Missile outer surface quality intelligent total inspection method and device

Abstract

The application discloses a kind of missile outer surface quality intelligent total inspection method and device, wherein, the method comprises: constructing global measurement coordinate system;According to the measurement of optical tracker to the missile to be inspected, the overall pose of the missile to be inspected in global measurement coordinate system is obtained;According to the missile outer surface three-dimensional point cloud of each inspection section and the overall pose of the missile to be inspected in global measurement coordinate system, the point cloud splicing compensation value of each inspection section is obtained;According to the point cloud splicing compensation value of each inspection section, the point cloud splicing accuracy of each section of outer surface is calibrated, and the missile full outer surface three-dimensional model is reconstructed and generated;The problem image screened is mapped to the missile full outer surface three-dimensional model;According to the missile full outer surface three-dimensional model, the geometric dimension value, surface appearance and identification inspection result are obtained, and the geometric dimension value, surface appearance and identification inspection result are compared with the preset total inspection process requirement to obtain total inspection result.The application has the advantages of high-precision measurement, high measurement point density, high inspection efficiency, strong flexible combination expansion capability.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of missile assembly inspection, and particularly relates to a missile external surface quality intelligent assembly inspection method and device. BACKGROUND

[0002] As a typical large-size product, the structure size of a missile is generally in the range of several meters to tens of meters, and the assembly precision is generally required to be in the sub-millimeter and angle second order. Meanwhile, missile external surface quality assembly inspection has the characteristics of many measured parameters and high detection precision requirements. Carrying out missile assembly external surface quality assembly inspection, completing the measurement of overall geometric size and shape error, and external surface quality are the key to ensuring the overall assembly quality of the missile and the basic prerequisite for ensuring the delivery of the missile product assembly.

[0003] The missile external surface quality assembly inspection content is mainly divided into three types: (1) geometric size measurement: overall length, front and rear body diameter, wing span size, horizontal measurement (head angle, rudder installation angle / angle); (2) surface appearance quality: coating appearance is smooth and bright without exposed bottom, paint surface is smooth without peeling and other defects, parts are free of missing, wrong assembly and damage scratches, and identification is free of damage, skew and font spacing color error; (3) identification text quality: identification content is correct without missing.

[0004] The existing geometric size measurement is still generally manual operation of laser tracker or articulated arm, or even manual fixture for local measurement, and the missile key geometric size is calculated by part of feature point coordinate values. The laser tracker measurement is generally disturbed by obstruction and light interruption, and the single-station spherical coordinate measurement mode can only realize sequential measurement of single-point space coordinates in principle, the view angle is limited, the function is single, there are problems such as low sampling density, inconsistency between feature point coordinate value measurement state and theoretical state, and the missile accurate size cannot be obtained. On the other hand, the surface appearance quality and identification text quality inspection have long followed manual visual inspection, after inspection, the camera is held and aligned for shooting, the on-site assembly inspection returns to the office for repeated piece-by-piece review, manual knocking and recording, and the management is marked, which has caused the problems of inconsistent inspection execution standards, strong dependence on personnel experience, missed judgment, cumbersome and inefficient appearance record picture quality backtracking, and large process labor intensity.

[0005] In recent years, the missile structure integration and large-scale development are rapidly developing, the product size is continuously increasing, and the measurement range requirement is becoming larger and larger. Meanwhile, the performance of the missile is increasingly strict to the aerodynamic shape and surface process design, and the quality assurance requirement is also increasing. The aerospace manufacturing enterprises have higher requirements for the measurement space range, measurement accuracy and measurement efficiency of the assembly general inspection technology. In view of the problems and disadvantages of the traditional operation method, the digital measurement and inspection technology for the missile is also proposed and gradually applied. However, the existing public patents and paper documents describe the digital level measurement and other measurement methods, which are still based on a small number of process control point coordinate measurements, and lack of comprehensive and full-range measurement perception of the missile entity shape, pose and mutual relationship. In addition, the current research is mostly for the improvement of single inspection type or the upgrading of size measurement method or surface quality detection means. However, the diversity of the missile shape surface quality general inspection content and the complexity of the inspection requirement make it difficult for the existing single measurement system to independently complete the inspection task. The digital transformation of the local inspection content or the combination of multiple independent measurement / inspection systems cannot effectively improve the general inspection process efficiency, and at the same time, it will introduce new problems such as coordination of multiple inspection systems, conversion and recording of multi-source quality data. SUMMARY

[0006] The technical problem solved by the present application is to overcome the shortcomings of the prior art and provide a missile shape surface quality intelligent general inspection method and device, which has the advantages of large-scale, high-precision measurement, high measurement point density, high inspection efficiency and strong flexible combination expansion capability.

