Assembly system and assembly method

Abstract

The assembly system comprises a sensing detection module which is composed of a proximity sensor array unit and a force sensitive detection unit, the proximity sensor array unit is arranged on the periphery of a to-be-installed area of an assembly object, and the force sensitive detection unit is arranged at the tail end of an execution operation module; the data processing and fusion module is used for processing the data from the perception detection module, generating space deviation parameters and mechanical state information of the assembly object relative to the installation reference, and outputting the space deviation parameters and the mechanical state information to the decision control module; the decision control module is used for generating a control instruction for executing the operation module based on the information from the data processing and fusion module; and the execution operation module acts based on the control instruction from the decision control module and feeds back the state information to the decision control module. And the real-time, accurate and rapid regulation and control of the clearance and the step difference of the parts in the assembly process can be realized. On the premise of guaranteeing the flight safety and performance, the assembly efficiency of the windshield and the passenger observation window is greatly improved.

Description

Technical Field

[0001] This invention relates to an assembly system and assembly method, and more particularly to an assembly system and assembly method for aircraft windshields and passenger observation windows. Background Technology

[0002] In aircraft manufacturing, the windshield is a crucial component, and its assembly precision plays a decisive role in the aircraft's aerodynamic performance, sealing, structural strength, and even flight safety. Therefore, accurately controlling the assembly gaps and step differences of the windshield is of great significance. Once the assembly gaps and step differences of the windshield become uneven or exceed tolerances, it may cause a series of serious problems: not only will it lead to poor windshield sealing, causing air pressure fluctuations and increased noise in the cabin, but it may also adversely affect the aircraft's aerodynamic shape, increase flight drag, and in extreme cases, even endanger flight safety.

[0003] From the perspective of the structural safety of the windshield itself, the assembly gaps and step differences of the windshield often pose a threat to its integrity. (1) Uneven stress. During flight, the windshield is subjected to significant air pressure. If the assembly gap and step difference exceed the tolerance, the windshield will not be able to bear the stress evenly after installation. Excessive local pressure will cause the windshield material to exceed its bearing limit, resulting in cracks. As the number of flights increases, the cracks continue to expand, eventually leading to windshield damage; (2) Vibration effects. Aircraft vibrate during flight. Under normal assembly clearance, a certain buffer space is formed between the windshield and the window frame to reduce vibration transmission. Excessive clearance will change the vibration characteristics of the windshield, making it more prone to resonance with the vibration frequency of the aircraft. Long-term resonance will cause fatigue of the windshield material, reduce its strength, and thus lead to cracks or damage; (3) Sealing failure: Proper assembly clearance is key to a good seal for the windshield. Excessive clearance will lead to poor sealing, allowing external moisture, dust, and other contaminants to enter the gap between the windshield and the window frame, corroding the edge structure. In low-temperature environments, moisture may also freeze and expand, further damaging the connection between the windshield and the window frame and increasing the risk of windshield breakage.

[0004] Similarly, uneven or out-of-tolerance gaps and step differences in the passenger observation window assembly can also cause similar problems.

[0005] The following technical methods are currently used in the field of controlling assembly gaps and step differences in aircraft windshields.

[0006] (1) Detection technology

[0007] Contact measurement relies on manual point-by-point operation using tools such as feeler gauges and dial indicators, which is inefficient and prone to errors. Furthermore, for complex structural parts of the windshield, the measuring tools are difficult to reach, resulting in a limited measurement range and failing to meet the requirements for high-precision and comprehensive windshield inspection.

[0008] Non-contact optical measurement (such as laser ranging, 3D scanning, etc.) has high accuracy but the equipment is expensive and is easily affected by environmental interference.

[0009] Machine vision measurement, which uses cameras to capture windshield images and analyzes them using image processing algorithms, is fast, but reflections and stains on the windshield surface can affect image analysis and processing. Furthermore, the accuracy and stability of current algorithms need to be further improved to better adapt to diverse windshield assembly scenarios.

[0010] (2) Regulation technology

[0011] Traditional manual mechanical adjustments rely on experience, making it difficult to control precision. Automated equipment, driven by servo motors, operates according to preset programs, significantly improving precision and efficiency. However, it struggles to adapt to dynamic factors such as temperature changes and component deformation, thus failing to achieve precise dynamic adjustment.

