Screw locking system and method based on tool intelligence and air conditioner outdoor unit production line

Abstract

The invention provides a screw locking system and method based on tool intelligence and an air conditioner outdoor unit production line, and relates to the technical field of electric appliance assembly. The screw locking system based on the body intelligence comprises a base; the pressing executing mechanism is used for applying pressing force to the top cover of the air conditioner outdoor unit, so that the top cover is aligned with the screw hole position of the shell; the locking execution mechanism is used for executing screw locking operation; the visual perception module is used for acquiring three-dimensional visual information of the air conditioner outdoor unit; the control module is configured to determine a locking track of the locking execution mechanism for the current air conditioner outdoor unit based on the three-dimensional visual information obtained by the visual perception module; and the pressing executing mechanism is controlled to apply pressing force, and the locking executing mechanism is controlled to execute screw locking along the locking track. According to the intelligent screw locking system and method based on the tool body and the air conditioner outdoor unit production line, the problems that when an air conditioner outdoor unit is assembled, the screw locking efficiency is low, the labor intensity is high, and the quality consistency is poor are solved.

Description

Screw fastening system, method and air conditioner outdoor unit production line based on embodied intelligence Technical Field

[0001] This application relates to the field of electrical assembly technology, and in particular to a screw fastening system, method and air conditioner outdoor unit production line based on embodied intelligence. Background Technology

[0002] In the production and assembly process of air conditioner outdoor units, the screw fastening between the top cover and the outer casing is one of the key processes.

[0003] The top cover and outer casing of an air conditioner outdoor unit are typically thin-walled sheet metal parts with a certain degree of flexibility. During handling, loading, and assembly, they are prone to bending or elastic deformation, causing temporary positional deviations in the screw mounting holes on the top cover and outer casing during assembly, resulting in misalignment of the holes. Traditional automated equipment cannot reliably complete the fastening operation, and misalignment or slippage is prone to occur. Therefore, this process has long relied on manual hole alignment and fastening.

[0004] However, manual screw fastening suffers from problems such as low efficiency, high labor intensity, and poor quality consistency. Summary of the Invention

[0005] This application provides a screw fastening system, method, and air conditioner outdoor unit production line based on embodied intelligence to solve the problems of low efficiency, high labor intensity, and poor quality consistency in screw fastening during air conditioner outdoor unit assembly.

[0006] In a first aspect, this application provides a screw fastening system based on embodied intelligence, comprising:

[0007] Base;

[0008] A clamping actuator is disposed on the base and is used to apply clamping force to the top cover of the outdoor unit of the air conditioner so that the top cover is aligned with the screw holes of the outer casing;

[0009] A screw fastening actuator is disposed on the base and is used to perform screw fastening operations;

[0010] A visual perception module is used to acquire three-dimensional visual information of the outdoor unit of the air conditioner;

[0011] The control module is communicatively connected to the clamping actuator, the locking actuator, and the vision perception module. The control module is configured to:

[0012] Based on the three-dimensional visual information obtained by the visual perception module, the locking trajectory of the locking actuator for the current outdoor unit of the air conditioner is determined;

[0013] The clamping actuator is controlled to apply the clamping force, and the locking actuator is controlled to perform screw locking along the locking trajectory.

[0014] As an optional implementation, the clamping actuator includes a first robotic arm, the end of which is provided with a force-controlled gripper;

[0015] The locking actuator includes a second robotic arm, the end of which is equipped with an electric screwdriver.

[0016] As an optional implementation, the embodied intelligence-based screw fastening system also includes a rotary table;

[0017] The rotating platform is mounted on the base and is communicatively connected to the control module. The rotating platform is used to support and drive the outdoor unit of the air conditioner to rotate to multiple locking angles.

[0018] As an optional implementation, the control module controls the clamping actuator to operate in force control mode, applying a controllable clamping force according to a preset pressure threshold.

[0019] As an optional implementation, the control module is further configured to:

[0020] Obtain the model identification information of the outdoor unit of the air conditioner;

[0021] If the model identification information corresponds to a known model, the corresponding locking trajectory is retrieved from the pre-stored trajectory database;

[0022] If the model identification information corresponds to an unknown model, then a new locking trajectory is calculated and stored based on the three-dimensional visual information obtained by the visual perception module.

[0023] As an optional implementation, the visual perception module includes:

[0024] The camera is fixed to the top of the base;

[0025] A follow-up camera is mounted on the locking actuator;

[0026] The control module generates the three-dimensional visual information based on the images acquired by the fixed camera and the servo camera.

[0027] As an optional implementation, the control module is also configured to: after calculating the new locking trajectory, perform trajectory feasibility verification in a simulation environment.

[0028] As an optional implementation, the embodied intelligence-based screw fastening system also includes a barcode reader;

[0029] The barcode reader is mounted on the base and is communicatively connected to the control module. The barcode reader is used to obtain the model identification information of the outdoor unit of the air conditioner.

[0030] Secondly, this application provides an air conditioner outdoor unit production line, including an infeed station, an outfeed station, and a screw fastening system based on embodied intelligence based on any one of the above.

[0031] The delivery station is used to transport the outdoor unit of the air conditioner to the assembly station of the screw fastening system based on embodied intelligence;

[0032] The delivery station is used to deliver the air conditioner outdoor unit with completed screw fastening from the assembly station of the embodied intelligence-based screw fastening system.

[0033] Thirdly, this application provides a screw fastening method based on embodied intelligence, applicable to any of the aforementioned screw fastening systems based on embodied intelligence, or any of the aforementioned air conditioner outdoor unit production lines, comprising:

[0034] Position the outdoor unit of the air conditioner at the assembly station;

[0035] The three-dimensional visual information of the outdoor unit of the air conditioner is obtained through the visual perception module;

[0036] Based on the three-dimensional visual information, the control module determines the current locking trajectory of the outdoor unit of the air conditioner;

[0037] The control clamping actuator applies clamping force to the top cover of the air conditioner outdoor unit;

[0038] With the clamping force applied to the top cover, the control locking actuator performs the screw locking operation along the locking trajectory.

