A method for on-line fast detection of multi-hole coordinates based on laser triangulation

Abstract

The application provides a kind of porous position coordinate on-line fast detection method based on laser triangulation, comprising: controlling detection fixture clamping mechanism to perform positioning and clamping action to the stamping part to be measured;Obtain a plurality of clamping coordinate points of the physical clamping action of the detection fixture clamping mechanism on the stamping part;Based on clamping coordinate points, a plurality of independent detection control partitions are divided;For each detection control partition, set a primary reference hole position and a plurality of following target hole positions;Control sensor to scan primary reference hole position first to extract initial center coordinates, and calculate local deformation deviation value by comparing theoretical coordinates;According to the deformation deviation value, dynamically adjust the moving speed and photoelectric sampling frequency parameters when scanning the following target hole position.The application adjusts the bottom scanning parameters and performs spatial vector compensation, filters the clamping deformation measurement error, and considers the overall on-line detection flow cycle at the same time.

Description

Technical Field

[0001] This invention relates to the field of optical three-dimensional precision measurement technology, and in particular to a rapid online detection method for multi-hole position coordinates based on laser triangulation. Background Technology

[0002] In industrial production, accurately obtaining the three-dimensional coordinates of multiple holes in stamped parts is a fundamental step in ensuring the quality of subsequent mechanical assembly. Non-contact sensors based on laser triangulation, combined with multi-axis servo modules, perform spatial sweeping. Due to their fast response and lack of contact friction with the workpiece, they have been applied in the online automatic mapping process for multiple hole dimensions. Before performing the inspection operation, the workpiece needs to be clamped and fixed using a base positioning pin and pneumatic pressure block, thereby limiting physical movement of the workpiece and standardizing the spatial geometric reference for measurement.

[0003] When a fixture applies a vertical clamping load to a thin-walled metal sheet, the mechanical force creates a localized area of ​​physical deformation diffusion around the contact area. Limited by the sheet's elastic modulus and yield strength, the actual spatial coordinates of the holes near the clamping point will deviate from their theoretical positions in the original 3D design model due to the compressive force. If this elastic deformation displacement caused by the physical positioning process is not effectively handled, the calculated hole coordinate parameters will contain systematic assembly errors.

[0004] Existing online inspection control logic typically sets a fixed sensor movement speed and photoelectric sampling frequency, sequentially traversing all the holes to be tested. When high-speed and low-frequency grasping commands are uniformly issued to match high-paced production lines, the collected sparse point cloud is insufficient to reconstruct the edge contours of holes experiencing localized pressure deformation. If a low-speed, slow-scanning and high-frequency, dense-scanning configuration is adopted for all holes, it will lead to redundant edge computing power and prolong the inspection throughput time of a single product. Balancing data acquisition accuracy and overall scanning speed while overcoming local displacement deviations caused by clamping and squeezing is a crucial engineering detail that needs to be considered in the current engineering applications of photoelectric inspection equipment. Summary of the Invention

[0005] The purpose of this invention is to provide an online rapid detection method for multi-hole coordinates based on laser triangulation, in order to solve the technical problems mentioned in the background art, such as the deviation in coordinate extraction caused by local deformation of stamped parts due to clamping and pressure, and the difficulty in balancing the production line detection cycle and local spatial mapping accuracy when using globally fixed scanning parameters.

[0006] This invention provides a rapid online detection method for multi-hole position coordinates based on laser triangulation, comprising: The control fixture clamping mechanism performs positioning and clamping actions on the stamping part to be tested; The control and drive module, carrying the laser triangulation sensor, moves it above the multi-hole position to be measured. The laser triangulation sensor is controlled to emit a laser beam toward the multi-aperture to be measured and to receive the reflected beam. The reflected beam is converted into contour point cloud data based on triangulation logic; The method further includes: Obtain multiple clamping coordinate points where the clamping mechanism of the inspection fixture applies physical clamping action to the stamping part to be tested; Multiple independent detection and control zones are divided based on each of the clamping coordinate points; For each of the aforementioned detection and control zones, a primary reference hole position and several follow-up target hole positions are set; The laser triangulation sensor is controlled to scan the primary reference hole position first and extract the initial center coordinates; By comparing the initial center coordinates with the theoretical center coordinates, the local deformation deviation value is calculated. Based on the local deformation deviation value, the moving sweep speed parameter and photoelectric sampling frequency parameter of the laser triangulation sensor are dynamically adjusted when scanning the target hole position.

[0007] Optionally, the steps for controlling the clamping mechanism of the inspection fixture to perform positioning and clamping actions on the stamping part to be tested include: The bottom positioning cylinder of the clamping mechanism of the inspection fixture pushes the base positioning pin through the reference positioning hole of the stamping part to be tested; The control side-tilting cylinder drives the movable pressure rod to move downward, and the pressure block of the fixture clamping mechanism presses the stamping part to be tested onto the support pad of the fixture clamping mechanism. Confirm that all cylinders have reached the end of their stroke lock and complete the physical locking action.

[0008] Optionally, the step of controlling the drive module to move the laser triangulation sensor above the multi-hole location to be measured includes: Read the standard 3D digital model parameters from the input system; By avoiding the spatial interference area occupied by the clamping mechanism of the fixture, a global movement trajectory connecting all holes is planned; According to the global movement trajectory, displacement control commands are continuously sent to the drive module.

