Intelligent positioning method and system for self-locking cable-stayed buckle hanging system assembly joint

Abstract

The application discloses an intelligent positioning method and system for a self-locking cable buckle hanging system assembly node, and the method comprises the following steps: obtaining three-dimensional point cloud data of a region to be assembled, identifying node characteristics of the region to be assembled according to the three-dimensional point cloud data, wherein the node characteristics comprise a geometric center and a surface normal vector set; constructing a node local coordinate system according to the node characteristics, and calculating a target homogeneous pose of the region to be assembled; controlling an execution mechanism to perform translation and attitude adjustment on a buckle, so that the current pose of the buckle converges to the target homogeneous pose; measuring the actual pose of the buckle, and performing error compensation according to the error between the actual pose and the target homogeneous pose until the error is less than an error threshold, so that the intelligent positioning of the assembly node is completed.

Description

Technical Field

[0001] This invention belongs to the field of bridge assembly and construction technology, and more specifically, relates to an intelligent positioning method and system for assembly nodes of a self-locking cable-stayed system. Background Technology

[0002] In the construction of long-span arch bridges, the cable-stayed bridge construction technique (also known as "cable-stayed cable-assisted construction" or "cable-stayed system construction") has become a common technical approach in current engineering practice. This technique typically relies on cable-stayed towers, cables, and anchoring systems set up on the arch foot or temporary towers. The cables provide an adjustable temporary force path, providing crucial vertical and horizontal construction support before the arch rib segments can form a self-balancing system, thereby ensuring the stability and alignment accuracy of the arch rib segments during installation.

[0003] However, traditional arch bridge cable-stayed construction relies heavily on total stations or levels for discrete measurements of key nodes. This data has a low update frequency and cannot reflect the dynamic deformation of arch rib segments under wind loads, temperature gradients, and cable tension disturbances in real time. Especially in cable-stayed systems with complex force coupling, a single measurement method is insufficient to capture real-time changes in spatial orientation, easily leading to accumulated deviations in the arch rib alignment.

[0004] Secondly, in existing projects, the setting and adjustment of cable tension values ​​are generally carried out by on-site engineers based on experience or static calculation results, lacking a closed-loop control mechanism that can provide real-time feedback on arch rib deformation, cable force changes, and docking errors. Once external disturbances are significant (wind, construction loads, etc.), manual judgment is difficult to respond to in a timely manner, easily leading to low efficiency in node attitude adjustment and even docking difficulties.

[0005] Therefore, there is an urgent need for a technical solution that can intelligently guide the actuator to complete high-precision positioning and compensate for translation and rotation errors in real time. Summary of the Invention

[0006] To address the above technical problems, this invention proposes an intelligent positioning method for assembly nodes of a self-locking inclined fastening system, comprising:

[0007] Acquire 3D point cloud data of the region to be assembled, and identify the node features of the node to be assembled based on the 3D point cloud data. The node features include: geometric center and set of surface normal vectors.

[0008] Based on the node characteristics, a local coordinate system for the node is constructed, and the target homogeneous pose of the node to be assembled is calculated. The actuator is controlled to translate and adjust the attitude of the fastener so that the current pose of the fastener is similar to the target homogeneous pose.

[0009] The actual pose of the fastener is measured, and error compensation is performed based on the error between the actual pose and the target homogeneous pose until it is less than the error threshold, thereby completing the intelligent positioning of the assembly node.

[0010] Furthermore, identifying the node features of the node to be assembled based on the 3D point cloud data includes: denoising and downsampling the 3D point cloud data to obtain downsampled 3D point cloud data;

[0011] The average value of the coordinates of all points in the downsampled 3D point cloud data is used as the coordinates of the geometric center of the node to be assembled.

[0012] Principal component analysis was used to obtain the set of surface normal vectors from the downsampled 3D point cloud data;

[0013] The average direction of all surface normal vectors in the set of surface normal vectors is taken as the principal normal vector.

[0014] Furthermore, constructing a local coordinate system for a node based on its characteristics includes:

[0015]

[0016] in, For the first The x-axis of the local coordinate system of the nodes to be assembled. For the first The y-axis of the local coordinate system of the nodes to be assembled. For the first The z-axis of the local coordinate system of the nodes to be assembled. For the first The principal normal vector of each node to be assembled. For the first The reference tangential direction of each node to be assembled. It is the cross product of vectors.

[0017] Furthermore, calculating the target homogeneous pose of the node to be assembled includes:

[0018]

[0019]

[0020]

[0021] in, For the first The target rotation matrix of the local coordinate system of each node to be assembled relative to the global coordinate system. For the first The target homogeneous pose of the nodes to be assembled. For the first The target translation vector of each node to be assembled. For from the first The known offset of the geometric center of each node to be assembled from the design assembly datum point. This is the transpose symbol.

[0022] Furthermore, error compensation is performed based on the error between the actual pose and the target homogeneous pose until it is less than the error threshold. This includes calculating the error between the actual pose and the target homogeneous pose, specifically:

[0023] The actual position of the fastener is:

[0024]

[0025] in, For the first The actual pose of the node to be assembled. For the first The actual rotation matrix of the local coordinate system of each node to be assembled relative to the global coordinate system. For the first The actual translation vector of each node to be assembled;

[0026] Calculate the translation error between the actual pose and the target homogeneous pose:

[0027]

[0028] in, For the first Translation error between the actual pose of the node to be assembled and the homogeneous pose of the target;

[0029] Calculate the rotation error between the actual pose and the target homogeneous pose:

[0030]

[0031] in, For the first The rotation error between the actual pose of the node to be assembled and the homogeneous pose of the target. This is an operation to convert an antisymmetric matrix into an axis-angle vector.

