Skylight anti-pinch function detection method and device

Abstract

The invention discloses a skylight anti-pinch function detection method and device, and relates to the technical field of skylight detection, and the method comprises the steps: constructing a skylight anti-pinch object library; determining a preset to-be-detected skylight and a closing speed controllable mode of the preset to-be-detected skylight; generating a first test scene set based on the closing speed controllable mode and a skylight anti-pinch object library; configuring different closing distances under each test scene in the first test scene set by taking the calibration distance as a constraint to generate a second test scene set; testing a preset to-be-detected skylight by using the second test scene set to generate an anti-pinch function test result set; and based on the anti-pinch function test result set, performing sensitivity evaluation on an anti-pinch system of the preset to-be-detected skylight, and generating an anti-pinch function detection result. According to the invention, the technical problems of insufficient sunroof anti-pinch test scenes and incomplete test coverage in the prior art are solved, and the technical effect of improving the detection comprehensiveness and accuracy of the sunroof anti-pinch function is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of skylight detection, and in particular to a method and device for detecting an anti-pinch function of a skylight. Background Art

[0002] With the rapid development of the automobile industry, sunroofs, as an important configuration of vehicles, have been widely used in various models. As users' requirements for safety, comfort and intelligent functions continue to increase, the anti-pinch function of the sunroof has become a key part of the vehicle safety system. However, with the increasing complexity of vehicle design and functions, the detection and verification of the sunroof anti-pinch system is also facing increasing challenges. Traditional sunroof anti-pinch function detection methods often have problems such as limited test scenarios, incomplete coverage, and inability to simulate complex environments, making it difficult to fully evaluate the anti-pinch function of the sunroof. Summary of the invention

[0003] The present application provides a method and device for detecting the anti-pinch function of a skylight, which are used to solve the technical problems of the prior art in that the applicability is limited and the degree of automation is low when scanning documents.

[0004] In view of the above problems, the present application provides a method and device for detecting the anti-pinch function of a sunroof.

[0005] In a first aspect of the present application, a method for detecting an anti-pinch function of a sunroof is provided, the method comprising: Collect skylight anti-pinch accident samples to perform feature clustering of accident objects and build a skylight anti-pinch object library; determine a preset skylight to be inspected and a controllable closing speed mode of the preset skylight to be inspected; perform traversal matching of different speeds and different objects based on the controllable closing speed mode and the skylight anti-pinch object library to generate a first test scenario set; determine the calibration distance of the preset skylight to be inspected from a fully open state to a closed state, and use the calibration distance as a constraint to configure different closing distances in each test scenario in the first test scenario set to generate a second test scenario set; test the preset skylight to be inspected with each test scenario in the second test scenario set, and record the triggering characteristics of the corresponding anti-pinch system to generate an anti-pinch function test result set; perform a sensitivity evaluation on the anti-pinch system of the preset skylight to be inspected based on the anti-pinch function test result set to generate an anti-pinch function detection result.

[0006] A second aspect of the present application provides a sunroof anti-pinch function detection device, the device comprising: A skylight anti-pinch object library construction module is used to collect skylight anti-pinch accident samples to perform feature clustering of accident objects and construct a skylight anti-pinch object library; a skylight to be inspected determination module is used to determine a preset skylight to be inspected and a controllable closing speed mode of the preset skylight to be inspected; a first test scenario generation module is used to perform traversal matching of different speeds and different objects based on the controllable closing speed mode and the skylight anti-pinch object library to generate a first test scenario set; a second test scenario generation module is used to determine the calibration distance of the preset skylight to be inspected from the fully open state to the closed state, and generate a second test scenario set for different closing distance configurations in each test scenario in the first test scenario set with the calibration distance as a constraint; a test result acquisition module is used to test the preset skylight to be inspected with each test scenario in the second test scenario set, and record the triggering characteristics of the corresponding anti-pinch system to generate an anti-pinch function test result set; a detection result generation module is used to perform sensitivity evaluation on the anti-pinch system of the preset skylight to be inspected based on the anti-pinch function test result set to generate an anti-pinch function detection result.

[0007] One or more technical solutions provided in this application have at least the following technical effects or advantages: The present application collects skylight anti-pinch accident samples to cluster the features of accident objects and construct a skylight anti-pinch object library; determines a preset skylight to be inspected and a controllable closing speed mode of the preset skylight to be inspected; performs traversal matching of different speeds and different objects based on the controllable closing speed mode and the skylight anti-pinch object library to generate a first test scene set; determines the calibration distance of the preset skylight to be inspected from the fully open state to the closed state, and uses the calibration distance as a constraint to configure different closing distances in each test scene in the first test scene set to generate a second test scene set; tests the preset skylight to be inspected with each test scene in the second test scene set, and records the triggering features of the corresponding anti-pinch system to generate an anti-pinch function test result set; performs sensitivity evaluation on the anti-pinch system of the preset skylight to be inspected based on the anti-pinch function test result set to generate an anti-pinch function detection result. The present invention solves the technical problems of insufficient skylight anti-pinch test scenes and incomplete test coverage in the prior art, and achieves the technical effect of improving the comprehensiveness and accuracy of skylight anti-pinch function detection by constructing a skylight anti-pinch object library, generating a multi-scenario test set and performing sensitivity evaluation. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0009] Figure 1 A schematic diagram of a method for detecting the anti-pinch function of a sunroof provided in an embodiment of the present application; Figure 2 A schematic structural diagram of a sunroof anti-pinch function detection device provided in an embodiment of the present application.