[0009] Optionally, the step of receiving the reflected beam and converting it into contour point cloud data includes: The reflected light beam is captured using a photosensitive pixel array located inside the laser triangulation sensor. Extract the pixel coordinates of the energy peak formed by the reflected beam on the photosensitive pixel array; By combining the laser emission angle parameters and the receiving optical path baseline parameters, the energy peak pixel coordinates are converted into spatial depth coordinates through triangulation logic, and the contour point cloud data is generated.

[0010] Optionally, the step of extracting the hole coordinates based on the contour point cloud data includes: Calculate the depth gradient parameters of the contour point cloud data; The coordinates of the points where the depth gradient parameter crosses the set jump limit are marked as the set of hole edge points; Random sampling approximation processing logic is applied to the set of edge points of the hole to remove error noise; The least squares fitting logic is used to perform spatial circle fitting calculation on the retained edge points of the hole, and the coordinates of the center of the fitted spatial circle are extracted as the coordinates of the hole.

[0011] Optionally, the step of dividing the area into multiple independent detection and control zones based on each of the clamping coordinate points includes: Read the sheet thickness parameters, material elastic modulus parameters, and vertical clamping load calculated by air pressure of the stamping part to be tested; Based on the plate thickness parameters, material elastic modulus parameters, and vertical clamping load, the physical influence radius of the outward diffusion of the physical clamping force causing minute deformation is calculated. Multiple influence envelope circles are generated on the surface of the stamping part to be tested, with each clamping coordinate point as the center and the physical influence radius as the radius. All holes located within the same influence envelope are grouped into the same detection control zone.

[0012] Optionally, for each of the detection control zones, the step of setting a primary reference hole position and several follower target hole positions includes: Within each detection control zone, extract the hole position that is physically closest to the clamping coordinate point; Set the nearest hole position as the primary reference hole position; All holes in the detection control zone, except for the primary reference hole, are set as the target holes.

[0013] Optionally, the step of comparing the initial center coordinates with the theoretical center coordinates to calculate the local deformation deviation includes: The theoretical center coordinates of the primary reference hole position are retrieved from the pre-acquired standard three-dimensional digital model parameters; Calculate the distance difference between the initial center coordinates and the theoretical center coordinates along each three-dimensional coordinate axis. The sum of squares and square roots of the distance differences along all three-dimensional coordinate axes are used to obtain the comprehensive spatial offset, which is then used as the local deformation deviation value.

[0014] Optionally, the step of dynamically adjusting the moving sweep speed parameter and photoelectric sampling frequency parameter of the laser triangulation sensor when scanning the target hole position based on the local deformation deviation value includes: Pre-set the allowable deformation judgment threshold; When the local deformation deviation value is determined to be less than or equal to the allowable deformation determination threshold, high-speed operation parameters are sent to the drive module and low-frequency sampling parameters are sent to the laser triangulation sensor. When the local deformation deviation value is determined to be greater than the allowable deformation determination threshold, low-speed operation parameters are sent to the drive module and high-frequency sampling parameters are sent to the laser triangulation sensor.

[0015] Optionally, after sending low-speed operating parameters to the drive module and high-frequency sampling parameters to the laser triangulation sensor, the following steps are also included: The high-density point cloud data obtained by high-frequency sampling is used to reconstruct the high-precision edge contour of the target hole. The local deformation deviation value is used as a spatial compensation vector; The spatial compensation vector is superimposed on the coordinate calculation process of all subsequent target hole positions to compensate for the systematic clamping deformation error within the detection and control zone.

[0016] The present invention has achieved the following beneficial effects: This invention acquires multiple clamping coordinate points where the fixture applies physical clamping force to the stamped part and divides them into independent detection and control zones. Priority scanning is performed on the primary reference hole positions within each zone, and the local deformation deviation caused by clamping pressure is objectively quantified. The system dynamically adjusts the sweeping speed and photoelectric sampling frequency parameters of subsequent scans following the target hole positions based on the magnitude of this deformation deviation. In zones where deformation is within a safe tolerance, a high-speed, low-frequency scanning mode is used to maintain the detection progress rhythm of the automated production line. In areas where deformation exceeds the limit, a low-speed, high-frequency, dense sampling mode is automatically switched to obtain high-density edge point cloud data, and the deviation value is extracted as a spatial compensation vector to inversely compensate for the physical displacement error caused by local deformation. This feedback control strategy, based on adaptive adjustment of underlying operating parameters according to the local pressure physical state, optimizes the hardware computational load distribution, overcomes the contradiction between computational redundancy and measurement distortion inherent in single scanning modes, and balances the spatial accuracy of multi-hole automatic measurement and the overall scanning station efficiency while effectively filtering out clamping displacement deviations.

[0017] Other features and advantages of the invention will be set forth in the following description, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in the written description and the accompanying drawings.

[0018] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0019] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings: Figure 1 This is a schematic diagram of the hardware composition of the online rapid detection system for multi-hole position coordinates in an embodiment of the present invention; Figure 2 This is a flowchart illustrating the online rapid detection method for multi-hole position coordinates in an embodiment of the present invention. Figure 3 This is a flowchart of the method for detecting and controlling partitioning in an embodiment of the present invention; Figure 4 This is a flowchart of the method for extracting and processing hole position coordinates in an embodiment of the present invention; Figure 5 This is a flowchart illustrating the determination of dynamically adjusting the moving sweep speed and photoelectric sampling frequency parameters in an embodiment of the present invention. Figure 6 This is a flowchart illustrating the process of handling interference (grayscale adjustment) caused by local physical material variations in an embodiment of the present invention. Detailed Implementation

[0020] The preferred embodiments of the present invention will be described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are for illustration and explanation only and are not intended to limit the present invention.