[0039] This application provides a screw fastening system, method, and air conditioner outdoor unit production line based on embodied intelligence. The screw fastening system based on embodied intelligence includes a base; a clamping actuator disposed on the base for applying clamping force to the top cover of the air conditioner outdoor unit to align the screw holes of the top cover with the outer casing; a fastening actuator disposed on the base for performing screw fastening operations; a vision perception module for acquiring three-dimensional visual information of the air conditioner outdoor unit; and a control module communicatively connected to the clamping actuator, the fastening actuator, and the vision perception module. The control module is configured to: determine the fastening trajectory of the fastening actuator for the current air conditioner outdoor unit based on the three-dimensional visual information acquired by the vision perception module; control the clamping actuator to apply clamping force; and control the fastening actuator to perform screw fastening along the fastening trajectory.

[0040] The embodied intelligence-based screw fastening system provided in this application, through the setting of an independent clamping actuator, can apply clamping force to the easily deformable top cover before fastening, effectively correcting or stabilizing temporary deformation, and aligning the screw holes of the top cover and the outer shell, facilitating automated fastening. Simultaneously, the vision perception module can acquire three-dimensional visual information to allow the control module to determine the fastening trajectory, achieving precise positioning and movement of the fastening actuator under complex working conditions. This embodied intelligence-based screw fastening system solves the problem of inaccurate hole alignment due to deformation of the top cover and outer shell, thus preventing automated fastening. It can replace traditional manual hole alignment and fastening, significantly improving assembly efficiency, consistency, and reliability. Attached Figure Description

[0041] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0042] Figure 1 is a schematic diagram of the usage status of the screw fastening system based on embodied intelligence provided in the embodiment of this application;

[0043] Figure 2 is a schematic flowchart of the screw fastening method based on embodied intelligence provided in the embodiments of this application.

[0044] Explanation of reference numerals in the attached figures:

[0045] 100. Base;

[0046] 200. Tighten the actuator;

[0047] 210. The first robotic arm;

[0048] 220. Force-controlled gripper;

[0049] 300. Lock-in execution agency;

[0050] 310. Second robotic arm;

[0051] 320. Electric screwdriver;

[0052] 400. Visual perception module;

[0053] 410. Fixed camera;

[0054] 420. Follow-up camera;

[0055] 500. Rotary table;

[0056] 600. Code reader;

[0057] 700. Nail feeder;

[0058] 810. Send to workstation;

[0059] 820. Deliver to workstation;

[0060] 90. Air conditioner outdoor unit;

[0061] 901. Top cover;

[0062] 902. Outer shell.

[0063] The accompanying drawings illustrate specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through reference to particular embodiments. Detailed Implementation

[0064] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the protection scope of the embodiments of this application.

[0065] In the embodiments of this application, the terms "upper," "lower," "inner," "middle," "outer," "front," and "rear," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. These terms are mainly for better describing the embodiments of this application and their implementations, and are not intended to limit the indicated device, element, or component to having a specific orientation, or to be constructed and operated in a specific orientation. Furthermore, some of the above terms may be used to indicate other meanings besides orientation or positional relationship; for example, the term "upper" may also be used in some cases to indicate a certain dependency or connection relationship. For those skilled in the art, the specific meaning of these terms in the embodiments of this application can be understood according to the specific circumstances.

[0066] Furthermore, the terms "set up," "connect," and "fix" should be interpreted broadly. For example, "connection" can be a fixed connection, a detachable connection, or an integral structure; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, or it can be an internal connection between two devices, components, or parts. Those skilled in the art can understand the specific meaning of the above terms in the embodiments of this disclosure according to the specific circumstances.

[0067] The terms "first," "second," "third," "fourth," etc. (if present) in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that embodiments of the present application described herein can be implemented, for example, in orders other than those illustrated or described herein.

[0068] In this application, the terms "exemplarily" or "for example" are used to indicate examples, illustrations, or descriptions. Any embodiment or design described as "exemplarily" or "for example" in this application should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of terms such as "exemplarily" or "for example" is intended to present the relevant concepts in a specific manner.

[0069] As can be seen from the background technology, the screw fastening between the top cover and the side panel is one of the key processes in the production and assembly of the air conditioner outdoor unit casing.

[0070] The top cover and side panels of air conditioner outdoor units are typically thin-walled sheet metal parts with a certain degree of flexibility. During handling, loading, and assembly, they are prone to bending or elastic deformation due to external forces, causing temporary positional deviations in the screw mounting holes on the top cover and side panels during assembly. This dynamic hole-aligning problem caused by the flexible deformation of the components makes it impossible for traditional automated equipment to reliably complete the fastening operation, easily leading to misalignment or slippage.

[0071] To enhance adaptability, some solutions apply embodied intelligence technology to the aforementioned automated equipment to build a closed loop of perception, decision-making, and execution, enabling the equipment to adapt to dynamic environments. However, in high-precision assembly scenarios such as screw fastening of air conditioner outdoor unit top covers, embodied intelligence systems often rely on pre-taught fixed work trajectories. They lack effective online sensing and adaptive adjustment mechanisms for real-time hole position offsets caused by the flexible deformation of sheet metal parts, making it difficult to guarantee the reliability and consistency of the fastening process.

[0072] Secondly, with the diversification of market demand, production lines often need to produce multiple models of air conditioner outdoor units. These different models vary in size, top cover structure, and screw hole distribution. Faced with this multi-model mixed production situation, existing embodied intelligent systems lack the ability to quickly learn and generate trajectories online. Technicians still need to frequently perform complex robotic arm trajectory teaching and program debugging, resulting in insufficient equipment flexibility, an inability to adapt to rapid production changes and mixed production needs, and increased downtime and debugging costs.

[0073] Therefore, the process of tightening the screws on the top cover of the air conditioner outdoor unit still relies on manual operation for a long time, which has problems such as low efficiency, high labor intensity and poor quality consistency.