[0012] In summary, traditional aircraft windshield and passenger observation window assembly relies on manual operation and simple measuring tools, with assembly workers using their experience to position and adjust components. In particular, the windshield structure is complex, involving multiple layers of glass, frames, and seals, making it difficult for manual assembly to guarantee consistency in minute gaps and fine step differences. This traditional, inefficient assembly method severely restricts aircraft production efficiency and cannot meet the ever-increasing order demands of the aviation manufacturing industry.

[0013] The current technology has the following problems.

[0014] (1) Conflict between detection accuracy and efficiency. In the existing technology, although contact measurement is relatively simple to operate, it is inefficient and has large errors, making it difficult to meet the requirements of high accuracy. Although non-contact optical measurement and machine vision measurement have advantages in accuracy or efficiency, the former is expensive and easily affected by environmental interference, while the latter is affected by many factors of image interference. Neither can take both detection accuracy and efficiency into account, making it difficult to meet the needs of large-scale, high-efficiency assembly and inspection of aircraft windshields and passenger observation windows.

[0015] (2) Lack of real-time dynamism in control. Neither traditional manual mechanical adjustment nor automated equipment control can cope with the dynamic changes in the assembly process. Manual adjustment relies on experience and cannot control accurately in real time. Although automated equipment has program settings, it cannot sense component deformation caused by factors such as temperature and stress in real time, and cannot dynamically adjust assembly gaps and step differences, making it difficult to ensure the long-term stability and reliability of the assembly.

[0016] (3) The current detection and control technologies are independent of each other and have not formed an effective closed-loop system. The detected gap and step difference data cannot be fed back to the control link in a timely and accurate manner, and the control link cannot adjust the strategy in real time based on the detection results, making it difficult to achieve coordinated optimization control of the assembly gap and step difference of the windshield and passenger observation window.

[0017] Modern aircraft place extremely high demands on the assembly of windshields and passenger observation windows. To address these issues, there is an urgent need for an intelligent, precise, and rapid assembly method and system for aircraft windshields and passenger observation windows. Summary of the Invention

[0018] This invention addresses the aforementioned technical problems by providing an assembly system and method capable of real-time, precise, and rapid adjustment of component gaps and step differences during the assembly of aircraft windshields and passenger observation windows. This significantly improves the assembly efficiency of windshields and passenger observation windows while ensuring flight safety and performance, meeting the demands of large-scale, high-quality production in the aviation industry.

[0019] One aspect of the present invention relates to an assembly system, comprising: a sensing and detection module, the sensing and detection module comprising a proximity sensor array unit and a force-sensitive detection unit, the proximity sensor array unit being disposed on the outer periphery of the area to be installed of the assembly object, and the force-sensitive detection unit being disposed at the end of an execution operation module; a data processing and fusion module, the data processing and fusion module processing data from the sensing and detection module to generate spatial deviation parameters and mechanical state information of the assembly object relative to an installation reference, and outputting them to a decision control module; a decision control module, the decision control module generating control commands for the execution operation module based on the spatial deviation parameters and mechanical state information from the data processing and fusion module; and an execution operation module, the execution operation module performing actions based on the control commands from the decision control module and feeding back state information to the decision control module.

[0020] The assembly system based on this structure can detect the position and force information of the assembled object in real time and accurately, and process this information to generate spatial deviation parameters and mechanical state information of the assembled object relative to the installation reference. The execution module performs actions based on control commands from the decision control module and feeds back the state information to the decision control module, forming a complete control closed loop of "execution-feedback-re-detection".

[0021] In addition, the sensing and detection module also includes a data acquisition and conversion device. The detection data of the proximity sensor array unit and the detection results of the force-sensitive detection unit are transmitted to the data acquisition and conversion device via shielded cables. After being processed into standard digital signals by the data acquisition and conversion device, they are transmitted to the data processing and fusion module.

[0022] According to this structure, the detection data of the proximity sensor array unit and the detection results of the force-sensitive detection unit can be processed into standard digital signals through the data acquisition and conversion device, and then the standard digital signals are transmitted to the data processing and fusion module.

[0023] Based on this structure, the signal transmission between the sensing and detection module and the data processing and fusion module is optimized, and the wiring can be simplified.

[0024] Ideally, the proximity sensor array unit uses capacitive proximity sensors and is evenly distributed on the outer periphery of the area to be installed on the assembly object.