[0021] It is understood that the following describes in detail the online rapid detection method for multi-hole position coordinates based on laser triangulation of the present invention, in conjunction with specific application scenarios and embodiments.

[0022] like Figure 1As shown, the system hardware for implementing this detection method mainly includes a support frame, a fixture clamping mechanism, a drive module, a laser triangulation sensor, and a central logic control unit. The fixture clamping mechanism includes a bottom positioning cylinder and a side-tilting cylinder, with an electromagnetic reversing valve and a magnetic position detection switch connected in series in the cylinder circuit. The drive module contains a three-axis servo-driven Cartesian coordinate module, driven by an AC servo motor and equipped with a photoelectric encoder. The laser triangulation sensor is mounted on the moving end of the drive module. The central logic control unit communicates with each hardware module and is responsible for coordinating data interaction and computation.

[0023] The online rapid detection method for multi-hole position coordinates of the present invention, such as Figure 2 As shown, the specific steps include: Step S1: Control the clamping mechanism of the inspection fixture to perform positioning and clamping actions on the stamping part to be tested.

[0024] Specifically, the bottom positioning cylinder of the clamping mechanism pushes the base positioning pin through the reference positioning hole of the stamping part under test. The central logic control unit sends a control command to the lower-level module to activate the solenoid valve corresponding to the bottom positioning cylinder of the clamping mechanism. The pneumatic circuit is inflated, pushing the piston outward and causing the base positioning pin to translate along a set axis. The base positioning pin penetrates the reference positioning hole of the stamping part under test, restricting the translational freedom of the stamping part in the horizontal plane.

[0025] The side-tilting cylinder drives the movable pressure rod downward, using the clamping block of the fixture clamping mechanism to press the stamped part to be tested onto the support pad of the fixture clamping mechanism. The central logic control unit sends a trigger command to the side-tilting cylinder, which drives the movable pressure rod to rotate around the pivot, causing the clamping block mounted at the front end of the movable pressure rod to move downward and press the stamped part to be tested firmly onto the support pad of the fixture, restricting the vertical and rotational degrees of freedom of the stamped part to be tested.

[0026] Once all cylinders have reached their stroke locking endpoint, the physical fixing action is complete. The central logic control unit reads the status signals fed back from the magnetic position detection switches of each cylinder. After determining that the signal status meets the set filtering time condition, it confirms that the mechanical clamping and physical positioning of the stamping part under test is complete.

[0027] Step S2: Obtain multiple clamping coordinate points where the clamping mechanism of the inspection fixture applies physical clamping action to the stamping part to be tested.

[0028] Specifically, the spatial layout coordinates of the clamping mechanism on the two-dimensional projection plane are read, and the centroids of the pressure block and base positioning pins that actually make physical extrusion contact with the surface of the stamped part are extracted and recorded as the corresponding clamping coordinates.

[0029] like Figure 3 As shown, in step S3, multiple independent detection and control zones are divided based on each of the clamping coordinate points.

[0030] Specifically, the plate thickness parameters, material elastic modulus parameters, and vertical clamping load calculated from the air pressure of the stamped part under test are read. The plate thickness data, material tensile yield strength, and elastic modulus parameters corresponding to the stamped part under test are extracted from the system's process database. Based on the plate thickness parameters, material elastic modulus parameters, and vertical clamping load, the physical influence radius of the outward diffusion of the physical clamping force causing minute deformation is calculated. The vertical clamping load is calculated by retrieving the static pressure data from the pneumatic circuit, and then substituted into the deflection equation using the plate thickness parameters and elastic modulus parameters to derive the physical influence radius corresponding to the set safe deformation tolerance.

[0031] The specific calculation derivation process is as follows: First, the exact calculation formula for the vertical clamping load is: ,in, For vertical compression load. The static holding pressure data of the pneumatic circuit was retrieved. This corresponds to the effective force-bearing area of ​​the piston in the cylinder. The preset pneumatic transmission efficiency coefficient is used (ranging from 0.85 to 0.95). Secondly, to meet the high computational efficiency requirements of online rapid detection of multi-hole positions, the system does not employ computationally intensive finite element mesh iteration, but instead constructs an approximate algebraic equation for radial deformation attenuation based on the local compression characteristics of the thin plate: ,in, Radial distance The deformation at the point is taken as the set safety deformation tolerance (critical deformation) when solving for the physical influence radius. The radial distribution distance from the clamping coordinate point, The elastic modulus of the material. For plate thickness parameters, The system presets the local support stiffness constant of the gauge. The physical deformation attenuation constant of the material. The base of the natural logarithm is used. The defined safety deformation tolerance (its exact range is defined as 10% to 15% of the current stamped part's hole position tolerance zone width, with specific physical values ​​between 0.01mm and 0.03mm) is substituted into the equation as the critical deformation. The analytical expression is obtained by solving the inverse logarithm. The absolute value of the radial distance obtained from this solution is directly extracted and locked by the system as the physical influence radius. The local support stiffness constant of the fixture is also specified. With the physical deformation attenuation constant of the material This method is designed for specific fixture structures and specific sheet materials. It involves pre-extracting local compressive strain cloud maps through finite element simulation analysis (CAE) for offline solution, or obtaining the data by performing local extrusion deformation tests on physical objects using a standard universal testing machine during the debugging phase. These data are then pre-set as curing parameters in the system's process database.