[0007] The technical problem solved by the present application is to overcome the shortcomings of the prior art and provide a missile shape surface quality intelligent general inspection method and device, which has the advantages of large-scale, high-precision measurement, high measurement point density, high inspection efficiency and strong flexible combination expansion capability. A missile shape surface quality intelligent general inspection method comprises the following steps: receiving a general inspection task and calling standard inspection data, setting multiple optical trackers on the outer periphery of a missile support frame, calibrating the multiple optical trackers, and constructing a global measurement coordinate system; transporting a missile to be inspected, which is fixed on the missile support frame, to a measurement area, and obtaining the overall pose of the missile to be inspected in the global measurement coordinate system according to the measurement of the optical tracker; setting the motion path of a mobile robot, the point cloud scanning track of a laser scanner driven by a mechanical arm, and the photographing position point of a monocular / dual camera in each inspection section; completing the data acquisition of the missile shape surface three-dimensional point cloud, the overall surface image and the identification image of each inspection section in a predetermined order by using the mobile robot, the laser scanner and the monocular / dual camera; obtaining the point cloud splicing compensation value of each inspection section according to the missile shape surface three-dimensional point cloud of each inspection section and the overall pose of the missile to be inspected in the global measurement coordinate system; calibrating the point cloud splicing accuracy of each section of the shape surface according to the point cloud splicing compensation value of each inspection section, and reconstructing a missile full shape surface three-dimensional model; mapping the problem image screened out according to the standard inspection data to the missile full shape surface three-dimensional model; obtaining the geometric dimension value, the surface appearance and the identification inspection result according to the missile full shape surface three-dimensional model; and comparing the geometric dimension value, the surface appearance and the identification inspection result with the preset general inspection process requirement to obtain a general inspection result.

[0008] In the intelligent general inspection method for the missile outer surface quality, the general inspection task calls standard inspection data, including inspection process, missile design three-dimensional model, coating surface appearance standard image data set, part installation standard image data set, and identification standard image data set.

[0009] In the intelligent general inspection method for the missile outer surface quality, the overall pose of the missile to be inspected in the global measurement coordinate system is obtained according to the measurement of the optical tracker, including: measuring the static three-dimensional pose of the vertex of the head of the missile to be inspected, the target centers of the targets on both sides of the missile support frame in the global measurement coordinate system by using the optical tracker, fitting and calculating the axial normal vector of the missile, and combining the missile design three-dimensional model to determine the overall pose of the missile to be inspected in the global measurement coordinate system.

[0010] In the intelligent general inspection method for the missile outer surface quality, the data acquisition of the missile outer surface three-dimensional point cloud, the overall surface image, and the identification image of each inspection section is completed in a predetermined order by using the mobile robot, the laser scanner, and the monocular / dual-camera, and the point cloud splicing compensation value of each inspection section is obtained according to the missile outer surface three-dimensional point cloud of each inspection section and the overall pose of the missile to be inspected in the global measurement coordinate system, including: step S31: according to the set motion path, the mobile robot moves to the first inspection section, the laser scanner continuously and dynamically scans the outer surface point cloud data around the missile circular surface, when the target is scanned to the boundary of the first inspection section, the dual-camera identifies the target and calculates the three-dimensional pose of the target center relative to the dual-camera coordinate system, the three-dimensional pose of the target center relative to the dual-camera coordinate system is converted through the laser scanner coordinate system to obtain the three-dimensional pose of the target center in the global measurement coordinate system, the three-dimensional pose of the target center in the global measurement coordinate system is compared with the static three-dimensional pose of the target center in the global measurement coordinate system to obtain a deviation value, and the deviation value is set as the point cloud splicing compensation value of the first inspection section; after the scanning is completed, the mobile robot and the mechanical arm drive the monocular camera to collect the overall appearance image of the outer surface of the first inspection section; if there are part and identification inspection contents in this inspection section, the monocular camera and the dual-camera are simultaneously used to identify and photograph the relevant areas; step S32: when all the contents are collected, move to the next inspection section, and repeat step S31 until all the inspection sections are collected.

[0011] In the intelligent general inspection method for the missile outer surface quality, the problem images are mapped to the three-dimensional model of the overall outer surface of the missile according to the standard inspection data, including: matching and comparing the coating surface appearance standard image data set, the part installation standard image data set, and the identification standard image data set, screening out the problem images, and mapping the problem images to the outer surface of the entity model based on the point cloud three-dimensional pose associated by the data fusion algorithm.