[0025] The proximity sensor array unit employs a 360° distributed layout of proximity sensors, covering multi-directional displacement and angular deviations for blind-spot-free detection. Simultaneously, the collaborative operation of multiple sensors compensates for the visual blind spots and measurement errors of a single sensor, cross-validating data from multiple angles, thus improving the accuracy and reliability of the detection results and laying a solid data foundation for subsequent precise adjustments.

[0026] In addition, the data acquisition and conversion device includes an integrated signal amplifier, a filter, and an analog-to-digital converter circuit.

[0027] The data acquisition and conversion device based on this structure can amplify and filter the detection data from the proximity sensor array unit and the detection results from the force-sensitive detection unit, and convert them into digital signals.

[0028] Ideally, the data processing and fusion module includes: a dedicated interface for acquiring data from the sensing and detection module; an adaptive noise filtering circuit for noise feature extraction and adaptive threshold adjustment; a Kalman filter operation unit for iteratively optimizing data accuracy based on minimum mean square error; a weighted fusion algorithm module for generating spatial deviation parameters and mechanical state information of the assembly object relative to the installation datum and standardizing and encapsulating them based on the coupling correlation of displacement-pressure data; and an output port for outputting the data from the weighted fusion algorithm module to the decision control module.

[0029] Based on the data processing and fusion module of this structure, environmental interference signals can be dynamically suppressed and a multi-dimensional fusion model can be constructed.

[0030] Furthermore, after receiving the installation deviation parameters and stress state information of the assembly object output from the data processing and fusion module, the decision control module generates the translation distance and rotation angle of the execution operation module based on PID and MPC algorithms by the central control chip.

[0031] Based on the decision control module of this structure, an intelligent decision control device can be constructed to achieve efficient calculation and real-time transmission of control commands.

[0032] Ideally, the decision control module also includes an online safety threshold detection circuit and a graded emergency braking logic module. The online safety threshold detection circuit monitors the force state information output by the data processing and fusion module in real time. When the force state information is exceeded, the emergency braking logic module executes an abnormal operating condition response.

[0033] By setting up an online safety threshold detection circuit and a graded emergency braking logic module, the system can prevent the forces on the assembled objects from exceeding the safety threshold, thus ensuring system safety.

[0034] In addition, the execution module includes a robotic arm, a flexible gripping unit, and a servo drive system. Each joint of the robotic arm is equipped with an absolute encoder and a servo motor. A flexible gripping unit is connected to the end of the robotic arm, and the flexible gripping unit has a built-in pressure detection sensor 4021.

[0035] The execution module of this structure is equipped with a flexible gripping unit, and a pressure detection sensor is installed within the flexible gripping unit, which can detect the force on the assembly object in real time.

[0036] Ideally, the flexible gripping unit includes a vacuum suction cup and a silicone buffer, and the vacuum suction cup is configured to dynamically adjust the adsorption pressure.

[0037] The flexible gripping unit of this structure can adapt to different assembly objects, adjust the adsorption pressure, and ensure the safety of the assembly objects from a physical perspective.

[0038] Another aspect of the present invention relates to an assembly method, comprising: a sensing and detection step, wherein position information of an assembly object is detected by a proximity sensor array unit and force on the assembly object is detected by a force-sensitive detection unit; a data processing and fusion step, wherein the data acquired by the sensing and detection step is processed to generate spatial deviation parameters and mechanical state information of the assembly object relative to an installation reference; a decision control step, wherein control commands are generated based on the spatial deviation parameters and mechanical state information from the data processing and fusion step; and an execution operation step, wherein an action is performed based on the control commands from the decision control step, and status information is fed back. Attached Figure Description

[0039] Figure 1 This is a block diagram illustrating the structure of the assembly system according to an embodiment of the present invention.

[0040] Figure 2(a) is a schematic diagram showing the distribution array of the main windshield proximity sensors in the assembly system of an embodiment of the present invention.

[0041] Figure 2(b) is a schematic diagram showing the distribution array of the side windshield proximity sensors of the assembly system according to an embodiment of the present invention.

[0042] Figure 3 This is a block diagram illustrating the structure of the sensing and detection module of the assembly system according to an embodiment of the present invention.

[0043] Figure 4 This is a block diagram illustrating the structure of the data processing and fusion module of the assembly system according to an embodiment of the present invention.

[0044] Figure 5 This is a block diagram illustrating the structure of the decision control module of the assembly system according to an embodiment of the present invention.