[0010] Explanation of the reference numerals: skylight anti-pinch object library construction module 11, skylight to be inspected determination module 12, first test scenario generation module 13, second test scenario generation module 14, test result acquisition module 15, detection result generation module 16. DETAILED DESCRIPTION

[0011] The present application provides a method and device for detecting the anti-pinch function of the sunroof, aiming to solve the technical problems in the prior art of insufficient anti-pinch test scenarios and incomplete test coverage of the sunroof. By constructing a sunroof anti-pinch object library, generating a multi-scenario test set and performing sensitivity evaluation, the technical effect of improving the comprehensiveness and accuracy of the sunroof anti-pinch function detection is achieved.

[0012] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0013] It should be noted that any variations of the terms "include" and "have" are intended to cover non-exclusive inclusions. For example, a process, method, apparatus, product or server that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or modules that are not explicitly listed or inherent to these processes, methods, products or devices.

[0014] Embodiment 1, as Figure 1 As shown, the present application provides a method for detecting the anti-pinch function of a sunroof, the method comprising: Step S100: Collect skylight anti-pinch accident samples to perform feature clustering of accident objects and build a skylight anti-pinch object library.

[0015] In an embodiment of the present application, first, pre-stored sunroof anti-pinch accident samples are obtained from a preset database, and then feature analysis is performed on the accident objects in the sunroof anti-pinch accident samples to extract key attributes such as hardness, elasticity, shape and size, and generate an accident object feature set; then, a similarity analysis is performed on every two object features in the accident object feature set, and features with a similarity greater than a preset threshold (such as 0.8) are clustered to form multiple categories of accident object features; finally, corresponding anti-pinch test objects (such as rubber balls simulating fingers and metal blocks simulating tools) are configured based on these clustering results to construct a sunroof anti-pinch object library covering real scenes.

[0016] Furthermore, in the method provided in the embodiment of the application, samples of skylight anti-pinch accidents are collected to perform feature clustering of accident objects, and a skylight anti-pinch object library is constructed, which also includes: The features of the accident objects in the sunroof anti-pinch accident samples are analyzed to generate an accident object feature set; similarity analysis is performed on every two accident object features in the accident object feature set, and accident object features with similarities greater than a preset similarity threshold are clustered to generate multiple categories of accident object features; corresponding anti-pinch test objects are configured with the multiple categories of accident object features to generate a sunroof anti-pinch object library.

[0017] In an embodiment of the present application, the features of the accident object in the skylight anti-pinch accident sample are first analyzed, and the key physical properties, including hardness, elasticity, shape and size, are extracted by calling the accident object information in a preset database. The acquired physical properties are integrated to generate a feature set of the accident object.

[0018] Next, perform similarity analysis on every two accident object features in the accident object feature set, and use methods such as Euclidean distance or cosine similarity to calculate the similarity value between object features. For example, two strip objects are similar in shape and size, and their similarity value is high. According to the preset similarity threshold (such as 0.8), the object features with similarity greater than the threshold are clustered. Use K-means clustering or hierarchical clustering algorithm to classify object features with high similarity into one category, and generate multiple categories of accident object features.

[0019] Finally, based on the multi-category accident object features generated by clustering, technicians in this field configure the corresponding anti-pinch test objects according to the hardness, size, thickness and other attributes in the object features. For example, for soft objects, rubber balls are used to simulate fingers to match their low hardness and high elasticity characteristics; for hard objects, metal blocks are used to simulate tools to match their high hardness and low elasticity characteristics; for irregular shapes, 3D printing technology is used to make models to match their unique shape and size characteristics. After being standardized and organized, these test objects together constitute the sunroof anti-pinch object library, covering the main types of obstacles in real accidents, and providing comprehensive and diverse scenario support for subsequent anti-pinch function tests.

[0020] Step S200: determining a preset skylight to be inspected and a controllable closing speed mode of the preset skylight to be inspected.

[0021] Furthermore, in the method provided in the embodiment of the application, determining a preset skylight to be inspected and a controllable closing speed mode of the preset skylight to be inspected further includes: Determine the opening and closing control mode of the preset sunroof to be inspected, wherein the opening and closing control mode includes one of an automatic mode, a manual mode or an automatic and manual mixed mode; if the opening and closing control mode is an automatic mode, connect to the sunroof opening and closing automatic control terminal to collect optional closing speeds and generate a first speed mode set; generate the closing speed controllable mode with the first speed mode set.

[0022] In an embodiment of the present application, a technical expert first determines a preset skylight to be inspected, and obtains an opening and closing control mode of the preset skylight to be inspected from a skylight product database, wherein the opening and closing control mode includes an automatic mode, a manual mode, or an automatic and manual mixed mode.

[0023] When the opening and closing control mode of the sunroof to be inspected is preset to the automatic mode, the sunroof opening and closing automatic control terminal (such as the vehicle-mounted central control system) is connected to collect the optional closing speeds supported by it. For example, the optional range of the sunroof closing speed (such as 0.5cm / s, 1cm / s, 2cm / s, etc.) is obtained through the vehicle-mounted central control system, and these speed values ​​are sorted into a first speed mode set.

[0024] Finally, based on the first speed pattern set, a closing speed controllable pattern is generated. The closing speed controllable pattern allows the tester to flexibly adjust the closing speed of the sunroof in different test scenarios to simulate a variety of practical application scenarios. In the subsequent anti-pinch test, the closing speed of the sunroof can be adjusted according to the closing speed controllable pattern to ensure the triggering response of the anti-pinch system at different speeds.

[0025] Furthermore, in the method provided in the embodiment of the application, after determining the opening and closing control mode of the preset skylight to be inspected, the method further includes: If the opening and closing control mode is a manual mode, the structural parameters of the preset skylight to be inspected are collected, wherein the structural parameters include skylight size parameters, opening and closing track structural parameters and sealing strip structural parameters; digital modeling is performed with the structural parameters, closing speed simulation under different pulling forces is performed according to a predetermined manual force range, and a second speed mode set is generated; the closing speed controllable mode is generated with the second speed mode set.

