A robot-based profiling needle-punching control system and method

Abstract

The application discloses a kind of based on robot's profiling needling control system, including master station and slave station, the master station includes main controller, profile thickness detection unit and needle execution unit;The slave station includes needle density detection unit and robot control unit, and master station is connected with slave station communication by bus.The application establishes profile thickness detection unit, needle execution unit, and the change of profile thickness is judged online, realizes the quick replacement and use of regionalization difference combination needle plate, compared with the existing fixed needle plate mode, meets the needle plate quick adjustment demand of various complex special-shaped prefabricated body automatic profiling needling, and improves needle dispersion, needle depth consistency and the applicability of needle plate, effectively improves the quality of needle prefabricated body;Profiling movement is realized by robot carrying needle density detection unit, realizes prefabricated body regionalization online needle density detection, discrete state detection, effectively solves the problem of quality online detection in the process of complex special-shaped prefabricated body automatic profiling needling.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of needle-punched preforms, in particular to a profiling needle-punching control system and method based on a robot. BACKGROUND

[0002] The needle-punched preform establishes interlayer short fiber connection through the needle-punching process, and has the advantages of high automation degree, low cost, short delivery cycle, etc., and has been gradually applied in new energy, rail transportation, aerospace and other fields. In order to meet the requirements of performance improvement of application components and adapt to the changes of complex environment, needle-punched preforms gradually develop towards large size and complex special-shaped structure, and in the preparation process, needle depth, needle density, needle replacement frequency and the like all affect the consistency and uniformity of the preform quality. The traditional flat plate needle-punching machine cannot meet the rapid and high-quality preparation requirements of the preform.

[0003] The existing patent No. 202211663441.0 discloses a kind of needle-punched preform robot needle-punching trajectory planning method and storage medium, by constructing seven-axis linkage robot needle-punching program conversion module and output executable program, realize rotary table rotation while robot needle-punching, can meet the efficient motion synthesis requirements of needle-punching point pose;The existing patent No. 202210812610.6 discloses a kind of hybrid switching double-station needle-punching robot, which can realize high-quality and efficient needle-punching forming of complex spatial curved surface preform, reduce the space occupation of workbench, and the equipment is relatively simple and easy to program.

[0004] The existing technology has carried out a lot of research on robot control needle-punching trajectory and multi-station device for complex special-shaped and rotary needle-punched preforms, but the existing needle-punching system does not study the needle plate for the change of profile thickness, only reduces the number of needle arrangement by reducing the traditional needle plate actuator, which will cause the preform surface to appear obvious horizontal stripes or diagonal stripes and clumping phenomenon after needle-punching according to the set profiling needle-punching trajectory;At the same time, the large size and complex shape of the preform make it difficult for manual detection to detect the dispersion state of needle-punching density in real time online, and the consistency of product quality fluctuates.

[0005] Therefore, in order to improve the needle-punched preform quality and meet the development needs of large size and complex special-shaped needle-punched preforms, the present application proposes a profiling needle-punching control system and method based on a robot, which realizes high-quality, high-precision and rapid preparation of complex special-shaped needle-punched preforms. SUMMARY

[0006] The application aims to provide a robot-based directional needling control system and method, which realizes regionalized differential combination needling, online detection of needling quality, effectively improves the preparation efficiency and needling quality of large-size irregular preform needling on the basis of ensuring high-precision profile copying.

[0007] To solve the above technical problems, the technical scheme of the application is as follows:

[0008] In the first aspect, the application provides a robot-based profile needling control system, which comprises a master station and a slave station.

[0009] The master controller is configured to receive and process the preform profile thickness detected by the profile thickness detection unit in real time, judge the change of the preform profile thickness, calculate and output the number of needling plate combinations of the needling execution unit, perform pneumatic control on the needling execution unit, output the profile motion path switching signal of the robot to the robot control unit, control the needling execution unit to complete the needling motion after receiving the needle target trajectory point motion in place signal of the robot control unit, and output the abnormal warning signal of the needling execution unit and the needle density detection unit in real time.

[0010] The profile thickness detection unit is configured to detect the preform profile thickness before the needling motion.

[0011] The needling execution unit is configured to increase or decrease the number of needling plates in the combination needling plate through the multi-region negative pressure adsorption mode according to the number of needling plate combinations output by the master controller, and perform the needling motion.

[0012] The needle density detection unit is configured to detect and count the discrete state of the needle density and the needle trace distribution after the needling motion in the set region range in combination with the profile motion path of the robot, and send the statistical result to the master controller.

[0013] The robot control unit is configured to call the profile motion path of the robot that meets the requirements of the needle density and the needling plate combination according to the profile motion path switching signal of the robot output by the master controller, and control the robot to move to the needle target trajectory point.

[0014] Preferably, the profile thickness unit comprises a sensor mounting plate and five displacement sensors S1, S2, S3, S4 and S5; the displacement sensors are arranged on the same reference surface of the sensor mounting plate, the displacement sensors are adjustably connected with the sensor mounting plate, and two groups of displacement sensors S1 and S2, S3 and S4 are arranged on the vertical extension lines of the two side lengths of the combined needle plate, respectively, and the displacement sensor S5 is arranged on the vertical extension line of the center line of the combined needle plate.

[0015] Preferably, the needle punching execution unit comprises a combined needle plate and a plurality of air cylinders; the combined needle plate is composed of a plurality of sheet-shaped needle plates, the front wall of the sheet-shaped needle plate is provided with a plurality of needle holes for mounting the needles; the two sides of the combined needle plate are pressed by air cylinders which are uniformly distributed and equal in number.

[0016] Preferably, the needle punching execution unit further comprises a needle punching servo motor driving unit, the needle punching servo motor driving unit comprises a needle punching servo driver, a needle punching servo motor, a needle punching servo motor control shaft and a negative pressure suction mechanism, the input end of the needle punching servo driver is connected with the main controller through a bus, the output end of the needle punching servo driver is connected with the input end of the needle punching servo motor, one end of the negative pressure suction mechanism is connected with the output end of the needle punching servo motor through the needle punching servo motor control shaft, the other end of the negative pressure suction mechanism is connected with the rear wall of the combined needle plate, and the needle punching servo motor drives the negative pressure suction mechanism and the combined needle plate to execute the needle punching motion under the control of the needle punching servo driver.