[0045] Figure 6 This is a block diagram illustrating the structure of the execution operation module of the assembly system according to an embodiment of the present invention.

[0046] (Symbol Explanation)

[0047] 2 Main windshield

[0048] 3 Side windshields

[0049] 100 Perception and Detection Module

[0050] 101 Proximity Sensor Array Unit

[0051] 102 Force-sensitive detection unit

[0052] 1021 Pressure Sensor

[0053] 200 Data Processing and Fusion Module

[0054] 201 Dedicated Interface

[0055] 202 Adaptive Noise Filtering Circuit

[0056] 203 Kalman Filter Operation Unit

[0057] 204 Weighted Fusion Algorithm Module

[0058] 205 Output Port

[0059] 300 Decision Control Module

[0060] 301 PID Control Algorithm Calculation Unit

[0061] 302 MPC Algorithm Module

[0062] 303 Safety Threshold Online Detection Circuit

[0063] 304 Graded Emergency Braking Logic Module

[0064] 305 Device Configuration Data Input Port

[0065] 306 Central Control Chip

[0066] 307 Command Output Port

[0067] 308 Status Feedback Interface

[0068] 400 Execution Operation Module

[0069] 401 robotic arm

[0070] 402 Flexible Gripping Unit

[0071] 403 Servo Drive System

[0072] 4011 Joint

[0073] 4022 Status Feedback Interface

[0074] 4023 Vacuum Suction Cup

[0075] 4024 silicone cushioning

[0076] 500 Data Acquisition and Conversion Device

[0077] 501 Integrated Signal Amplifier

[0078] 502 filter

[0079] 503 Analog-to-Digital Converter Circuit

[0080] L shielded cable

[0081] P Controller Area Network Bus Detailed Implementation

[0082] Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings. In the drawings, the same symbols are used to denote the same components, and repeated descriptions are sometimes omitted. Furthermore, in the drawings, dimensions and shapes are sometimes exaggerated to facilitate understanding of the invention. In addition, in the following detailed description, directional terms such as "upper," "lower," "left," "right," "front," and "rear" are used for illustrative purposes and not for limitation. In this specification, the plane direction of the surface to be mounted is used as the X / Y axis plane, and the direction perpendicular to the X / Y plane is used as the Z-axis direction.

[0083] In this specification, the assembly of an aircraft windshield is used as an example (the same applies to passenger observation windows), but the assembly system of this invention can also be applied to the installation of other equipment, such as the installation of car windows.

[0084] Figure 1 This is a block diagram illustrating the structure of the assembly system according to an embodiment of the present invention. For example... Figure 1 As shown, the assembly system of this invention includes a sensing and detection module 100, a data processing and fusion module 200, a decision control module 300, and an execution module 400. The modules are physically connected via industrial bus communication protocols and dedicated control lines, and through logical interactions such as data transmission, command response, and status feedback, a fully automated process is achieved from position detection, data processing, control decision-making to precise execution. Through the collaborative cooperation of these modules, a closed-loop architecture of "detection-processing-decision-execution" is formed, enabling automated and precise fine-tuning of the installation of the main windshield, side windshields, etc.

[0085] First, the structure of the perception and detection module 100 will be explained. Figure 3 This is a block diagram illustrating the structure of the sensing and detection module of the assembly system according to an embodiment of the present invention.

[0086] The sensing and detection module 100 adopts a composite multidimensional sensing device, which consists of a distributed proximity sensor array unit 101 and a force-sensitive detection unit 102. In addition, it also includes a data acquisition and conversion device 500 to acquire and convert data from the signals received from the proximity sensor array unit 101 and the force-sensitive detection unit 102.

[0087] Figure 2(a) is a schematic diagram showing the distribution array of the main windshield proximity sensors in the assembly system according to an embodiment of the present invention. Figure 2(b) is a schematic diagram showing the distribution array of the side windshield proximity sensors in the assembly system according to an embodiment of the present invention. As shown in Figures 2(a) and 2(b), the proximity sensor array units 101 are uniformly distributed on the outer perimeter frame of the installation area of ​​the main windshield 2 and the side windshield 3, which are the objects to be assembled. In Figure 2(a), the proximity sensor 2 a ~2 i The outer perimeter frame, evenly distributed in the installation area of ​​the main windshield 2, includes proximity sensor 3 in Figure 2(b). a ~3 j The outer perimeter frame is evenly distributed in the installation area of ​​the side windshield 3.

