Wire harness measurement method and device, computer device and storage medium

Abstract

The present application relates to the field of computer vision and industrial automation measurement, and discloses a wire harness measurement method and device, computer equipment and storage medium, comprising: acquiring point cloud data of a wire harness to be measured; based on the point cloud data, identifying components in the wire harness to be measured, and determining the identities of the components and the topological relationship between the components; performing feature analysis on the overlapping area of the point cloud data of adjacent frames to correct the pose of the point cloud data based on the features of the overlapping area; calculating the length of a flexible component based on the corrected point cloud data, and determining the total length of the wire harness to be measured based on the topological relationship between the components, the length of the flexible component and the preset length of a rigid component. The present application identifies the components and determines the topological relationship between the components, thereby clarifying the attribution relationship between the components, and combining the actual measured length of the flexible component with the preset length of the rigid component for accumulation, so as to eliminate the error caused by splicing and improve the accuracy of length measurement.

Description

Technical Field

[0001] This invention relates to the fields of computer vision and industrial automation measurement technology, specifically to wire harness measurement methods, devices, computer equipment, and storage media. Background Technology

[0002] With the increasing demands for reliability in electrical systems from high-end manufacturing sectors such as automotive and aerospace, wire harnesses, as core components connecting various electronic devices, are crucial for the assembly quality and signal transmission stability of the entire vehicle or machine due to their length accuracy. Complex wire harness assemblies typically consist of multiple flexible tubing such as corrugated tubes and self-winding tubes, bundled with cable ties and secured with branched protective shells to form a tree-like structure with multiple branches. In production and quality inspection, accurately measuring the length of each branch path is a critical process to ensure product quality.

[0003] However, when multiple corrugated pipes or cables are secured together with cable ties, the pipe surfaces adhere to each other and may even be deformed by compression. Furthermore, rigid components such as forked protective shells in the cable harness may cause physical splitting of the pipes. Related technologies struggle to identify the ownership of each pipe segment and distinguish between different pipes at the binding points, resulting in inaccurate output cable harness lengths. Therefore, accurately calculating the cable harness length is a pressing issue that needs to be addressed. Summary of the Invention

[0004] In view of this, the present invention provides a wire harness measurement method, apparatus, computer equipment, and storage medium to solve the problem of inaccurate wire harness length measurement.

[0005] In a first aspect, the present invention provides a wire harness measurement method, the method comprising: Acquire point cloud data of the harness under test; Based on the point cloud data, the components in the harness under test are identified, and the identifiers corresponding to each component and the topological relationships between the components are determined. The components include at least flexible components and rigid components. Feature analysis is performed on the overlapping region of point cloud data in adjacent frames to correct the pose of the point cloud data based on the features of the overlapping region; The length of the flexible component is calculated based on the corrected point cloud data, and the total length of the test harness is determined based on the topological relationship between the components, the length of the flexible component, and the preset length of the rigid component.

[0006] In one optional implementation, acquiring the point cloud data of the harness to be tested includes: The acquisition device is controlled to move along a preset path, and during the movement, multiple frames of raw data of the wire harness under test are acquired based on a preset step distance. Point cloud data of the wire harness under test is extracted based on the multiple frames of raw data.

[0007] In one optional implementation, the step of identifying components in the harness under test based on the point cloud data, and determining the identifiers corresponding to each component and the topological relationships between the components, includes: Based on the point cloud data, the flexible components in the test harness are distinguished, and an identifier is set for each flexible component. If a physical split of a rigid component or a single pipe is detected, a logical node is set to construct the topological relationship between the components based on the logical node.

[0008] In one optional implementation, distinguishing the flexible components in the test harness based on the point cloud data includes: Point cloud data is divided based on a preset distance threshold. If the spatial distance between different point clouds is greater than the preset distance threshold, the different point clouds are divided into point cloud clusters corresponding to different flexible components. If there is physical contact between point cloud clusters, calculate the normal vector of each point in the point cloud cluster, and extract the contact boundary at the physical contact point based on the distribution characteristics of the normal vector. Starting from the end point of each point cloud cluster, the expansion extends to the neighborhood and stops when it reaches the contact boundary, thus obtaining target point cloud clusters corresponding to different flexible components.

[0009] In one optional implementation, the step of setting a logical node to construct the topological relationship between components based on the logical node if a physical breakage of a rigid component or a single pipe is detected includes: If a physical split of a rigid component or a single tube is detected, the location of the rigid component and the location of the physical split are set as logical nodes. Identify the identifier of the flexible component associated with the logical node; Based on the connection relationships of the logical nodes and the identifiers of the flexible components, the topological relationships between the components are determined.

[0010] In one optional implementation, the step of performing feature analysis on the overlapping regions of point cloud data from adjacent frames to correct the pose of the point cloud data based on the features of the overlapping regions includes: Feature analysis is performed on the overlapping regions of point cloud data from adjacent frames; If the overlapping region has a preset rigid feature, then feature matching is performed on the point cloud data of adjacent frames based on the feature points in the overlapping region to obtain a transformation matrix; If the overlapping region does not have a preset rigid feature, the first point cloud of the rigid component is compared with the preset standard point cloud model to obtain the transformation matrix; The pose of the point cloud data is corrected based on the transformation matrix.

[0011] In one optional implementation, calculating the length of the flexible component based on the corrected point cloud data includes: Extract the point cloud data corresponding to the flexible component from the corrected point cloud data; Extract the central axis of the point cloud data corresponding to the flexible component to determine the central axis of the flexible component; Calculate the length of the flexible component based on its central axis.

[0012] In one optional implementation, determining the total length of the harness to be tested based on the topological relationship between the components, the length of the flexible component, and the preset length of the rigid component includes: Based on the topological relationship between the components, the path corresponding to the end of each branch in the harness under test is determined, and the path includes at least one flexible component and / or a rigid component. Based on the topological relationship between the components, the length of each flexible component and the preset length of each rigid component on the path are obtained; The measured lengths of each flexible component along the path are summed with the preset lengths of each rigid component to obtain the total length of the wire harness to be tested.