[0012] In the intelligent general inspection method for the missile outer surface quality, the geometric dimension value includes the overall length, the front and rear body diameter, and the wing span dimension.

[0013] In the intelligent overall inspection method for the missile outer surface quality, the optical tracker tracks the mark points on the laser scanner in the measurement space, and when at least three mark points on the scanner are photographed by two optical trackers, the three-dimensional coordinate values of the mark points are calculated in real time to determine the three-dimensional space pose of the laser scanner in the global measurement coordinate system.

[0014] In the intelligent overall inspection method for the missile outer surface quality, the algorithm calculation process of the reconstruction includes point cloud noise reduction, point cloud filtering, point cloud splicing and surface reconstruction.

[0015] In the intelligent overall inspection method for the missile outer surface quality, the number of optical trackers is determined according to the envelope space involved in the missile geometric size measurement and the visibility analysis that the mark points on the laser scanner can be photographed by at least two optical trackers.

[0016] An intelligent overall inspection device for a missile outer surface quality comprises: a first module configured to receive an overall inspection task call standard inspection data and construct a global measurement coordinate system; a second module configured to obtain the overall pose of a missile to be inspected in the global measurement coordinate system; set the motion path of a mobile robot in each inspection section, the point cloud scanning track of a laser scanner driven by a mechanical arm, and the photographing position point of a monocular / dual-camera; a third module configured to obtain a point cloud splicing compensation value of each inspection section according to the three-dimensional point cloud of the missile outer surface of each inspection section and the overall pose of the missile to be inspected in the global measurement coordinate system; a fourth module configured to calibrate the point cloud splicing accuracy of each section of the outer surface according to the point cloud splicing compensation value of each inspection section, and reconstruct a three-dimensional model of the entire outer surface of the missile; and a fifth module configured to obtain geometric size values, surface appearance and identification inspection results according to the three-dimensional model of the entire outer surface of the missile, compare the geometric size values, surface appearance and identification inspection results with preset overall inspection process requirements, and obtain an overall inspection result.

[0017] Compared with the prior art, the present application has the following beneficial effects:

[0018] (1) The present application integrates digital inspection technologies such as large-size precision measurement and surface quality inspection, integrates motion control and measurement control into an automatic equipment, interconnects an industrial internet platform, and constructs an overall inspection mode of digital multi-system collaborative geometric measurement and quality inspection, realizes a one-stop solution of large-size product outer surface quality multi-inspection content full coverage, process automatic acquisition, and result intelligent evaluation, and fills the gap of intelligent overall inspection of missile outer surface quality.

[0019] (2) The multi-sensor cooperative combination measurement system constructed by the application, and the omnidirectional mobile platform and the mechanical arm selected as the executing mechanism of the measurement system, show sufficient flexibility and combination expansion capability: the measurement range has sufficient scalability, and can adapt to large size shape features; meanwhile, high measurement accuracy, measurement point density and measurement efficiency are ensured in the full range, and significant changes of normal vector and curvature are adapted. Thus, the bottleneck contradiction that a single device cannot simultaneously consider the range of measurement, measurement resolution, and complete on-site large space long distance overall measurement positioning and limited space near distance high precision, high resolution topographic data acquisition is solved.

[0020] (3) The missile outer surface quality intelligent general inspection system designed by the application completes the integration of general inspection collection, general inspection processing, general inspection quality evaluation and other processes, and realizes independent analysis, rapid execution and rapid evaluation of general inspection.

[0021] (4) The application effectively solves the problems of low measurement accuracy, strong dependence on personnel experience, high labor intensity, difficulty in tracing inspection data and other disadvantages of the existing general inspection method, and has the effects of large scale, high precision, high efficiency and low cost. BRIEF DESCRIPTION OF DRAWINGS

[0022] Various other advantages and benefits will become apparent to those of ordinary skill in the art upon reading the following detailed description of the preferred embodiments. The accompanying drawings are included to provide a description of the preferred embodiments and are not intended to limit the scope of the application. Moreover, like reference numerals designate like parts throughout the several views in the drawings. In the drawings:

[0023] Figure 1 is a general composition diagram of the missile outer surface quality intelligent general inspection system provided by the embodiment of the application;

[0024] Figure 2 is a hardware layout diagram of the equipment layer of the missile outer surface quality intelligent general inspection system provided by the embodiment of the application;

[0025] Figure 3 is a composition diagram of each measurement sensor carried by the mechanical arm provided by the embodiment of the application;

[0026] Figure 4 is a flowchart of the missile outer surface quality intelligent general inspection method provided by the embodiment of the application. DETAILED DESCRIPTION

[0027] Exemplary embodiments of the present disclosure will be described in greater detail below with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it is understood that the present disclosure can be embodied in various forms without being limited by the embodiments set forth herein. Rather, these embodiments are provided so that the present disclosure will be thorough and complete, and will fully convey the scope of the present disclosure to those skilled in the art. Note that the embodiments in the present disclosure and the features in the embodiments can be combined with each other without conflict. The present disclosure will be described in detail below with reference to the accompanying drawings and in conjunction with the embodiments.

