An aero-engine blind cavity nut automatic tightening robot system and method

Abstract

The present application belongs to the technical field of aero-engine bearing cavity component assembly, and particularly relates to an automatic tightening robot system and method for a blind cavity nut of an aero-engine. The robot system comprises a clamping mechanical arm and a visual positioning system for identifying a positioning hole on the surface of an aero-engine bearing cavity component. The output end of the clamping mechanical arm is connected with a blind cavity wrench. The lower end of the blind cavity wrench is provided with a sleeve for sleeving the nut. The sleeve is provided with a spring pin for fixing the nut. The output end of the clamping mechanical arm is further fixed with a positioning pin for cooperating with the positioning hole on the surface of the aero-engine bearing cavity component. The lower end of the blind cavity wrench is further provided with a pinhole camera for observing the relative position of the nut and the bolt. A force sensor is installed on the flange at the end of the clamping mechanical arm. The present application provides an automatic tightening robot system and method for a blind cavity nut of an aero-engine. The target bolt is accurately positioned based on 3D vision and a positioning pin. The soft automatic assembly operation is realized based on force control guidance.

Description

Technical Field

[0001] This invention belongs to the field of aero-engine bearing cavity component assembly technology, and specifically relates to an automatic tightening robot system and method for aero-engine blind cavity nuts. Background Technology

[0002] The internal structure of aero-engine bearing cavities is complex, containing numerous bolt arrays within confined blind cavities. These fasteners present several challenges in actual assembly: limited space, blind spots, and high tightening torque, leading to low efficiency, high error rates, and poor consistency in manual operations. Therefore, achieving high-precision, high-consistency automated bolt tightening assembly in confined blind cavity environments has become a critical technical challenge that urgently needs to be overcome in the current aerospace assembly field.

[0003] Bolt tightening methods can be divided into manual, electric, pneumatic or hydraulic drive.

[0004] 1) Manual mechanical torque wrench: Manual mechanical torque wrenches generally employ a ratchet structure, whose core components include a ratchet, pawl, and return spring, providing both unidirectional rotation and bidirectional locking. During clockwise operation, the ratchet allows free rotation and applies torque; conversely, it locks the movement, ensuring efficient force transmission. This design enables the operator to apply torque to threaded connections stably and continuously.

[0005] AtlasCopco's SWR series of preset torque wrenches employs this ratchet design. Once the preset torque value is reached, the wrench features an automatic slip mechanism that triggers disengagement at the critical torque point, allowing the transmission system to idle and preventing overtightening due to continuous force.

[0006] Digital torque wrenches are equipped with a digital display screen, allowing users to more intuitively view and set torque values, while also greatly simplifying the operation process. When the actual torque reaches the preset value, the digital torque wrench will emit an audible sound or other means to alert the user, ensuring accuracy and convenience of operation. AtlasCopco's MWR-KIT intelligent mechanical electronic wrench integrates a torque sensor to monitor dynamic torque values ​​during tightening. It supports seamless integration with the PowerFocus 6000 control system, enabling unified management of different series of intelligent electric and manual tools. 2) Electric torque wrench: As industrial assembly tasks become increasingly complex, power tools, with their stable output and intelligent control capabilities, are widely used in connecting large structural components. For example, the constant torque electric wrench launched by Shanghai Shenmo Electric Co., Ltd. features automatic torque control and is widely used in high-strength structures such as bridges, towers, chemical equipment, and large machinery assembly to meet the demands for high-precision and high-strength connections.

[0007] 3) Pneumatic torque wrench: Pneumatic tools demonstrate excellent stability in high-intensity, continuous operations. Shanghai Paite Industrial Co., Ltd.'s pneumatic torque wrenches utilize imported Japanese heavy-duty motors, providing stable output over extended periods and making them suitable for installing and removing heavy-duty nuts. Their reaction-free structure allows operators to maintain a relaxed posture during high-intensity work, while preventing nut damage due to impact, providing reliable assurance for the safe assembly of large industrial equipment.

[0008] 4) Automated fastener tightening system: Automated bolt / nut tightening systems can be divided into robotic tightening systems and automated tightening equipment according to their implementation methods.

[0009] 4.1) Robotic tightening system Represented by ABB's CRB15000 collaborative robot, collaborative automated tightening systems have been widely applied in assembly lines for engines, automobiles, and other industries. The CRB15000 boasts high force control precision and human-robot collaboration capabilities, allowing it to complete tightening tasks while sharing space with operators without the need for safety barriers. This robot is particularly suitable for production line scenarios with limited space and high levels of human intervention. During production, the robot can consistently tighten each bolt according to pre-set torque and tightening strategies, while torque sensors provide real-time feedback on the tightening status, ensuring that every connection meets quality requirements and effectively improving product reliability and production efficiency.

[0010] In the field of domestically produced collaborative robots, ROKAE robots, which combine high-sensitivity force control strategies and task-level intelligent path planning capabilities, can complete automatic bolt assembly tasks for components such as engine blocks and transmission housings.

[0011] In addition to collaborative robots, some companies have also developed dedicated automated tightening robot systems suitable for standardized production lines. Zowell's PTR5512 three-axis screw robot, based on a SCARA structure, features high-speed, high-repeatability motion performance and integrates a high-precision torque control module.