[0017] Preferably, the needle punching execution unit further comprises a diffuse reflection photoelectric sensor S6, the diffuse reflection photoelectric sensor S6 is arranged on the end mechanism of the industrial robot, and is used for detecting the completion of the needle punching motion in real time.

[0018] Preferably, the needle density detection unit comprises a visual detection device and an image algorithm processing platform; the visual detection device comprises a camera, a lens and a coaxial white light source, and is used for collecting images under constant focal length conditions; the image algorithm processing platform is mounted on a visual detection industrial PC, and is used for counting the discrete state of the needle density and the needle punching track distribution, and feeding back the counting results to the main controller in real time.

[0019] Preferably, the robot control unit comprises a six-degree-of-freedom industrial robot and an external extension shaft controlled preform core mold rotating motion platform; the end mechanism of the six-degree-of-freedom industrial robot is sequentially connected with the needle density detection unit, the needle punching execution unit and the profile thickness unit from top to bottom.

[0020] In the second aspect, a robot-based profiling needle punching control method is provided, comprising the following steps:

[0021] Step 1: setting the parameters of the needle punching motion;

[0022] Step 2: divide the preform into multiple different regions, set the initial state of the unit layer Ci = 1, region M = 1, and 1 ≤ Ci ≤ C_num, 1 ≤ M ≤ N, wherein C_num is the total number of needle unit layers, and N is the total number of regions in each unit layer;

[0023] Step 3: the robot control unit controls the robot to move from the initial safe position to the initial point of the unit layer Ci region M in step 2;

[0024] Step 4: detect the preform profile thickness of region M by the profile thickness detection unit;

[0025] Step 5: the main controller receives and processes the preform profile thickness in step 4 in real time through the analog input module, judges the change of the preform profile thickness, and calculates the number of needle plate combinations of the needle execution unit;

[0026] Step 6: based on the number of needle plate combinations calculated in step 5, the main controller controls the needle execution unit pneumatically, increases or decreases the sheet needle plate in the combined needle plate, and then tightens the sheet needle plate through the air cylinder to form a new combined needle plate;

[0027] Step 7: based on the new combined needle plate formed in step 6, according to the needle density requirement of region M, the main controller outputs the robot profiling motion path switching signal to the robot controller of the robot control unit, the robot controller calls the robot profiling motion path that meets the needle density and combined needle plate requirements, then controls the robot to move to the needle target trajectory point, and feeds back the needle target trajectory point movement to the main controller. In place signal, record the actual needle target trajectory point movement times plus one in region M of the current combined needle plate; the main controller controls the needle execution unit to execute the needle movement according to the needle depth requirement of region M, after the needle movement is in place, the diffuse reflection photoelectric sensor S6 detects that the needle movement is completed to the in-place state, and feeds back the needle movement completion to the in-place signal to the main controller and the needle density detection unit, and records the needle execution times plus one in the unit layer Ci and the needle execution times plus one of the current combined needle plate in region M;

[0028] Step 8: set the allowed movement times of the combined needle plate in region M, judge the actual needle target trajectory point movement times of the current combined needle plate in region M and the needle execution times of the robot in the unit layer Ci, and the specific implementation is as follows:

[0029] Step 8.1: judge the actual needle target trajectory point movement times of the current combined needle plate in region M;

[0030] If the actual needle target trajectory point movement times is greater than or equal to the allowed movement times, it means that the preform profile thickness of region M needs to be detected again, and the combined needle plate needs to be updated, so the actual needle target trajectory point movement times is cleared, and steps 4-8 are executed again.

[0031] If the actual number of movements of the target trajectory point is less than the allowed number of movements, it indicates that the current combination of needle plates can continue to be used, and the steps 7-8 are continued to be executed in a loop;

[0032] Step 8.2: Determine the number of needle puncture executions of the robot in the unit layer Ci;

[0033] If the number of needle puncture executions is greater than or equal to the needle replacement frequency, it indicates that the needle is worn out, and the main controller outputs a needle replacement warning signal;

[0034] If the number of needle puncture executions is less than the needle replacement frequency, it indicates that the needle is in good condition and can continue to be used, and the steps 7-8 are continued to be executed in a loop;

[0035] Step 9: After receiving the needle puncture movement completion signal fed back in step 7, the detection area range of the needle density detection unit is set according to the change of the thickness of the preform profile in step 5, the preform image is collected under a constant focal length state by a visual detection device combined with the profiling motion path of the robot, the collected image is processed by an image algorithm processing platform, the needle density and the dispersion state of the needle puncture track distribution in the area range are detected and counted in real time online, and the statistical results are fed back to the main controller in real time, so as to determine the needle puncture quality of the preform;

[0036] If the needle density deviates or the needle puncture track appears stripes, diagonal lines, or clumping phenomenon, it indicates that the needle puncture quality of the preform is abnormal, and the main controller outputs an abnormal warning in real time;

[0037] If the needle density does not deviate or the needle puncture track does not appear stripes, diagonal lines, or clumping phenomenon, it indicates that the needle puncture quality of the preform is normal, and the needle puncture movement is continued to be executed;

[0038] Step 10: Based on the needle puncture movement completion signal in step 7, it is determined whether the current needle puncture target trajectory point is the last one in the region M in the unit layer Ci;

[0039] If the current needle puncture point is the last one, it indicates that the needle puncture movement in the region M has been completed, and step 11 is executed;

[0040] If the current needle puncture point is not the last one, it indicates that the needle puncture movement in the region M has not been completed, and steps 7-10 are executed in a loop;

[0041] Step 11: Determine whether the region M is the last needle puncture execution movement in the unit layer Ci;

[0042] If the region M is the last needle puncture execution movement, it indicates that all regions in the unit layer Ci have completed the needle puncture movement, and step 12 is executed;

[0043] If region M is not the last needle-punching motion, it means that there are still regions in unit layer Ci that have not completed the needle-punching motion. Then the count of region M is incremented by one, and steps 7-11 are executed repeatedly.

[0044] Step 12: Determine if cell layer Ci is the last cell layer;

[0045] If unit layer Ci is the last one, it means that all unit layers of the prefabricated body have completed the needle-punching motion. The robot controller controls the robot to move to the initial safe position and feeds back the prefabricated body contouring needle-punching completion signal through the bus.