[0026] In an embodiment of the present application, if the opening and closing control mode of the preset skylight to be inspected is the manual mode, the structural parameters of the preset skylight to be inspected are collected, including skylight size parameters (such as length, width, thickness), opening and closing track structure parameters (such as track length, track material, friction coefficient) and sealing strip structure parameters (such as sealing strip material, compression performance). These parameters are obtained through the technical documents, design drawings or actual measurements of the skylight. For example, the size of a certain model of skylight is 500mm×300mm, the track length is 400mm, the sealing strip is made of rubber, and the friction coefficient is 0.3. This step provides basic data for subsequent digital modeling and simulation through the collection of structural parameters.

[0027] Next, based on the collected structural parameters, the skylight is digitally modeled using computer-aided design software (such as SolidWorks or AutoCAD). During the modeling process, the size, track structure, and sealing strip characteristics of the skylight are accurately restored to ensure that the physical properties of the model are consistent with the actual skylight. For example, the friction coefficient of the track is set to 0.3 in the model, and the compression performance of the sealing strip is the standard value of the rubber material. This step builds a high-precision virtual model of the skylight through digital modeling to provide support for subsequent simulation analysis.

[0028] Then, based on the digital model, the closing speed simulation under different pulling forces is performed according to the predetermined manual force range (such as 5N, 10N, and 15N). Finite element analysis (FEA) software (such as ANSYS or Abaqus) is used to simulate the process of the user pushing the sunroof to close under different pulling forces, and the closing speed of the sunroof is calculated. For example, under a pulling force of 5N, the closing speed of the sunroof is 1cm / s; under a pulling force of 15N, the closing speed is 3cm / s. The simulation results are organized into a second speed mode set, covering the closing speed values ​​under different pulling forces. This step generates closing speed data in manual mode through simulation analysis, providing a basis for the subsequent controllable closing speed mode.

[0029] Finally, based on the second speed mode set, a controllable closing speed mode is generated. By mapping the closing speed value obtained by simulation with the actual operating characteristics of the sunroof, it is ensured that the sunroof can achieve a controllable closing speed according to the pulling force applied by the user in manual mode.

[0030] Furthermore, in the method provided in the embodiment of the application, after determining the opening and closing control mode of the preset skylight to be inspected, the method further includes: If the opening and closing control mode is an automatic-manual mixed mode, the first speed mode set and the second speed mode set are merged and clustered according to a preset speed consistency threshold, one of the speed modes in each clustering result is retained to generate a merged speed mode set; the closing speed controllable mode is generated by the merged speed mode set.

[0031] In the embodiment of the present application, if the opening and closing control mode of the sunroof to be inspected is preset as an automatic-manual mixed mode, that is, the sunroof supports both automatic mode and manual mode, first merge the first speed mode set (closing speed options in automatic mode) and the second speed mode set (closing speed simulation results in manual mode). For example, the first speed mode set includes 0.5cm / s, 1cm / s, and 2cm / s, and the second speed mode set includes 1cm / s, 2cm / s, and 3cm / s. The speed set obtained after merging is {0.5, 1, 2, 3}cm / s. This step integrates the closing speed options in automatic and manual modes through data merging, providing a basis for subsequent cluster analysis.

[0032] Next, the merged velocity pattern set is clustered according to a preset velocity consistency threshold (e.g., 0.2 cm / s). The purpose of the clustering algorithm is to group similar velocity values ​​into one category to reduce redundancy and retain the most representative velocity patterns. For example, if the velocity consistency threshold is 0.2 cm / s, 0.9 cm / s and 1.0 cm / s are classified into the same category, while 1.0 cm / s and 1.3 cm / s are classified into different categories. Through clustering, only one representative velocity value is retained in each category, such as retaining 1.0 cm / s from {0.9, 1.0} cm / s. This step generates a streamlined and representative merged velocity pattern set, such as {0.5, 1, 2, 3} cm / s, through cluster analysis.

[0033] Finally, a controllable closing speed pattern is generated based on the combined speed pattern set. By mapping the combined speed value with the sunroof control module, it is ensured that the sunroof can achieve a controllable closing speed according to the user's selection or applied pulling force in the automatic and manual hybrid mode.

[0034] Step S300: performing traversal matching of different speeds and different objects based on the controllable closing speed mode and the sunroof anti-pinch object library to generate a first test scene set.

[0035] In the embodiment of the present application, firstly, based on the controllable closing speed mode (i.e., the set of closing speeds supported by the sunroof in different control modes), a match is performed with the sunroof anti-pinch object library (including the types of objects that may be pinched by the sunroof and their physical properties, such as diameter, material, hardness, etc.). For example, the controllable closing speed mode includes 0.5cm / s, 1cm / s, and 2cm / s, while the anti-pinch object library includes rubber rods, plastic rods, and metal rods with diameters of 5mm, 10mm, and 15mm, respectively. This step provides basic data for subsequent test scenario generation by associating the closing speed with the object characteristics.

[0036] Next, we perform traversal matching according to the combination of different speeds and different objects. For example, we match a closing speed of 0.5 cm / s with a 5 mm rubber rod, a 10 mm plastic rod, and a 15 mm metal rod to generate a set of test scenarios. We then match a closing speed of 1 cm / s with the above objects to generate another set of test scenarios. This step ensures that each closing speed and each object type are covered through systematic combination matching, thereby generating comprehensive test scenarios.

[0037] Finally, the traversal and matching results are organized into the first test scene set.