[0074] In view of this, embodiments of this application provide a screw fastening system, method, and air conditioner outdoor unit production line based on embodied intelligence. The screw fastening system based on embodied intelligence includes a base; a clamping actuator disposed on the base for applying clamping force to the top cover of the air conditioner outdoor unit to align the screw holes of the top cover with the outer casing; a fastening actuator disposed on the base for performing screw fastening operations; a vision perception module for acquiring three-dimensional visual information of the air conditioner outdoor unit; and a control module communicatively connected to the clamping actuator, the fastening actuator, and the vision perception module. The control module is configured to: determine the fastening trajectory of the fastening actuator for the current air conditioner outdoor unit based on the three-dimensional visual information acquired by the vision perception module; control the clamping actuator to apply clamping force; and control the fastening actuator to perform screw fastening along the fastening trajectory.

[0075] To address the issue of automatic locking of the air conditioner outdoor unit's top cover and outer casing due to their flexibility and tendency to undergo temporary deformation, leading to hole misalignment, this application addresses this problem by incorporating a clamping actuator. This actuator applies pressure to the top cover before locking, correcting deformation and aligning the holes, thus creating favorable conditions for automated locking. Simultaneously, a vision perception module acquires three-dimensional visual information, which the control module uses to determine the locking trajectory, achieving precise positioning and movement even under conditions of deviation.

[0076] To address the issue of insufficient equipment flexibility and the need for frequent manual teaching caused by the mixed production of multiple models on the production line, this application, through the collaboration of visual perception and control modules, can dynamically determine the locking trajectory applicable to different models based on the acquired three-dimensional visual information, without the need for tedious manual teaching and debugging for each model. It can adapt to different product models and improve the flexibility of the production line.

[0077] Therefore, this application can not only replace manual labor and improve assembly efficiency and quality consistency, but also adapt to the needs of rapid production changeover and mixed production, reducing downtime and long-term maintenance costs.

[0078] The technical solution of this application will be described in detail below with reference to the accompanying drawings and specific embodiments. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments.

[0079] Referring to Figure 1, the first aspect of this application provides a screw fastening system based on embodied intelligence, including a base 100, a clamping actuator 200, a fastening actuator 300, a vision perception module 400, and a control module.

[0080] Among them, the base 100 is the main structure and installation foundation of the entire embodied intelligence-based screw fastening system, which can provide stable support and positioning for other functional modules.

[0081] The clamping actuator 200 is disposed on the base 100 and is used to apply a clamping force toward the lower housing 902 to the top cover 901 of the outdoor unit 90 of the air conditioner before the screw fastening operation begins.

[0082] It should be noted that the outdoor unit 90 of an air conditioner typically has a box-shaped outer casing 902 formed by side panels and a bottom panel, with an opening at the top. The top cover 901 is a separate component used to close this opening. Since the top cover 901 and the outer casing 902 are mostly made of thin-walled sheet metal, they are prone to elastic deformation due to external forces during handling and loading, which can cause positional deviations between the mounting holes of the top cover 901 and the corresponding screw holes on the outer casing 902, resulting in misalignment.

[0083] The clamping force applied by the clamping actuator 200 can cause the top cover 901 to shift or deform and recover through mechanical action, thereby correcting its temporary deformation and making the top cover 901 and the outer shell 902 fit tightly together, realizing the physical alignment of the screw holes between the two, and establishing stable assembly conditions for subsequent automated fastening.

[0084] For example, the clamping actuator 200 may include a first robotic arm 210 with multiple degrees of freedom. The end of the first robotic arm 210 may be equipped with a force-controlled gripper 220 or a pressure head capable of force sensing and force control, so as to accurately control the magnitude of the clamping force and avoid damage to the top cover 901 due to excessive pressure or alignment failure due to insufficient pressure.

[0085] The tightening actuator 300 is also mounted on the base 100 and is used to perform the screw tightening operation. After the clamping actuator 200 completes the hole alignment, the tightening actuator 300 moves to a position directly opposite each screw hole, inserts the screw into the hole, and completes the tightening according to the set torque.

[0086] For example, the locking actuator 300 may include a second robotic arm 310, the end of which may be integrated with an electric screwdriver 320 (e.g., an electric screwdriver) and a matching automatic screw feeding device. The electric screwdriver 320 can output a precise tightening torque according to control commands.

[0087] Specifically, the embodied intelligence-based screw fastening system can also integrate an automatic screw feeding unit. This screw feeding unit may include a screw feeder 700, a screw feeding tube, and a bit clamping mechanism at the end of the fastening actuator 300. When planning the fastening trajectory, the control module will synchronously coordinate the start and stop of the screw feeding unit to ensure that the screw has been accurately fed into the bit of the electric screwdriver 320 before each fastening action is performed.

[0088] The visual perception module 400 is used to acquire three-dimensional visual information of the outdoor unit 90 of the air conditioner to be assembled. For example, the visual perception module 400 can use optical imaging and three-dimensional reconstruction technology to non-contactly measure key information such as the position, orientation, dimensions, and screw hole coordinates of the outdoor unit 90 in actual space, thereby reflecting the true state of the outdoor unit 90 due to individual differences, positioning errors, or deformation. This three-dimensional visual information can provide a data foundation for subsequent trajectory planning.

[0089] The control module is communicatively connected (e.g., wired or wireless) to the clamping actuator 200, the locking actuator 300, and the vision perception module 400, and is used to coordinate the actions of each component and execute the control algorithm.

[0090] Specifically, the control module can receive and process the three-dimensional visual information from the visual perception module 400, determine the locking trajectory of the locking actuator 300 for the current air conditioner outdoor unit 90, then control the pressing actuator 200 to apply the pressing force, and control the locking actuator 300 to perform screw locking along the locking trajectory.

[0091] Specifically, the control module receives and processes the three-dimensional visual information collected by the visual perception module 400, and determines the movement trajectory of the locking actuator 300 for the current air conditioner outdoor unit 90 based on this information; then it controls the pressing actuator 200 to apply the pressing force, and controls the locking actuator 300 to perform the screw locking operation along the planned trajectory.

[0092] For example, the control module can identify the spatial coordinates of all the screw holes to be locked using image processing and path planning algorithms, and generate a complete motion trajectory of the locking actuator 300 from the starting position to each hole, perform locking, and then move to the next hole.