[0032] Multiple influence envelope circles are generated on the surface of the stamping part under test, centered on each of the clamping coordinate points and with the physical influence radius as the radius. All holes within the same influence envelope circle are assigned to the same detection control zone. When it is determined that multiple influence envelope circles overlap in the spatial projection plane, Boolean union logic is executed to merge them into a connected envelope region, and the holes falling within this connected envelope region are uniformly assigned to a single detection control zone.

[0033] Step S4: For each of the detection control zones, set the primary reference hole position and several follow target hole positions.

[0034] Specifically, within each detection control zone, the hole position with the closest physical distance to the clamping coordinate point is extracted. The spatial linear Euclidean distance between the theoretical center coordinates of each hole position to be tested within the zone and the corresponding clamping coordinate point is calculated, and the hole position with the minimum distance value is selected. This closest hole position is set as the primary reference hole position. All other holes within the detection control zone, except for the primary reference hole position, are set as the following target hole positions.

[0035] Step S5: The control drive module moves the laser triangulation sensor above the multi-hole position to be measured.

[0036] Specifically, the standard three-dimensional digital model parameters are read from the input system. Avoiding the spatial interference area occupied by the fixture clamping mechanism, a global movement trajectory connecting all holes is planned.

[0037] To ensure efficient online detection cycles, the system does not employ heuristic pathfinding probabilistic algorithms, which are computationally intensive and prone to getting trapped in local optima. Instead, it generates obstacle avoidance trajectories based on a defined 3D spatial geometric envelope boundary: First, it extracts the highest absolute mechanical coordinates of the Z-axis of each cylinder component in the fully locked state of the fixture clamping mechanism, adds a preset mechanical motion safety physical clearance (limited to 15mm to 20mm), and generates a globally unified, set safe obstacle-crossing elevation surface within the system. Subsequently, when crossing different detection holes, the system forcibly inserts a dimensionality reduction and decoupling interpolation action, first controlling the servo Z-axis to vertically lift over the safe obstacle-crossing elevation surface, then performing a two-dimensional linear translation in the XY plane to cross directly above the target hole, and finally vertically descending to the sensor's rated working focal length position. This gate-shaped explicit electromechanical geometry effectively avoids physical interference and collisions with relatively low spatial computational overhead.

[0038] According to the global movement trajectory, displacement control commands are continuously sent to the drive module. The drive module parses the displacement control commands and drives the AC servo motor to operate. In the current detection and control partition execution sequence, the drive module, carrying the laser triangulation sensor, is first controlled to move directly above the primary reference hole position.

[0039] Step S6: Control the laser triangulation sensor to prioritize scanning the primary reference hole position and extract the initial center coordinates. During this process, control the laser triangulation sensor to emit a laser beam towards the multi-hole position to be measured and receive the reflected beam.

[0040] Specifically, the laser triangulation sensor emits a linear laser beam from its internal source, which is projected onto the primary reference aperture and diffusely reflected. The reflected beam is captured by a photosensitive pixel array within the laser triangulation sensor, and the energy peak pixel coordinates formed by the reflected beam on the array are extracted. Combined with preset internal optical parameters (including laser emission angle parameters and receiving optical path baseline parameters), the energy peak pixel coordinates are converted into spatial depth coordinates using triangulation logic, and the resulting contour point cloud data is generated.

[0041] The trigonometric conversion logic is solved directly based on a defined optical geometric perspective mapping equation: .in, This is the absolute spatial depth coordinate for the calculated output; The extracted optical path baseline parameters (i.e., the physical structural distance between the center of the laser emission source and the center of the receiving optical path). The equivalent calibrated focal length for the receiving lens; This refers to the emission angle parameter between the laser emission optical axis and the principal optical axis of the receiving lens; This is the optical physical offset of the peak from the center of the photosensitive target surface (calculated by multiplying the extracted absolute pixel offset by the physical side length of a single photosensitive pixel). This explicit nonlinear geometric algebraic analytical formula rigorously and accurately maps the two-dimensional discrete pixel electrical signals to three-dimensional physical elevation, forming the absolute metric for all subsequent deformation deviation calculations.

[0042] like Figure 4 As shown, in step S7, the hole coordinates are extracted based on the contour point cloud data.

[0043] Specifically, the depth gradient parameters of the contour point cloud data are calculated. The coordinates of the points where the depth gradient parameters cross a set jump limit are marked as the set of hole edge points.

[0044] The specific feature extraction and physical determination logic is as follows: Calculate the depth gradient parameter of the contour point cloud data, which refers to extracting the absolute depth difference between two adjacent contour spatial sampling points on the elevation Z-axis along the laser scanning direction. .in, The sequence number of the sampling points in the contour space is incremented. and The first The and the first The absolute depth of each sampling point on the Z-axis elevation. The set jump limit is physically mapped to 0.5 to 0.8 times the plate thickness parameter. The physical basis for this determination is that when the laser spot crosses the solid plate surface of the stamped part and falls into the open area inside the multi-hole position to be measured, it usually produces a step-like spatial depth jump that is close to or even equal to the thickness of the entire solid plate. Therefore, when the system determines the local depth gradient... ( When the thickness parameter is specified, the edge point extraction and marking action is triggered. This rule effectively filters out minor deep wave noise interference caused by surface undulations, shallow scratches, or oxide peeling.