[0046] If unit layer Ci is not the last one, it means that there are still unit layers in the prefabricated body that have not completed the needle-piercing movement. Then, the count of unit layer Ci is incremented by one, region M is reassigned to 1, and steps 3-12 are executed in a loop.

[0047] Preferably, in step 5, the calculation formula for determining the change in the thickness of the precast body surface is as follows:

[0048] Δh i =|h5-h i |,1≤i≤4

[0049]

[0050] The number of needle plates combined is calculated using the following formula:

[0051]

[0052] Where A is the distance between displacement sensors S1, S2, S3, and S4 and displacement sensor S5 respectively, B is the width of the sheet-like needle plate, Δh0 is the maximum allowable needle depth fluctuation, Δh is the average deviation of the surface thickness, and Δh i h is the profile thickness deviation value, h5 is the profile thickness detected by displacement sensor S5, h i Let be the thickness of the profile detected by the i-th displacement sensor, and n be the number of sheet-like needle plates.

[0053] Preferably, step 6 further includes the following steps:

[0054] Step 6.1: The main controller adjusts the electro-proportional valve through the analog output module to control the negative pressure in the needle acupuncture unit;

[0055] Step 6.2: The main controller adjusts the pneumatic solenoid valve b to disconnect through the digital output module, thereby controlling the cylinder in the needle puncture execution unit to release the combined needle plate;

[0056] Step 6.3: The main controller adjusts the suction and disconnection of the pneumatic electromagnetic valve a through the digital output module, controls the adsorption state of the sheet needle plate corresponding to the region in the needle execution unit, and completes the combination of the sheet needle plate;

[0057] If the sheet needle plate needs to be increased in the corresponding region, the pneumatic electromagnetic valve a is adjusted to be sucked, the negative pressure gas flows into the negative pressure adsorption mechanism, the adsorption force is generated, and the sheet needle plate is adsorbed;

[0058] If the sheet needle plate needs to be reduced in the corresponding region, the pneumatic electromagnetic valve a is adjusted to be disconnected, the negative pressure gas stops flowing into the negative pressure adsorption mechanism, and the sheet needle plate is released;

[0059] Step 6.4: The main controller adjusts the suction of the pneumatic electromagnetic valve b through the digital output module, controls the cylinder in the needle execution unit to compress the sheet needle plate in step 6.3, and forms a new combined needle plate.

[0060] Compared with the prior art, the beneficial effects of the present application are: 1. By establishing the profile thickness detection unit and the needle execution unit, the change of the profile thickness is predicted online, the rapid replacement and use of the regionalized differential combined needle plate are realized, compared with the existing fixed needle plate method, the rapid adjustment demand of the needle plate for automatic profiling needle punching of various complex shaped preforms is met, the needle punching discreteness, needle depth consistency and the applicability of the needle plate are improved, and the quality of the needle punched preform is effectively improved; 2. By profiling movement of the robot carrying the needle density detection unit, the preform regional online needle density detection and discrete state detection are realized, and the problem of online quality detection in the automatic profiling needle punching process of the complex shaped preform is effectively solved; 3. The present application has the functions of needle replacement, needle density and motion mechanism abnormality early warning, and can realize the integrated preparation of the complex shaped needle punched preform with high quality and high precision. BRIEF DESCRIPTION OF DRAWINGS

[0061] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the description of the embodiments of the present application. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creating laborious work.

[0062] Figure 1 The structural block diagram of the profiling needle punching control system based on the robot of the present application;

[0063] Figure 2 The layout structure diagram of the profiling needle punching control system based on the robot and the preform of the present application;

[0064] Fig. 3 is a principle diagram of the profile thickness detection unit and the sheet needle plate combination of the present application;

[0065] In the figure 3(a), the combined needle plate is composed of odd piece-shaped needle plate; in the figure 3(b), the combined needle plate is composed of even piece-shaped needle plate.

[0066] Figure 4 The preform surface needle trace distribution image collected by the needle density detection unit of the application;

[0067] Figure 5 The flow chart of the profiled needle punching control method based on the robot of the application;

[0068] In the figure, 1 is a sensor mounting plate, 2 is a combined needle plate, 3 is a cylinder, 4 is a piece-shaped needle plate, 5 is a needle hole, 6 is a visual detection device, 7 is a visual detection industrial PC, 8 is a six-degree-of-freedom industrial robot, 9 is a preform core mold rotating motion platform, and 10 is a preform. DETAILED DESCRIPTION

[0069] Further detailed description of the application is given below in combination with the drawings. The terminal technical solutions of the embodiments of the application are clearly and completely described. Obviously, the described embodiments are part of the embodiments of the application, rather than all the embodiments. Based on the embodiments of the application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the application.

[0070] In this embodiment, the industrial robot with the model KUKA KR210 R2700 is taken as an example.

[0071] As shown in the figure 3(a) and the figure 3(b), the combined needle plate is composed of piece-shaped needle plates. Figure 1 As shown in the figure 3(a) and the figure 3(b), the combined needle plate is composed of piece-shaped needle plates.

[0072] The main controller is used for receiving and processing the preform 10 profile thickness detected in the profile thickness detection unit in real time, judging the change of the preform 10 profile thickness, calculating and outputting the needle plate combination number of the needle punching execution unit; pneumatically controlling the needle punching execution unit, and outputting the profiled motion path switching signal of the robot to the robot control unit, controlling the needle punching execution unit to complete the needle punching motion after receiving the needle punching target trajectory point motion in place signal of the robot control unit; and outputting the abnormal warning signal of the needle punching execution unit and the needle density detection unit in real time.

[0073] In this embodiment, the main controller is a Siemens programmable controller S7-1200 with motion control function, including extended digital input / output module (DI / DO), analog input / output module (AI / AO), supporting Modbus TCP, Profinet or Ethernet multiple industrial bus communication. The main controller interacts with the human-machine interface through the Ethernet bus to realize the input of control parameters for the preform profiling needling, and communicates with the robot control unit and the needle serving motor in the needle serving execution unit through the Profinet bus, and communicates with the visual detection industrial PC in the needle density detection unit through the Modbus TCP bus.