[0088] Capacitive proximity sensors are selected as proximity sensors to meet the requirements of measurement accuracy and sampling frequency, thus achieving the desired precision and response speed. These proximity sensors 2 a ~2 i 3 a ~3 jThe X / Y axis plane deviation (horizontal direction) is detected, the tilt angle θ is calculated through the height difference, and the distance data of the Z axis (perpendicular to the mounting surface) is collected simultaneously. Then, the proximity sensor array unit 101 transmits this data, i.e. the position information of the assembled object, to the data acquisition and conversion device 500 via the shielded cable L.

[0089] The force-sensitive detection unit 102 is integrated into the stress concentration area of ​​the flexible gripper at the end of the robotic arm 401 of the execution module 400, and has a built-in pressure sensor 1021. The detection result of the pressure sensor 1021 is also transmitted to the data acquisition and conversion device 500 via the shielded cable L.

[0090] The data acquisition and conversion device 500 includes an integrated signal amplifier 501, a filter 502, and an analog-to-digital converter 503. Since the integrated signal amplifier 501, filter 502, and analog-to-digital converter 503 employ existing structures, detailed descriptions of them are omitted here.

[0091] According to the sensing and detection module 100 of this structure, the proximity sensor array unit 101 collects in real time the translation parameters and tilt angle (θ) data of the windshield and its mounting reference surface in the X, Y, and Z axes. The force-sensitive detection unit 102 simultaneously detects the pressure changes when the robotic arm 401 contacts the main windshield 2 and the side windshield 3. All detection data are processed into standard digital signals by the data acquisition and conversion device 500 and transmitted to the data processing and fusion module 200 through the controller local area network bus P.

[0092] The structure of the data processing and fusion module 200 will be described next.

[0093] Figure 4 This is a block diagram illustrating the structure of the data processing and fusion module of the assembly system according to an embodiment of the present invention. For example... Figure 4 As shown, the data processing and fusion module 200 includes a dedicated interface 201, an adaptive noise filtering circuit 202, a Kalman filter operation unit 203, a weighted fusion algorithm module 204, and an output port 205, thereby constructing an intelligent data processing device.

[0094] After the data processing and fusion module 200 collects the raw data from the sensing and detection module 100 through the dedicated interface 201, the adaptive noise filtering circuit 202 first dynamically suppresses environmental interference signals through noise feature extraction and adaptive threshold adjustment. Then, a multi-dimensional state-space model is established through the Kalman filter operation unit 203, and the data accuracy is iteratively optimized based on the minimum mean square error. Finally, the weighted fusion algorithm module 204 constructs a multi-dimensional fusion model based on the coupling correlation of displacement-pressure data, calculates and generates the spatial deviation parameters and mechanical state information of the main windshield 2 and side windshields 3 relative to the installation reference, and outputs them to the decision control module 300 after standardization and encapsulation.

[0095] Figure 5 This is a block diagram illustrating the structure of the decision control module of the assembly system according to an embodiment of the present invention. For example... Figure 5 As shown, the decision control module 300 includes a proportional-integral-derivative (PID) control algorithm operation unit 301, a model predictive control (MPC) algorithm module 302, an online safety threshold detection circuit 303, a graded emergency braking logic module 304, a device configuration data input port 305, a central control chip 306, an instruction output port 307, and a status feedback interface 308, thereby constructing an intelligent decision control device to realize efficient operation and real-time transmission of control instructions.

[0096] The operating mechanism of the decision control module 300 is as follows.

[0097] After receiving the installation deviation parameters and stress status information of the main windshield 2 and side windshields 3 from the data processing and fusion module 200, the decision control module 300 uses the central control chip 306 to collaboratively calculate and generate adjustment parameters such as the translation distance and rotation angle of the robotic arm 401, forming a refined control command sequence. Simultaneously, the online safety threshold detection circuit 303 monitors the status variables output by the data processing and fusion module 200 in real time. When a safety threshold is exceeded, the graded emergency braking logic module 304 is triggered to execute an abnormal operating condition response.