[0045] Random sampling approximation processing logic is applied to the set of borehole edge points to eliminate error noise. Three spatially non-collinear basic verification coordinate points are randomly selected from the set of borehole edge points, and substituted into algebraic equations to obtain the initial spatial center coordinates and the initial reference radius. The absolute radial straight-line distance from each remaining verification coordinate point in the set of borehole edge points to the initial spatial center coordinates is calculated, and the difference between this absolute radial straight-line distance and the initial reference radius is calculated. If the difference falls within a predetermined tolerance distance limit, it is accumulated and recorded by the counting module as an intra-point valid reference point. The system repeatedly performs multiple sampling and verification iterations, extracting the iteration batch containing the maximum number of intra-point valid reference points. The intra-point valid reference point data within this batch is retained for subsequent fitting calculations, while noise points exceeding the tolerance distance limit constant are discarded.

[0046] The specific geometric steps for obtaining spatial coordinates by substituting into algebraic equations are as follows: Based on three randomly selected non-collinear verification coordinate points, substitute them into the standard planar circle equation system. Eliminate the quadratic parameters containing coordinate square terms by subtracting each equation pairwise, reducing the dimension to a system of two linear algebraic equations in two variables. Then, use Cramer's rule to solve for the unique intersection coordinates of the coefficient matrix, thereby extracting the initial spatial circle center coordinates and the initial reference radius. Furthermore, the exact value of the tolerance distance limit constant is logically defined as follows: ,in, Let be the tolerance distance limit constant. This represents the single-pixel physical spatial resolution limit of the laser triangulation sensor at its current working distance (e.g., 0.015mm). This is the maximum allowable physical size of the punching tear burr in the current material in the stamping process database (e.g., 0.05mm). Limiting this range to between 0.05mm and 0.10mm effectively accommodates the discrete fluctuations in electrical signals of underlying optoelectronic devices while accurately identifying abnormal point clouds caused by punching breakage as out-of-place noise.

[0047] The least squares fitting logic is used to perform spatial circle fitting calculation on the retained edge points of the hole, and the center coordinates of the fitted spatial circle are extracted as the hole coordinates. These coordinates are then written into a register and recorded as the initial center coordinates.

[0048] Step S8: Compare the initial center coordinates with the theoretical center coordinates to calculate the local deformation deviation value.

[0049] Specifically, the theoretical center coordinates of the primary reference hole position are retrieved from the pre-acquired standard three-dimensional digital model parameters. The distance difference between the initial center coordinates and the theoretical center coordinates in each three-dimensional coordinate axis direction is calculated. The sum of the squares and the square root of the distance differences in all three-dimensional coordinate axis directions are performed to obtain the comprehensive spatial offset, which is used as the local deformation deviation value.

[0050] like Figure 5 As shown, in step S9, the moving sweep speed parameter and photoelectric sampling frequency parameter of the laser triangulation sensor are dynamically adjusted according to the local deformation deviation value when scanning the target hole position.

[0051] Specifically, a permissible deformation threshold is preset. The central logic control unit retrieves the maximum positional drift tolerance of a specific hole position, subtracts the detection and positioning machining tolerance, and obtains the remaining usable tolerance bandwidth. This tolerance bandwidth is multiplied by a preset safety factor constant and fixed as the permissible deformation threshold.

[0052] The safety factor constant is preset by the system to a dimensionless exact value between 0.65 and 0.80. The engineering basis for establishing this range is that, in addition to deducting the static assembly tolerance of the positioning datum itself, a buffer margin of 20% to 35% must be reserved for the unavoidable mechanical lag and dynamic tracking error of the electromechanical servo drive module during high-speed global sweep. This ensures that only when the local deformation caused by the physical force applied by the fixture actually occupies (or substantially consumes) more than 65% of the available bandwidth does the system recognize a physical risk of causing the final hole position to exceed tolerance. This allows for early and precise triggering of the dynamic intervention mechanism of speed reduction and high-frequency sampling, avoiding the system frequently falling into unnecessary low-speed operation due to overly sensitive threshold settings, thus disrupting the overall production line cycle.

[0053] When the local deformation deviation value is determined to be less than or equal to the allowable deformation determination threshold, high-speed operating parameters are sent to the driving module, and low-frequency sampling parameters are sent to the laser triangulation sensor. The central logic control unit sends high-speed operating parameters containing the maximum physical translation limit to the driving module, controlling the laser triangulation sensor to cross the target aperture; simultaneously, it sends low-frequency sampling parameters to the laser triangulation sensor to reduce the shutter trigger frequency of the photosensitive pixel array and extend the single exposure integration time.

[0054] When the local deformation deviation value is determined to be greater than the allowable deformation judgment threshold, low-speed operation parameters are sent to the drive module, and high-frequency sampling parameters are sent to the laser triangulation sensor. The central logic control unit sends low-speed operation parameters to the servo drive node to reconstruct the operation curve and reduce the drive frequency. Simultaneously, high-frequency sampling parameters are sent to the laser triangulation sensor to increase the high-frequency pulse working clock frequency of its emission source, shorten the single exposure integration time, and continuously refresh the frame rate to read the charge array of the photosensitive focal plane, transferring multiple frames of two-dimensional digital grayscale image matrix to memory.

[0055] Step S10: Reconstruct the high-precision edge contour of the following target hole position using the high-density point cloud data obtained by high-frequency sampling. Use the local deformation deviation value as a spatial compensation vector. Superimpose the spatial compensation vector into the coordinate calculation process of all subsequent following target holes to compensate for the systematic clamping deformation error within the detection control zone.

[0056] Specifically, the high-density two-dimensional digital grayscale image matrix acquired by following the target hole positions is converted into a contour point cloud data set through triangulation perspective conversion logic. After extracting edge scattered points and denoising, the original scan space coordinates of each target hole position are calculated through least squares fitting logic.