[0038] Step S400: determining a calibration distance of the preset sunroof to be tested from a fully open state to a closed state, and generating a second test scenario set based on different closing distance configurations in each test scenario in the first test scenario set with the calibration distance as a constraint.

[0039] In the embodiment of the present application, the calibration distance of the preset sunroof to be inspected from the fully open state to the closed state is first determined. The calibration distance refers to the total travel distance of the sunroof from fully open to fully closed, which is obtained through the design parameters of the sunroof or actual measurement. For example, the calibration distance of a certain model of sunroof is 400mm.

[0040] Next, with the calibration distance as a constraint, the calibration distance is divided according to the predetermined interval step to obtain multiple distance nodes. The predetermined interval step refers to the distance interval set according to the test requirements. For example, with a step of 100mm, the calibration distance of 400mm is divided into four distance nodes of 100mm, 200mm, 300mm, and 400mm. This step provides multiple test points of closed positions for each test scenario through systematic division.

[0041] Then, the distance from the skylight to the object is configured for each test scene in the first test scene set using a plurality of distance nodes. Finally, the configured test scenes are organized into a second test scene set.

[0042] Furthermore, in the method provided in the embodiment of the application, taking the calibration distance as a constraint, generating a second test scene set for different closing distance configurations in each test scene in the first test scene set, further comprising: The calibrated distance is divided according to a predetermined interval step to obtain a plurality of distance nodes; the distance from the skylight to the object is configured for each test scene in the first test scene set using the plurality of distance nodes to generate the second test scene set.

[0043] In the embodiment of the present application, the calibrated distance is first divided according to a predetermined interval step to obtain multiple distance nodes. The predetermined interval step refers to the distance interval set according to the test requirements. For example, with a step length of 100 mm, the calibrated distance of 400 mm is divided into four distance nodes of 100 mm, 200 mm, 300 mm and 400 mm. This step provides multiple test points of the closed position for each test scenario through systematic division, ensuring that different stages of the sunroof closing process are covered.

[0044] Next, multiple distance nodes are used to configure the distance between the skylight and the object for each test scene in the first test scene set. For example, for the scene "closing speed is 0.5 cm / s, and the clamped object is a 5mm rubber rod" in the first test scene set, the distance between the skylight and the object is configured to be 100mm, 200mm, 300mm, and 400mm, respectively, to generate 4 sub-scenes. This step generates more detailed test scenes by combining each test scene with multiple distance nodes, ensuring that the trigger conditions of the anti-pinch function are tested in different closing positions.

[0045] Finally, the configured test scenarios are organized into a second test scenario set.

[0046] Step S500: testing the preset skylight to be inspected with each test scenario in the second test scenario set, and recording the triggering characteristics of the corresponding anti-pinch system to generate an anti-pinch function test result set.

[0047] In an embodiment of the present application, a preset skylight to be inspected is first tested with each test scene in a second test scene set, and data during the test is collected by a high-speed camera to obtain a test monitoring data set; then, the trigger feature of the anti-pinch system is identified based on the test monitoring data set to obtain each trigger feature corresponding to each test scene, wherein any trigger feature includes the contact time between the skylight and the object in the test scene, the speed difference before and after contact with the object, and the time point when the anti-pinch system is triggered in reverse.

[0048] Finally, the anti-pinch function test result set is generated according to the trigger features corresponding to each test scenario.

[0049] Furthermore, in the method provided in the embodiment of the application, the preset sunroof to be inspected is tested with each test scenario in the second test scenario set, and the triggering characteristics of the corresponding anti-pinch system are recorded to generate an anti-pinch function test result set, which also includes: After configuring the actual scene with each test scene in the second test scene set, the preset skylight to be inspected is tested for closing, and data in the test process is collected by a high-speed camera to obtain a test monitoring data set; based on the test monitoring data set, the trigger feature of the anti-pinch system is identified to obtain each trigger feature corresponding to each test scene, wherein any trigger feature includes the time when the skylight contacts the object in the test scene, the speed difference before and after contact with the object, and the time point when the anti-pinch system is triggered in reverse; the anti-pinch function test result set is generated with each trigger feature corresponding to each test scene.

[0050] In the embodiment of the present application, based on the second test scene set (a set of test scenes covering different closing speeds, clamping object types and closing distances generated by calibrating the distance and configuring the object distance), firstly, actual scene configuration is performed for each test scene. Specifically, according to the test requirements, a calibration tool (such as a laser rangefinder) is used to accurately set the distance between the skylight and the clamping object, and the closing speed of the skylight is adjusted (by controlling the motor parameters). Subsequently, a closing test is performed on the preset skylight to be tested (a skylight model with anti-pinch function test requirements selected from the database) to ensure that the test scene is consistent with the actual use scene.

[0051] During the closing test, the motion data of the sunroof is collected in real time by a high-speed camera (such as a Phantom series high-speed camera). The high-speed camera records the position, speed and contact of the sunroof with the clamped object at a high frame rate (such as 1000 frames per second). The collected data includes the motion trajectory of the sunroof, the time point of the contact moment, and the speed change before and after the contact. After preprocessing (such as denoising and time synchronization), these data are generated into a test monitoring data set.

[0052] Next, based on the test monitoring data set, the triggering features of the anti-pinch system are identified. Specifically, first, the image processing algorithm (such as OpenCV) is used to accurately identify the time point when the sunroof contacts the object in the test scene; secondly, the speed difference before and after the sunroof contacts the object is calculated through data analysis tools (such as MATLAB). The specific method is to perform curve fitting on the motion trajectory of the sunroof, extract the speed values ​​before and after the contact, and calculate their difference; finally, the time point when the anti-pinch system triggers the reverse direction is identified, and the response time of the anti-pinch system is determined by detecting the sudden change in the direction of the sunroof movement (such as from the closing direction to the opening direction). These triggering features (contact time, speed difference, reverse triggering time point) together reflect the response performance of the anti-pinch system in different test scenarios.