[0074] The profile thickness detection unit is used to detect the profile thickness of the preform 10 before the needle serving movement.

[0075] In this embodiment, the profile thickness unit includes a sensor mounting plate 1 and five displacement sensors S1, S2, S3, S4 and S5; the displacement sensor is a Panasonic high-precision laser ranging sensor HG-C1100, with a range of 100±35mm, a detection accuracy of 0.07mm, and an output of 4-20mA analog signal, the displacement sensors S1, S2, S3, S4 and S5 are arranged on the sensor mounting plate 1 within a range of 100cm 2 The displacement sensors S1, S2, S3, S4 and S5 are adjustably connected to the sensor mounting plate 1 on the same reference surface within the range, and the two groups of displacement sensors S1 and S2, S3 and S4 are arranged on the vertical extension lines of the two side lengths of the combined needle plate 2, respectively, and the displacement sensor S5 is arranged on the vertical extension line of the center line of the combined needle plate 2, the distance A between the displacement sensors S1, S2, S3, S4 and the displacement sensor S5 is initially 60mm, and the distance A will be adjusted according to the width of the combined needle plate 2 to meet the detection of the profile thickness within the range of 10-100cm 2

[0076] The needle serving execution unit is used to increase or decrease the number of needle plates in the combined needle plate 2 through multi-region negative pressure adsorption according to the number of needle plate combinations output by the main controller, and execute the needle serving movement.

[0077] In this embodiment, the needle serving execution unit includes a combined needle plate 2 and eight air cylinders 3; the combined needle plate 2 is at least one sheet needle plate 4 with a width B of 20mm, and at most is composed of seven sheet needle plates 4, the front wall of the sheet needle plate 4 is provided with at least five needle holes 5 for installing the needles, and the number of needle holes 5 can be adjusted according to actual needs; the two sides of the combined needle plate 2 are pressed tightly by four air cylinders 3 which are evenly distributed and equal in number.

[0078] ​The main controller adjusts the increase and decrease of the seven sheet-shaped needle plates 4 through seven-area negative pressure adsorption, and controls the close combination of each sheet-shaped needle plate 4 through pneumatic pressure, so as to ensure the stability of the combined needle plate 2 in the needling movement.

[0079] In the embodiment, the needling execution unit further comprises a needling servo motor driving unit, which comprises a needling servo driver, a needling servo motor, a needling servo motor control shaft and a negative pressure adsorption mechanism. The input end of the needling servo driver is connected with the main controller through a bus, the output end of the needling servo driver is connected with the input end of the needling servo motor, one end of the negative pressure adsorption mechanism is connected with the output end of the needling servo motor through the needling servo motor control shaft, and the other end of the negative pressure adsorption mechanism is connected with the rear wall of the combined needle plate. The needling servo motor drives the negative pressure adsorption mechanism and the combined needle plate to perform the needling movement under the control of the needling servo driver, so that the needling depth can be accurately controlled.

[0080] In the embodiment, the needling execution unit further comprises a diffuse reflection photoelectric sensor S6, which is arranged on the end mechanism of the industrial robot and is used for detecting the in-place state of the needling movement in real time, so as to facilitate the main control system to count the needling frequency. The diffuse reflection photoelectric sensor S6 can adjust the range of the sensing interval according to the change of the needle depth.

[0081] When the number of the sheet-shaped needle plates 4 in the combined needle plate 2 is odd, the center point of the sheet-shaped needle plate 4 in the middle coincides with the TCP of the robot, the TCP (Tool Center Point) is the tool center point of the robot, which is used for determining the position and direction of the tool in space, and the sheet-shaped needle plates 4 added or reduced are respectively located on the two sides of the sheet-shaped needle plate 4 in the middle, so as to ensure that the TCP of the robot does not change under the combination of different numbers of sheet-shaped needle plates 4; when the number of the sheet-shaped needle plates 4 in the combined needle plate 2 is even, the center point of the joint of the two sheet-shaped needle plates 4 in the middle coincides with the TCP of the robot, and the sheet-shaped needle plates 4 added or reduced are respectively located on the two sides of the two sheet-shaped needle plates 4 in the middle, so as to ensure that the TCP of the robot does not change under the combination of different numbers of sheet-shaped needle plates 4.

[0082] The needle density detection unit is used for detecting the needle density and the discrete state of the needle trace distribution after the needling movement is completed in the set area range combined with the profiling motion path of the robot, and sending the statistical result to the main controller.

[0083] In the embodiment, the needle density detection unit comprises a visual detection device 6 and an image algorithm processing platform. The visual detection device 6 comprises a 2000 million pixel area array CMOS camera, a low distortion telecentric lens and a 100mm coaxial white light source. The visual detection device 6 is arranged on the end mechanism of the industrial robot, and is used for detecting the in-place state of the needling movement in real time. 2In the detection area range, the image is collected under the condition of constant focal length combined with the profiling motion path of the robot; the image algorithm processing platform is carried on the visual detection industrial PC 7, which is used for counting the discrete state of the needle density and the needle track distribution, and the statistical result is fed back to the main controller in real time through the bus, so that the main controller judges the needle punching quality of the preform 10.

[0084] The robot control unit is used for calling the profiling motion path of the robot meeting the requirements of the needle density and the needle plate combination according to the profiling motion path switching signal of the robot output by the main controller, and controlling the robot to move to the needle punching target track point.

[0085] In the embodiment, the robot control unit includes a six-degree-of-freedom industrial robot 8 and an external extension shaft controlled preform core mold rotating motion platform 9; the end mechanism of the six-degree-of-freedom industrial robot 8 is sequentially connected with a needle density detection unit, a needle punching execution unit and a profile thickness unit from top to bottom, realizing the integration of needle punching control and needle density detection.

[0086] Combined with the number and the needle arrangement mode of the needle density and the sheet-shaped needle plate combination, the moving step distance of the combined needle plate 2 in the needle punching execution unit is determined, and then based on the three-dimensional graph of the profile of the preform 10, the robot profiling motion path with different moving step distances is constructed by using the robot trajectory programming software and stored in the robot control unit, so that the robot control unit can call the robot profiling motion path meeting the requirements of the needle density and the combined needle plate in time.