[0012] Jiangsu Taizhi Technology Co., Ltd. has proposed a device for assembling nuts in the blind cavity of aero-engines. By integrating an adjustable phase angle tightening assembly and a vision observation module onto a robotic arm, the device achieves one-time automated assembly of nuts, retaining rings, and snap rings. The device is compact and highly modular, and can be adapted to various nut specifications by replacing the floating torsion shaft and the anti-torsion fixing shaft, exhibiting good versatility and flexibility.

[0013] Beijing Information Science and Technology University and Aerospace Intelligent Manufacturing (Beijing) Engineering Technology Co., Ltd. have jointly proposed a nut tightening mechanism and robot for use in the blind cavity of aero-engines. Based on the design concept of a tool changer disc and modular tool head, the system enables rapid loading and unloading and automatic tightening, significantly reducing the labor intensity of workers and improving tightening accuracy and assembly efficiency. Its detachable structure gives the robot good task adaptability, making it suitable for multi-station, highly repetitive blind zone tightening tasks.

[0014] 4.2) Automated tightening equipment Beijing University of Aeronautics and Astronautics has developed an automatic nut tightening device and method to address the complex spatial structure inside the compressor disk cavity of aero-turbofan engines. The device employs an L-shaped tightening structure, combined with multi-degree-of-freedom actuators including lifting, deflection, expansion / contraction, and circumferential rotation, along with a nut supply system, to achieve fully automatic nut positioning and tightening in deep blind cavity locations. This system is particularly suitable for areas with limited visibility and tight spacing in low-bypass-ratio engines, demonstrating excellent engineering adaptability.

[0015] Zhejiang University has proposed a hydraulically driven fastener tightening and assembly system and method for blind cavities in aero-engines. The system includes a five-degree-of-freedom nut tightening mechanism and a nut feeding mechanism. Through coordinated operation at both ends, it achieves precise tightening within the confined space of the blind cavity using multi-stage wheel discs. Its hydraulic system, in conjunction with a control unit, enables real-time monitoring and feedback control of the tightening torque, ensuring the visibility and consistency of assembly quality. This makes it particularly suitable for high-precision aero-engine assembly tasks.

[0016] Although a comprehensive product system for bolt tightening has been developed both domestically and internationally, encompassing manual, electric, and pneumatic tools as well as system-level automated tightening equipment, significant shortcomings remain in several key aspects when facing bolt assembly tasks in the blind cavity environment of aero-engines: ① Poor versatility, with limited adaptability to different blind cavity models; ② Insufficient torque transmission capability, making it difficult to meet high-torque tightening requirements; ③ Lack of positioning for the spatial orientation of bolt arrays; ④ Lack of a force control adjustment assembly mechanism that can simulate manual tightening processes, leading to easy interference or docking failure between the tool tip and the bolt, affecting operational stability and reliability. Summary of the Invention

[0017] In order to solve the above-mentioned problems in the existing technology, the purpose of this invention is to provide an automatic tightening robot system and method for blind cavity nuts of aero-engines, which achieves precise positioning of target bolts based on 3D vision and positioning pins, and achieves compliant automatic assembly operation based on force control guidance.

[0018] The technical solution adopted in this invention is as follows: An automatic tightening robot system for blind cavity nuts in aero-engines includes a gripping robotic arm and a vision positioning system for identifying positioning holes on the surface of aero-engine bearing cavity components. The output end of the gripping robotic arm is connected to a blind cavity wrench, and the lower end of the blind cavity wrench is provided with a sleeve for fitting the nut. The output end of the gripping robotic arm is also fixed with a positioning pin for engaging with the positioning hole on the surface of the aero-engine bearing cavity components. The lower end of the blind cavity wrench is also equipped with a pinhole camera for observing the relative position of the nut and bolt. A force sensor is installed on the flange at the end of the gripping robotic arm.

[0019] To achieve the positioning of bolts that cannot be directly observed within a blind cavity, this invention designs a composite positioning scheme integrating 3D machine vision and automatic borehole probing physical perception. The system first uses a visual positioning system (high-precision 3D binocular vision sensor) to acquire 3D point cloud data of the upper surface of the aero-engine bearing cavity. Through feature extraction and matching algorithms, it identifies the predefined positioning holes in the bearing cavity design drawings and calculates their spatial pose. If visual recognition is difficult due to the cavity's depth, surface reflection, or oil contamination, the system can automatically switch to a manual guidance mode to insert the positioning pin into the positioning hole or probe.

[0020] After obtaining the actual pose of one or more positioning holes, the system automatically calculates the three-dimensional spatial pose of each bolt in the entire bolt array in the blind cavity in the robot's base coordinate system through coordinate transformation based on the known geometric relationship of the engine structure (i.e., the fixed relative position, angle and depth dimensions between the positioning holes and each bolt in the blind cavity), providing accurate target points for subsequent assembly.

[0021] Addressing the core challenges of blind cavity assembly in aero-engines, such as extremely limited space and complete visual obstruction, traditional rigid assembly strategies relying on absolute positioning accuracy are prone to jamming or docking failure due to even minor positional deviations. To address this, this invention proposes a compliant automated assembly strategy centered on force feedback and mimicking human touch. This strategy does not pursue one-time, error-free absolute positioning, but rather guides the assembly system from initial contact to a mechanically stable, ideal docking state through active force control interaction, thereby reliably compensating for all unknown positioning and attitude errors. Force control technology is used to transition the contact state between the nut and bolt from unstable to stable.