[0028] Figure 4 is a flow chart of a missile external shape quality intelligent general inspection method provided by an embodiment of the present disclosure. As shown in Figure 4 , the method comprises:

[0029] receiving general inspection task calling standard inspection data, setting a plurality of optical trackers on a missile support frame, calibrating the plurality of optical trackers, and constructing a global measurement coordinate system;

[0030] transporting the missile to be inspected fixed on the missile support frame to a measurement area, obtaining the overall pose of the missile to be inspected in the global measurement coordinate system according to the measurement of the optical tracker on the missile to be inspected, and setting the motion path of the mobile robot, the point cloud scanning track of the laser scanner driven by the mechanical arm, and the monocular / dual camera shooting position point of each inspection section;

[0031] completing the data acquisition of the missile external shape three-dimensional point cloud, overall surface image, and identification image of each inspection section by the mobile robot, laser scanner, and monocular / dual camera in a preset order, and obtaining the point cloud splicing compensation value of each inspection section according to the missile external shape three-dimensional point cloud of each inspection section and the overall pose of the missile to be inspected in the global measurement coordinate system;

[0032] calibrating the point cloud splicing accuracy of each section of the external shape according to the point cloud splicing compensation value of each inspection section, and reconstructing to generate a missile full external shape three-dimensional model; and mapping the problem image screened out according to the standard inspection data to the missile full external shape three-dimensional model;

[0033] obtaining the geometric dimension value, surface appearance, and identification inspection result according to the missile full external shape three-dimensional model, comparing the geometric dimension value, surface appearance, and identification inspection result with the preset general inspection process requirement to obtain a general inspection result.

[0034] Specifically, the method comprises the following steps:

[0035] Step S1: general inspection preparation: the console 4 receives the general inspection task, automatically retrieves the inspection process, missile design three-dimensional model, coating surface appearance standard image dataset, component installation standard image dataset, and identification standard image dataset according to the missile product code to be inspected. A plurality of optical trackers are arranged in order around the missile support frame, the optical trackers are calibrated, and a global measurement coordinate system is constructed.

[0036] Step S2: general inspection planning: the missile 7 to be inspected is transported to the measurement area and fixed on the missile support frame 5, as shown in Figure 2 The optical tracker 1 is used to measure the static three-dimensional pose of the missile head vertex, the target centers of the target markers 6 on both sides of the support frame 5 in the global measurement coordinate system, the missile axis vector is calculated by fitting, the overall pose of the missile in the global measurement coordinate system is determined in combination with the missile design three-dimensional model, and the inspection sections are further set according to the arrangement and layout of the target markers 6 on both sides of the support frame 5. The motion path of the mobile robot 3, the scanning trajectory of the laser scanner 10 in each inspection section driven by the mechanical arm 2, the photographing position of the monocular camera 8, and the photographing position of the binocular camera 9 are simulated and planned, as shown in Figure 3 The laser scanner 10, the monocular camera 8, and the binocular camera 9 are connected with the mechanical arm 2. The binocular camera 9 is located above the laser scanner 10, and the monocular camera 8 is located above the binocular camera 9.

[0037] Step S3: general inspection acquisition: according to the planned motion path, the mobile robot moves to the first inspection section, the laser scanner continuously and dynamically scans the outer surface to obtain the point cloud data, when the target marker is scanned to the boundary of the first inspection section, the binocular camera identifies the target marker and calculates the three-dimensional pose of the target center relative to the binocular camera coordinate system, and further continuously obtains the three-dimensional pose of the target center in the global measurement coordinate system through the laser scanner coordinate system conversion. The deviation is compared with the static three-dimensional pose of the corresponding target center obtained in step S2, and the deviation value is set as the point cloud splicing compensation value of the outer surface of the first inspection section. After scanning is completed, the mobile robot and the mechanical arm drive the monocular camera to collect the overall appearance image of the outer surface of the first inspection section. If there are component and identification inspection contents in this inspection section, the monocular camera and the binocular camera are simultaneously used to identify and photograph the relevant area. When all the contents are collected, move to the next inspection section and repeat the above process until all the inspection sections are collected.