[0013] The wire harness measurement method provided in this embodiment includes acquiring point cloud data of the wire harness to be tested; identifying components in the wire harness based on the point cloud data, and determining the identifiers corresponding to each component and the topological relationships between the components; performing feature analysis on the overlapping areas of point cloud data in adjacent frames to correct the pose of the point cloud data based on the features of the overlapping areas; calculating the length of the flexible components based on the corrected point cloud data, and determining the total length of the wire harness to be tested based on the topological relationships between the components, the length of the flexible components, and the preset length of the rigid components. This method identifies components and determines the topological relationships between them, thereby clarifying the attribution relationships between components. It then corrects the point cloud pose by performing feature analysis on the overlapping areas of adjacent frames, and finally accumulates the measured lengths of the flexible components and the preset lengths of the rigid components. This eliminates errors caused by long-size splicing and improves the accuracy of length measurement.

[0014] In a second aspect, the present invention provides a wire harness measuring device, the device comprising: The point cloud data acquisition module is used to acquire the point cloud data of the harness under test. The topology determination module is used to identify the components in the harness under test based on the point cloud data, and determine the identifiers corresponding to each component and the topology relationships between the components. The components include at least flexible components and rigid components. The feature analysis module is used to perform feature analysis on the overlapping region of point cloud data in adjacent frames, so as to correct the pose of the point cloud data based on the features of the overlapping region. The length calculation module is used to calculate the length of the flexible component based on the corrected point cloud data, and to determine the total length of the wire harness to be tested based on the topological relationship between the components, the length of the flexible component and the preset length of the rigid component.

[0015] Thirdly, the present invention provides a computer device, comprising: a memory and a processor, the memory and the processor being communicatively connected to each other, the memory storing computer instructions, and the processor executing the computer instructions to perform the wire harness measurement method of the first aspect or any corresponding embodiment described above.

[0016] Fourthly, the present invention provides a computer-readable storage medium storing computer instructions for causing a computer to perform the wire harness measurement method of the first aspect or any corresponding embodiment thereof. Attached Figure Description

[0017] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0018] Figure 1 This is a schematic flowchart of a wire harness measurement method according to an embodiment of the present invention; Figure 2 This is a schematic diagram of the pipe topology separation process according to an embodiment of the present invention; Figure 3 This is a schematic diagram illustrating the establishment of topological relationships according to an embodiment of the present invention; Figure 4 This is a schematic diagram of the point cloud stitching principle according to an embodiment of the present invention; Figure 5 This is a structural block diagram of a wire harness measuring device according to an embodiment of the present invention; Figure 6 This is a schematic diagram of the hardware structure of a computer device according to an embodiment of the present invention. Detailed Implementation

[0019] 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.

[0020] According to an embodiment of the present invention, a wire harness measurement method embodiment is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0021] This embodiment provides a method for measuring wire harnesses. Figure 1 This is a flowchart of a wire harness measurement method according to an embodiment of the present invention, such as... Figure 1 As shown, the process includes the following steps: Step S101: Obtain the point cloud data of the harness to be tested.

[0022] The wire harness under test is a tree-like structure with multiple branches, formed by bundling multiple flexible tubes with cable ties and securing them with branched protective shells. The flexible tubes can be corrugated pipes, sub-coils, etc., and wires are threaded inside the flexible tubes. The tubes themselves are the main object to be measured. The wire harness under test may also include rigid components set along the path of the flexible tubes, such as connectors, brackets, cable ties, branched protective shells, etc. These rigid components play a role in fixing, connecting, or protecting the wire harness.

[0023] Point cloud data is a collection of discrete points in space, each point containing three-dimensional spatial coordinate information, used to characterize the surface morphology and spatial position of the wire harness under test. The point cloud data of the wire harness under test is obtained by scanning and acquiring data using acquisition devices such as structured light cameras and laser scanners.

[0024] Step S102: Based on point cloud data, identify the components in the harness to be tested, and determine the corresponding identifier of each component and the topological relationship between each component.

[0025] The components include at least flexible components and rigid components. Flexible components refer to tubular parts in the wire harness that have bendable and deformable characteristics, such as corrugated tubing, self-winding tubing, and heat shrink tubing, with flexible and deformable surface morphology. Rigid components refer to parts in the wire harness that have a fixed shape and are not easily deformed, such as connectors, brackets, cable ties, and branch protection shells.

[0026] Identifying components within a harness under test involves distinguishing which points in the point cloud data belong to flexible components, which belong to rigid components, and which specific component they belong to. Semantic parsing of the point cloud data determines the component type of each point, and points belonging to the same component are aggregated and assigned the same instance label. Specifically, the point cloud data is input into a trained instance segmentation network, which outputs the semantic category probability and instance mask for each point. Based on the mask, point cloud clusters corresponding to different component individuals are extracted. Furthermore, based on the geometric characteristics of each point cloud cluster (e.g., aspect ratio, surface curvature distribution, degree of local normal variation, etc.), they are classified as flexible or rigid components. Each identified component is assigned a unique identifier to distinguish it from other components.

[0027] Establish the connection structure and hierarchical relationship between components, and clarify the connection methods between each flexible component and between flexible components and rigid components in physical space, that is, determine the topological relationship between each component.

[0028] By analyzing the spatial distribution of point cloud clusters of each component, we determine whether there are physical connections or intersections between components, and establish associations between components at the locations where connections or intersections exist. Based on this, we identify spatial intersection points on the component connection paths, treating these intersection points as logical nodes in the topology, and recording the components and connection relationships associated with each logical node. By traversing and organizing all components and their connection relationships, we establish a tree-like topology of the harness under test, representing the topological relationships between components. This tree-like topology includes a root node (the beginning of the trunk), edges (path segments), nodes (branching points), and leaf nodes (the ends of branches).

[0029] Step S103: Perform feature analysis on the overlapping region of point cloud data in adjacent frames, and correct the pose of the point cloud data based on the features of the overlapping region.

[0030] Since the length of the wire harness under test is usually large, the acquisition device cannot cover the entire wire harness structure within a single acquisition range. Therefore, it is necessary to acquire multiple frames of point cloud data in segments along a preset path, and then register and stitch them together to form a complete global point cloud. During this process, there is a certain spatial overlap between two frames of point cloud data acquired by the acquisition device at adjacent acquisition positions. The overlapping area corresponds to the observation results of the same physical part of the wire harness under test from different viewpoints. However, the pose estimation of the acquisition device inevitably has errors during its movement. If the original poses of adjacent frames are directly used for stitching, the errors will accumulate frame by frame along the scanning path, causing spatial drift in the global point cloud of long-distance wire harnesses and affecting measurement accuracy. Therefore, it is necessary to use the spatial features in the overlapping area to correct the inter-frame pose.