[0022] Based on stable three-point contact, the gripping robotic arm drives the blind cavity wrench to rotate at low speed in stages, completing the pre-tightening of the nut. The gripping robotic arm maintains its position and locks, allowing the operator to use a manual torque wrench in conjunction with the gripping robotic arm to complete the final tightening. The gripping robotic arm also possesses compliance, performing compliant follow-the-end task to avoid interference.

[0023] As a preferred embodiment of the present invention, the system of the present invention further includes a control system, wherein the visual positioning system, the gripping robotic arm, the pinhole camera, and the force sensor are all electrically connected to the control system; the visual positioning system and / or the positioning pin identify the position of the positioning hole, and the control system calculates the theoretical working point position of all target bolts in the blind cavity based on the actual pose of the identified positioning hole and the geometric constraint relationship of the aero-engine bearing cavity component; the control system uses the relative position of the nut and bolt fed back by the pinhole camera and / or the visual positioning system to precisely position and fine-tune the blind cavity wrench so that the nut in the sleeve is aligned with the bolt.

[0024] In a preferred embodiment of the present invention, the blind cavity wrench includes a horizontal operating section, a vertical section, and a horizontal extension section. The horizontal operating section is connected to the output end of the clamping robotic arm. A sleeve is disposed at the end of the horizontal extension section, and a pinhole camera is mounted on the side of the horizontal extension section. Two pinhole cameras are used, one on each side of the horizontal extension section.

[0025] In a preferred embodiment of the present invention, an operating lever is provided at the end of the horizontal operating end, and the operating lever is provided with a force-applying interface for cooperating with a manual torque wrench. The operator connects the manual torque wrench to the force-applying interface on the blind-cavity wrench and applies force to the specified torque to achieve a threaded connection between the nut and the bolt.

[0026] In a preferred embodiment of the present invention, the lower section of the sleeve has a polygonal inner ring shape. The lower section of the sleeve mates with the nut, and a spring pin for tightening the nut is provided on the inner wall of the upper end of the sleeve. When the blind cavity wrench is moved by the clamping robotic arm, the sleeve at the end of the blind cavity wrench moves until the nut is placed, and then the nut is inserted into the sleeve by force control technology. To improve efficiency, the nut can also be manually inserted into the sleeve. The polygonal structure inside the sleeve mates with the nut, so the nut will not rotate relative to the sleeve, and thus the nut can rotate accordingly when the blind cavity wrench is tightened. After the nut is inserted into the sleeve, the spring pin can tighten the nut, preventing the nut from falling out of the sleeve during the movement of the nut, ensuring that the blind cavity wrench can smoothly carry the nut.

[0027] As a preferred embodiment of the present invention, the system of the present invention further includes a mobile trolley, a gripping robotic arm and a control system mounted on the mobile trolley, a support frame mounted on the mobile trolley, and a visual positioning system mounted at the end of the support frame.

[0028] An automatic tightening method for blind cavity nuts in aero engines includes the following steps: S1: Initial positioning: The vision positioning system scans the upper surface of the bearing cavity component of the aero-engine, identifies and calculates the three-dimensional pose of the positioning hole; if vision fails, the positioning pin of the output end of the gripping robotic arm is inserted into the positioning hole by manual guidance. Based on the actual pose of the identified positioning holes, and combined with the geometric constraint relationship of the aero-engine bearing cavity component, the theoretical working point pose of all target bolts in the blind cavity of the aero-engine bearing cavity component is calculated. S2: Automated assembly cycle for single bolts: The control system plans and controls the blind cavity wrench to move from a safe position along a pre-calculated collision-free path to directly above the theoretical position of the target bolt; The nut is placed into the sleeve by manual or automatic feeding device and secured by spring pin; the nut is precisely positioned and finely adjusted by feedback from a pinhole camera or visual positioning system so that the nut in the sleeve is roughly aligned with the bolt below. The blind cavity wrench is slowly fed along the axis of the target bolt until the force sensor detects a clear contact force signal, indicating that the assembly pair has entered a single-point contact state. The control system actively adjusts the posture of the blind cavity wrench according to the direction and magnitude of the real-time contact force, so that the contact state evolves from single-point contact to two-point contact. The adjustment continues until the force mode fed back by the force sensor indicates that the assembly pair has formed a stable three-point contact and the posture deviation is within the allowable range. Under the stable state of three-point contact, the gripping robotic arm drives the sleeve to rotate at a set rotation angle without interference, pre-tightening the nut onto the target bolt; The robot maintains its position and issues a prompt, allowing the operator to connect the manual torque wrench to the force-applying interface on the blind-cavity wrench and apply force to the specified torque.

[0029] As a preferred embodiment of the present invention, in step S1, before initial positioning, system preparation is performed: Based on the model of the aero-engine bearing cavity component to be assembled, load the corresponding three-dimensional digital model and bolt array process documents; replace the socket on the blind cavity wrench that matches the current nut specification; install and fix the aero-engine bearing cavity component.