[0038] Step S4: Total inspection calculation fusion: based on three-dimensional point cloud splicing reconstruction, multi-type surface defect detection, identification recognition detection and other algorithms, the collected point clouds and images of each inspection section are calculated and processed simultaneously. The point cloud splicing compensation value of each inspection section is used to calibrate the point cloud splicing accuracy of the outer surface, and a full-outer-surface three-dimensional entity model of the missile is reconstructed and output. The standard image dataset of the coating surface appearance, the standard image dataset of the component installation, and the standard image dataset of the identification are matched and compared, the problem images are screened out, and the problem images are mapped to the entity model outer surface based on the corresponding point cloud three-dimensional pose associated by the data fusion algorithm.

[0039] Step S5: Total inspection quality evaluation: based on the full-outer-surface three-dimensional entity model of the missile, the full length, the front and rear body diameter, the wing surface span size, etc. are calculated according to various size formulas, the geometric size, surface appearance, and identification text inspection results are compared with the total inspection process requirements one by one to determine the results, the total inspection results such as size, outer surface defect, and text error are summarized to the three-dimensional display of the entity model, the problem area is highlighted in red, a total inspection quality report of the missile outer surface is generated simultaneously, and uploaded to the industrial internet platform.

[0040] The optical tracker tracks the identification points on the laser scanner in the measurement space. When at least three identification points on the scanner are photographed by two optical trackers, the three-dimensional coordinate values of the identification points are calculated in real time to determine the three-dimensional space pose of the laser scanner in the global measurement coordinate system. Through the unified conversion of the local outer surface measurement data of the laser scanner to the global measurement coordinate system, the splicing of the local measurement data in different poses is realized. The number of optical trackers is determined according to the envelope space involved in the missile geometric size measurement and the visibility analysis that the identification points on the laser scanner can be photographed by at least two optical trackers, and can be dynamically and conveniently adjusted according to the spatial size change of the missile.

[0041] The support frame can be selected from fixed supports or movable frame cars, and the target can also be adjusted in space layout and quantity as needed. The mobile robot has the functions of omnidirectional movement, active obstacle avoidance and early warning.

[0042] The calculation process of the three-dimensional point cloud splicing reconstruction algorithm includes point cloud noise reduction, point cloud filtering, point cloud splicing, and surface reconstruction processes.

[0043] The multi-type surface defect detection algorithm adopts deep learning defect detection technology, and the calculation process includes defect image dataset processing, defect detection model training, and defect detection processes.

[0044] The deep learning text detection algorithm adopts deep learning text detection technology, and the calculation process includes identification text region extraction, identification character recognition, and identification information system matching processes.

[0045] The data fusion algorithm calculation process includes fusion and unification of local measurement coordinate system and global measurement coordinate system, fusion of image data and point cloud data, and measurement uncertainty estimation after fusion.

[0046] The coating paint appearance standard image data set, the part installation standard image data set, and the identification standard image data set are formed by image acquisition, labeling, and accumulation of the qualified missile through the step S3.

[0047] As shown in Figure 1 The missile outer surface quality intelligent general inspection system includes a device layer, an algorithm layer, and an application layer. The device layer includes a laser scanner, an optical tracker, a monocular camera, a binocular camera, a mechanical arm, a mobile robot, a control console, a support frame, and a target. The algorithm layer includes a measurement trajectory planning algorithm, a three-dimensional point cloud stitching reconstruction algorithm, a multi-type surface defect detection algorithm, an identification recognition detection algorithm, and a multi-quality data fusion evaluation algorithm. The application layer includes a general inspection task management module, a general inspection calibration module, a general inspection motion monitoring module, a general inspection acquisition control module, and a quality data analysis module.

[0048] Specifically, the algorithm layer and the device layer interface interact to perform real-time measurement data acquisition and execute control instruction issuance.

[0049] Specifically, the application layer interacts with the industrial internet platform to perform general inspection task issuance reception and quality report uploading, and interconnects with the factory management system.

[0050] General inspection preparation: the control console starts running the missile outer surface quality intelligent general inspection system and completes system self-inspection, and is ready for inspection. When the general inspection task management module receives the general inspection task issued by the industrial internet platform, the inspection process, the missile design three-dimensional model, the coating paint appearance standard image data set, the part installation standard image data set, and the identification standard image data set are automatically retrieved according to the product code of the missile to be inspected. A plurality of optical trackers are arranged in order around the missile support frame, the general inspection calibration module is run to calibrate the optical trackers, and a global measurement coordinate system is constructed.