[0031] Feature analysis of overlapping regions in point cloud data from adjacent frames involves extracting and evaluating spatial geometric features within these regions to determine whether they contain sufficient feature information for reliable registration. Spatial geometric features refer to geometric attributes in the point cloud that reflect the local shape characteristics of an object, such as curvature variations, normal distribution, and local surface morphology. If the overlapping region contains rich geometric features, a reliable point correspondence can be established between adjacent frames, thereby enabling inter-frame pose transformation.

[0032] By calculating the spatial transformation relationship between adjacent frames, the point cloud data of each frame is accurately aligned to a unified global coordinate system, thereby correcting the pose of the point cloud data and making the overlapping parts of adjacent frames coincide in space, thus eliminating or reducing the registration error accumulated frame by frame.

[0033] Step S104: Calculate the length of the flexible component based on the corrected point cloud data, and determine the total length of the harness to be tested based on the topological relationship between the components, the length of the flexible component, and the preset length of the rigid component.

[0034] After pose correction, the point cloud data of each frame has been precisely aligned to the complete point cloud data of a unified global coordinate system. Based on the point cloud data of each flexible component, the actual spatial length of the flexible component along its extension direction is calculated. Since the flexible component is tubular, its length should be measured along the central axis of the tube. The length of the flexible component can be determined by extracting the central axis of the flexible component and calculating its spatial arc length.

[0035] The preset length of a rigid component is its theoretical length. Since the geometry of a rigid component is fixed and its shape is regular, the preset length of the rigid component can be stored in the database in advance.

[0036] Based on the topological relationships between the components, the lengths of each branch path of the harness under test are calculated, and the total length of the harness under test represents the total length of each branch path. For each branch end, starting from its corresponding leaf node, it is necessary to backtrack along the tree topology to the root node (the beginning of the trunk), and accumulate the measured lengths of all flexible components along the path with the preset lengths of all rigid components in the order of the path to obtain the total length corresponding to that branch end.

[0037] The wire harness measurement method provided in this embodiment includes acquiring point cloud data of the wire harness to be tested; identifying components in the wire harness based on the point cloud data, and determining the identifiers corresponding to each component and the topological relationships between the components; performing feature analysis on the overlapping areas of point cloud data in adjacent frames to correct the pose of the point cloud data based on the features of the overlapping areas; calculating the length of the flexible components based on the corrected point cloud data, and determining the total length of the wire harness to be tested based on the topological relationships between the components, the length of the flexible components, and the preset length of the rigid components. This method identifies components and determines the topological relationships between them, thereby clarifying the attribution relationships between components. It then corrects the point cloud pose by performing feature analysis on the overlapping areas of adjacent frames, and finally accumulates the measured lengths of the flexible components and the preset lengths of the rigid components. This eliminates errors caused by long-size splicing and improves the accuracy of length measurement.

[0038] This embodiment provides a wire harness measurement method, which includes the following steps: Step S201: Obtain the point cloud data of the harness to be tested.

[0039] Specifically, step S201 includes: Step S2011: Control the acquisition device to move along a preset path, and acquire multiple frames of raw data of the test harness based on a preset step distance during the movement.

[0040] like Figure 2 As shown, the wire harness under test is a tree-like structure with multiple branches, formed by binding multiple flexible tubes with cable ties and fixing them with a branched protective shell. The flexible tubes can be corrugated pipes, sub-coiled pipes, etc., and wires are threaded inside the flexible tubes. The tubes themselves are the main objects to be measured. The acquisition device is a three-dimensional measuring device used to acquire the surface spatial information of the wire harness under test. The acquisition device can be a structured light camera, laser scanner, etc. The preset path is the motion trajectory pre-set by the acquisition device during the scanning process. The motion trajectory can cover the complete spatial distribution range of the wire harness under test. For example, it can move along a linear guide rail or be carried by a robotic arm along a planned path. The preset step distance is the spatial interval between two adjacent acquisition positions on the motion path. The preset step distance can be pre-set according to the measurement accuracy requirements and the field of view of the acquisition device.

[0041] The acquisition device is controlled to move along a preset path, and acquisition is triggered at preset step intervals during the movement to obtain multiple frames of raw data arranged in spatial order, covering the complete structure of the harness under test. Each frame of raw data includes three-dimensional point cloud coordinate information and the device's spatial pose information at the time of acquisition, where the device pose information is used for subsequent inter-frame registration and global stitching.

[0042] Step S2012: Extract point cloud data of the wire harness to be tested based on multiple frames of raw data.

[0043] Because the raw data may contain interference information from non-test objects such as workbench reflections, environmental noise, and outliers, directly using the raw data will affect the accuracy of subsequent component identification and length measurement. Therefore, it is necessary to extract the valid point cloud belonging to the test harness from multiple frames of raw data.

[0044] like Figure 2 As shown, the collected point cloud data is input into the instance segmentation network. The instance segmentation network can simultaneously identify point sets belonging to different object instances in the point cloud and output corresponding masks. The mask is a segmentation marker that identifies which points in the point cloud belong to a specific instance. Specifically, the instance segmentation network extracts rigid component masks and flexible tube masks in parallel from the original point cloud. The rigid component mask is used to mark the point cloud regions of rigid components such as connectors, brackets, cable ties, and bifurcated protective shells; the flexible tube mask is used to mark the point cloud regions of flexible tubes such as corrugated pipes and self-winding pipes.

[0045] After mask extraction for both rigid and flexible components, an environmental noise stripping operation is performed. Environmental noise stripping involves determining the background region—the portion of the point cloud that does not belong to the rigid or flexible components—based on the inverse operation of the mask. These background point clouds are then removed. Specifically, a pass-through filter can be used to set a retention range in the spatial coordinate direction, removing point clouds that extend beyond the spatial region of the test harness. A statistical filter is then used to perform statistical analysis on the neighborhood of each point, removing outliers that deviate from the mean by more than a set standard deviation, thereby eliminating interference from workbench reflections, dust noise, and other sources.