[0087] As shown in Figure 4 The application also provides a profiling needle punching control method based on a robot, which includes the following steps:

[0088] Step 1: setting the control parameters of the needle punching motion in the human-computer interaction interface; including the number of needle punching unit layers C_num, the needle density and the needle depth of the top, the leading edge and the different areas of the curved surface of each unit layer, the frequency of replacing the needles, and the frequency of replacing the combined needle plates in different areas.

[0089] Step 2: dividing the preform into multiple different areas in the human-computer interaction interface, setting the unit layer Ci=1 and the area M=1 in the initial state, and 1≤Ci≤C_num and 1≤M≤N, wherein C_num is the total number of needle punching unit layers, and N is the total number of areas of each unit layer; for example, area 1 is the top, area 2 is the leading edge, area 3 is curved surface 1, and area N is curved surface T.

[0090] Step 3: The robot control unit controls the robot to move from the initial safe position to the initial point of the unit layer Ci area M in step 2, specifically: the main controller outputs a trigger digital signal to the robot control unit to move the robot to the needling target trajectory point, the robot controller KRC4 in the robot control unit receives the digital signal and controls the robot to move to the needling target trajectory point, and feeds back the needling target trajectory point movement to the main controller; the initial point is the first needling target trajectory point of the unit layer Ci area M;

[0091] Step 4: Detect the preform profile thickness of area M by the profile thickness detection unit;

[0092] In order to ensure the consistency of the needle depth of different thickness areas during the execution of the needling movement, the five displacement sensors S1, S2, S3, S4 and S5 in the profile thickness detection unit complete the profile thickness detection before the needling plate combination needs to be updated, and the detection results are recorded as h1, h2, h3, h4 and h5.

[0093] Step 5: The main controller receives and processes the preform profile thickness h1, h2, h3, h4 and h5 in step 4 in real time through the analog input module, calculates the profile thickness deviation value Δh of the four displacement sensors S1-S4 relative to the center displacement sensor S5 i , judges the change of the preform profile thickness, i.e. the profile thickness difference average Δh, and calculates the number of needle plate combinations of the needling execution unit;

[0094] The calculation formula for judging the change of the preform profile thickness is as follows:

[0095] Δh i = |h5-h i |, 1≤i≤4

[0096]

[0097] The calculation formula for the number of needle plate combinations is as follows:

[0098]

[0099] Wherein, A is the distance between displacement sensors S1, S2, S3, S4 and displacement sensor S5, B is the width of the sheet needle plate, Δh0 is the maximum value of the allowable needle depth fluctuation, which is set in the human-computer interaction interface, Δh is the profile thickness deviation average, h5 is the profile thickness detected by displacement sensor S5, h i is the profile thickness detected by the ith displacement sensor, and n is the number of sheet needle plates.

[0100] Step 6: Based on the number of needle plate combinations calculated in step 5, the main controller pneumatically controls the needle punching execution unit, increases or decreases the sheet needle plate in the combined needle plate, and then presses the sheet needle plate through the cylinder to form a new combined needle plate, including the following steps:

[0101] Step 6.1: The main controller adjusts the electric proportional valve through the analog output module to control the size of the negative pressure in the needle punching execution unit;

[0102] Step 6.2: The main controller adjusts the pneumatic electromagnetic valve b to be disconnected through the digital output module to control the cylinder to release the combined needle plate in the needle punching execution unit;

[0103] Step 6.3: The main controller adjusts the suction and disconnection of pneumatic electromagnetic valve a through the digital output module to control the suction state of the sheet needle plate in the corresponding area in the needle punching execution unit, and completes the combination of the sheet needle plate;

[0104] If the sheet needle plate needs to be added in the corresponding area, adjust the suction of pneumatic electromagnetic valve a, and the negative pressure gas flows into the negative pressure suction mechanism to generate suction force to suck the sheet needle plate;

[0105] If the sheet needle plate needs to be reduced in the corresponding area, adjust the pneumatic electromagnetic valve a to be disconnected, and the negative pressure gas stops flowing into the negative pressure suction mechanism to release the sheet needle plate;

[0106] Step 6.4: The main controller adjusts the suction of pneumatic electromagnetic valve b through the digital output module to control the cylinder in the needle punching execution unit to press the sheet needle plate in step 6.3 to form a new combined needle plate.

[0107] Step 7: Based on the new combined needle plate formed in step 6, according to the needle density requirements of area M, the main controller outputs the robot profiling motion path switching signal to the robot controller of the robot control unit, and after the robot controller calls the robot profiling motion path that meets the needle density and combined needle plate requirements, controls the robot to move to the needle target trajectory point, and feeds back the needle target trajectory point motion to the main controller. In place signal, record the actual needle target trajectory point movement number of the current combined needle plate in area M plus one; after receiving the request to execute needle punching motion signal output by the robot controller, the main controller controls the needle punching execution unit to execute needle punching motion according to the needle depth requirements of area M, and completes the needle punching; after the needle punching motion is in place, the diffuse reflection photoelectric sensor S6 detects the needle punching motion completion to place state, and feeds back the needle punching motion completion to place signal to the main controller and the needle density detection unit, and records the needle punching execution number of the unit layer Ci plus one and the needle punching execution number of the current combined needle plate in area M plus one;

[0108] Step 8: In the human-computer interaction interface, set the number of allowed movements of the combined needle plate in region M, judge the number of actual movements of the combined needle plate in region M to the target trajectory point and the number of needle puncture executions of the robot in the unit layer Ci, and the specific implementation is as follows:

[0109] Step 8.1: Judge the number of actual movements of the combined needle plate in region M to the target trajectory point.

[0110] If the number of actual movements of the combined needle plate in region M to the target trajectory point is greater than or equal to the number of allowed movements, it means that the preform profile thickness in region M needs to be re-detected, and the combined needle plate needs to be updated. Then, the number of actual movements of the combined needle plate in region M to the target trajectory point is cleared, and steps 4-8 are executed again.

[0111] If the number of actual movements of the combined needle plate in region M to the target trajectory point is less than the number of allowed movements, it means that the combined needle plate can continue to be used, and steps 7-8 are executed again.

[0112] Step 8.2: Judge the number of needle puncture executions of the robot in the unit layer Ci.