[0057] The central logic control unit extracts the distance difference of the primary reference hole position in each three-dimensional coordinate axis direction from the aforementioned calculation steps, combines and encapsulates it into three-dimensional vector data covering the offset components of the horizontal axis, vertical axis and elevation axis, and sets it as the spatial compensation vector of the current detection control zone.

[0058] In the coordinate compensation calculation process, the lateral coordinate components of the original scanned spatial coordinates of each following target hole position within the current detection and control zone are added with the inverse of the lateral axis offset component in the spatial compensation vector; the longitudinal coordinate components are added with the inverse of the longitudinal axis offset component; and the elevation coordinate components are added with the inverse of the elevation axis offset component. Through algebraic inverse cancellation operations, the common spatial position drift components induced by clamping physical stress are eliminated, and the final compensated absolute coordinates of each following target hole position are output.

[0059] After all coordinate compensation calculations within a detection control zone are completed, the system clears the current zone cache and controls the drive module to carry the laser triangulation sensor into the adjacent sequential detection control zone, sequentially triggering the primary reference hole position scanning, deviation value calculation, parameter adjustment response, and superimposed compensation vector operation.

[0060] Understandably, during the dynamic adjustment of the moving sweep speed parameters and photoelectric sampling frequency parameters, the servo driver transmits the raw position pulse signal generated by the photoelectric encoder of the AC servo motor to the hardware counter module. The hardware counter module accumulates the feed pulse quantity and compares it with the pre-configured interval trigger position nodes. When the machine coordinates cross the node value set in the interval trigger position node list, a trigger command level signal is generated and sent directly to the external trigger pin of the laser triangulation sensor, forcing the photosensitive pixel array to wake up and perform a single exposure. When high-frequency sampling parameters are issued, the distributed node data with shortened spacing is added to the list. Combined with the low-speed operation parameters issued from the underlying layer, the frequency and number of hardware trigger pulses generated match the image extraction frame rate allowed by the laser sensor, ensuring that the extracted spatial slices have a reliably equidistant, absolutely mechanical three-dimensional coordinate reference.

[0061] The specific underlying pulse node data generation and electromechanical conversion logic is as follows: the central logic control unit pre-retrieves the mechanical lead parameters of the ball screw in the Cartesian coordinate module. The total absolute pulse count per revolution of the photoelectric encoder for AC servo motors Calculate the pulse equivalent coefficient representing the physical displacement of a single pulse. ,in, This is the pulse equivalent coefficient. Combined with the set high-resolution spatial equidistant scanning step size. (Physical values ​​are limited to the range of 0.02mm to 0.05mm), using the formula The absolute mechanical pulse increment constant required for two consecutive exposure triggers is calculated. Among them, This is a function for rounding to the nearest integer. The system starts with the original absolute pulse coordinates of the machine corresponding to the edge of the target following hole, and follows... A one-dimensional discrete pulse sequence array is generated for the step size, and this array is directly overwritten into the high-speed comparison register of the hardware counter as the distributed node data. This conversion accurately transforms the abstract software high-frequency sampling requirements into pure hardware-level absolute pulse position triggering, overcoming the impact of bus communication latency and achieving high-precision equidistant point cloud capture.

[0062] like Figure 6As shown, further, when dealing with interference from local physical material variations, during the process of scanning the primary reference aperture and extracting the initial center coordinates, the central logic control unit performs histogram statistical calculations on the acquired two-dimensional digital grayscale image matrix to calculate the average grayscale value within the effective light spot connected pixel region. Based on the proportional difference between this average grayscale value and the set photosensitive range, the corresponding exposure time parameters and laser diode source reference current parameters are pre-calculated and configured. When it is determined that the target detection surface has high reflectivity, the laser source reference current parameter is lowered; when it is determined that the target detection surface has low reflectivity, the exposure time parameter is extended within the time limit allowed by the predetermined sampling period.

[0063] The specific decoupling adjustment corresponding mapping calculation rule is as follows: the reference gray value of the ideal photosensitive center is pre-calibrated in the system. (Set the value to 128 in 8-bit grayscale), and configure the reference exposure time for the laser triangulation sensor. With reference laser source Operating current. Extract the average grayscale value within the effectively connected pixels. And calculate the deviation ratio coefficient from the benchmark value. When judged When the target surface has high reflectivity, keep the exposure time constant and directly apply the formula. The laser source reference current is linearly and proportionally reduced to decrease the luminous flux from the physical light source, thereby suppressing the expansion of high-brightness spots; among which... The reduced operating current of the target laser source; when determining When the target surface has light-absorbing or low-reflectivity characteristics, keep the reference current constant and directly apply the formula. Extending the exposure time parameter simultaneously triggers underlying hardware-based mandatory security interception logic: where, This refers to the extended final target exposure time parameter. It controls the extended final time. The absolute value must not exceed 85% of the current underlying optoelectronic external trigger clock cycle.

[0064] After the final absolute coordinates of all inspection zones are calculated, the central logic control unit packages the coordinate data according to the specified format protocol and reports it to the external manufacturing execution system. The central logic control unit sends a reset command to the input / output module, controlling the side-tilting cylinder to exhaust air and perform a retraction action, causing the pressure block to flip upwards and move away from the stamped part to be tested; it also controls the bottom positioning cylinder of the fixture clamping mechanism to exhaust air and perform a retraction action, causing the base positioning pin to disengage from the reference positioning hole interference. The stamped part to be tested is released from the force field constraint and returns to a free state, and is then moved out of the inspection station by the external automated logistics mechanism. After receiving the message, the external system data server performs an over-limit comparison and issues commands to drive the mechanical sorting execution mechanism.