[0030] As a preferred embodiment of the present invention, in step S2, after the tightening action is completed, the end pinhole camera immediately captures an image of the current assembly status of the nut, and automatically determines whether the nut is installed in place, whether there is obvious tilting or omission through image processing algorithm; at the same time, the system checks whether the recorded final torque value is within the qualified range; the verification result is recorded and fed back in real time.

[0031] As a preferred embodiment of the present invention, the method of the present invention further includes the following steps: S3: Task Completion and Data Archiving: Follow the sequence specified in the process document to repeatedly execute step S2 until the assembly of all bolts in the blind cavity of the aero-engine bearing cavity component is completed; Once all assembly is complete, the control system generates an assembly process report, summarizing information such as the final torque, verification image, and assembly timestamp of each bolt, and uploads it to the database for archiving and quality traceability and analysis. The gripper arm returns to its initial standby position, ready for the next operation.

[0032] The beneficial effects of this invention are as follows: 1. This invention employs 3D vision integration or manual guidance to insert a positioning pin into a positioning hole, thereby acquiring the orientation of the positioning hole in the bearing cavity. Then, based on the geometric relationship between the positioning hole and the bolt array within the blind cavity, the orientation of the bolt array is automatically calculated. Thus, this invention enables precise positioning of the target bolt, facilitating the accurate delivery of the nut to the target bolt position via a gripping robotic arm and a blind cavity wrench, ensuring automated nut assembly.

[0033] 2. The hand-held robotic arm of this invention is responsible for locating the blind cavity wrench. It employs force control technology based on impedance control principles. By recording expert operation data or simulating nut-tightening tasks in a physical simulator, it learns the adjustment rules of impedance parameters during the assembly process. This allows the robot to automatically adjust control parameters during the pre-contact, alignment, and tightening stages under a behavior cloning strategy. Force control technology adjusts the orientation of the threaded fastener to achieve two-point contact with the bolt; it transitions from two-point contact to stable three-point contact, reducing the deflection angle between the threaded fastener and the bolt; finally, a tightening torque is applied manually, and the gripping robotic arm performs a compliant following motion to avoid interference, tightening the threaded fastener along the bolt. This invention improves the accuracy and efficiency of the automatic bolt assembly process.

[0034] 3. This invention features a high-torque blind cavity wrench with visual monitoring. By changing the tool head, it can be adapted to multiple blind cavity structures, achieving high versatility for different blind cavity models. Two high-definition pinhole cameras are installed on the bottom edge of the blind cavity wrench to observe the installation status of nuts inside the blind cavity, solving problems such as misinstallation, omission, and collision that may occur during bolt assembly in narrow blind cavities. Attached Figure Description

[0035] Figure 1 This is a schematic diagram of the structure of the present invention; Figure 2 This is a schematic diagram of a blind-cavity wrench; Figure 3 This is a partial structural diagram of a blind-cavity wrench; Figure 4This is an assembly diagram of the present invention and the aero-engine bearing cavity component; Figure 5 This is a structural schematic diagram of an aero-engine bearing cavity component; Figure 6 This is a cross-sectional view of an aero-engine bearing cavity component; Figure 7 This is a schematic diagram of the contact state between the bolt and the nut.

[0036] In the diagram: 1-Clamping robotic arm; 2-Aircraft engine bearing cavity component; 3-Blind cavity wrench; 4-Nut; 5-Sleeve; 6-Pinhole camera; 7-Control system; 8-Mobile trolley; 9-Vision positioning system; 11-Mounting rod; 12-Positioning pin; 21-Positioning hole; 22-Bolt; 31-Horizontal operating section; 32-Vertical section; 33-Horizontal extension section; 34-Operating lever; 51-Spring pin; 81-Support frame; 341-Force application interface. Detailed Implementation

[0037] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0038] Therefore, the following detailed description of the embodiments of the invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the invention without inventive effort are within the scope of protection of the invention. It should be noted that, unless otherwise specified, the embodiments and features described in the embodiments of the invention can be combined with each other.

[0039] like Figures 1-6 As shown, this embodiment of an automatic tightening robot system for blind cavity nuts of an aero-engine includes a clamping robotic arm 1 and a vision positioning system 9 for identifying positioning holes 21 on the surface of an aero-engine bearing cavity component 2. The output end of the clamping robotic arm 1 is connected to a blind cavity wrench 3. The lower end of the blind cavity wrench 3 is provided with a sleeve 5 for fitting a nut 4. The output end of the clamping robotic arm 1 is also equipped with a positioning pin 12 for engaging with the positioning holes 21 on the surface of the aero-engine bearing cavity component 2. The lower end of the blind cavity wrench 3 is also equipped with a pinhole camera 6 for observing the relative position of the nut 4 and the bolt 22. A force sensor is installed on the flange at the end of the clamping robotic arm 1.

[0040] The system of the present invention also includes a control system 7, a visual positioning system 9, a clamping robotic arm 1, a pinhole camera 6, and a force sensor, all of which are electrically connected to the control system 7. The visual positioning system 9 and / or the positioning pin 12 identify the position of the positioning hole 21. Based on the actual pose of the identified positioning hole 21 and the geometric constraint relationship of the aero-engine bearing cavity component 2, the control system 7 calculates the theoretical working point pose of all target bolts 22 in the blind cavity. The control system 7 uses the relative position of the nut 4 and the bolt 22 fed back by the pinhole camera 6 and / or the visual positioning system 9 to precisely position and fine-tune the blind cavity wrench 3 so that the nut 4 in the sleeve 5 is aligned with the bolt 22.