[0093] In practice, the control module, following the workflow, first controls the clamping actuator 200 to move to a predetermined position and applies clamping force in force control mode. After confirming that the clamping state meets the requirements, it then controls the locking actuator 300 to move sequentially to each screw hole position, triggering the electric screwdriver 320 to complete the screw picking, alignment, screwing in, and tightening. Throughout the process, the control module ensures that the actions of the clamping actuator 200 and the locking actuator 300 are coordinated in timing and space, for example, maintaining the continuous application of clamping force throughout the locking process and ensuring that the movement path of the locking actuator 300 avoids the clamping actuator 200.

[0094] The embodied intelligence-based screw fastening system provided in this embodiment applies a clamping force to the top cover 901 before fastening via the clamping actuator 200 to correct deformation and align holes, creating the necessary conditions for automated fastening. Simultaneously, the vision perception module 400 acquires three-dimensional visual information, and the control module plans the fastening trajectory based on this information, achieving precise positioning and automated operation even under assembly deviation conditions. Furthermore, through the collaborative work of the vision perception and control modules, this embodied intelligence-based screw fastening system can dynamically adapt the fastening trajectory of different product models based on real-time acquired three-dimensional information, eliminating the need for manual teaching and debugging for each product. This significantly improves the embodied intelligence-based screw fastening system's flexibility in handling mixed production of multiple models. This embodied intelligence-based screw fastening system not only replaces traditional manual operations, improving assembly efficiency and quality consistency, but also effectively adapts to rapid production changeovers and mixed production needs, reducing equipment downtime and long-term maintenance costs.

[0095] Referring to Figure 1, in some embodiments, the embodied intelligence-based screw fastening system also includes a rotary table 500. The rotary table 500 is disposed on the base 100 and is communicatively connected to the control module, used to support and drive the air conditioner outdoor unit 90 to rotate to multiple fastening angles.

[0096] Specifically, the tabletop of the rotary table 500 is used to place and fix the outdoor unit 90 of the air conditioner to be assembled. The rotary table 500 typically includes a rotary drive device (such as a servo motor, reducer, etc.) and a slewing bearing structure, which can perform precise angular rotation under the command of the control module.

[0097] For example, the rotary table 500 may be integrated with a clamping mechanism (such as a pneumatic clamp). When the outdoor unit 90 of the air conditioner is fed in by the conveyor line and positioned on the table surface of the rotary table 500, the clamping mechanism is activated to reliably fix the outdoor unit 90 of the air conditioner and prevent it from shifting during rotation or locking.

[0098] The rotary table 500 can drive the air conditioner outdoor unit 90 to rotate as a whole, so that the screw holes distributed in different positions around the top cover 901 are sequentially brought into the working range of the locking actuator 300.

[0099] Since the screws around the top cover 901 of the outdoor unit 90 of the air conditioner are usually distributed circumferentially along its edge, if the locking actuator 300 itself is relied upon to cover all the holes by a large range of movement, it may require the locking actuator 300 to have a large working space or a complex trajectory, which is inefficient and may cause interference.

[0100] By rotating the rotary table 500, any side of the outdoor unit 90 that needs to be fastened can be rotated to a fixed position facing the fastening actuator 300. This allows the fastening actuator 300 to complete the fastening of all screws on that side by making a small, precise movement within a relatively fixed spatial position, thereby simplifying the movement trajectory of the fastening actuator 300 and improving positioning accuracy and work cycle.

[0101] Furthermore, the angular control accuracy of the rotary table 500 directly affects the alignment accuracy of the screw fastening. Therefore, the rotary table 500 can use an angle encoder for position feedback to ensure that it can accurately stop at each preset or calculated fastening angle according to the instructions issued by the control module, providing a stable angular reference for subsequent visual precision positioning and screw fastening.

[0102] In summary, this application, through the integration of the rotary table 500, can achieve efficient and orderly fastening of the circumferential screws on the top cover 901 of the outdoor unit 90 of the air conditioner, avoiding the efficiency loss and accuracy reduction caused by the long-distance and large-range movement of the fastening actuator 300, so that a single fastening actuator 300 can efficiently complete the fastening task of the entire top cover 901.

[0103] In some embodiments, the control module controls the clamping actuator 200 to operate in force control mode, applying a controllable clamping force according to a preset pressure threshold.

[0104] Specifically, force control mode means that the control module can control the magnitude of the clamping force applied to the top cover 901 of the outdoor unit 90 of the air conditioner, rather than simply controlling the end of the clamping actuator 200 to reach a certain fixed position.

[0105] In force control mode, the control module can acquire the clamping force feedback signal in real time through the force sensor built into the clamping actuator 200. The control module compares the real-time feedback force value with one or more preset pressure thresholds, and dynamically adjusts the output of the clamping actuator 200 through a closed-loop control algorithm (such as impedance control, admittance control, etc.) so that the actual applied clamping force is stabilized within the desired threshold range.

[0106] The preset pressure threshold can be pre-set or calibrated experimentally based on factors such as the material properties, structural rigidity, deformation characteristics, and safety assembly requirements of the top cover 901 and the outer casing 902 of the air conditioner outdoor unit 90. This preset pressure threshold needs to be sufficiently large to effectively overcome the temporary elastic deformation of the top cover 901, ensuring a tight fit with the outer casing 902 and reliable hole alignment. Simultaneously, the preset pressure threshold cannot be too large to avoid damaging the sheet metal surfaces of the top cover 901 or the outer casing 902, or causing irreversible plastic deformation. Specifically, the control module can store differentiated pressure threshold parameters for different models of air conditioner outdoor units 90, so that the corresponding parameters can be automatically called upon based on the identified model during execution.

[0107] The force control mode can adaptively compensate for individual differences in the outdoor unit itself and minor deviations in material loading and positioning. Since each air conditioner outdoor unit 90 may have slight differences in deformation and initial posture, if simple position control is used, the fixed downward displacement may result in insufficient clamping force (holes are not fully aligned) or excessive force (risk of damage).

[0108] In this embodiment, the force control mode automatically adapts to these differences by applying a specific target force to ensure a consistent and reliable clamping alignment effect under various conditions.

