspection method further includes: A second collision model is configured at the grille-type part; The grid-like parts are divided into protruding parts and peripheral parts, and the corresponding cross-sectional lines are extracted; Discrete sampling is performed on the cross-sectional line to obtain a discrete point set. For the discrete points of the protruding part, the nearest neighbor points are searched in the discrete point set of the surrounding parts to form discrete point pairs. Valid discrete point pairs whose included normal vector angle satisfies the preset perpendicular condition are selected from the discrete point pairs; Based on the distance between the effective discrete point pairs, the minimum gap and minimum sheet thickness of the grid-like parts are calculated, and the inspection results for the grid-like parts are output in combination with preset compliance requirements.

8. The method for inspecting protrusions in a vehicle passenger compartment as described in claim 1, characterized in that, When the three-dimensional geometric model to be tested includes knob-like parts, the protrusion inspection method further includes: A third collision model is configured at the knob-like part, and the third collision model is a ball head model configured with a sliding pressure head of a preset size; Extract the rotation axis and the surface to be inspected from the knob-like part; The third collision model is controlled to rotate and scan along the rotation axis at preset angle intervals, and collision interference detection is performed during the scanning process; Record the interference-free limit position of the third collision model during the rotational scanning process, and extract the maximum protrusion of the surface to be inspected at the limit position and the cross-sectional area at the preset offset position, so as to output the inspection results for the knob-type parts in combination with the preset compliance requirements.

9. The method for inspecting protrusions in a vehicle passenger compartment as described in claim 1, characterized in that, The step of comparing the radius of curvature of each effective contact point with a preset threshold and outputting the inspection result of the protrusions in the vehicle passenger compartment based on the comparison result includes: The preset threshold includes a first threshold and a second threshold, and the first threshold is less than the second threshold; If the radius of curvature is less than the first threshold, the corresponding effective contact point is determined to be unqualified. If the radius of curvature is greater than or equal to the first threshold and less than or equal to the second threshold, then the corresponding effective contact point is determined to be in a dangerous state. If the radius of curvature is greater than the second threshold, the corresponding effective contact point is determined to be qualified; Based on the classification of non-compliance, dangerous conditions, and compliance, the inspection results of protrusions in the vehicle's passenger compartment are output.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the method for inspecting protrusions in a vehicle passenger compartment as described in any one of claims 1 to 9.

Images