[0046] After semantic stripping and denoising of a single frame, the point cloud data of all frames are transformed into a unified global coordinate system based on the device spatial pose information carried by each frame, so as to obtain global point cloud data that completely covers the wire harness under test, i.e., the point cloud data of the wire harness under test.

[0047] Step S202: Based on point cloud data, identify the components in the harness to be tested, and determine the corresponding identifier of each component and the topological relationship between each component.

[0048] Specifically, step S202 includes: Step S2021: Based on point cloud data, distinguish the flexible components in the harness under test and set the corresponding identifier for each flexible component.

[0049] The point clouds belonging to different flexible components are separated, so that each flexible component corresponds to an independent subset of point clouds, thereby distinguishing the flexible components in the harness under test. Since the harness under test contains multiple parallel flexible tubes, which may be spatially separated or physically contacted or even deformed due to cable ties, it is necessary to distinguish the different tubes based on the spatial distribution characteristics of the point clouds. After distinguishing the individual flexible components, a corresponding identifier is set for each flexible component.

[0050] In some optional implementations, step S2021 above includes: Step a1: Divide the point cloud data based on a preset distance threshold. If the spatial distance between different point clouds is greater than the preset distance threshold, then divide the different point clouds into point cloud clusters corresponding to different flexible components.

[0051] The preset distance threshold is the upper limit of the spatial distance used to determine whether two points belong to the same flexible component. The preset distance threshold can be preset according to the point cloud density and the diameter of the flexible component.

[0052] A spatial index structure is constructed for the point cloud of the input flexible component. All points are marked as unvisited. Starting from any unvisited point as the initial seed point, a search is conducted within the spatial index structure for a set of neighboring points with a radius equal to a preset distance threshold, centered on the initial seed point. All unvisited points within this neighborhood with a distance less than the threshold are marked as belonging to the same cluster as the seed point and added to the growth queue of the current cluster. The next point is retrieved from the queue, and the neighborhood search process is repeated until no new points are added to the queue. After the growth of a tube cluster is completed, a temporary identifier is assigned to the cluster, and the next unvisited point is selected as the new seed point, repeating the above process until all points are marked. In this way, point cloud clusters belonging to different flexible components are defined.

[0053] Step a2: If there is physical contact between point cloud clusters, calculate the normal vector of each point in the point cloud cluster, and extract the contact boundary at the physical contact point based on the distribution characteristics of the normal vector.

[0054] When multiple flexible components are bundled together with fasteners such as cable ties, the surfaces of different tubes come into contact with each other and may even deform under pressure, causing the point clouds of the different tubes to be spatially connected and impossible to separate using only a distance threshold. A normal vector is the direction of the normal to the point cloud surface at a given point, used to characterize the orientation of the local surface at that point. By calculating the normal vectors of points in each point cloud cluster and analyzing the spatial distribution characteristics of the normal vectors, the boundary positions at physical contact points can be located.

[0055] Specifically, a plane is fitted within the neighborhood of each point in the point cloud cluster. The covariance matrix of this fitted plane is then decomposed to obtain the normal vector of that point. The angle between the normal vectors of all neighboring points and the normal vector of the current point is calculated, and a local normal vector deviation index is defined (e.g., the average angle between the neighborhood normal vectors and the current point's normal vector). Simultaneously, the principal curvature value of the point cloud at that point is calculated. The principal curvature of the tube surface undergoes abrupt changes at the physical contact point, meaning the curvatures of the two tubes overlap in the contact area, forming a local depression or curvature abrupt change region. Curvature abrupt change thresholds and normal vector deviation thresholds are set. Points that simultaneously satisfy both the local curvature value exceeding the curvature abrupt change threshold and the local normal vector deviation index exceeding the normal vector deviation threshold are marked as candidate valley line points. Connectivity analysis is performed on the candidate valley line points, aggregating spatially adjacent candidate points and extracting them as continuous valley lines, which represent the physical contact boundary between the two contacting tubes. The above is the contact surface peeling algorithm provided in this embodiment, used to locate the separation boundary between multiple physically contacting tubes.

[0056] Step a3: Starting from the end point of each point cloud cluster, expand outwards to the neighborhood, and stop expanding when reaching the contact boundary to obtain target point cloud clusters corresponding to different flexible components.

[0057] End points refer to the points at both ends of the point cloud cluster along the extension direction of the tube body. They usually correspond to the starting or ending face of the tube body and can be determined by detecting the extreme points in the point cloud cluster along the direction of the tube body.

[0058] Specifically, for each pipe point cloud cluster obtained, the vertices at both ends along the pipe's direction are selected as initial seed points, and these seed points are marked as visited. The angle between the normal vectors of neighboring points and the current seed point is calculated. If the angle is less than a preset normal deflection threshold, the neighboring point and the current seed point are determined to belong to the same smooth continuous surface, and it is also checked whether the neighboring point belongs to the extracted valley point set. If the neighboring point satisfies both the condition that the angle between its normal vectors is less than the preset normal deflection threshold and it does not belong to the valley point set, then the neighboring point is included in the current pipe cluster, marked as visited, and added to the growth queue to continue expanding outward; conversely, if the angle between the normal vectors of a neighboring point is greater than or equal to the preset normal deflection threshold, then the location is determined to be a discontinuous region or abrupt bend on the pipe surface, and growth in that direction is stopped; if the neighboring point belongs to the valley point set, then the physical contact boundary of another pipe has been reached, and growth in that direction is stopped. New seed points are sequentially taken from the queue, and the above growth process is repeated until the queue is empty, and the growth of the current pipe cluster is completed. Assign a globally unique topological identifier to each grown point cloud cluster as the identifier for that flexible component. Repeat the above process until all tube clusters have been grown and identified, resulting in target point cloud clusters corresponding to different flexible components.

[0059] In step S2022, if a physical split of a rigid component or a single pipe is detected, a logical node is set to construct the topological relationship between the components based on the logical node.

[0060] Physical splitting of a single tube refers to the phenomenon where a flexible tube branches into two or more sub-tubes in space, such as when a wire harness splits from a main trunk into multiple branches after passing through a branching protective shell. A logical node is a data structure unit in a topology used to mark specific spatial locations, representing the starting point, merging point, or turning point of a branch.

[0061] By setting logical nodes, continuous wire harness paths are logically segmented at physical branching or intersection points, ensuring that each segment has a clearly defined start and end node. Using logical nodes as connection hubs, flexible and rigid components are organized into a tree structure according to their physical connection order, clarifying the hierarchical relationships and connection directions of each component, thereby determining the topological relationships between the components.