[0051] General inspection planning: the missile to be inspected is transported to the measurement area and fixed on the support frame. The general inspection calibration module is run to measure the static three-dimensional pose of the missile head vertex and the target centers of the two target markers on the support frame in the global measurement coordinate system using the optical tracker, fit and calculate the missile axis normal vector, run the general inspection motion monitoring module, determine the overall pose of the missile in the global measurement coordinate system in combination with the missile design three-dimensional model, further set the inspection sections according to the arrangement and layout of the two target markers on the support frame, and simulate and plan the motion path of the mobile robot, the scanning trajectory of the laser scanner in each inspection section driven by the mechanical arm, and the photographing position point of the monocular / binary camera.

[0052] Total inspection collection: run the total inspection collection control module, follow the planned motion path, the mobile robot moves to the first inspection section, the laser scanner continuously and dynamically scans the missile camber surface to obtain the outer surface point cloud data, when the target is scanned to the boundary of the first inspection section, the binocular camera identifies the target and calculates the target center relative to the binocular camera coordinate system three-dimensional pose, further continuously through the laser scanner coordinate system conversion to obtain the three-dimensional pose of the target center in the global measurement coordinate system, compare the deviation with the corresponding target center static three-dimensional pose obtained in step S1, set the deviation value as the outer surface point cloud splicing compensation value of the first inspection section; after scanning, the mobile robot and mechanical arm drive the monocular camera to collect the overall appearance image of the outer surface of the first inspection section. If there are parts and identification inspection contents in this inspection section, the monocular camera and binocular camera are used to identify and photograph the relevant area at the same time. When all the contents are collected, move to the next inspection section and repeat the above process until all the inspection sections are collected.

[0053] Total inspection calculation fusion: run the quality data analysis module, based on three-dimensional point cloud splicing reconstruction, multi-type surface defect detection, identification detection algorithm, synchronous calculation and processing of each inspection section point cloud and image collected. Use the point cloud splicing compensation value of each inspection section to calibrate the outer surface point cloud splicing accuracy, reconstruct and output the missile full outer surface three-dimensional entity model; match and compare the coating appearance standard image data set, part installation standard image data set and identification standard image data set, screen out problem images, and run multi-quality data fusion evaluation algorithm to map the problem images to the entity model outer surface.

[0054] Total inspection quality evaluation: in the quality data analysis module, based on the missile full outer surface three-dimensional entity model, calculate the full length, front and rear body diameter, wing span size, etc. according to various size formulas, synchronize the geometric size, surface appearance, identification text inspection results with the total inspection process requirements one by one to determine the results, and collect the total inspection results size, outer surface defect, text error, etc. to the entity model three-dimensional display, the problem area is highlighted in red, the quality data analysis module synchronously generates the missile outer surface total inspection quality report, and uploads it to the industrial internet platform.

[0055] The embodiment also provides a missile external surface quality intelligent general inspection device, which comprises: a first module, which is used for receiving general inspection task calling standard inspection data and constructing a global measurement coordinate system; a second module, which is used for obtaining the overall pose of a to-be-inspected missile in the global measurement coordinate system; setting the motion path of each inspection section mobile robot, the point cloud scanning track of a laser scanner driven by a mechanical arm, and the monocular / dual-camera camera shooting position point; a third module, which is used for obtaining each inspection section point cloud splicing compensation value according to the missile external surface three-dimensional point cloud of each inspection section and the overall pose of the to-be-inspected missile in the global measurement coordinate system; a fourth module, which is used for calibrating the external surface section point cloud splicing accuracy according to the point cloud splicing compensation value of each inspection section, and reconstructing a missile full external surface three-dimensional model; and a fifth module, which is used for obtaining geometric dimension value, surface appearance and identification inspection results according to the missile full external surface three-dimensional model, comparing the geometric dimension value, surface appearance and identification inspection results with preset general inspection process requirements to obtain general inspection results.

[0056] The application fuses and applies digital inspection technologies such as large-size precision measurement and surface quality inspection, integrates motion control and measurement control integrated automation equipment, interconnects an industrial internet platform, and builds a general inspection mode of digital multi-system collaborative geometric measurement and quality inspection, so that a one-stop solution scheme of large-size product external surface quality multi-inspection content full coverage, process automatic acquisition and result intelligent evaluation is realized, and the blank of missile external surface quality intelligent general inspection is filled.