[0041] The system of this invention also includes a mobile trolley 8, a gripping robotic arm 1, and a control system 7 mounted on the mobile trolley 8. A support frame 81 is mounted on the mobile trolley 8, and a vision positioning system 9 is mounted at the end of the support frame 81. A blind-cavity wrench 3 is mounted on the end effector of the gripping robotic arm 1. The blind-cavity wrench 3 can coaxially transmit the tightening torque of the operator to the threaded fastener without loss. The mobile trolley 8, carrying the gripping robotic arm 1 and the vision positioning system 9, is manually pushed to the working position. The vision positioning system 9 is used to identify the positioning hole 21 on the upper surface of the bearing cavity and calculate the position of the bolt 22 based on the bearing cavity design drawings.

[0042] To achieve the positioning of bolts 22 that cannot be directly observed within the blind cavity, this invention designs a composite positioning scheme integrating three-dimensional machine vision and automatic borehole probing physical perception. The system first uses a visual positioning system 9 (high-precision 3D binocular vision sensor) to acquire three-dimensional point cloud data of the upper surface of the aero-engine bearing cavity. Through feature extraction and matching algorithms, it identifies the predefined positioning hole 21 in the bearing cavity design drawings and calculates its spatial pose. If visual recognition is difficult due to the cavity's depth, surface reflection, or oil contamination, the system can automatically switch to manual guidance to insert the positioning pin 12 into the positioning hole 21 or to borehole probing mode.

[0043] After obtaining the actual pose of one or more positioning holes 21, the system automatically calculates the three-dimensional spatial pose of each bolt 22 in the entire blind cavity array in the robot base coordinate system through coordinate transformation based on the known geometric relationship of the engine structure (i.e., the fixed relative position, angle and depth dimensions between the positioning hole 21 and each bolt 22 in the blind cavity), providing accurate target points for subsequent assembly.

[0044] Addressing the core challenges of blind cavity assembly in aero-engines, such as extremely limited space and complete visual obstruction, traditional rigid assembly strategies relying on absolute positioning accuracy are prone to jamming or docking failure due to minute positional deviations. To address this, this invention proposes a compliant automated assembly strategy centered on force feedback and mimicking human touch. This strategy does not pursue one-time, error-free absolute positioning, but rather guides the assembly system from initial contact to a mechanically stable, ideal docking state through active force control interaction, thereby reliably compensating for all unknown positioning and attitude errors. Force control technology is used to transition the contact state between nut 4 and bolt 22 from unstable to stable.

[0045] Based on stable three-point contact, the gripping robotic arm 1 drives the blind cavity wrench 3 to rotate at low speed in stages, completing the pre-tightening of the nut 4. The gripping robotic arm 1 maintains its position and locks, and the operator uses a manual torque wrench in cooperation with the gripping robotic arm 1 to complete the final tightening. The gripping robotic arm 1 also has compliance, performing compliant follow-up tasks to avoid interference.

[0046] Specifically, the blind cavity wrench 3 includes a horizontal operating section 31, a vertical section 32, and a horizontal extension section 33. The horizontal operating section 31 is connected to the output end of the gripping robotic arm 1. The sleeve 5 is disposed at the end of the horizontal extension section 33, and the pinhole camera 6 is mounted on the side of the horizontal extension section 33. There are two pinhole cameras 6, which are respectively disposed on both sides of the horizontal extension section 33.

[0047] The horizontal operating end is provided with an operating lever 34, which has a force-applying interface 341 for cooperating with a manual torque wrench. The operator connects the manual torque wrench to the force-applying interface 341 on the blind cavity wrench 3 and applies force to the specified torque to achieve the threaded connection between the nut 4 and the bolt 22.

[0048] The output end of the clamping robotic arm 1 is also fixed with a mounting rod 11. The other end of the mounting rod 11 is fixed with a positioning pin 12 for engaging with the positioning hole 21 on the surface of the aero-engine bearing cavity component 2. The positioning pin 12 is set downwards.

[0049] The lower section of the sleeve 5 has a polygonal inner ring shape. The lower section of the sleeve 5 mates with the nut 4. A spring pin 51 for tightening the nut 4 is provided on the inner wall of the upper end of the sleeve 5. When the blind cavity wrench 3 is moved by the clamping robotic arm 1, the sleeve 5 at the end of the blind cavity wrench 3 moves until the nut 4 is placed. Then, through force control technology, the nut 4 is inserted into the sleeve 5. To improve efficiency, the nut 4 can also be manually placed into the sleeve 5. The polygonal structure inside the sleeve 5 mates with the nut 4, so the nut 4 will not rotate relative to the sleeve 5, and thus the nut 4 can rotate accordingly when the blind cavity wrench 3 is tightened. After the nut 4 is inserted into the sleeve 5, the spring pin 51 can tighten the nut 4, preventing the nut 4 from falling out of the sleeve 5 during the movement of the nut 4, and ensuring that the blind cavity wrench 3 can smoothly carry the nut 4.