[0109] In actual operation, when the control module executes the clamping action according to the plan, it sends a force control command to the clamping actuator 200. After the clamping actuator 200 moves to a predetermined position close to the top cover 901, it switches to force control mode and begins to contact the top cover 901 at a slower speed. Once the contact force reaches the set contact detection threshold, the control module confirms that contact has been established, and then controls the clamping actuator 200 to continue outputting pressure until the clamping force reaches and stabilizes at the preset pressure threshold.

[0110] During the subsequent tightening process, the control module can instruct the clamping actuator 200 to maintain force control mode to keep a constant clamping force, ensuring that the hole alignment does not change due to vibration or stress relaxation throughout the entire tightening cycle. Only when all screws at this angle have been tightened, and the turntable 500 needs to be rotated or the outdoor unit 90 of the air conditioner removed, does the control module instruct the clamping actuator 200 to release force control and lift.

[0111] Therefore, by implementing force control mode operation of the clamping actuator 200 through the control module, this embodied intelligence-based screw fastening system can apply precise, controllable, and adaptive clamping force, improving the reliability, safety, and adaptability to differences in the air conditioner outdoor unit 90 throughout the assembly process.

[0112] In some embodiments, the control module is also configured to perform intelligent locking trajectory management function, specifically including: obtaining the model identification information of the outdoor unit 90 of the air conditioner; if the model identification information corresponds to a known model, then calling the corresponding locking trajectory from the pre-stored trajectory database; if the model identification information corresponds to an unknown model, then calculating and storing a new locking trajectory based on the three-dimensional visual information obtained by the visual perception module 400.

[0113] For example, the embodied intelligence-based screw fastening system can obtain this information in a variety of ways. For instance, a code reader 600 (such as a QR code or barcode scanner) is installed on the base 100. The code reader 600 is communicatively connected to the control module and is used to automatically scan the identification code containing model information affixed to the housing of the air conditioner outdoor unit 90 to obtain the model identification information of the air conditioner outdoor unit 90 and send it to the control module.

[0114] Alternatively, the control module can receive model data corresponding to the current workstation's outdoor air conditioning unit 90 from the production line main control system (MES) via the communication interface. After obtaining the model identification information, the control module will compare it with the internally stored model-track database.

[0115] If the comparison is successful, it can be confirmed that the current model is a known model (i.e., the corresponding locking trajectory data exists in the database), and the control module will directly call the complete locking trajectory corresponding to the model from the database.

[0116] The locking trajectory data can include the precise motion path sequence, attitude, and speed of the locking actuator 300 for all screw holes of this model in the global coordinate system, as well as the action commands of the electric screwdriver 320. Calling pre-stored locking trajectories allows the system to be deployed to production faster and with higher reliability, eliminating the need for repeated visual modeling and trajectory calculations, thus greatly improving the production cycle time and stability of known product models.

[0117] If the comparison fails, meaning the model identification information corresponds to an unknown model (e.g., a newly imported product or a model not included in the database), the control module will activate the self-learning mode. In this mode, the control module can first coordinate the visual perception module 400 and the rotary table 500 to perform a comprehensive 3D scan of the air conditioner outdoor unit 90 to obtain its complete 3D point cloud model.

[0118] Furthermore, the control module runs built-in algorithms, such as a trajectory generation algorithm, which can automatically identify the edge contours of the top cover 901 and the outer shell 902 based on 3D point cloud data, and accurately extract the center 3D coordinates of all screw holes through image processing technology (such as edge detection, feature matching, template recognition, etc.).

[0119] Furthermore, the trajectory generation algorithm can automatically plan a motion trajectory that starts from the starting point, passes through all screw holes in sequence and performs the locking action, and finally returns to a safe position based on the distribution of screw holes, the working space of the locking actuator 300, kinematic constraints and anti-collision rules.

[0120] It should be noted that the pre-stored trajectory database is not static, but a dynamically learning and expanding knowledge base. During initial deployment, the trajectory database can be empty, or initial data can be established through offline programming or by teaching the first batch of standard model air conditioner outdoor units 90. The trajectory database maintenance mechanism works as follows: whenever an unknown model is processed and a new locking trajectory is successfully generated, that locking trajectory is automatically associated with the corresponding model identifier and stored in the trajectory database, thus accumulating data. Furthermore, the control module can provide a database management interface, allowing for the editing, optimization, or deletion of existing trajectories to ensure their continued effectiveness and optimization.

[0121] Specifically, through the above configuration, this embodied intelligence-based screw fastening system possesses both high-efficiency production and flexible adaptability. For mature models, the system can invoke and execute a fast mode; for new models, the system can sense, calculate, learn, and execute an intelligent mode to respond. This setup not only reduces the manual teaching and downtime debugging time required for product changeovers, lowering maintenance costs, but also enables the same production line to seamlessly produce multiple products of different specifications, improving the intelligence level of the production system and overall economic efficiency.

[0122] In some embodiments, the visual perception module 400 includes a fixed camera 410 and a follower camera 420. The fixed camera 410 is fixed to the top of the base 100. The follower camera 420 is disposed on the locking actuator 300. The control module generates three-dimensional visual information for trajectory planning based on the images acquired by the fixed camera 410 and the follower camera 420.

[0123] Specifically, the fixed camera 410 can be mounted on a fixed bracket above or around the base 100, with its field of view vertically downward covering the entire rotating stage 500 and the air conditioner outdoor unit 90 supported thereon.

[0124] For example, the fixed camera 410 can be a high-resolution 2D industrial camera for rapid identification and coarse positioning. For instance, a top-view image captured by the fixed camera 410 can provide information on the planar position (X, Y coordinates) and horizontal rotation angle (yaw angle about the Z-axis) of the air conditioner outdoor unit 90 on the worktable, as well as the approximate outline and dimensions of the top cover 901. This two-dimensional information provides the control module with the initial pose of the air conditioner outdoor unit 90, which can be used to guide the locking actuator 300 or the follow-up camera 420 to quickly approach the target area, improving initial positioning efficiency.