[0057] The multi-sensor collaborative combination measurement system built in the application selects an omnidirectional mobile platform and a mechanical arm as an execution mechanism of the measurement system, and shows sufficient flexibility and combination expansion capability: the measurement range has sufficient scalability, and can adapt to large-size shape features; meanwhile, high measurement precision, measurement point density and measurement efficiency are ensured in the full range, and significant changes of normal vectors and curvatures are adapted. Therefore, the bottleneck contradiction that a single device cannot simultaneously complete on-site large-space long-distance overall measurement positioning and limited-space short-distance high-precision and high-resolution topographic data acquisition in terms of measurement range and measurement resolution is solved.

[0058] The missile external surface quality intelligent general inspection system designed in the application completes general inspection acquisition, general inspection processing, general inspection quality evaluation and other process integration, and realizes general inspection autonomous analysis, rapid execution and rapid evaluation.

[0059] The application effectively solves the drawbacks of low measurement precision, strong dependence on personnel experience, large labor intensity, difficult inspection data tracing and other drawbacks of the existing general inspection method, and has the effects of large scale, high precision, high efficiency and low cost.

[0060] Although the present application has been disclosed with reference to the preferred embodiments, it is not intended to limit the present application, and any person skilled in the art can make possible changes and modifications to the technical solutions of the present application using the disclosed methods and technical contents without departing from the spirit and scope of the present application. Therefore, any simple modification, equivalent change and modification made to the above embodiments according to the technical essence of the present application without departing from the technical solutions of the present application shall fall within the protection scope of the technical solutions of the present application.

Claims

1. An intelligent general inspection method for missile outer surface quality, characterized in that The method comprises the following steps: Receiving total inspection task calling standard inspection data, setting multiple optical trackers on the periphery of the missile support frame, calibrating the multiple optical trackers, and constructing a global measurement coordinate system; Transferring the missile to be inspected fixed on the missile support frame to a measurement area, obtaining the overall pose of the missile to be inspected in the global measurement coordinate system according to the measurement of the optical tracker on the missile to be inspected; Setting the motion path of the mobile robot, the point cloud scanning track of the laser scanner driven by the mechanical arm, and the monocular / dual-camera photographing position point of each inspection section; Using the mobile robot, the laser scanner, and the monocular / dual-camera to complete the data acquisition of the missile outer surface three-dimensional point cloud, overall surface image, and identification image of each inspection section in a preset order, obtaining the point cloud splicing compensation value of each inspection section according to the missile outer surface three-dimensional point cloud of each inspection section and the overall pose of the missile to be inspected in the global measurement coordinate system; Calibrating the point cloud splicing accuracy of each section of the outer surface according to the point cloud splicing compensation value of each inspection section, and reconstructing to generate a missile full-outer-surface three-dimensional model; Mapping the screened problem image to the missile full-outer-surface three-dimensional model according to the standard inspection data; Obtaining the geometric dimension value, surface appearance, and identification inspection result according to the missile full-outer-surface three-dimensional model, comparing the geometric dimension value, surface appearance, and identification inspection result with the preset total inspection process requirement to obtain the total inspection result; Using the mobile robot, the laser scanner, and the monocular / dual-camera to complete the data acquisition of the missile outer surface three-dimensional point cloud, overall surface image, and identification image of each inspection section in a preset order, obtaining the point cloud splicing compensation value of each inspection section according to the missile outer surface three-dimensional point cloud of each inspection section and the overall pose of the missile to be inspected in the global measurement coordinate system comprises: Step S31: According to the set motion path, the mobile robot moves to the first inspection section, the laser scanner continuously and dynamically scans the outer surface point cloud data around the missile arc surface, when the scanning reaches the target at the boundary of the first inspection section, the binocular camera identifies the target and calculates the three-dimensional pose of the target center relative to the binocular camera coordinate system, the three-dimensional pose of the target center relative to the binocular camera coordinate system is converted through the laser scanner coordinate system to obtain the three-dimensional pose of the target center in the global measurement coordinate system, the three-dimensional pose of the target center in the global measurement coordinate system is compared with the static three-dimensional pose of the target center in the global measurement coordinate system to obtain a deviation value, and the deviation value is set as the point cloud splicing compensation value of the first inspection section; after the scanning is completed, the mobile robot and the mechanical arm drive the monocular camera to collect the overall appearance image of the outer surface of the first inspection section; if there are part and identification inspection contents in this inspection section, the monocular camera and the binocular camera are simultaneously used to identify and photograph the relevant area for collection; Step S32: When all the contents are collected, move to the next inspection section, and repeat step S31 until all the inspection sections are collected.