[0062] In some optional implementations, step S2022 above includes: Step b1: If a physical split of a rigid component or a single pipe is detected, the location of the rigid component is set as a logical node and the location of the physical split is set as a logical node.

[0063] Rigid components can be detected based on the geometric features of point clouds. Rigid components have fixed dimensions and regular geometric shapes. They can be identified by recognizing point cloud regions that conform to the geometric model of the rigid component. Alternatively, they can be detected using semantic segmentation, that is, by using the rigid component mask output by the instance segmentation network to obtain its spatial location and boundary. The location of the detected rigid component is set as a logical node.

[0064] The centerline of the tube is detected, and the bifurcation point of the centerline is analyzed. The bifurcation point is the position where the centerline splits into two independent branches. If a bifurcation point is detected at the center point, it indicates that a single tube has a physical split. The position of the bifurcation point is then set as a logical node.

[0065] Step b2: Determine the identifier of the flexible component associated with the logical node.

[0066] Detect the flexible components connected to each configured logical node and obtain the identifiers of these flexible components. For example, if a bifurcated protective shell is connected to the ends of three flexible tubes, obtain the identifiers of each of the three flexible tubes and record their connection directions at that node.

[0067] Step b3: Based on the connection relationship of logical nodes and the identification of flexible components, determine the topological relationship between each component.

[0068] The connection relationships between logical nodes represent the connection paths formed by flexible components between logical nodes. Specifically, it indicates which two logical nodes are connected by which flexible component, and whether that flexible component belongs to a common segment or a private segment within that path. After obtaining the identifiers of all logical nodes and their associated flexible components, the topological relationships between the components can be established. Specifically, a flexible component connects two logical nodes, forming an edge in the topological relationship; logical nodes, as endpoints of the edges, form nodes in the topological relationship.

[0069] like Figure 3 As shown, the topology of the harness under test extends along the wiring direction with the root node at the starting end of the trunk 301. When the branch protection shell 302 is detected, it is a rigid component, and its location is marked as a logical node, serving as the logical boundary for the branch's initiation. The branch protection shell divides the trunk into two branches: branch 1 and branch 2. During the scanning process, the data acquisition frame sequence is obtained frame by frame along the scanning path, and the component identification results in each frame are associated with the topology nodes to complete the topological relationship between the components.

[0070] Step S203: Perform feature analysis on the overlapping region of point cloud data in adjacent frames, and correct the pose of the point cloud data based on the features of the overlapping region.

[0071] Specifically, step S203 includes: Step S2031: Perform feature analysis on the overlapping regions of point cloud data from adjacent frames.

[0072] Analyzing spatial geometric features within overlapping regions involves examining point clouds. Spatial geometric features refer to the geometric attributes in a point cloud that reflect the local shape characteristics of an object, such as corners, edges, and ridges—local morphologies with significant geometric recognizability. Point clouds of objects with significant rigidity features (e.g., wire harness brackets, clips, cable ties, connectors, etc.) have regular planar, cylindrical, or square cross-sections, stable curvature distributions, and distinct, repeatable features such as corners and edges on their surfaces, providing rich registration constraints in overlapping regions. Objects without significant rigidity features (e.g., corrugated pipes, self-winding pipes, bare cables, etc.) have surfaces with continuous corrugations or flexible curves, no fixed corners or planes, and unstable curvature distributions.

[0073] like Figure 4 As shown, the overlapping areas of adjacent frames k and k+1 are analyzed to determine whether the overlapping areas contain preset rigid features. Preset rigid features are geometric structural features corresponding to rigid components, such as sharp corners, edges, plane intersections, cylindrical end faces, and other local shapes with significant geometric recognizability.

[0074] Step S2032: If the overlapping region has a preset rigid feature, then feature matching is performed on the point cloud data of adjacent frames based on the feature points in the overlapping region to obtain the transformation matrix.

[0075] When a pre-defined rigid feature exists in the overlapping region, significant feature points are extracted from the overlapping region, such as SIFT (Scale Invariant Feature Transform), ISS (Intrinsic Shape Signature), or Harris 3D corner points. These feature points exhibit repeatability and geometric stability under different viewpoints. Based on a spatial index structure, fast feature descriptor matching is performed on feature points in adjacent frames to obtain an initial set of matched point pairs. The RANSAC (Random Sample Consensus) algorithm is used to remove mismatched point pairs, retaining the geometrically consistent matching set to improve the robustness of the matching. A least-squares problem is constructed for the filtered matched feature point pairs to find the rigid body transformation that minimizes the error between the two point clouds. The rotation matrix and translation vector are calculated using SVD (Singular Value Decomposition) to obtain a 6-DOF inter-frame transformation matrix. This transformation matrix is ​​used to transform the point cloud of the current frame to the coordinate system of the previous frame, achieving spatial alignment between the two frames.

[0076] Step S2033: If there is no preset rigid feature in the overlapping area, the first point cloud of the rigid component is compared with the preset standard point cloud model to obtain the transformation matrix.

[0077] When the overlapping region does not have a preset rigid feature, for example, when the overlapping region only contains flexible components with smooth surfaces and repetitive textures such as bellows, pose correction is performed. Figure 4 As shown, any identified rigid component within the scan segment, such as rigid component A, is searched, and its local point cloud is extracted as the first point cloud set. This first point cloud set is compared with a preset standard model in the global CAD prior database. The preset standard model is the point cloud data obtained by converting the CAD standard model corresponding to rigid component A. By registering the local rigid point cloud with its corresponding CAD standard model (e.g., using the ICP algorithm), the transformation matrix from the local point cloud to the CAD model in the current frame is solved. The theoretical pose of the CAD model is known, representing the absolute pose of rigid component A in the global coordinate system. By performing calculations using this theoretical pose matrix and the registration matrix from the point cloud to the CAD model, the global pose of the acquisition device in this frame can be deduced.

[0078] Step S2034: Correct the pose of the point cloud data based on the transformation matrix.