[0113] If the number of needle puncture executions is greater than or equal to the needle replacement frequency, it means that the needle is worn out, and the main controller outputs a needle replacement warning signal. The worker receives the abnormal warning in the human-computer interaction interface, stops the system operation, replaces the needle, and then continues to run the system, thereby ensuring the quality of the preform needle puncture.

[0114] If the number of needle puncture executions is less than the needle replacement frequency, it means that the needle is in good condition and can continue to be used, and steps 7-8 are executed again.

[0115] Step 9: After receiving the needle puncture movement completion signal fed back in step 7, set the detection area range S of the needle density detection unit according to the change of the preform profile thickness in step 5, S∈[10~100]cm 2 , through the visual detection device, combined with the profiling motion path of the robot, the surface image of the preform after the needle puncture is collected under the condition of constant focal length, the collected image is processed by using the image algorithm processing platform, the dispersion state of the needle density and the needle puncture track distribution in the area range is detected and counted in real time, and the statistical result is fed back to the main controller in real time through the bus, so as to judge the quality of the preform needle puncture.

[0116] If the needle density deviates or the needle puncture track appears stripes, diagonal lines, and clumping phenomenon, it means that the quality of the preform needle puncture is abnormal, and the main controller outputs an abnormal warning in real time. The worker receives the abnormal warning in the human-computer interaction interface, modifies the needle puncture target trajectory point in the robot trajectory programming software, and locally adjusts the position of the needle puncture target trajectory point, thereby ensuring the quality of the preform needle puncture.

[0117] If the needle density does not deviate or the needle track does not appear striped, diagonal or clumping, it indicates that the preform needle punching quality is normal, and the needle punching movement continues to be executed;

[0118] The needle density is counted, including the following steps:

[0119] Step 9.11: The image algorithm processing platform performs gray processing on the image collected by the visual detection device, simplifies the image data and improves the processing speed.

[0120] Step 9.12: The image algorithm processing platform performs adaptive threshold segmentation and morphological processing on the detection area, highlights the features of the detection area, and according to the image gray characteristics obtained after the gray processing in step 9.11, uses the optimal threshold when the inter-class variance is maximum to divide the image into two parts: background and needle track.

[0121] Step 9.13: The image algorithm processing platform performs template feature matching by calculating the area of the detection area and combining the needle track features; the pixel size of the detection area is obtained by measuring the distance, the pixel distance is converted into actual physical distance according to the calibration file, and the area of the region is calculated; the number of needle tracks in the detection area is counted, and the needle density per unit area is calculated.

[0122] The dispersion state of the needle track distribution is counted, including the following steps:

[0123] Step 9.21: The image algorithm processing platform reduces the size of the original image collected by the visual detection device from x pix *y pix to X pix *Y pix , retains the basic information of structure and brightness, and discards the image differences caused by different sizes and proportions, and the calculation formula is as follows:

[0124] Y pix =F y *y pix

[0125] X pix =F x *x pix

[0126] Wherein, F x , F y are the scaling coefficients along the x-axis and y-axis of the image, x pix is the pixel of the original image in the x direction, y pix is the pixel of the original image in the y direction, X pix is the pixel of the reduced image in the X direction, and Y pix is the pixel of the reduced image in the Y direction.

[0127] Step 9.22: The reduced image is grayed to obtain the gray value of each pixel.

[0128] Gray m = 0.299 * R m + 0.587 * G m + 0.114 * B m , 1≤m≤p

[0129] wherein Gray m is the gray value of the mthpixel, R m , G m , B m are the gray values of the red, green, and blue color channels corresponding to the mthpixel, respectively, 0.299, 0.587, and 0.114 are the color weights corresponding to the red, green, and blue colors; and p is the number of pixel statistics.

[0130] Step 9.23: The average pixel gray value AVG Gray of the reduced image is calculated.

[0131]

[0132] wherein Gray m is the gray value of the mthpixel, and p is the number of pixel statistics.

[0133] Step 9.24: The gray value of each pixel of the jthreduced image is compared with the average pixel gray value AVG Gray , and the comparison result q jm is represented by 0 or 1; when the gray value of the mthpixel is greater than or equal to the average pixel gray value, q jm is 1; otherwise, q jm is 0, which can be expressed as follows:

[0134]

[0135] wherein q jm is the comparison result of the mthpixel of the jthreduced image.

[0136] Step 9.25: The comparison results in step 9.24 are combined to form a p-bit gray contrast statistical sequence Q j .

[0137] Q j = {q j1 , q j2 , q j3 , …, q jm , …, q jp}

[0138] Step 9.26: Obtain the contrast similarity P of two reduced images by comparing the gray contrast statistical sequence Q of adjacent reduced images j and Q (j+1) , and the formula is as follows:

[0139]

[0140] Wherein, r m is the similarity statistical value of the mth pixel point of the adjacent reduced images, r0 is the initial value of 0, p is the pixel point statistical quantity, q jm is the comparison result of the mth pixel point of the jth reduced image, q (j+1)m is the comparison result of the mth pixel point of the j+1th reduced image.

[0141] If P is greater than k*P, k is the discrete deviation coefficient, k∈[0,1], it indicates that the distribution of the needle mark is more discrete, and if P is smaller, it indicates that the similarity of adjacent images is high, and the discrete degree of the needle mark distribution is not high; on the contrary, the greater P is, the lower the similarity of adjacent images is, and the higher the discrete degree of the needle mark distribution is.

[0142] Step 10: Based on the needle puncture motion completion signal in step 7, it is judged whether the current needle puncture target trajectory point is the last one in the unit layer Ci area M;

[0143] If the current needle puncture point is the last one, it indicates that the needle puncture motion in the area M has been completed, and step 11 is executed;

[0144] If the current needle puncture point is not the last one, it indicates that the needle puncture motion in the area M has not been completed, and steps 7-10 are executed in a loop;

[0145] Step 11: It is judged whether the area M is the last needle puncture execution motion in the unit layer Ci;

[0146] If the area M is the last needle puncture execution motion, it indicates that all areas in the unit layer Ci have completed the needle puncture motion, and step 12 is executed;

[0147] If the area M is not the last needle puncture execution motion, it indicates that there are areas in the unit layer Ci that have not completed the needle puncture motion, and the count of the area M is increased by one, and steps 7-11 are executed in a loop;

[0148] Step 12: It is judged whether the unit layer Ci is the last unit layer;

[0149] If the unit layer Ci is the last one, it indicates that all unit layers of the prefabricated body have completed the needle puncture motion, the robot controller controls the robot to move to the initial safe position, and feeds back the prefabricated body profiling needle puncture completion signal through the bus;

[0150] If the unit layer Ci is not the last one, it means that the preform still has unit layers to be completed with the needling movement, then the count of the unit layer Ci is increased by one, the area M is re-assigned as 1, and the step 3-12 is executed in a loop.