[0050] Specifically, such as Figure 7 As shown, the possible contact states between nut 4 and bolt 22 mainly include single-point contact outside the hole ( Figure 7 (a) Single-point contact inside the hole ( Figure 7 (b) Single-point contact of the hole wall ( Figure 7 (c) Two-point contact outside the hole ( Figure 7 (d) Two-point contact inside the hole ( Figure 7 (e) Three-point contact ( Figure 7 (f) etc. Specifically, the assembly process is defined as an ordered transition of contact states: 1) Initial single-point contact: After the initial positioning under visual guidance, unstable point contact is formed between the parts, and the contact force is concentrated at one point, which easily generates torque.

[0051] 2) Transition to two-point contact: Under force control guidance, micro-attitude adjustments are made to develop the contact point into two points, and the system obtains initial attitude constraints.

[0052] 3) Convergence to stable three-point contact: Further optimize the posture to form a balanced triangular force support. At this point, the axes of the parts are basically aligned, and the system is in a mechanically stable "self-attraction domain," possessing the geometric conditions to achieve final precise docking.

[0053] This invention proposes a phased assembly strategy based on force control feedback, involving adjustment and tightening: such as Figure 7 As shown, the 4 sets of nuts and 22 bolts are introduced into single-point contact through active tilting, and then gradually transitioned to three-point stable contact outside the hole through impedance adjustment, and then the coaxial alignment and torque tightening are completed.

[0054] Operational process: Alignment, application of force, and following. This compliant strategy is specifically embodied in the following operational process: 1) Compliant Alignment Stage: Under geometric impedance control, the gripping robotic arm 1 carries the nut 4 (or sleeve 5) toward the target bolt 22 (or nut 4). Once the force sensor detects contact, the compliant alignment procedure is initiated. Based on the contact force feedback, the control system 7 actively performs a small-amplitude spiral search, so that the contact state changes from a single point, through two points, and finally stabilizes at a three-point contact state, completing the alignment task.

[0055] 2) Pre-tightening and Cooperative Force Application Stage: Based on stable three-point contact, the gripping robotic arm 1 drives the blind cavity wrench 3 to rotate at low speed in stages, completing the pre-tightening of the nut 4. The gripping robotic arm 1 maintains its position and locks, and the operator uses a manual torque wrench to cooperate with the gripping robotic arm 1 to complete the final tightening. The gripping robotic arm 1 also has compliance, performing compliant follow-up tasks to avoid interference.

[0056] 3) Cycle and completion: The control system 7 repeatedly performs the alignment-pre-tightening-tightening cycle on the bolt array 22 in the blind cavity according to the cross-symmetrical process sequence until all assembly tasks are completed.

[0057] The automatic tightening method for the blind cavity nut of the aircraft engine in this embodiment includes the following steps: S1: System Preparation and Initial Positioning S11: Based on the model of bearing cavity component 2 of the aircraft engine to be assembled, load the corresponding three-dimensional digital model and bolt 22 array process file (including bolt 22 coordinates, tightening sequence, target torque value, etc.). S12: Replace with a blind-cavity wrench 3 and socket 5 that match the current nut 4 specification; S13: The aircraft engine bearing cavity component 2 is installed and fixed on a chuck-rotation system or a special tooling; S14: Identification of Positioning Hole 21: The vision positioning system 9 scans the upper surface of the aero-engine bearing cavity component 2, identifies and calculates the three-dimensional pose of the positioning hole 21. If vision fails, a manual teaching-guided positioning program is initiated, and the positioning pin 12 at the output end of the gripping robotic arm 1 is inserted into the positioning hole 21.

[0058] S15: Bolt 22 array pose calculation: Based on the actual pose of the identified positioning hole 21, combined with the geometric constraint relationship in the component CAD model, the theoretical working point pose of all target bolts 22 in the blind cavity is calculated.

[0059] S2: Automated assembly cycle for single bolt 22 (executed one bolt at a time according to process sequence): S21: Collision-free path movement: The control system 7 plans and controls the blind cavity wrench 3 through the gripping robotic arm 1 to move from a safe position along a pre-calculated collision-free path to directly above the theoretical position of the bolt 22. This path planning takes into account spatial constraints such as the blind cavity inlet size and the shape of the cavity inner wall.

[0060] S22: Visual coarse alignment: Nut 4 is placed into sleeve 5 by manual or automatic feeding device and tightened by spring pin 51; using feedback from end pinhole camera 6 or visual positioning system 9, the final stage of precise positioning fine adjustment is performed so that nut 4 in sleeve 5 or empty sleeve 5 is roughly aligned with bolt 22 below or pre-installed nut 4.

[0061] S23: Force-controlled compliant docking: a. Contact sensing: Control the blind cavity wrench 3 to slowly feed along the axis of bolt 22 until the force sensor detects a clear contact force signal, indicating that the assembly pair has entered a single-point contact state.

[0062] b. State Guidance and Adjustment: Activate the compliant control algorithm. Based on the direction and magnitude of the real-time contact force, actively adjust the posture of the blind cavity wrench 3 (such as performing slight tilting, rotation, or translation search) to evolve the contact state from single-point contact to two-point contact.