[0079] If the transformation matrix is ​​obtained through feature matching under the condition that the overlapping region has preset rigid features in step S2032, then the transformation matrix is ​​the relative pose transformation matrix between two adjacent frames, used to transform the current frame point cloud to the coordinate system of the previous frame. The transformation matrix calculated in step S2032 is applied to each point in the current frame point cloud data, so that all points in the current frame are transformed to the coordinate system of the previous frame, achieving precise alignment of the two adjacent frames in the overlapping region. By performing the above operation frame by frame, the point clouds of each frame are transformed to a unified coordinate system in sequence, completing the pose correction based on direct geometric registration.

[0080] If the transformation matrix is ​​obtained by forcibly aligning CAD anchor points in step S2033 when there are no preset rigid features in the overlapping area, then this transformation matrix is ​​the global pose matrix of the acquisition device obtained by back-reasoning the theoretical pose of the rigid component's CAD model, representing the absolute global pose of the acquisition device at the current frame. The back-reasoned global pose of the acquisition device replaces the original pose estimation value obtained by accumulating frame by frame, and the theoretical pose of the CAD model is used as the absolute reference system to determine the accurate position of the current frame in the global coordinate system.

[0081] After correcting the pose of each frame, the point cloud data of all frames are converted to a unified global coordinate system based on the corrected pose information of each frame. This ensures that the corresponding points in the overlapping areas of adjacent frames coincide precisely, resulting in complete point cloud data in the unbiased global coordinate system, i.e., the corrected point cloud data.

[0082] Step S204: Calculate the length of the flexible component based on the corrected point cloud data, and determine the total length of the harness to be tested based on the topological relationship between the components, the length of the flexible component, and the preset length of the rigid component.

[0083] Specifically, step S204 includes: Step S2041: Extract the point cloud data corresponding to the flexible component from the corrected point cloud data.

[0084] From the corrected point cloud data, all identifiers belonging to the flexible components are filtered out, and then the point cloud data corresponding to the flexible components is extracted.

[0085] Step S2042: Extract the central axis of the point cloud data corresponding to the flexible component and determine the central axis of the flexible component.

[0086] When extracting the central axis of the point cloud data corresponding to the flexible component, different extraction methods can be selected according to the wrapping material of the flexible component. For example, for a pipe with a periodic corrugated surface and fluctuation noise (e.g., a corrugated pipe), the point cloud can be sliced ​​along the wiring direction, the centroid of the point cloud in each section can be calculated, and the centroids of each section can be connected to obtain a smooth central axis point set. For a pipe with a smooth surface but which may deform due to binding and pressure, the discrete points can be guided to converge toward the geometric centerline through the gravity function, thereby extracting the central axis point set.

[0087] After obtaining the discrete point set of the central axis, the discrete point set is spatially sorted according to the pipe's orientation to ensure a monotonically distributed distribution along the pipe's centerline. Outliers are removed (e.g., by statistical filtering to eliminate points with a neighborhood distance exceeding the mean of 3σ), and interpolation is performed to fill in sparse regions, ensuring uniform point density. A chord length parameterization method is used to calculate the parameters corresponding to each point based on the Euclidean distance between the point sets, constructing third- or fifth-order B-spline basis functions. Least-squares fitting is then used to solve for the spline control points, generating a continuous spatial parametric curve equation C(t). Third-order B-splines are suitable for smooth, continuous pipes with low computational cost, while fifth-order B-splines are suitable for corrugated pipes or curved sections with drastic curvature changes.

[0088] Step S2043: Calculate the length of the flexible component based on its central axis.

[0089] The measured length of the flexible component is calculated by integrating the arc length of the spatial parameter curve equation C(t) along its central axis. The length L of the flexible component can be determined by the following formula:

[0090] Where L represents the length of the flexible component, and C(t) represents the spatial parameter curve equation of the central axis of the flexible component generated after spline fitting.

[0091] Step S2044: Determine the path corresponding to the end of each branch in the harness under test based on the topological relationship between the components.

[0092] The path includes at least one flexible component and / or a rigid component. The path is all the components and their connection order that are traversed from the root node of the topology to a certain leaf node. For each branch end, starting from the leaf node, backtracking along the tree topology to the root node, all the components traversed on the path and their connection order are recorded in sequence, thereby obtaining the complete path corresponding to the branch end.

[0093] Specifically, since the tree-like topology records the component identifier associated with each logical node and the connection direction between components, starting from any leaf node, the process traces upwards level by level according to the parent connection relationship recorded by the node, passing through each intermediate node and edge in sequence until the root node is reached. During this process, for each edge traversed, the component identifier and component type (flexible or rigid component) corresponding to that edge are recorded; for each node traversed, the spatial location information of the node is recorded. After the traversal is complete, the complete component sequence from the start to the end of the branch is obtained, i.e., the path corresponding to the end of the branch.

[0094] Step S2045: Based on the topological relationship between the components, obtain the length of each flexible component and the preset length of each rigid component on the path.

[0095] Obtain the length of each component on the path, specifically the length of the flexible component and the preset length of the rigid component.

[0096] Step S2046: The measured lengths of each flexible component on the path are added together with the preset lengths of each rigid component to obtain the total length of the wire harness to be tested.

[0097] Based on the obtained length values ​​of each component on any branch path, the lengths of all flexible components and the preset lengths of rigid components are summed in order along the path to obtain the total length of the branch path (the total length corresponding to the branch ends). For a harness with multiple branch ends, the total length of each branch end is calculated separately, and the measurement result of the total length of each branch path is output.

[0098] The total length of the harness under test represents the total length of each branch path. The total length of any branch path can be determined by the following formula:

[0099] in, This represents the total length of the nth branch path; It represents the sum of the lengths of the common path segments traversed by the nth branch path, that is, the lengths of all path segments from the trunk starting point to the branch fork node (including the lengths of flexible components and the preset lengths of rigid components). It represents the sum of the path segment lengths of the branch part of the nth branch path, that is, the length of the exclusive path segment from the fork node to the end of the branch (including the measured length of the flexible component and the preset length of the rigid component).

[0100] During the calculation process, common path segments and private path segments are automatically distinguished based on the topological relationship of the components. Common path segments are counted only once in each branch to avoid duplicate calculations.