[0098] Figure 6 This is a block diagram illustrating the structure of the execution operation module of the assembly system according to an embodiment of the present invention. For example... Figure 6 As shown, the execution module 400 employs a six-axis servo actuator, including a robotic arm 401, a flexible gripping unit 402, and a servo drive system 403. Each joint 4011 of the robotic arm 401 is equipped with a high-precision absolute encoder 4012 and a servo motor 4013. Furthermore, the flexible gripping unit 402 is connected to the end of the robotic arm 401 via a quick-change interface. The flexible gripping unit 402 has a built-in pressure detection sensor 4021. Additionally, the flexible gripping unit 402 includes a vacuum suction cup 4023 and a silicone buffer 4024. The vacuum suction cup 4023 is configured to dynamically adjust the suction pressure.

[0099] By employing this execution module, a repeatability accuracy of ±0.03mm can be achieved, supporting force control mode with a force control range of 0-80N and an accuracy of ±0.3N. Furthermore, the suction force of the vacuum suction cup is adjustable from 80-250N, and the vacuum suction cup and the robotic arm are connected via a 3-DOF flexible joint, allowing for ±2° rotational deviation compensation. The robotic arm controller communicates with the sensor system via a bus, ensuring low command response latency.

[0100] The operating mechanism of the execution module 400 is as follows.

[0101] After receiving control commands from the decision control module 300, the execution module 400 causes the servo drive system 403 to control the joints 4011 of the robotic arm 401, enabling linear translation and fine-tuning of the attitude angles of the main windshield 2 and side windshield 3 in the X, Y, and Z axes. The flexible gripping unit 402 stably holds the main windshield 2 and side windshield 3 with a constant vacuum suction force and feeds back real-time pressure data to the sensing and detection module 100 via the status feedback interface 4022.

[0102] The execution module 400 receives control commands and executes actions via the servo control bus, while simultaneously feeding back status information such as position and pressure to the decision control module 300. After each adjustment, the status feedback interface 4022 triggers the sensing and detection module 100 to start a new round of data acquisition, forming a complete control closed loop of "execution-feedback-re-detection".

[0103] The assembly system based on the above structure has the following advantages.

[0104] 1. Distributed, multi-dimensional perception for accurate detection across the entire domain.

[0105] Breaking through the limitations of traditional single-sensor or local detection, by setting up a distributed proximity sensor array unit, a multi-dimensional and comprehensive windshield / passenger observation window installation deviation detection system can be constructed. This system can accurately capture installation posture errors, including minute deviations such as micro-displacement and small-angle tilt, eliminating sealing failures and structural hazards at the source.

[0106] Traditional windshield / passenger observation window installations rely on vision or a single sensor, which can only detect positional deviations in one direction. This easily leads to missed vertical offsets and angular tilts, resulting in problems such as loose fit and poor sealing. The assembly system of this invention utilizes a 360° blind-spot-free inspection by employing a 360° distributed layout of proximity sensors to cover multi-directional displacement and angular deviations. Simultaneously, the collaborative work of multiple sensors compensates for the visual blind spots and measurement errors of a single sensor, cross-validating data from multiple angles, improving the accuracy and reliability of the detection results, and laying a solid data foundation for subsequent precise adjustments.

[0107] 2. Flexible and safe glass adaptation

[0108] Focusing on the technical challenges of glass's brittleness and difficulty in adapting its specifications, we have constructed a safe adaptation system that combines flexible gripping with precise force control. This system can cover diverse glass installation scenarios, achieving zero damage during installation and full compatibility across all scenarios.

[0109] Traditional installation methods are prone to causing glass to bump and break, and the compatibility with different glass sizes is poor. The assembly system of this invention has achieved a dual breakthrough in both hardware and algorithm.

[0110] In terms of hardware, a flexible gripping unit is set at the end of the robotic arm. This flexible gripping unit adopts a design of vacuum suction cup + silicone buffer. The vacuum suction cup is designed to dynamically adjust the adsorption pressure (to adapt to glass of different weights), and the silicone buffer is used to prevent rigid impact, thus ensuring the safety of the glass from a physical perspective.

[0111] In terms of algorithms, the execution module receives control commands and executes actions via the servo control bus, while simultaneously feeding back status information such as position and pressure to the decision control module. After each adjustment, the perception and detection module is triggered via the status feedback interface to start a new round of data acquisition, forming a complete control closed loop of "execution-feedback-re-detection".