[0125] The servo camera 420 can be mounted at or near the end of the locking actuator 300 (such as the second robotic arm 310) and moves with the second robotic arm 310. Exemplarily, the servo camera 420 can be a 3D camera, such as a 3D sensor based on structured light, binocular stereo vision, or laser triangulation principles. The servo camera 420 can perform 3D measurement and precise positioning before final locking.

[0126] Before the locking action is performed, the control module controls the locking actuator 300, carrying the follower camera 420, to move to a position roughly above the target hole based on the coarse positioning information provided by the fixed camera 410. Then, the follower camera 420 performs a local 3D scan. Based on the acquired high-precision three-dimensional point cloud, the control module can calculate the precise three-dimensional coordinates and orientation of the hole and fine-tune the end position of the locking actuator 300 accordingly to achieve precise alignment between the electric screwdriver 320 and the screw hole.

[0127] In the workflow of the embodied intelligence-based screw fastening system, the visual perception module 400 collaborates with the rotary table 500 and the control module. For example, when scanning and modeling an unknown model, the control module can control the rotary table 500 to rotate uniformly for one revolution or in increments. During this time, the fixed camera 410 can capture images from different angles from the top to assist in contour stitching or feature recognition; simultaneously, the control module controls the fastening actuator 300 to carry a follow-up 3D camera and move it to multiple key viewpoints (such as each side of the housing 902 and the top cover 901) to perform local 3D scanning of the screw hole distribution area.

[0128] By integrating the angle encoder data of the rotary table 500 with the three-dimensional point cloud data acquired by the servo camera 420 at different angles, the control module can construct a complete three-dimensional coordinate model of the screw holes and their angle distribution at the connection between the outer shell 902 and the top cover 901.

[0129] During the screw fastening execution phase, the control module sequentially controls the rotary table 500 to rotate to each target angle based on known or calculated screw hole angle information. After rotating to and positioning at each angle, the control module first controls the clamping actuator 200 to apply clamping force to the top cover 901. For each screw hole at that angle, the fastening actuator 300, carrying a follow-up 3D camera, moves to approximately above the hole to perform final precise positioning measurements and attitude adjustments, and then executes the fastening operation. After all screws at that angle are fastened, the system moves to the next angle and repeats this process.

[0130] This application achieves efficient and high-precision visual perception by configuring a hybrid vision system that combines a fixed camera 410 and a servo camera 420. It can meet the different needs of rapid identification, coarse positioning and fine three-dimensional measurement, and provides reliable visual assurance for the system to achieve accurate and flexible automated locking under complex working conditions. At the same time, it has good economy and practicality.

[0131] In some embodiments, the control module is further configured to: after calculating a new locking trajectory based on the three-dimensional visual information acquired by the visual perception module 400, perform trajectory feasibility verification in a simulation environment. Trajectory feasibility verification helps ensure the safety, reliability, and executability of the new locking trajectory.

[0132] When the control module generates a new locking trajectory for the unknown model of the outdoor air conditioner 90, this trajectory is only theoretically calculated based on geometric coordinates, but has not yet been run in an actual physical system. Directly using the new locking trajectory to control the locking actuator 300 may pose risks. For example, the trajectory may spatially interfere with the clamping actuator 200, the rotary table 500, or the outdoor air conditioner 90 itself, or it may require the locking actuator 300 to move to its workspace boundary or even beyond, or pass through its kinematic singularity, leading to speed loss.

[0133] To avoid the aforementioned risks, the control module can have a built-in or connected simulation environment. This simulation environment can be a digital virtual system that establishes precise 3D models of key components of the screw fastening system based on embodied intelligence, including standard models of the base 100, fastening actuator 300, clamping actuator 200, and rotary table 500, as well as a specific 3D model of the current model air conditioner outdoor unit 90 reconstructed from current 3D visual information. All models' dimensions, joint range of motion, speed and acceleration limits, and other parameters are consistent with the actual physical equipment.

[0134] During the verification process, the control module inputs the newly generated locking trajectory into the simulation environment. The virtual locking actuator 300 in the simulation environment will perform motion simulation in virtual space according to this trajectory. The control module monitors in real time whether collisions occur, whether the workspace of the robotic arm is exceeded or joint limits are not reached, whether excessive motion performance is required or singularities are encountered, and whether the timing and position logic of the electric screwdriver 320's action commands are reasonable.

[0135] If the simulation verification passes without reporting any errors or warnings, the control module determines that the new trajectory is feasible and officially stores it in the trajectory database, marking it as verified, and it can then be used in actual production. If the simulation verification fails, for example, if a potential collision is detected, the control module will record the collision point and related components, and may automatically trigger a trajectory correction algorithm to optimize the original trajectory, and then perform simulation verification again until a safe and feasible trajectory is generated. This setup ensures that even with a brand-new air conditioner outdoor unit model 90 being used for the first time, the system can automatically generate safe and reliable operating instructions, avoiding serious accidents such as equipment collisions, damage to the air conditioner outdoor unit model 90, or production interruptions that may occur in actual production, thus ensuring the safety of equipment and personnel.

[0136] For example, the control module can be an integrated industrial controller or a hierarchical distributed system. For instance, the control module can be composed of an upper-level industrial computer (responsible for vision processing, trajectory planning, database management, and human-machine interaction) and lower-level programmable logic controllers (PLCs) and multi-axis motion controllers (precisely controlling the motion and force control of the rotary table 500, clamping actuator 200, and locking actuator 300 respectively), all connected via an industrial network (such as EtherCAT or PROFINET). This architecture ensures both the capability to perform complex algorithm calculations and the accuracy and reliability of real-time motion control.

[0137] Furthermore, the control module can also be configured to monitor multiple key parameters in real time. For example:

[0138] The force sensor monitors whether the clamping force is within the set range to prevent clamping failure or overload; the current or torque feedback of the locking actuator 300 monitors the screw tightening process and identifies faults such as stripped threads, floating locks, or missed screws; the position and status feedback of each actuator monitors whether the process is executed correctly in sequence.

[0139] Once any anomaly is detected, the control module will immediately interrupt the current operation, execute a safety stop, issue an alarm, and record the fault information and location according to the preset strategy. For some recoverable anomalies (such as a stripped screw), the control module can also attempt to retry or record and skip the error to continue completing the remaining tasks, ensuring the continuity and maintainability of the production line.