[0053] Finally, the trigger features corresponding to each test scenario are systematically organized to generate an anti-pinch function test result set. Specifically, the trigger features (contact time, speed difference, reverse trigger time point) of each test scenario are stored in the form of structured data, and data visualization tools (such as Tableau) are used to generate charts to intuitively display the anti-pinch performance under different test scenarios.

[0054] Furthermore, in the method provided in the embodiment of the application, the trigger feature identification of the anti-pinch system is performed based on the test monitoring data set to obtain each trigger feature corresponding to each test scenario, and further includes: Based on multiple frames of images in the test monitoring data set, by identifying the position relationship of the skylight in adjacent image frames, the change relationship of the skylight position over time is determined; based on the position change data and time interval data in the change relationship of the skylight position over time, the moving speed is calculated to generate a moving speed time series; based on the multiple frames of images in the test monitoring data set, the time when the skylight contacts the object in the test scene is located; based on the time when the skylight contacts the object in the test scene, the speed difference before and after the contact with the object is extracted in the moving speed time series; based on the multiple frames of images in the test monitoring data set, the time point when the skylight changes from a closed direction to an open direction is identified, and the time point when the anti-pinch system triggers the reverse direction is generated; based on the time when the skylight contacts the object in the test scene, the speed difference before and after the contact with the object, and the time point when the anti-pinch system triggers the reverse direction, each trigger feature corresponding to each test scene is generated.

[0055] In an embodiment of the present application, based on multiple frames of images in a test monitoring data set, each frame of the image is first analyzed using an image processing algorithm (such as edge detection and feature point matching technology in OpenCV). Specifically, the edge contour of the skylight is extracted using the Canny edge detection algorithm, and the position change of the skylight in adjacent frames is identified using the SIFT or ORB feature point matching algorithm. Combined with the timestamp of the image frame, the position coordinates of the skylight in each frame (such as the center point coordinates) are recorded to generate data on the relationship between the position of the skylight and time. This step provides basic position and time information for subsequent speed calculations.

[0056] Then, based on the data of the change relationship between the position of the sunroof and time, the speed calculation formula (speed = displacement / time) is used to calculate the movement speed of the sunroof. By extracting the position change (displacement) and time interval between adjacent frames, the instantaneous speed corresponding to each frame is calculated by the formula. The calculated speed value is combined with the timestamp to generate the movement speed time series (sequence data of the sunroof speed changing over time). This step converts the position change into speed information through mathematical calculation, providing key data for the subsequent analysis of the movement status of the sunroof.

[0057] Then, based on multiple frames of images in the test monitoring dataset, an object detection algorithm (such as YOLO) is used to locate the time point when the sunroof touches the object in the test scene. The specific steps include object detection for each frame, identifying the position of the sunroof and the clamped object, and determining the moment when the sunroof touches the object through image analysis technology (such as edge overlap detection). Combined with the timestamp of the image frame, the exact time point when the contact occurred is recorded.

[0058] Next, based on the time point when the skylight contacts the object, the speed values ​​before and after the contact are extracted from the moving speed time series. Specifically, according to the contact time point, the speed data of several frames before and after the contact are extracted from the speed time series, the average speed before and after the contact is calculated, and the difference is calculated to obtain the speed difference before and after the contact with the object.

[0059] Based on multiple frames of images in the test monitoring data set, a motion direction detection algorithm (such as optical flow method) is used to identify the time point when the sunroof changes from the closed direction to the open direction. By performing optical flow analysis on the image sequence, the motion vector of the sunroof is calculated, and the mutation point of the motion direction (such as from forward motion to reverse motion) is detected. Combined with the timestamp, the exact time point when the anti-pinch system triggers the reverse direction is recorded.

[0060] Finally, the trigger features corresponding to each test scenario are generated based on the time when the skylight contacts the object in the test scenario, the speed difference before and after contact with the object, and the time point when the anti-pinch system triggers the reverse direction. Specifically, the time point, speed difference, and reverse trigger time point obtained from the above analysis are stored in the form of structured data, and a complete set of trigger features is generated in combination with the configuration information of the test scenario (such as closing speed, clamped object type, etc.). This step provides comprehensive data support for the performance evaluation of the anti-pinch function through data integration and structured processing.

[0061] Step S600: Based on the anti-pinch function test result set, a sensitivity evaluation is performed on the anti-pinch system of the preset sunroof to be tested, and an anti-pinch function test result is generated.

[0062] In an embodiment of the present application, when evaluating the sensitivity of the anti-pinch system of a preset sunroof to be tested based on the anti-pinch function test result set, firstly, a sensitivity analysis is performed on the triggering features of each test scene to generate a corresponding sensitivity index. Then, a cluster analysis is performed on these sensitivity indexes to divide similar indexes into different clusters. Then, the concentrated values ​​of the test object features corresponding to each cluster are analyzed and associated with the average sensitivity value within the cluster. Finally, based on these analysis results, an anti-pinch function test result is generated to evaluate the overall performance and sensitivity of the sunroof anti-pinch system.

[0063] Furthermore, in the method provided in the embodiment of the application, based on the anti-pinch function test result set, the sensitivity of the anti-pinch system of the preset sunroof to be tested is evaluated to generate the anti-pinch function test result, and further includes: Based on each trigger feature corresponding to each test scene in the anti-pinch function test result set, trigger sensitivity analysis is performed respectively to generate each sensitivity index corresponding to each test scene; indicator consistency clustering analysis is performed on each sensitivity index to generate multiple cluster sensitivity indexes; the concentrated values ​​of the test object features in the test scenes corresponding to the multiple cluster sensitivity indexes are determined, and associated with the average values ​​corresponding to the multiple cluster sensitivity indexes to generate the anti-pinch function detection result.