[0063] c. Stable state achieved: Continue adjustments until the force mode feedback from the force sensor indicates that the assembly pair has achieved stable three-point contact and the attitude deviation is within the allowable range. At this point, nut 4 and bolt 22 or sleeve 5 and nut 4 are well aligned.

[0064] S24: Pre-tightening and final tightening: a. Pre-tightening: Under the stable state of three-point contact, the blind cavity wrench 3 clamps the mechanical arm 1 and drives the sleeve 5 to rotate at a low speed according to the set rotation angle that does not cause interference, thus pre-tightening the nut 4 onto the bolt 22.

[0065] b. High torque tightening: The clamping robotic arm 1 maintains its position and issues a prompt, and the operator connects the manual torque wrench to the force application interface 341 and applies force to the specified torque.

[0066] S25: Verification: After the tightening action is completed, the end pinhole camera 6 immediately captures an image of the current assembly status of the nut 4. The image processing algorithm automatically determines whether the nut 4 is properly installed, whether there is any obvious tilting, or whether it is missing. Simultaneously, the control system 7 checks whether the recorded final torque value is within the acceptable range. The verification results are recorded and fed back in real time.

[0067] S3: Task Completion and Data Archiving: S31: Perform the second stage in a cyclical manner according to the order specified in the process document (such as a cross-symmetrical order) until the assembly of all bolts 22 in the blind cavity is completed.

[0068] S32: After all assembly is completed, the system generates an assembly process report, which summarizes the final torque, verification image, assembly timestamp and other information of each bolt 22, and uploads it to the database for archiving, for quality traceability and analysis.

[0069] S33: The gripper arm 1 returns to its initial standby position, ready for the next operation.

[0070] Typically, the assembly clearance between threaded fasteners and bolts 22 is less than 0.3 mm. Most industrial robotic arms have a repeatability of less than ±0.03 mm, which meets the requirements for end-effector positioning accuracy in automated tightening. However, current target position measurement accuracy based on industrial cameras is typically at the millimeter level, making it impossible to directly utilize robot vision servo technology for automated assembly of bolt arrays within blind cavities. Fortunately, since there is always a stable three-point contact state between nut 4 and bolts 22, an automated tightening strategy based on force control technology can be designed using the concept of an environmental self-attraction domain.

[0071] The present invention has the following advantages: High versatility: It adopts a modular blind cavity wrench 3 design, which supports quick replacement of tool heads according to the blind cavity size and bolt 22 specification of different engine models. The overall system structure remains unchanged, which is convenient for promotion and application across multiple engine platforms.

[0072] High assembly precision: Through 3D vision positioning and force control feedback, the positioning accuracy is improved, the orientation error is reduced, and the assembly torque accuracy is increased, meeting the high reliability requirements of aero-engines.

[0073] Efficiency improvement: The assembly time of a single bearing cavity can be significantly shortened, improving efficiency and enabling continuous operation for extended periods, thus reducing labor costs.

[0074] Quality consistency: It can control the axial pressure distribution of circumferential connectors, and the success rate of tightening bolt 22 is relatively high, which significantly improves the stability of engine operation.

[0075] Intelligent error prevention: Automatic verification is achieved through the fusion of vision and force, reducing the rate of misassembly and omission, and reducing rework costs.

[0076] Compliant assembly with human-hand-like characteristics: The robotic arm can adaptively adjust force and posture, simulating the process of manually tightening bolts.

[0077] Capable of high-torque tightening: The system structure design has high rigidity and high-strength transmission capacity. The blind cavity wrench 3 path and force optimization design ensure that it can effectively tighten high-torque bolts 22 such as typical M8 / M10 nuts 4, taking into account both accuracy and strength requirements, and meeting the needs of heavy-load aviation conditions.

[0078] This invention is not limited to the above-described optional embodiments. Anyone can derive other various forms of products under the guidance of this invention. However, regardless of any changes made in their shape or structure, any technical solution that falls within the scope of the claims of this invention shall be protected by this invention.

Claims

1. An automatic tightening robot system for blind cavity nuts in aero-engines, characterized in that: The system includes a gripping robotic arm (1) and a vision positioning system (9) for identifying the positioning hole (21) on the surface of the aero-engine bearing cavity component (2). The output end of the gripping robotic arm (1) is connected to a blind cavity wrench (3). The lower end of the blind cavity wrench (3) is provided with a sleeve (5) for fitting a nut (4). The output end of the gripping robotic arm (1) is also fixed with a positioning pin (12) for engaging with the positioning hole (21) on the surface of the aero-engine bearing cavity component (2). The lower end of the blind cavity wrench (3) is also equipped with a pinhole camera (6) for observing the relative position of the nut (4) and the bolt (22). A force sensor is installed on the flange at the end of the gripping robotic arm (1).

2. The automatic tightening robot system for blind cavity nuts of aero-engines according to claim 1, characterized in that: It also includes a control system (7), a visual positioning system (9), a gripping robotic arm (1), a pinhole camera (6), and a force sensor, all of which are electrically connected to the control system (7); the visual positioning system (9) and / or the positioning pin (12) identify the position of the positioning hole (21), and the control system (7) calculates the theoretical working point position of all target bolts (22) in the blind cavity based on the actual position of the identified positioning hole (21) and the geometric constraint relationship of the aero-engine bearing cavity component (2); the control system (7) uses the relative position of the nut (4) and the bolt (22) fed back by the pinhole camera (6) and / or the visual positioning system (9) to precisely position and fine-tune the blind cavity wrench (3) so that the nut (4) in the sleeve (5) is aligned with the bolt (22).