[0140] It should be noted that the embodied intelligence-based screw fastening system provided in this application integrates a high-precision visual perception module 400, a force-controlled clamping actuator 200, a high-degree-of-freedom fastening actuator 300, and an intelligent control module to construct a complete embodied intelligence closed loop, which can fully leverage the advantages of embodied intelligence in real-time perception and autonomous decision-making.

[0141] Specifically, in response to the dynamic deviation of hole positions caused by component deformation, the control module identifies the deviation in real time through the vision perception module 400 and generates corresponding clamping force commands and trajectory compensation commands, which are executed collaboratively by the clamping actuator 200 and the locking actuator 300, realizing real-time interaction and autonomous adjustment during the assembly process.

[0142] For situations involving mixed production of multiple models, the control module has online learning capabilities. Specifically, for unknown models, it can autonomously perceive product characteristics through visual scanning, generate locking trajectories through trajectory planning algorithms, and store them as reusable locking trajectories after simulation verification. For known models, it can directly call the stored locking trajectories for efficient decision-making and execution.

[0143] Therefore, the screw fastening system based on embodied intelligence in this application embodiment not only effectively solves the dynamic hole-fitting problem caused by the flexible deformation of components through embodied intelligence closed loop, but also significantly improves the system's flexible adaptability to mixed production of multiple models by introducing an online learning mechanism, thereby improving the overall level of intelligence, production efficiency and reliability of air conditioner outdoor unit assembly operations.

[0144] The second aspect of this application provides an air conditioner outdoor unit production line, including an infeed station 810, an outfeed station 820, and a screw fastening system based on embodied intelligence provided in any of the above embodiments.

[0145] The feeding station 810 is located at the front end of the assembly station (i.e., the rotary table 500) of the embodied intelligence-based screw fastening system. It can be composed of a conveyor line (such as a roller conveyor or belt conveyor) and a positioning mechanism. It is used to automatically and orderly transport the air conditioner outdoor unit 90 to be assembled to the rotary table 500 and ensure that it arrives at the rotary table 500 with a preset posture and accuracy, thus preparing for subsequent automated fastening operations.

[0146] The delivery station 820 is located at the rear end of the assembly station of the embodied intelligence-based screw fastening system. It can also be composed of a conveyor line. It is used to receive and deliver the air conditioner outdoor unit 90 with the top cover 901 screw fastened from the rotary table 500, so that it can be transferred to the next production process or the off-line area.

[0147] The screw fastening system based on embodied intelligence has been described in detail in the above embodiments and will not be repeated here.

[0148] This air conditioner outdoor unit production line integrates an automated, embodied intelligence-based screw fastening system with front-end and back-end logistics conveying stations, achieving a fully automated closed loop for the assembly of the air conditioner outdoor unit 90 top cover 901 from loading, positioning, pressing, fastening, and unloading. This effectively improves production cycle time, assembly quality consistency, and overall intelligence level.

[0149] Referring to Figure 2, a third aspect of this application provides a screw fastening method based on embodied intelligence, applicable to the screw fastening system based on embodied intelligence provided in any of the above embodiments, or the air conditioner outdoor unit production line provided in any of the above embodiments, specifically including the following steps:

[0150] S101. Position the outdoor unit of the air conditioner at the assembly station;

[0151] S102. Obtain three-dimensional visual information of the outdoor unit of the air conditioner through the visual perception module;

[0152] S103. The control module determines the current locking trajectory of the outdoor unit of the air conditioner based on three-dimensional visual information.

[0153] S104. Control the clamping actuator to apply clamping force to the top cover of the air conditioner outdoor unit;

[0154] S105. Under the condition of clamping force acting on the top cover, control the locking actuator to perform screw locking operation along the locking trajectory.

[0155] The outdoor unit 90 of the air conditioner can be positioned at the assembly station automatically by the conveyor line of the feeding station 810, or it can be manually loaded and then precisely positioned and fixed by positioning fixtures (such as the clamping mechanism on the rotary table 500) to ensure that the outdoor unit 90 of the air conditioner remains stable in subsequent operations.

[0156] In practice, the conveyor line at station 810 transports the outdoor unit 90 of the air conditioner to the approximate area on the rotary table 500 for rough positioning. Subsequently, the clamping mechanism (such as pneumatic or electric grippers) on the rotary table 500 is activated to clamp the outer casing 902 of the outdoor unit 90 of the air conditioner from both sides or the bottom, achieving mechanical fixation and initial alignment.

[0157] To further eliminate feeding deviations, the control module can instruct the fixed camera 410 to quickly capture images after clamping, identify the features of the air conditioner outdoor unit 90, and calculate the minute offset of the air conditioner outdoor unit 90 relative to the theoretical coordinate system. This offset can serve as the coordinate system compensation benchmark for the entire subsequent visual recognition and trajectory execution, thereby achieving high-precision software positioning.

[0158] The visual perception module 400 acquires three-dimensional visual information of the air conditioner outdoor unit 90. Specifically, the control module coordinates the operation of the fixed camera 410 and the follow-up camera 420. The fixed camera 410 can quickly acquire the outline, model identification, and rough pose of the air conditioner outdoor unit 90 from the top; for cases requiring precise modeling (such as unknown models), the control module can further control the rotation of the turntable 500 and guide the follow-up camera 420 to scan from multiple angles, thereby acquiring high-precision point cloud data including the surface topography of the top cover 901 and the outer shell 902, the precise three-dimensional coordinates of the screw holes, and the normal direction.

[0159] When determining the locking trajectory of the current outdoor unit 90 of the air conditioner, the control module first identifies or reads the model identifier of the outdoor unit 90. If the model is known, the corresponding verified locking trajectory is directly retrieved from the pre-stored trajectory database; if the model is unknown, based on the acquired 3D visual information, an optimal motion path covering all screw holes is automatically calculated through a trajectory generation algorithm, and further feasibility verifications such as collision detection and reachability are performed in a simulation environment. After successful verification, the new locking trajectory is stored in the database.