[0101] This embodiment also provides a wire harness measuring device for implementing the above embodiments and preferred embodiments; details already described will not be repeated. As used below, the term "module" can refer to a combination of software and / or hardware that performs a predetermined function. Although the device described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0102] This embodiment provides a wire harness measuring device, such as... Figure 5 As shown, it includes: The point cloud data acquisition module is used to acquire the point cloud data of the harness under test. The topology determination module is used to identify the components in the harness under test based on the point cloud data, and determine the identifiers corresponding to each component and the topology relationships between the components. The components include at least flexible components and rigid components. The feature analysis module is used to perform feature analysis on the overlapping region of point cloud data in adjacent frames, so as to correct the pose of the point cloud data based on the features of the overlapping region. The length calculation module is used to calculate the length of the flexible component based on the corrected point cloud data, and to determine the total length of the wire harness to be tested based on the topological relationship between the components, the length of the flexible component and the preset length of the rigid component.

[0103] In some alternative implementations, the point cloud data acquisition module includes: The equipment control unit is used to control the acquisition device to move along a preset path and acquire multiple frames of raw data of the wire harness under test based on a preset step distance during the movement. The point cloud extraction unit is used to extract point cloud data of the wire harness under test based on the multiple frames of raw data.

[0104] In some optional implementations, the topology determination module includes: The component differentiation unit is used to differentiate the flexible components in the test harness based on the point cloud data, and to set an identifier for each flexible component. The topology determination unit is used to set up logical nodes if a physical split of a rigid component or a single pipe is detected, so as to construct the topology relationship between the components based on the logical nodes.

[0105] In some optional implementations, the component differentiation unit includes: A point cloud partitioning subunit is used to partition point cloud data based on a preset distance threshold. If the spatial distance between different point clouds is greater than the preset distance threshold, the different point clouds are partitioned into point cloud clusters corresponding to different flexible components. The boundary extraction subunit is used to calculate the normal vector of each point in the point cloud cluster if there is physical contact between point cloud clusters, and extract the contact boundary at the physical contact point based on the distribution characteristics of the normal vector. The component area molecular unit is used to expand into the neighborhood from the end point of each point cloud cluster, and to stop expanding when it reaches the contact boundary, so as to obtain target point cloud clusters corresponding to different flexible components.

[0106] In some optional implementations, the topology determination unit includes: The logic node setting subunit is used to set the location of the rigid component and the location of the physical split as logic nodes if a physical split of a rigid component or a single pipe is detected. The identifier determination subunit is used to determine the identifier of the flexible component associated with the logical node; The topology determination subunit is used to determine the topology relationships between components based on the connection relationships of the logical nodes and the identifiers of the flexible components.

[0107] In some optional implementations, the feature analysis module includes: The feature analysis unit is used to perform feature analysis on the overlapping regions of point cloud data from adjacent frames; The feature matching unit is used to perform feature matching on the point cloud data of adjacent frames based on the feature points in the overlapping region if a preset rigid feature exists in the overlapping region, so as to obtain a transformation matrix. The comparison unit is used to compare the first point cloud of the rigid component with a preset standard point cloud model if the overlapping region does not have a preset rigid feature, and to obtain a transformation matrix. The pose correction unit is used to correct the pose of the point cloud data based on the transformation matrix.

[0108] In some optional implementations, the length calculation module is also used for: Extract the point cloud data corresponding to the flexible component from the corrected point cloud data; Extract the central axis of the point cloud data corresponding to the flexible component to determine the central axis of the flexible component; Calculate the length of the flexible component based on its central axis.

[0109] In some optional implementations, the length calculation module is also used for: Based on the topological relationship between the components, the path corresponding to the end of each branch in the harness under test is determined, and the path includes at least one flexible component and / or a rigid component. Based on the topological relationship between the components, the length of each flexible component and the preset length of each rigid component on the path are obtained; The measured lengths of each flexible component along the path are summed with the preset lengths of each rigid component to obtain the total length of the wire harness to be tested.

[0110] Further functional descriptions of the above modules and units are the same as those in the corresponding embodiments described above, and will not be repeated here.

[0111] In this embodiment, the wire harness measuring device is presented in the form of a functional unit. Here, a unit refers to an ASIC (Application Specific Integrated Circuit) circuit, a processor and memory that execute one or more software or fixed programs, and / or other devices that can provide the above functions.

[0112] This invention also provides a computer device having the above-described wire harness measuring device.

[0113] Please see Figure 6 , Figure 6 This is a schematic diagram of the structure of a computer device provided in an optional embodiment of the present invention, such as... Figure 6 As shown, the computer device includes one or more processors 10, memory 20, and interfaces for connecting the components, including high-speed interfaces and low-speed interfaces. The components communicate with each other via different buses and can be mounted on a common motherboard or otherwise installed as needed. The processors can process instructions executed within the computer device, including instructions stored in or on memory to display graphical information of a GUI on external input / output devices (such as display devices coupled to the interfaces). In some alternative implementations, multiple processors and / or multiple buses can be used with multiple memories and multiple memory modules, if desired. Similarly, multiple computer devices can be connected, each providing some of the necessary operations (e.g., as a server array, a group of blade servers, or a multiprocessor system). Figure 6 Take a processor 10 as an example.

[0114] Processor 10 may be a central processing unit, a network processor, or a combination thereof. Processor 10 may further include a hardware chip. The hardware chip may be an application-specific integrated circuit (ASIC), a programmable logic device (PLD), or a combination thereof. The programmable logic device may be a complex programmable logic device (CAMP), a field-programmable gate array (FPGA), a general-purpose array logic (GDA), or any combination thereof.

[0115] The memory 20 stores instructions executable by at least one processor 10 to cause the at least one processor 10 to perform the method shown in the above embodiments.

[0116] The memory 20 may include a program storage area and a data storage area. The program storage area may store the operating system and applications required for at least one function; the data storage area may store data created based on the use of the computer device. Furthermore, the memory 20 may include high-speed random access memory and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some alternative embodiments, the memory 20 may optionally include memory remotely located relative to the processor 10, and these remote memories may be connected to the computer device via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0117] The memory 20 may include volatile memory, such as random access memory; the memory may also include non-volatile memory, such as flash memory, hard disk or solid-state drive; the memory 20 may also include a combination of the above types of memory.

[0118] The computer device also includes a communication interface 30 for communicating with other devices or communication networks.