2. The method of claim 1, wherein: The total inspection task calling standard inspection data comprises: inspection process, missile design three-dimensional model, coating paint appearance standard image data set, part installation standard image data set, and identification standard image data set.

3. The method of claim 1, wherein: The overall pose of the missile to be inspected in the global measurement coordinate system obtained according to the measurement of the optical tracker on the missile to be inspected comprises: The static three-dimensional pose of the vertex of the head of the missile to be inspected, the target centers of the targets on both sides of the missile support frame in the global measurement coordinate system is measured by using an optical tracker, the missile axis vector is calculated by fitting, and the overall pose of the missile to be inspected in the global measurement coordinate system is determined by combining the three-dimensional model of the missile.

4. The method of claim 1, wherein: According to the standard inspection data, the problem image is screened out and mapped to the three-dimensional model of the whole outer surface of the missile, which includes: Matching and comparing the standard image data set of the coating surface appearance, the standard image data set of the component installation, and the standard image data set of the identification, screening out the problem image, and mapping the problem image to the entity model outer surface based on the point cloud three-dimensional pose associated by the data fusion algorithm.

5. The method of claim 1, wherein: The geometric size value includes the overall length, the diameter of the front and rear body, and the wing surface span size.

6. The method of claim 1, wherein: The optical tracker tracks the identification points on the laser scanner in the measurement space. When at least three identification points on the scanner are photographed by two optical trackers, the three-dimensional coordinate values of the identification points are calculated in real time to determine the three-dimensional space pose of the laser scanner in the global measurement coordinate system.

7. The method of claim 1, wherein: The algorithm calculation process of reconstruction includes point cloud denoising, point cloud filtering, point cloud splicing, and surface reconstruction.

8. The method of claim 1, wherein: The number of optical trackers is determined according to the envelope space involved in the missile geometric size measurement and the visibility analysis that at least two optical trackers can photograph the identification points on the laser scanner.

9. An intelligent general inspection device for missile outer surface quality, characterized in that It includes: The first module is used for receiving the standard inspection data of the total inspection task and constructing the global measurement coordinate system; The second module is used for obtaining the overall pose of the missile to be inspected in the global measurement coordinate system; The motion path of the mobile robot in each inspection section, the point cloud scanning track of the laser scanner driven by the mechanical arm, and the photographing position point of the monocular / dual-camera are set; The third module is used for obtaining the point cloud splicing compensation value of each inspection section according to the three-dimensional point cloud of the missile outer surface of each inspection section and the overall pose of the missile to be inspected in the global measurement coordinate system; The fourth module is used for calibrating the point cloud splicing accuracy of each section of the outer surface according to the point cloud splicing compensation value of each inspection section, and reconstructing the three-dimensional model of the whole outer surface of the missile; According to the standard inspection data, the problem image is screened out and mapped to the three-dimensional model of the whole outer surface of the missile; The fifth module is used for obtaining the geometric size value, surface appearance, and identification inspection result according to the three-dimensional model of the whole outer surface of the missile, comparing the geometric size value, surface appearance, and identification inspection result with the preset total inspection process requirement to obtain the total inspection result. The data acquisition of the three-dimensional point cloud of the missile outer surface, the overall surface image, and the identification image of each inspection section is completed by using the mobile robot, the laser scanner, and the monocular / dual-camera in a preset order, and the point cloud splicing compensation value of each inspection section is obtained according to the three-dimensional point cloud of the missile outer surface of each inspection section and the overall pose of the missile to be inspected in the global measurement coordinate system, which includes: Step S31: According to the set motion path, the robot moves to the first inspection section, the laser scanner continuously and dynamically scans the missile arc surface to obtain the outer surface point cloud data, when the scanning reaches the target at the boundary of the first inspection section, the binocular camera identifies the target and calculates the three-dimensional pose of the target center relative to the binocular camera coordinate system, the three-dimensional pose of the target center relative to the binocular camera coordinate system is converted through the laser scanner coordinate system to obtain the three-dimensional pose of the target center in the global measurement coordinate system, the three-dimensional pose of the target center in the global measurement coordinate system is compared with the static three-dimensional pose of the target center in the global measurement coordinate system to obtain the deviation value, and the deviation value is set as the point cloud splicing compensation value of the first inspection section; after the scanning is completed, the mobile robot and the mechanical arm drive the monocular camera to collect the overall appearance image of the outer surface of the first inspection section; if there are parts and identification inspection contents in the inspection section, the monocular camera and the binocular camera are simultaneously started to identify the related area and collect the image; Step S32: After all the contents are collected, move to the next inspection section, repeat step S31 until all the inspection sections are collected.

Images