[0112] 3. Algorithm Control System

[0113] The assembly system of this invention uses an industrial PC running a real-time operating system. The processing and fusion module employs an extended Kalman filter algorithm to perform spatiotemporal synchronous correction on the proximity sensor data, outputting the center coordinates (Δx, Δy, Δz) of the aircraft's windshield / passenger observation window glass and the tilt angles (Δθx, Δθy) around the X and Y axes. The decision control module uses an incremental PID algorithm to generate translational adjustment amounts based on the deviation values ​​Δx, Δy, and Δz, which are then converted into rotation angles of each axis of the robotic arm through inverse kinematics calculation. Furthermore, an online safety threshold detection circuit is integrated, allowing setting of a force sensor threshold (set to 40N) and spatial boundary parameters (an 8mm extension beyond the edge of the installation area as a safety zone), immediately sending an emergency stop command upon triggering.

[0114] The structure of the assembly system has been described above. The assembly method of the assembly system will be described next.

[0115] 1) Step 1: Initialization and Calibration

[0116] After the assembly system is started, the vacuum suction cup 4023 of the robotic arm 401 moves the aircraft's windshield to the installation area to wait (waiting step). The assembly system's control system triggers the sensor calibration program, performs zero-point calibration on the proximity sensors using a standard calibration block, and stores the initial offset of each proximity sensor relative to the aircraft's windshield mounting reference surface window frame in a register (sensor calibration step). Simultaneously, the vacuum suction cup detects the negative pressure value (≥-90KPa) to confirm that the aircraft's windshield has been firmly gripped (negative pressure detection step).

[0117] 2) Step 2: Preliminary positioning

[0118] The robotic arm 401 moves the aircraft's windshield along a predetermined path to a designated position, for example, 150mm, directly in front of the installation area. At this point, distributed proximity sensors begin collecting data. The proximity sensors detect the horizontal distance (distance in the X and Y axes) between the edge of the aircraft's windshield and the installation area, as well as the vertical height difference between the upper and lower edges of the aircraft's windshield and the window frame of the installation reference surface (used to calculate the tilt angle θ) (proximity sensor detection step).

[0119] The data processing and fusion module 200 completes the filtering process within 4ms and outputs the initial deviation of the distance from the mounting surface and the tilt angle θ (data processing and output steps).

[0120] 3) Step 3: Closed-loop adjustment

[0121] The decision control module 300 generates adjustment instructions (adjustment steps) based on the initial deviation: a) Z-axis direction: Drive the robotic arm closer to the mounting surface, and switch to low-speed mode when Δz≤8mm; b) X-axis and Y-axis directions: Calculate the adjustment amount according to the PID algorithm, first compensate Δx, and then correct Δy synchronously; c) Tilt correction: θ is adjusted by the rotary joint at the end of the robotic arm, and the proximity sensor provides real-time feedback on the deviation changes during the process.

[0122] 4) Step 4: Fine-tuning and confirmation

[0123] When the distance between the aircraft's windshield and the mounting surface is ≤1.5mm, the assembly system enters fine-tuning mode, reducing the adjustment speed and increasing the data refresh rate of the proximity sensor to continuously monitor the deviation value. When the installation conditions are met, the control system sends a "in place" signal. The robotic arm 401 triggers the vacuum suction cup 4023 to release pressure while maintaining suction to prevent slippage. Simultaneously, the installation auxiliary mechanism is activated, and the windshield is installed and fixed through the automatic riveting device.

[0124] 5) Step 5: Safe Exit

[0125] After confirming that the aircraft windshield is securely installed, the gripper at the end of the robotic arm releases and withdraws along a safe path, maintaining a distance of ≥60mm from the aircraft windshield. The assembly system records the installation data, including the number of adjustments, maximum deviation, and time taken, before proceeding to the next work cycle.

[0126] The control system and method described in this disclosure can also be implemented using a dedicated computer, which is provided by comprising a processor and memory, the processor being programmed to perform one or more functions embodied in the computer program. Alternatively, the control unit and method described in this disclosure can be implemented using a dedicated computer, which is provided by comprising a processor composed of one or more dedicated hardware logic circuits. Alternatively, the control unit and method described in this disclosure can be implemented using one or more dedicated computers, which are constituted by a combination of a processor and memory programmed to perform one or more functions and a processor composed of one or more hardware logic circuits. Furthermore, the computer program can also be stored on a computer-readable, non-transitory tangible recording medium as instructions to be executed by the computer.

[0127] The embodiments and variations of the present invention have been described above. However, it should be understood that this disclosure is not limited to the above embodiments and structures. This disclosure also includes various variations and modifications within the equivalent scope. In addition, various combinations and methods, and further combinations and methods that include only one element or more or less thereof, also fall within the scope and spirit of this disclosure.