[0160] When a clamping force is applied to the top cover 901 of the outdoor unit 90 of the air conditioner, the control module controls the clamping actuator 200 to move to a predetermined position above the top cover 901, and then switches to force control mode to apply a controllable clamping force according to a preset pressure threshold. This force control process achieves closed-loop control through real-time force feedback, ensuring that the clamping force is sufficient to correct temporary deformation of the top cover 901, make it fit tightly against the outer casing 902 and align the screw holes, while avoiding damage to the top cover 901 of the outdoor unit 90 of the air conditioner.

[0161] During the screw fastening operation, the control module controls the fastening actuator 300 to move along the fastening trajectory. For each target screw hole, the fastening actuator 300 can first use the follow-up camera 420 for final precise positioning and fine-tuning, and then guide the electric screwdriver 320 to be vertically aligned with the hole position to complete the automatic feeding, screwing, and tightening of the screw. Furthermore, the rotary table 500 can be rotated in increments to cooperate with the fastening actuator 300 to sequentially fasten screws at various angles. Throughout the fastening process, the clamping actuator 200 typically maintains a constant clamping force to ensure stable alignment.

[0162] This embodied intelligence-based screw fastening method, through the synergy of the above steps, achieves automation, intelligence, and flexibility in the fastening of the 901 screws on the 90 top cover of the air conditioner outdoor unit. It solves the problem of hole alignment caused by deformation of thin-walled sheet metal parts, and can adapt to different product models, significantly improving assembly efficiency, accuracy, and the rapid changeover capability of the production line.

[0163] Finally, it should be noted that those skilled in the art, upon considering the specification and practicing the application disclosed herein, will readily conceive of other embodiments of the present application. The embodiments of this application are intended to cover any variations, uses, or adaptations of the embodiments of this application that follow the general principles of the embodiments of this application and include common knowledge or customary technical means in the art not disclosed in the embodiments of this application. The specification and embodiments are to be considered exemplary only, and the true scope and spirit of the embodiments of this application are indicated by the following claims.

[0164] It should be understood that the embodiments of this application are not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from their scope. The scope of the embodiments of this application is limited only by the appended claims.

Claims

1. A screw fastening system based on embodied intelligence, characterized in that, include: Base (100); clamping actuator (200), disposed on the base (100), for applying clamping force to the top cover (901) of the outdoor unit (90) of the air conditioner so that the top cover (901) is aligned with the screw holes of the outer casing (902); screw fastening actuator (300), disposed on the base (100), for performing screw fastening operation; vision perception module (400), for acquiring three-dimensional vision information of the outdoor unit (90); control module, communicatively connected to the clamping actuator (200), the screw fastening actuator (300) and the vision perception module (400), the control module being configured to: determine the screw fastening trajectory of the screw fastening actuator (300) for the current outdoor unit (90) based on the three-dimensional vision information acquired by the vision perception module (400); control the clamping actuator (200) to apply the clamping force, and control the screw fastening actuator (300) to perform screw fastening along the screw fastening trajectory.

2. The screw fastening system based on embodied intelligence according to claim 1, characterized in that, The clamping actuator (200) includes a first robotic arm (210), the end of which is provided with a force-controlled gripper (220); the locking actuator (300) includes a second robotic arm (310), the end of which is provided with an electric screwdriver (320).

3. The screw fastening system based on embodied intelligence according to claim 1, characterized in that, It also includes a rotating platform (500); the rotating platform (500) is disposed on the base (100) and is communicatively connected to the control module. The rotating platform (500) is used to carry and drive the outdoor unit (90) of the air conditioner to rotate to multiple locking angles.

4. The screw fastening system based on embodied intelligence according to claim 1, characterized in that, The control module controls the clamping actuator (200) to operate in force control mode, applying a controllable clamping force according to a preset pressure threshold.

5. The screw fastening system based on embodied intelligence according to any one of claims 1 to 4, characterized in that, The control module is also configured to: obtain the model identification information of the outdoor unit (90) of the air conditioner; if the model identification information corresponds to a known model, then call the corresponding locking trajectory from the pre-stored trajectory database; if the model identification information corresponds to an unknown model, then calculate and store a new locking trajectory based on the three-dimensional visual information obtained by the visual perception module (400).

6. The screw fastening system based on embodied intelligence according to claim 5, characterized in that, The visual perception module (400) includes: a fixed camera (410) fixed to the top of the base (100); a follow-up camera (420) disposed on the locking actuator (300); the control module generates the three-dimensional visual information based on the images acquired by the fixed camera (410) and the follow-up camera (420).

7. The screw fastening system based on embodied intelligence according to claim 5, characterized in that, The control module is also configured to: after calculating the new locking trajectory, perform trajectory feasibility verification in a simulation environment.

8. The screw fastening system based on embodied intelligence according to claim 5, characterized in that, It also includes a barcode reader (600); the barcode reader (600) is disposed on the base (100) and is communicatively connected to the control module, and the barcode reader (600) is used to obtain the model identification information of the outdoor unit (90) of the air conditioner.

9. An air conditioner outdoor unit production line, characterized in that, The system includes an infeed station (810), an outfeed station (820), and a screw fastening system based on embodied intelligence as described in any one of claims 1 to 8; the infeed station (810) is used to transport the outdoor unit (90) of the air conditioner to the assembly station of the screw fastening system based on embodied intelligence; the outfeed station (820) is used to send the outdoor unit (90) of the air conditioner with screw fastening completed from the assembly station of the screw fastening system based on embodied intelligence.

10. A screw fastening method based on embodied intelligence, applied to the screw fastening system based on embodied intelligence as described in any one of claims 1 to 8, or the air conditioner outdoor unit production line as described in claim 9, characterized in that, include: Position the outdoor unit (90) of the air conditioner at the assembly station; The three-dimensional visual information of the outdoor unit (90) of the air conditioner is obtained through the visual perception module (400); Based on the three-dimensional visual information, the control module determines the current locking trajectory of the outdoor unit (90) of the air conditioner; controls the clamping actuator (200) to apply clamping force to the top cover (901) of the outdoor unit (90); under the condition that the clamping force is applied to the top cover (901), controls the locking actuator (300) to perform screw locking operation along the locking trajectory.

Images