[0064] In the embodiment of the present application, based on the trigger characteristics (such as contact time, speed difference and reverse trigger time point) corresponding to each test scenario in the anti-pinch function test result set, a trigger sensitivity analysis is performed using a statistical analysis method to generate a sensitivity index corresponding to each test scenario. Specifically, for each test scenario, the time difference between the reverse trigger time point and the contact time is calculated ( = reverse trigger time - contact time), and the speed difference before and after contact ( Then, according to the preset threshold range, t and To classify, for example, <50ms and The test scene with >0.2cm / s is marked as "high sensitivity", and the test scene with 50ms≤ ≤100ms and 0.1cm / s≤ ≤0.2cm / s is marked as "medium sensitivity". >100ms and <0.1cm / s is marked as "low sensitivity". Finally, a corresponding sensitivity index (such as high, medium, and low) is generated for each test scene, and the specific and value.

[0065] Next, we perform index consistency clustering analysis on each sensitivity index generated to generate multiple cluster sensitivity indexes. Specifically, we convert the sensitivity index of each test scenario (such as and ) as the feature vector, and use the index consistent clustering algorithm (such as K-means clustering) to cluster the feature vector. By selecting an appropriate number of clusters (such as K=3), the test scene is divided into several clusters (such as high sensitivity cluster, medium sensitivity cluster and low sensitivity cluster), and the number of clusters in each cluster is calculated. and The average value of is taken as the central value of the sensitivity index of the cluster. This step classifies the test scenarios according to the response mode through consistent clustering of indicators, revealing the performance distribution law of the anti-pinch system in different scenarios.

[0066] Then, the concentrated value of the test object features in the test scene corresponding to each cluster of sensitivity indicators is determined. For each cluster of test scenes, the physical properties of the clamped object (such as hardness, size, and surface roughness) are extracted, and the median of each property (such as hardness median, size median, surface roughness median) is calculated as the concentrated value of the test object features of the cluster. This step quantifies the typical physical properties of the clamped object in each cluster of test scenes through concentrated value calculation, providing data support for subsequent association analysis.

[0067] Finally, the sensitivity indexes of multiple clusters are associated with the central values ​​of the test object features to generate the anti-pinch function detection results. Specifically, the central value of the sensitivity index of each cluster ( and The average value) is associated with the corresponding test object feature set value (such as hardness median, size median, surface roughness median), and the association results are displayed in tabular form.

[0068] In the embodiments of the present application, in summary, the embodiments of the present application have at least the following technical effects: The present application collects skylight anti-pinch accident samples to cluster the features of accident objects and construct a skylight anti-pinch object library; determines a preset skylight to be inspected and a controllable closing speed mode of the preset skylight to be inspected; performs traversal matching of different speeds and different objects based on the controllable closing speed mode and the skylight anti-pinch object library to generate a first test scene set; determines the calibration distance of the preset skylight to be inspected from the fully open state to the closed state, and uses the calibration distance as a constraint to configure different closing distances in each test scene in the first test scene set to generate a second test scene set; tests the preset skylight to be inspected with each test scene in the second test scene set, and records the triggering features of the corresponding anti-pinch system to generate an anti-pinch function test result set; performs sensitivity evaluation on the anti-pinch system of the preset skylight to be inspected based on the anti-pinch function test result set to generate an anti-pinch function detection result. The present invention solves the technical problems of insufficient skylight anti-pinch test scenes and incomplete test coverage in the prior art, and achieves the technical effect of improving the comprehensiveness and accuracy of skylight anti-pinch function detection by constructing a skylight anti-pinch object library, generating a multi-scenario test set and performing sensitivity evaluation.

[0069] Embodiment 2 is based on the same inventive concept as a method for detecting the anti-pinch function of a sunroof in the above embodiment. Figure 2 As shown, the present application provides a sunroof anti-pinch function detection device, and the device and method embodiments in the present application are based on the same inventive concept. The device includes: A skylight anti-pinch object library construction module 11 is used to collect skylight anti-pinch accident samples to perform feature clustering of accident objects and construct a skylight anti-pinch object library; a skylight to be inspected determination module 12 is used to determine a preset skylight to be inspected and a controllable closing speed mode of the preset skylight to be inspected; a first test scenario generation module 13 is used to perform traversal matching of different speeds and different objects based on the controllable closing speed mode and the skylight anti-pinch object library to generate a first test scenario set; a second test scenario generation module 14 is used to determine the calibration distance of the preset skylight to be inspected from the fully open state to the closed state, and generate a second test scenario set for different closing distance configurations in each test scenario in the first test scenario set with the calibration distance as a constraint; a test result acquisition module 15 is used to test the preset skylight to be inspected with each test scenario in the second test scenario set, and record the triggering characteristics of the corresponding anti-pinch system to generate an anti-pinch function test result set; a detection result generation module 16 is used to perform sensitivity evaluation on the anti-pinch system of the preset skylight to be inspected based on the anti-pinch function test result set to generate an anti-pinch function detection result.

[0070] Furthermore, the device is also used to achieve the following functions: The features of the accident objects in the sunroof anti-pinch accident samples are analyzed to generate an accident object feature set; similarity analysis is performed on every two accident object features in the accident object feature set, and accident object features with similarities greater than a preset similarity threshold are clustered to generate multiple categories of accident object features; corresponding anti-pinch test objects are configured with the multiple categories of accident object features to generate a sunroof anti-pinch object library.