[0128] For example, in the above embodiment, the detection data of the proximity sensor array unit 101 and the detection results of the force-sensitive detection unit 102 are transmitted to the data acquisition and conversion device 500 via the shielded cable L. The data acquisition and conversion device 500 processes the data into standard digital signals, which are then transmitted to the data processing and fusion module 200 via the controller local area network bus P. However, the present invention is not limited to this; the data acquisition and conversion device 500 may be omitted, and the detection data of the proximity sensor array unit 101 and the detection results of the force-sensitive detection unit 102 may be directly sent to the data processing and fusion module 200.

Claims

1. An assembly system, characterized by It comprises: a perception detection module (100) composed of a proximity sensor array unit (101) arranged on the outer periphery of the installation area of the assembly object and a force-sensitive detection unit (102) arranged at the end of the execution operation module; a data processing and fusion module (200) that processes data from the perception detection module, generates spatial deviation parameters and mechanical state information of the assembly object relative to the installation reference, and outputs to the decision control module; a decision control module (300) that generates control instructions for the execution operation module based on spatial deviation parameters and mechanical state information from the data processing and fusion module; and an execution operation module (400) that acts based on control instructions from the decision control module and feeds back state information to the decision control module.

2. The assembly system of claim 1, wherein the perception detection module further comprises a data acquisition and conversion device (500), the detection data of the proximity sensor array unit and the detection results of the force-sensitive detection unit are transmitted to the data acquisition and conversion device via shielded cable (L), and after being processed into standard digital signals by the data acquisition and conversion device, they are transmitted to the data processing and fusion module.

3. The assembly system of claim 1 or 2, wherein the proximity sensor array unit uses capacitive proximity sensors and is uniformly arranged on the outer periphery of the installation area of the assembly object.

4. The assembly system of claim 2, wherein the data acquisition and conversion device comprises an integrated signal amplifier (501), a filter (502), and an analog-to-digital conversion circuit (503).

5. The assembly system of claim 1 or 2, wherein the data processing and fusion module comprises: a dedicated interface (201) that acquires data from the perception detection module; an adaptive noise filtering circuit (202) that extracts noise characteristics and adjusts adaptive thresholds; a Kalman filtering operation unit (203) that optimizes data accuracy based on least mean square error iteration; a weighted fusion algorithm module (204) that generates spatial deviation parameters and mechanical state information of the assembly object relative to the installation reference based on the coupling correlation of displacement-pressure data and standardizes the packaging; and an output port (205) that outputs data from the weighted fusion algorithm module to the decision control module.

6. The assembly system of claim 5, wherein after receiving installation deviation parameters and force state information about the assembly object output from the data processing and fusion module, the decision control module generates the translation distance and rotation angle of the execution operation module based on PID and MPC algorithms by a central control chip (306).

7. The assembly system of claim 6, wherein The decision control module further comprises a safety threshold online detection circuit (303) and an emergency braking logic module (304), the safety threshold online detection circuit monitors the force state information output by the data processing and fusion module in real time, and when it is detected that the safety threshold is exceeded, the emergency braking logic module executes an abnormal working condition response.

8. The assembly system of claim 1 or 2, wherein, The execution operation module comprises a mechanical arm (401), a flexible gripping unit (402) and a servo drive system (403), an absolute value encoder (4012) and a servo motor (4013) are arranged at each joint (4011) of the mechanical arm, a flexible gripping unit (402) is connected to the end of the mechanical arm, and a pressure detection sensor (4021) is built in the flexible gripping unit.

9. The assembly system of claim 8, wherein, The flexible gripping unit comprises a vacuum suction cup (4023) and a silica gel buffer (4024), and the vacuum suction cup is arranged to be able to dynamically adjust the adsorption pressure.

10. An assembly method applied to the assembly system of claim 1, characterized in that, Comprise: a perception detection step, in which the position information of the assembly object is detected by a proximity sensor array unit, and the force of the assembly object is detected by a force-sensitive detection unit; a data processing and fusion step, in which the data obtained by the perception detection step is processed to generate the spatial deviation parameters and mechanical state information of the assembly object relative to the installation reference; a decision control step, in which the spatial deviation parameters and mechanical state information from the data processing and fusion step are used to generate control instructions; and an execution operation step, in which actions are taken based on the control instructions from the decision control step, and state information is fed back.

Images