[0151] The parts of the present application not involved are the same as or realized by using the prior art.

[0152] The above is a further detailed description of the present application in combination with specific embodiments, and the specific implementation of the present application should not be limited to these descriptions. For ordinary skilled persons in the technical field of the present application, several simple deductions or substitutions can be made without departing from the concept of the present application, and all of them should be considered as falling within the protection scope of the present application.

Claims

1. A robot-based contour needle puncture control system, characterized in that: It includes a master station and a slave station. The master station includes a main controller, a surface thickness detection unit, and a needle punching execution unit. The slave station includes a needle density detection unit and a robot control unit. The master station communicates with the slave station via a bus. The main controller is used to receive and process the preform surface thickness detected by the surface thickness detection unit in real time, determine the change in the preform surface thickness, calculate and output the number of needle plate combinations of the needle-punching execution unit; perform pneumatic control on the needle-punching execution unit, and output the robot's contour motion path switching signal to the robot control unit; after receiving the needle-punching target trajectory point movement arrival signal from the robot control unit, it controls the needle-punching execution unit to complete the needle-punching motion; and output abnormal warning signals of the needle-punching execution unit and the needle density detection unit in real time. The surface thickness detection unit is used to detect the surface thickness of the preform before the needle piercing motion. The acupuncture execution unit is used to increase or decrease the number of needle plates in the combined needle plates according to the number of needle plate combinations output by the main controller, and to perform acupuncture movements. The needle density detection unit is used to detect and statistically analyze the discrete state of needle density and needle trace distribution after the needle puncture movement is completed within a set area, combined with the robot's contour motion path, and send the statistical results to the main controller. The robot control unit is used to call the robot's contour motion path that meets the requirements of needle density and needle plate combination according to the robot's contour motion path switching signal output by the main controller, and control the robot to move to the needle puncture target trajectory point.

2. The robot-based contour needle puncture control system according to claim 1, characterized in that: The surface thickness detection unit includes a sensor mounting plate and five displacement sensors S1, S2, S3, S4 and S5. All displacement sensors are set on the same reference surface of the sensor mounting plate. The displacement sensors are tunably connected to the sensor mounting plate. Two sets of displacement sensors S1 and S2, and S3 and S4 are respectively set on the vertical extension lines of the side lengths of the combined needle plate. The displacement sensor S5 is set on the vertical extension line of the center line of the combined needle plate.

3. The robot-based contour needle puncture control system according to claim 1, characterized in that: The acupuncture execution unit includes a combined needle plate and several cylinders; the combined needle plate is composed of several sheet-like needle plates, and the front wall of the sheet-like needle plates has several needle holes for installing needles; the two sides of the combined needle plate are pressed together by cylinders that are evenly distributed and equal in number.

4. The robot-based contour needle puncture control system according to claim 3, characterized in that: The acupuncture execution unit also includes an acupuncture servo motor drive unit, which includes an acupuncture servo driver, an acupuncture servo motor, an acupuncture servo motor control shaft, and a negative pressure adsorption mechanism. The input end of the acupuncture servo driver is connected to the main controller via a bus, and the output end of the acupuncture servo driver is connected to the input end of the acupuncture servo motor. One end of the negative pressure adsorption mechanism is connected to the output end of the acupuncture servo motor via the acupuncture servo motor control shaft, and the other end of the negative pressure adsorption mechanism is connected to the rear wall of the combined needle plate. Under the control of the acupuncture servo driver, the acupuncture servo motor drives the negative pressure adsorption mechanism and the combined needle plate to perform acupuncture movements.

5. The robot-based contour needle puncture control system according to claim 3, characterized in that: The needle-piercing execution unit also includes a diffuse reflection photoelectric sensor S6, which is installed on the end mechanism of the industrial robot and is used to detect the completion status of the needle-piercing motion in real time.

6. The robot-based contour needle puncture control system according to claim 1, characterized in that: The needle density detection unit includes a visual inspection device and an image algorithm processing platform. The visual inspection device includes a camera, a lens, and a coaxial white light source, used to acquire images under constant focal length. The image algorithm processing platform is mounted on a visual inspection industrial control PC, used to statistically analyze the discrete state of needle density and needle prick / needle mark distribution, and to feed the statistical results back to the main controller in real time.

7. The robot-based contour needle puncture control system according to claim 1, characterized in that: The robot control unit includes a six-degree-of-freedom industrial robot and a preform core mold rotation platform controlled by its external extended axes; the end effector of the six-degree-of-freedom industrial robot is connected from top to bottom to a needle density detection unit, a needle punching execution unit, and a surface thickness detection unit.