[0065] Furthermore, before the equipment is put into the formal online surveying and mapping automated cycle, the system pre-executes a benchmark mapping calibration step. A standard solid metal template is stably placed on the fixture support block and pneumatically tightened. The drive module and laser sensor are forced to use extreme low-speed operation parameters and high-frequency sampling parameters to perform global detection of the preset calibration hole positions on the standard template. The control system calculates and extracts the initial measurement coordinates of each benchmark calibration hole position. Spatial alignment is then performed by calculating the homogeneous transformation matrix between the initial measurement coordinate group and the theoretical center coordinate point set.

[0066] Considering the time-sensitive requirements of online calibration, to avoid time consumption, the calculation of the homogeneous transformation matrix did not employ a point cloud iterative approximation registration algorithm (such as the ICP algorithm). Instead, a closed-form exact algebraic solution based on singular value decomposition (SVD) was used: First, the centroids of the measured coordinate group and the theoretical center coordinate point set in three-dimensional space were calculated separately. The corresponding points of the two coordinate groups were then subtracted from their respective centroids to achieve zero-mean decentering. Next, the two decentered point set matrices were multiplied to construct the cross-covariance matrix. Perform singular value decomposition on it to obtain Using the formula The optimal 3×3 orthogonal rotation matrix directly used to correct spatial attitude errors ;in, Cross covariance matrix The left singular vector matrix; It is a diagonal matrix containing singular values; It is the transpose of the right singular vector matrix. It is a right singular vector matrix. This is the transpose of the left singular vector matrix; then, using the theoretical centroid coordinates, subtract the values ​​passed through... The 3×1 spatial translation vector is then derived by inversely calculating the centroid coordinates after matrix transformation. Finally, and By combining and assembling the components, a 4×4 standard homogeneous transformation matrix with clearly defined algebraic relationships for absolute rigid body physical transformations is generated. This algebraic operation provides a deterministic analytical solution to overcome assembly tolerance errors.

[0067] This spatial transformation matrix internally encompasses and solidifies the spatial translational transformation offsets and spatial rotational Euler angle offset components along the three-dimensional coordinate axes. In subsequent actual batch part inspection operations, the extracted parameters are subjected to inverse algebraic alignment operations using this transformation matrix, effectively filtering out systematic basis offset constants introduced by underlying mechanism assembly tolerances, etc.

[0068] The underlying hardware configuration incorporates redundant interlocking protection and interrupt control logic. A photoelectric power-off induction switch is installed at the axial limit end of the linear sliding guide. When the drive module malfunctions and the slider blocks the optical path of the photoelectric induction switch, the hardware relay circuit instantly cuts off the servo motor power supply, simultaneously triggering a non-maskable protection interrupt command, sending a brake electromagnetic braking signal to the motor, and blocking bus communication. Simultaneously, a pressure transmitter sensor module is installed in the main air supply pipeline of the pneumatic mechanism to continuously monitor the transient static pressure value of the supplied air.

[0069] The safety holding pressure limit differs from conventional empirical constants; it is dynamically calculated based on the physical clamping mechanics equations of the stamped part. The system pre-extracts the minimum normal rigid clamping mechanical load required for the stamped part under test to overcome its own elastic rebound stress and resist the micro-vibrations of high-speed electromechanical scanning. The total effective cross-sectional area of ​​the piston in the compression cylinder. and the flow resistance loss coefficient of the pneumatic pipeline network (System fixed value set to 0.85 to 0.90), through physical statics formulas The critical pressure holding value was calculated (wherein) (This refers to the 0.05MPa pipeline anti-vibration back pressure constant that is fixedly superimposed on the system), where... This is the critical pressure holding value. The derived absolute physical pressure value is written into the sensing module as the set safe pressure holding limit. The minimum normal rigid clamping mechanical load is also considered. The specific method of obtaining the data is as follows: During the benchmark calibration stage before the equipment is put into production, a thin film pressure sensor is installed at the contact interface between the gauge support pad and the standard sample, and the drive module is controlled to perform extreme high-frequency sweeping conditions. The minimum normal force peak value that can completely offset the elastic rebound of the plate and suppress the mechanical micro-vibration displacement is actually measured and extracted.

[0070] When the static pressure of the gas supply drops below the set safe pressure limit, the pressure sensing module flips its level and generates a high-level abnormal alarm signal. Upon detecting this alarm signal, the central logic control unit suspends the subsequent spatial scan command timing queue, freezes the detection schedule, and the system enters a standby response hibernation state until the main gas supply pressure returns to a safe range within the rated operating pressure parameters.

[0071] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A rapid online detection method for multi-hole position coordinates based on laser triangulation, comprising: The control fixture clamping mechanism performs positioning and clamping actions on the stamping part to be tested; The control and drive module, carrying the laser triangulation sensor, moves it above the multi-hole position to be measured. The laser triangulation sensor is controlled to emit a laser beam toward the multi-aperture to be measured and to receive the reflected beam. The reflected beam is converted into contour point cloud data based on triangulation logic; The method is characterized in that it further includes: Obtain multiple clamping coordinate points where the clamping mechanism of the inspection fixture applies physical clamping action to the stamping part to be tested; Multiple independent detection and control zones are divided based on each of the clamping coordinate points; For each of the aforementioned detection and control zones, a primary reference hole position and several follow-up target hole positions are set; The laser triangulation sensor is controlled to scan the primary reference hole position first and extract the initial center coordinates; By comparing the initial center coordinates with the theoretical center coordinates, the local deformation deviation value is calculated. Based on the local deformation deviation value, the moving sweep speed parameter and photoelectric sampling frequency parameter of the laser triangulation sensor are dynamically adjusted when scanning the target hole position.