[0119] This invention also provides a computer-readable storage medium. The methods described above according to embodiments of the invention can be implemented in hardware or firmware, or implemented as computer code that can be recorded on a storage medium, or implemented as computer code downloaded via a network and originally stored on a remote storage medium or a non-transitory machine-readable storage medium and then stored on a local storage medium. Thus, the methods described herein can be processed by software stored on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. The storage medium can be a magnetic disk, optical disk, read-only memory, random access memory, flash memory, hard disk, or solid-state drive, etc.; further, the storage medium can also include combinations of the above types of memory. It is understood that computers, processors, microprocessor controllers, or programmable hardware include storage components capable of storing or receiving software or computer code, which, when accessed and executed by the computer, processor, or hardware, implements the methods shown in the above embodiments.

[0120] A portion of this invention can be applied as a computer program product, such as computer program instructions, which, when executed by a computer, can invoke or provide the methods and / or technical solutions according to the invention through the operation of the computer. Those skilled in the art will understand that the forms in which computer program instructions exist in a computer-readable medium include, but are not limited to, source files, executable files, installation package files, etc. Correspondingly, the ways in which computer program instructions are executed by a computer include, but are not limited to: the computer directly executing the instructions, or the computer compiling the instructions and then executing the corresponding compiled program, or the computer reading and executing the instructions, or the computer reading and installing the instructions and then executing the corresponding installed program. Here, the computer-readable medium can be any available computer-readable storage medium or communication medium accessible to a computer.

[0121] Although embodiments of the invention have been described in conjunction with the accompanying drawings, those skilled in the art can make various modifications and variations without departing from the spirit and scope of the invention, and such modifications and variations all fall within the scope defined by the appended claims.

Claims

1. A method for measuring wire harnesses, characterized in that, The method includes: Acquire point cloud data of the harness under test; Based on the point cloud data, the components in the harness under test are identified, and the identifiers corresponding to each component and the topological relationships between the components are determined. The components include at least flexible components and rigid components. Feature analysis is performed on the overlapping region of point cloud data in adjacent frames to correct the pose of the point cloud data based on the features of the overlapping region; The length of the flexible component is calculated based on the corrected point cloud data, and the total length of the test harness is determined based on the topological relationship between the components, the length of the flexible component, and the preset length of the rigid component.

2. The method according to claim 1, characterized in that, The acquisition of point cloud data of the harness under test includes: The acquisition device is controlled to move along a preset path, and during the movement, multiple frames of raw data of the wire harness under test are acquired based on a preset step distance. Point cloud data of the wire harness under test is extracted based on the multiple frames of raw data.

3. The method according to claim 1, characterized in that, The step of identifying components in the harness under test based on the point cloud data, and determining the identifiers corresponding to each component and the topological relationships between the components, includes: Based on the point cloud data, the flexible components in the test harness are distinguished, and an identifier is set for each flexible component. If a physical split of a rigid component or a single pipe is detected, a logical node is set to construct the topological relationship between the components based on the logical node.

4. The method according to claim 3, characterized in that, The step of distinguishing the flexible components in the test harness based on the point cloud data includes: Point cloud data is divided based on a preset distance threshold. If the spatial distance between different point clouds is greater than the preset distance threshold, the different point clouds are divided into point cloud clusters corresponding to different flexible components. If there is physical contact between point cloud clusters, calculate the normal vector of each point in the point cloud cluster, and extract the contact boundary at the physical contact point based on the distribution characteristics of the normal vector. Starting from the end point of each point cloud cluster, the expansion extends to the neighborhood and stops when it reaches the contact boundary, thus obtaining target point cloud clusters corresponding to different flexible components.

5. The method according to claim 3, characterized in that, If a physical breakage of a rigid component or a single pipe is detected, a logical node is set to construct the topological relationship between the components based on the logical node, including: If a physical split of a rigid component or a single tube is detected, the location of the rigid component and the location of the physical split are set as logical nodes. Identify the identifier of the flexible component associated with the logical node; Based on the connection relationships of the logical nodes and the identifiers of the flexible components, the topological relationships between the components are determined.

6. The method according to claim 1, characterized in that, The step of performing feature analysis on the overlapping regions of point cloud data from adjacent frames, and correcting the pose of the point cloud data based on the features of the overlapping regions, includes: Feature analysis is performed on the overlapping regions of point cloud data from adjacent frames; If the overlapping region has a preset rigid feature, then feature matching is performed on the point cloud data of adjacent frames based on the feature points in the overlapping region to obtain a transformation matrix; If the overlapping region does not have a preset rigid feature, the first point cloud of the rigid component is compared with the preset standard point cloud model to obtain the transformation matrix; The pose of the point cloud data is corrected based on the transformation matrix.

7. The method according to claim 1, characterized in that, The calculation of the length of the flexible component based on the corrected point cloud data includes: Extract the point cloud data corresponding to the flexible component from the corrected point cloud data; Extract the central axis of the point cloud data corresponding to the flexible component to determine the central axis of the flexible component; Calculate the length of the flexible component based on its central axis.

8. The method according to claim 1, characterized in that, The determination of the total length of the harness under test based on the topological relationship between the components, the length of the flexible component, and the preset length of the rigid component includes: Based on the topological relationship between the components, the path corresponding to the end of each branch in the harness under test is determined, and the path includes at least one flexible component and / or a rigid component. Based on the topological relationship between the components, the length of each flexible component and the preset length of each rigid component on the path are obtained; The measured lengths of each flexible component along the path are summed with the preset lengths of each rigid component to obtain the total length of the wire harness to be tested.

9. A wire harness measuring device, characterized in that, The device includes: The point cloud data acquisition module is used to acquire the point cloud data of the harness under test. The topology determination module is used to identify the components in the harness under test based on the point cloud data, and determine the identifiers corresponding to each component and the topology relationships between the components. The components include at least flexible components and rigid components. The feature analysis module is used to perform feature analysis on the overlapping region of point cloud data in adjacent frames, so as to correct the pose of the point cloud data based on the features of the overlapping region. The length calculation module is used to calculate the length of the flexible component based on the corrected point cloud data, and to determine the total length of the wire harness to be tested based on the topological relationship between the components, the length of the flexible component and the preset length of the rigid component.

10. A computer device, characterized in that, include: A memory and a processor are communicatively connected, the memory storing computer instructions, and the processor executing the computer instructions to perform the wire harness measurement method according to any one of claims 1 to 8.

11. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions for causing the computer to perform the harness measurement method according to any one of claims 1 to 8.

Images