8. A robot-based contour needle puncture control method, characterized in that, Includes the following steps: Step 1: Set the parameters for the acupuncture motion; Step 2: Divide the prefabricated body into multiple different regions. Set the initial unit layer Ci = 1 and the region M = 1, and 1 ≤ Ci ≤ C_num, 1 ≤ M ≤ N, where C_num is the total number of needled unit layers and N is the total number of regions in each unit layer. Step 3: The robot control unit controls the robot to move from the initial safe position to the initial point of the unit layer Ci region M in step 2; Step 4: Detect the thickness of the precast body surface in region M using the surface thickness detection unit; Step 5: The main controller receives and processes the prefabricated body surface thickness in step 4 in real time through the analog input module, judges the change in the prefabricated body surface thickness, and calculates the number of needle plate combinations of the needle punching execution unit. Step 6: Based on the number of needle plate combinations calculated in Step 5, the main controller performs pneumatic control on the needle acupuncture execution unit to add or remove sheet needle plates in the combined needle plate, and then presses the sheet needle plates with a cylinder to form a new combined needle plate. Step 7: Based on the new combined needle plate formed in Step 6, according to the needle density requirement of region M, the main controller outputs a robot contouring motion path switching signal to the robot controller of the robot control unit. After the robot controller calls the robot contouring motion path that meets the needle density and combined needle plate requirements, it controls the robot to move to the needle puncture target trajectory point and sends a signal to the main controller that the needle puncture target trajectory point has been reached. The number of times the combined needle plate actually moves to the needle puncture target trajectory point in region M is incremented by one. According to the needle depth requirement of region M, the main controller controls the needle puncture execution unit to perform needle puncture motion. After the needle puncture motion is reached, the diffuse reflection photoelectric sensor S6 detects the completion of the needle puncture motion and sends a signal to the main controller and the needle density detection unit that the needle puncture motion has been reached. The number of needle puncture executions in unit layer Ci is incremented by one and the number of needle puncture executions of the combined needle plate in region M is incremented by one. Step 8: Set the allowed number of moves of the combined needle plate within region M, determine the actual number of moves of the combined needle plate to the needle target trajectory point within region M and the number of needle-piercing operations performed by the robot within unit layer Ci. The specific implementation is as follows; Step 8.1: Determine the actual number of times the current combined needle plate moves within region M to the target needle trajectory point; If the actual number of times the needle-piercing target trajectory point moves is greater than or equal to the allowed number of moves, it means that the thickness of the prefabricated body surface in region M needs to be re-detected and the combined needle plate updated. Then, the actual number of times the needle-piercing target trajectory point moves is cleared to zero, and steps 4-8 are executed again in a loop. If the actual number of times the target trajectory point is moved is less than the allowed number of moves, it means that the current combination of needle plates can continue to be used, and then continue to repeat steps 7-8. Step 8.2: Determine the number of times the robot performs needle penetration within unit layer Ci; If the number of needle punctures is greater than or equal to the needle replacement frequency, it indicates that the needle is worn, and the main controller will output a needle replacement warning signal. If the number of acupuncture treatments is less than the frequency of needle replacement, it means that the needle is intact and can continue to be used. Then, continue to repeat steps 7-8. Step 9: After receiving the acupuncture motion completion signal from Step 7, the detection area of ​​the needle density detection unit is set according to the change in the thickness of the preform surface in Step 5. The preform image is acquired by the vision detection device in combination with the robot's contouring motion path under a constant focal length. The acquired image is processed by the image algorithm processing platform to detect and count the discrete state of needle density and needle trace distribution within the area in real time. The statistical results are fed back to the main controller in real time to judge the needle puncture quality of the preform. If the needle density deviates or the needle marks show stripes, diagonal lines, or clumps, it indicates that the needle piercing quality of the precast body is abnormal, and the main controller will output an abnormality warning in real time. If there is no deviation in needle density or no stripes, diagonal lines, or lumps in the needle marks, it indicates that the needle puncture quality of the precast body is normal, and the needle puncture operation should continue. Step 10: Based on the acupuncture motion completion signal in Step 7, determine whether the current acupuncture target trajectory point is the last one in the unit layer Ci region M; If the current acupuncture point is the last one, it means that the acupuncture movement in region M has been completed, so proceed to step 11; If the current acupuncture point is not the last one, it means that the acupuncture movement in region M has not been completed, so the steps 7-10 are executed repeatedly. Step 11: Determine whether region M is the last needle-splitting motion executed within unit layer Ci; If region M is the last region to be needled, it means that all regions within unit layer Ci have completed needled movements, then proceed to step 12. If region M is not the last needle-punching motion, it means that there are still regions in unit layer Ci that have not completed the needle-punching motion. Then the count of region M is incremented by one, and steps 7-11 are executed repeatedly. Step 12: Determine if cell layer Ci is the last cell layer; If unit layer Ci is the last one, it means that all unit layers of the prefabricated body have completed the needle-punching motion. The robot controller controls the robot to move to the initial safe position and feeds back the prefabricated body contouring needle-punching completion signal through the bus. If unit layer Ci is not the last one, it means that there are still unit layers in the prefabricated body that have not completed the needle-piercing movement. Then, the count of unit layer Ci is incremented by one, region M is reassigned to 1, and steps 3-12 are executed in a loop.

9. The robot-based contour needle puncture control method according to claim 8, characterized in that, Step 5, the calculation formula for determining the change in the thickness of the precast body surface is as follows: Δh i =|h5-h i |,1≤i≤4 The number of needle plates combined is calculated using the following formula: Where A is the distance between displacement sensors S1, S2, S3, and S4 and displacement sensor S5 respectively, B is the width of the sheet-like needle plate, Δh0 is the maximum allowable needle depth fluctuation, Δh is the average deviation of the surface thickness, and Δh i h is the profile thickness deviation value, h5 is the profile thickness detected by displacement sensor S5, h i Let be the thickness of the profile detected by the i-th displacement sensor, and n be the number of sheet-like needle plates.

10. The robot-based contour needle puncture control method according to claim 8, characterized in that, Step 6 also includes the following steps: Step 6.1: The main controller adjusts the electro-proportional valve through the analog output module to control the negative pressure in the needle acupuncture unit; Step 6.2: The main controller adjusts the pneumatic solenoid valve b to disconnect through the digital output module, thereby controlling the cylinder in the needle puncture execution unit to release the combined needle plate; Step 6.3: The main controller adjusts the engagement and disengagement of the pneumatic solenoid valve a through the digital output module, and controls the adsorption state of the sheet-like needle plate in the corresponding area in the acupuncture execution unit to complete the combination of the sheet-like needle plate. If a sheet-like needle plate needs to be added to the corresponding area, adjust the pneumatic solenoid valve a to engage, and negative pressure gas flows into the negative pressure adsorption mechanism to generate adsorption force and adsorb the sheet-like needle plate. If the number of sheet-like needle plates needs to be reduced in the corresponding area, adjust the pneumatic solenoid valve a to disconnect, stop the flow of negative pressure gas into the negative pressure adsorption mechanism, and release the sheet-like needle plates; Step 6.4: The main controller adjusts the pneumatic solenoid valve b to engage through the digital output module, and controls the cylinder in the needle acupuncture unit to press the sheet-like needle plate in step 6.3 to form a new combined needle plate.

Images