[0071] Furthermore, the device is also used to achieve the following functions: Determine the opening and closing control mode of the preset sunroof to be inspected, wherein the opening and closing control mode includes one of an automatic mode, a manual mode or an automatic and manual mixed mode; if the opening and closing control mode is an automatic mode, connect to the sunroof opening and closing automatic control terminal to collect optional closing speeds and generate a first speed mode set; generate the closing speed controllable mode with the first speed mode set.

[0072] Furthermore, the device is also used to achieve the following functions: If the opening and closing control mode is a manual mode, the structural parameters of the preset skylight to be inspected are collected, wherein the structural parameters include skylight size parameters, opening and closing track structural parameters and sealing strip structural parameters; digital modeling is performed with the structural parameters, closing speed simulation under different pulling forces is performed according to a predetermined manual force range, and a second speed mode set is generated; the closing speed controllable mode is generated with the second speed mode set.

[0073] Furthermore, the device is also used to achieve the following functions: If the opening and closing control mode is an automatic-manual mixed mode, the first speed mode set and the second speed mode set are merged and clustered according to a preset speed consistency threshold, one of the speed modes in each clustering result is retained to generate a merged speed mode set; the closing speed controllable mode is generated by the merged speed mode set.

[0074] Furthermore, the device is also used to achieve the following functions: After configuring the actual scene with each test scene in the second test scene set, the preset skylight to be inspected is tested for closing, and data in the test process is collected by a high-speed camera to obtain a test monitoring data set; based on the test monitoring data set, the trigger feature of the anti-pinch system is identified to obtain each trigger feature corresponding to each test scene, wherein any trigger feature includes the time when the skylight contacts the object in the test scene, the speed difference before and after contact with the object, and the time point when the anti-pinch system is triggered in reverse; the anti-pinch function test result set is generated with each trigger feature corresponding to each test scene.

[0075] Furthermore, the device is also used to achieve the following functions: Based on multiple frames of images in the test monitoring data set, by identifying the position relationship of the skylight in adjacent image frames, the change relationship of the skylight position over time is determined; based on the position change data and time interval data in the change relationship of the skylight position over time, the moving speed is calculated to generate a moving speed time series; based on the multiple frames of images in the test monitoring data set, the time when the skylight contacts the object in the test scene is located; based on the time when the skylight contacts the object in the test scene, the speed difference before and after the contact with the object is extracted in the moving speed time series; based on the multiple frames of images in the test monitoring data set, the time point when the skylight changes from a closed direction to an open direction is identified, and the time point when the anti-pinch system triggers the reverse direction is generated; based on the time when the skylight contacts the object in the test scene, the speed difference before and after the contact with the object, and the time point when the anti-pinch system triggers the reverse direction, each trigger feature corresponding to each test scene is generated.

[0076] Furthermore, the device is also used to achieve the following functions: Based on each trigger feature corresponding to each test scene in the anti-pinch function test result set, trigger sensitivity analysis is performed respectively to generate each sensitivity index corresponding to each test scene; indicator consistency clustering analysis is performed on each sensitivity index to generate multiple cluster sensitivity indexes; the concentrated values ​​of the test object features in the test scenes corresponding to the multiple cluster sensitivity indexes are determined, and associated with the average values ​​corresponding to the multiple cluster sensitivity indexes to generate the anti-pinch function detection result.

[0077] Furthermore, the device is also used to achieve the following functions: The calibrated distance is divided according to a predetermined interval step to obtain a plurality of distance nodes; the distance from the skylight to the object is configured for each test scene in the first test scene set using the plurality of distance nodes to generate the second test scene set.

[0078] It should be noted that the above-mentioned sequence of the embodiments of the present application is only for description and does not represent the advantages and disadvantages of the embodiments. And the above-mentioned specific embodiments of this specification are described. The processes depicted in the accompanying drawings do not necessarily require the specific order and continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0079] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

[0080] This specification and drawings are merely exemplary illustrations of the present application and are deemed to cover any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, a person skilled in the art may make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application intends to include these modifications and variations.

Claims

1. A method for detecting the anti-pinch function of a sunroof, characterized in that: include: Collect skylight anti-pinch accident samples to perform feature clustering of accident objects and build a skylight anti-pinch object library; Determining a preset skylight to be inspected and a controllable closing speed mode of the preset skylight to be inspected; Based on the controllable closing speed mode and the sunroof anti-pinch object library, traversal matching of different speeds and different objects is performed to generate a first test scene set; Determine a calibration distance of the preset sunroof to be inspected from a fully open state to a closed state, and generate a second test scenario set for different closing distance configurations in each test scenario in the first test scenario set with the calibration distance as a constraint; Testing the preset skylight to be inspected with each test scenario in the second test scenario set, and recording the triggering characteristics of the corresponding anti-pinch system to generate an anti-pinch function test result set; Based on the anti-pinch function test result set, a sensitivity evaluation is performed on the anti-pinch system of the preset sunroof to be tested to generate an anti-pinch function test result.

2. A method for detecting the anti-pinch function of a sunroof as claimed in claim 1, characterized in that: Collect skylight anti-pinch accident samples to perform feature clustering of accident objects and build a skylight anti-pinch object library, including: Performing feature analysis on the accident object of the sunroof anti-pinch accident sample to generate an accident object feature set; Performing similarity analysis on every two accident object features in the accident object feature set, clustering accident object features whose similarity is greater than a preset similarity threshold, and generating multiple categories of accident object features; Corresponding anti-pinch test objects are configured according to the characteristics of the multiple types of accident objects to generate a skylight anti-pinch object library.