2. The online rapid detection method for multi-hole position coordinates based on laser triangulation as described in claim 1, characterized in that, The steps for controlling the clamping mechanism of the inspection fixture to perform positioning and clamping actions on the stamping part to be tested include: The bottom positioning cylinder of the clamping mechanism of the inspection fixture pushes the base positioning pin through the reference positioning hole of the stamping part to be tested; The control side-tilting cylinder drives the movable pressure rod to move downward, and the pressure block of the fixture clamping mechanism presses the stamping part to be tested onto the support pad of the fixture clamping mechanism. Confirm that all cylinders have reached the end of their stroke lock and complete the physical locking action.

3. The online rapid detection method for multi-hole position coordinates based on laser triangulation as described in claim 2, characterized in that, The steps of controlling the drive module to move the laser triangulation sensor above the multi-hole position to be measured include: Read the standard 3D digital model parameters from the input system; By avoiding the spatial interference area occupied by the clamping mechanism of the fixture, a global movement trajectory connecting all holes is planned; According to the global movement trajectory, displacement control commands are continuously sent to the drive module.

4. The online rapid detection method for multi-hole position coordinates based on laser triangulation as described in claim 3, characterized in that, The steps of receiving the reflected beam and converting it into contour point cloud data include: The reflected light beam is captured using a photosensitive pixel array located inside the laser triangulation sensor. Extract the pixel coordinates of the energy peak formed by the reflected beam on the photosensitive pixel array; By combining the laser emission angle parameters and the receiving optical path baseline parameters, the energy peak pixel coordinates are converted into spatial depth coordinates through triangulation logic, and the contour point cloud data is generated.

5. The online rapid detection method for multi-hole position coordinates based on laser triangulation as described in claim 4, characterized in that, The step of extracting hole coordinates from the contour point cloud data includes: Calculate the depth gradient parameters of the contour point cloud data; The coordinates of the points where the depth gradient parameter crosses the set jump limit are marked as the set of hole edge points; Random sampling approximation processing logic is applied to the set of edge points of the hole to remove error noise; The least squares fitting logic is used to perform spatial circle fitting calculation on the retained edge points of the hole, and the coordinates of the center of the fitted spatial circle are extracted as the coordinates of the hole.

6. The online rapid detection method for multi-hole position coordinates based on laser triangulation as described in claim 1, characterized in that, The step of dividing the area into multiple independent detection and control zones based on each of the clamping coordinate points includes: Read the sheet thickness parameters, material elastic modulus parameters, and vertical clamping load calculated by air pressure of the stamping part to be tested; Based on the plate thickness parameters, material elastic modulus parameters, and vertical clamping load, the physical influence radius of the outward diffusion of the physical clamping force causing minute deformation is calculated. Multiple influence envelope circles are generated on the surface of the stamping part to be tested, with each clamping coordinate point as the center and the physical influence radius as the radius. All holes located within the same influence envelope are grouped into the same detection control zone.

7. The online rapid detection method for multi-hole position coordinates based on laser triangulation as described in claim 6, characterized in that, For each of the aforementioned detection control zones, the steps of setting a primary reference hole position and several follower target hole positions include: Within each detection control zone, extract the hole position that is physically closest to the clamping coordinate point; Set the nearest hole position as the primary reference hole position; All holes in the detection control zone, except for the primary reference hole, are set as the target holes.

8. The online rapid detection method for multi-hole position coordinates based on laser triangulation as described in claim 7, characterized in that, The step of calculating the local deformation deviation by comparing the initial center coordinates with the theoretical center coordinates includes: The theoretical center coordinates of the primary reference hole position are retrieved from the pre-acquired standard three-dimensional digital model parameters; Calculate the distance difference between the initial center coordinates and the theoretical center coordinates along each three-dimensional coordinate axis. The sum of squares and square roots of the distance differences along all three-dimensional coordinate axes are used to obtain the comprehensive spatial offset, which is then used as the local deformation deviation value.

9. The online rapid detection method for multi-hole position coordinates based on laser triangulation as described in claim 8, characterized in that, The step of dynamically adjusting the moving sweep speed parameter and photoelectric sampling frequency parameter of the laser triangulation sensor when scanning the target hole position based on the local deformation deviation value includes: Pre-set the allowable deformation judgment threshold; When the local deformation deviation value is determined to be less than or equal to the allowable deformation determination threshold, high-speed operation parameters are sent to the drive module and low-frequency sampling parameters are sent to the laser triangulation sensor. When the local deformation deviation value is determined to be greater than the allowable deformation determination threshold, low-speed operation parameters are sent to the drive module and high-frequency sampling parameters are sent to the laser triangulation sensor.

10. The online rapid detection method for multi-hole position coordinates based on laser triangulation as described in claim 9, characterized in that, After sending low-speed operating parameters to the drive module and high-frequency sampling parameters to the laser triangulation sensor, the following steps are also included: The high-density point cloud data obtained by high-frequency sampling is used to reconstruct the high-precision edge contour of the target hole. The local deformation deviation value is used as a spatial compensation vector; The spatial compensation vector is superimposed on the coordinate calculation process of all subsequent target hole positions to compensate for the systematic clamping deformation error within the detection and control zone.

Images