3. The automatic tightening robot system for blind cavity nuts of aero-engines according to claim 1, characterized in that: The blind cavity wrench (3) includes a horizontal operating section (31), a vertical section (32) and a horizontal extension section (33). The horizontal operating section (31) is connected to the output end of the clamping robotic arm (1). The sleeve (5) is located at the end of the horizontal extension section (33), and the pinhole camera (6) is installed on the side of the horizontal extension section (33).

4. The automatic tightening robot system for blind cavity nuts of aero-engines according to claim 3, characterized in that: The end of the horizontal operating end is provided with an operating lever (34), and the operating lever (34) is provided with a force-applying interface (341) for cooperating with a manual torque wrench.

5. The automatic tightening robot system for blind cavity nuts of aero-engines according to claim 1, characterized in that: The lower section of the sleeve (5) has a polygonal inner ring shape. The lower section of the sleeve (5) is engaged with the nut (4). A spring pin (51) for tightening the nut (4) is provided on the inner wall of the upper end of the sleeve (5).

6. The automatic tightening robot system for blind cavity nuts of aero-engines according to claim 2, characterized in that: It also includes a mobile trolley (8), a gripping robotic arm (1) and a control system (7) mounted on the mobile trolley (8), a support frame (81) mounted on the mobile trolley (8), and a visual positioning system (9) mounted at the end of the support frame (81).

7. A method for automatically tightening blind cavity nuts in aircraft engines, using the automatic tightening robot system for blind cavity nuts in aircraft engines as described in claim 2, characterized in that: Includes the following steps: S1: Initial positioning: The visual positioning system (9) scans the upper surface of the aero-engine bearing cavity component (2), identifies and calculates the three-dimensional pose of the positioning hole (21); if the vision fails, the positioning pin (12) at the output end of the gripping robotic arm (1) is inserted into the positioning hole (21) by manual guidance. Based on the actual pose of the identified positioning hole (21), combined with the geometric constraint relationship of the aero-engine bearing cavity component (2), the theoretical working point pose of all target bolts (22) in the blind cavity of the aero-engine bearing cavity component (2) is calculated. S2: Single bolt (22) automated assembly cycle: The control system (7) plans and controls the blind cavity wrench (3) to move from a safe position along a pre-calculated collision-free path to directly above the theoretical position of the target bolt (22); Place the nut (4) into the sleeve (5) and tighten it; use the feedback from the pinhole camera (6) or the visual positioning system (9) to make precise positioning and fine adjustment so that the nut (4) in the sleeve (5) is roughly aligned with the bolt (22) below. The blind cavity wrench (3) is slowly fed along the axis of the target bolt (22) until the force sensor detects a clear contact force signal, indicating that the assembly pair has entered a single-point contact state; the control system (7) actively adjusts the posture of the blind cavity wrench (3) according to the direction and magnitude of the real-time contact force, so that the contact state evolves from single-point contact to two-point contact; the adjustment continues until the force mode fed back by the force sensor indicates that the assembly pair has formed a stable three-point contact and the posture deviation is within the allowable range; Under the stable state of three-point contact, the clamping robot arm (1) drives the sleeve (5) to rotate at a set rotation angle without interference, and pre-tightens the nut (4) on the target bolt (22); The robot maintains its position and issues a prompt, allowing the operator to connect the manual torque wrench to the force-applying interface (341) on the blind cavity wrench (3) and apply force to the specified torque.

8. The automatic tightening method for a blind cavity nut of an aero-engine according to claim 7, characterized in that: In step S1, before initial positioning, system preparation is performed: According to the model of the aircraft engine bearing cavity component (2) to be assembled, load the corresponding three-dimensional digital model and bolt (22) array process document; replace the sleeve (5) on the blind cavity wrench (3) that matches the current nut (4) specification; install and fix the aircraft engine bearing cavity component (2).

9. The automatic tightening method for a blind cavity nut of an aero-engine according to claim 7, characterized in that: In step S2, after the tightening action is completed, the end pinhole camera (6) immediately captures an image of the current assembly status of the nut (4). The image processing algorithm automatically determines whether the nut (4) is installed in place, whether there is obvious tilting or missing parts. At the same time, the system checks whether the recorded final torque value is within the qualified range. The verification results are recorded and fed back in real time.

10. The automatic tightening method for a blind cavity nut of an aero-engine according to claim 7, characterized in that: It also includes the following steps: S3: Task Completion and Data Archiving: Follow the sequence specified in the process document to perform step S2 repeatedly until the assembly of all bolts (22) in the blind cavity of the aero-engine bearing cavity component (2) is completed; After all assembly is completed, the control system (7) generates an assembly process report, summarizing the final torque, verification image, and assembly timestamp of each bolt (22), and uploads it to the database for archiving, for quality traceability and analysis. The gripper arm (1) returns to its initial standby position, ready to perform the next operation.

Images