3. A method for detecting the anti-pinch function of a sunroof as claimed in claim 1, characterized in that: Determining a preset skylight to be inspected and a controllable closing speed mode of the preset skylight to be inspected includes: Determining an opening and closing control mode of the preset sunroof to be inspected, wherein the opening and closing control mode includes one of an automatic mode, a manual mode, or an automatic and manual mixed mode; If the opening and closing control mode is the automatic mode, connect to the sunroof opening and closing automatic control terminal to collect optional closing speeds and generate a first speed mode set; The closing speed controllable pattern is generated with the first speed pattern set.

4. A method for detecting the anti-pinch function of a sunroof as claimed in claim 3, characterized in that: After determining the opening and closing control mode of the preset skylight to be inspected, the method further includes: If the opening and closing control mode is the manual mode, collecting the structural parameters of the preset skylight to be inspected, wherein the structural parameters include skylight size parameters, opening and closing track structural parameters and sealing strip structural parameters; Digital modeling is performed using the structural parameters, closing speed simulations are performed under different pulling forces according to a predetermined manual force range, and a second speed mode set is generated; The closing speed controllable mode is generated with the second speed mode set.

5. A method for detecting the anti-pinch function of a sunroof as claimed in claim 4, characterized in that: After determining the opening and closing control mode of the preset skylight to be inspected, the method further includes: If the opening and closing control mode is an automatic-manual hybrid mode, the first speed mode set and the second speed mode set are merged and clustered according to a preset speed consistency threshold, one of the speed modes in each clustering result is retained to generate a merged speed mode set; The closing speed controllable pattern is generated with the combined speed pattern set.

6. A method for detecting the anti-pinch function of a sunroof according to claim 1, characterized in that: The preset sunroof to be inspected is tested with each test scenario in the second test scenario set, and the triggering characteristics of the corresponding anti-pinch system are recorded to generate an anti-pinch function test result set, including: After configuring an actual scene with each test scene in the second test scene set, a closing test is performed on the preset skylight to be inspected, and data is collected during the test by a high-speed camera to obtain a test monitoring data set; Based on the test monitoring data set, trigger feature identification of the anti-pinch system is performed to obtain various trigger features corresponding to various test scenarios, wherein any trigger feature includes the time when the skylight contacts the object in the test scenario, the speed difference before and after contact with the object, and the time point when the anti-pinch system is triggered in the reverse direction; The anti-pinch function test result set is generated with each trigger feature corresponding to each test scenario.

7. A method for detecting the anti-pinch function of a sunroof as claimed in claim 6, characterized in that: Based on the test monitoring data set, trigger feature identification of the anti-pinch system is performed to obtain various trigger features corresponding to various test scenarios, including: Based on multiple frames of images in the test monitoring data set, the relationship between the positions of the skylights in adjacent image frames is identified to determine the relationship between the positions of the skylights over time. Calculate the moving speed based on the position change data and time interval data in the relationship between the position of the skylight and time, and generate a moving speed time series; Based on multiple frames of images in the test monitoring dataset, locate the time when the sunroof contacts the objects in the test scene; Based on the time when the skylight contacts the object in the test scene, the speed difference before and after the contact with the object is extracted in the moving speed time series; Based on multiple frames of images in the test monitoring data set, identify the time point when the sunroof changes from the closed direction to the open direction, and generate the time point when the anti-pinch system triggers the reverse direction; The trigger features corresponding to each test scenario are generated based on the time when the sunroof contacts the object in the test scenario, the speed difference before and after the contact with the object, and the time point when the anti-pinch system is triggered in the reverse direction.

8. A method for detecting the anti-pinch function of a sunroof as claimed in claim 7, characterized in that: Based on the anti-pinch function test result set, a sensitivity evaluation is performed on the anti-pinch system of the preset sunroof to be tested, and an anti-pinch function test result is generated, including: Perform trigger sensitivity analysis based on each trigger feature corresponding to each test scenario in the anti-pinch function test result set, and generate each sensitivity index corresponding to each test scenario; Performing index consistency cluster analysis on each of the sensitivity indicators to generate multiple cluster sensitivity indicators; The test object feature concentration values ​​in the test scenes corresponding to the multiple clusters of sensitivity indicators are determined, and are associated with the average values ​​corresponding to the multiple clusters of sensitivity indicators to generate the anti-pinch function detection result.

9. A method for detecting the anti-pinch function of a sunroof as claimed in claim 1, characterized in that: Taking the calibration distance as a constraint, generating a second test scene set for different closing distance configurations in each test scene in the first test scene set, including: Dividing the calibrated distance according to a predetermined interval step to obtain a plurality of distance nodes; The distance from the skylight to the object is configured for each test scene in the first test scene set using the multiple distance nodes to generate the second test scene set.

10. A sunroof anti-pinch function detection device, characterized in that: The device comprises: A skylight anti-pinch object library construction module is used to collect skylight anti-pinch accident samples to perform feature clustering of accident objects and build a skylight anti-pinch object library; A skylight to be inspected determining module, used to determine a preset skylight to be inspected and a controllable closing speed mode of the preset skylight to be inspected; A first test scenario generating module, configured to perform traversal matching of different speeds and different objects based on the controllable closing speed mode and the sunroof anti-pinch object library to generate a first test scenario set; A second test scenario generating module is used to determine a calibration distance of the preset sunroof to be tested from a fully open state to a closed state, and generate a second test scenario set for different closing distance configurations in each test scenario in the first test scenario set with the calibration distance as a constraint; A test result acquisition module, used to test the preset sunroof to be inspected with each test scenario in the second test scenario set, and record the triggering characteristics of the corresponding anti-pinch system to generate an anti-pinch function test result set; The detection result generating module is used to perform a sensitivity evaluation on the anti-pinch system of the preset sunroof to be tested based on the anti-pinch function test result set, and generate an anti-pinch function detection result.

Images