[0032]

[0033] in, For the first The pose error between the actual pose of the node to be assembled and the homogeneous pose of the target node.

[0034] Furthermore, according to the first The pose error between the actual pose and the target homogeneous pose of each node to be assembled Error compensation until it is less than the error threshold includes: outputting the displacement correction amount of the actuator in the form of a PD through a control law or compensation law, thereby reducing the pose error. Less than the error threshold.

[0035] Furthermore, when pose error When the error is less than the threshold, the self-locking mechanism is triggered to complete the locking and the locking status is monitored in real time. If locking fails, the displacement correction amount of the actuator is adjusted by the control law or compensation law.

[0036] This invention also proposes an intelligent positioning system for assembly nodes of a self-locking inclined buckle system, comprising:

[0037] The node feature extraction module is used to acquire the 3D point cloud data of the region to be assembled, and to identify the node features of the node to be assembled based on the 3D point cloud data. The node features include: geometric center and surface normal vector set.

[0038] The pose acquisition module is used to construct a local coordinate system of the node based on the node features, calculate the target homogeneous pose of the node to be assembled, and control the actuator to translate and adjust the attitude of the fastener so that the current pose of the fastener is similar to the target homogeneous pose.

[0039] The assembly module is used to measure the actual pose of the fastener and compensate for the error between the actual pose and the target homogeneous pose until the error is less than the error threshold, thereby completing the intelligent positioning of the assembly node.

[0040] Furthermore, identifying the node features of the node to be assembled based on the 3D point cloud data includes: denoising and downsampling the 3D point cloud data to obtain downsampled 3D point cloud data;

[0041] The average value of the coordinates of all points in the downsampled 3D point cloud data is used as the coordinates of the geometric center of the node to be assembled.

[0042] Principal component analysis was used to obtain the set of surface normal vectors from the downsampled 3D point cloud data;

[0043] The average direction of all surface normal vectors in the set of surface normal vectors is taken as the principal normal vector.

[0044] Furthermore, constructing a local coordinate system for a node based on its characteristics includes:

[0045]

[0046] in, For the first The x-axis of the local coordinate system of the nodes to be assembled. For the first The y-axis of the local coordinate system of the nodes to be assembled. For the first The z-axis of the local coordinate system of the nodes to be assembled. For the first The principal normal vector of each node to be assembled. For the first The reference tangential direction of each node to be assembled. It is the cross product of vectors.

[0047] In summary, the technical solutions conceived by this invention have the following beneficial effects compared with the prior art:

[0048] The technical solution of this invention can intelligently guide the actuator to complete high-precision positioning based on error feedback and PD control law with learning adjustment, and compensate for translation and rotation errors in real time to ensure that the assembly action reaches the set threshold. Finally, the fastener is locked through the self-locking mechanism, and the locking force and displacement status are monitored in real time to determine whether the locking is successful, so as to achieve safe, reliable and closed-loop automatic assembly control. Attached Figure Description

[0049] Figure 1 This is a flowchart of the method in Embodiment 1 of the present invention;

[0050] Figure 2 This is a system structure diagram of Embodiment 2 of the present invention. Detailed Implementation

[0051] To better understand the above technical solutions, the following will provide a detailed explanation of the technical solutions in conjunction with the accompanying drawings and specific implementation methods.

[0052] The method provided by this invention can be implemented in a terminal environment that may include one or more of the following components: a processor, a storage medium, and a display screen. The storage medium stores at least one instruction, which is loaded and executed by the processor to implement the method described in the following embodiments.

[0053] A processor may include one or more processing cores. The processor uses various interfaces and lines to connect various parts of the terminal, and performs various functions and processes data by running or executing instructions, programs, code sets or instruction sets stored in the storage medium, and by calling data stored in the storage medium.

[0054] Storage media can include random access memory (RAM) or read-only memory (ROM). Storage media can be used to store instructions, programs, code, code sets, or instructions.

[0055] The display screen is used to show the user interface of each application.

[0056] In addition, those skilled in the art will understand that the structure of the terminal described above does not constitute a limitation on the terminal. The terminal may include more or fewer components, or combine certain components, or have different component arrangements. For example, the terminal may also include radio frequency circuits, input units, sensors, audio circuits, power supplies, and other components, which will not be described in detail here.

[0057] Example 1

[0058] like Figure 1 As shown, this embodiment proposes an intelligent positioning method for assembly nodes of a self-locking inclined fastening system, including:

[0059] Step 101: Obtain the 3D point cloud data of the region to be assembled, and identify the node features of the node to be assembled based on the 3D point cloud data. The node features include: geometric center and set of surface normal vectors.

[0060] Specifically, identifying the node features of the node to be assembled based on the 3D point cloud data includes: denoising and downsampling the 3D point cloud data to obtain downsampled 3D point cloud data;

[0061] The average value of the coordinates of all points in the downsampled 3D point cloud data is used as the coordinates of the geometric center of the node to be assembled.

[0062] Principal component analysis was used to obtain the set of surface normal vectors from the downsampled 3D point cloud data;

[0063] The average direction of all surface normal vectors in the set of surface normal vectors is taken as the principal normal vector.

[0064] Step 102: Construct a local coordinate system for the node based on the node features, calculate the target homogeneous pose of the node to be assembled, and control the actuator (e.g., a robotic arm or a guiding mechanism) to translate and adjust the attitude of the fastener so that the current pose of the fastener is similar to the target homogeneous pose.

[0065] Specifically, constructing a local coordinate system for a node based on its characteristics includes:

[0066]

[0067] in, For the first The x-axis of the local coordinate system of the nodes to be assembled. For the first The y-axis of the local coordinate system of the nodes to be assembled. For the first The z-axis of the local coordinate system of the nodes to be assembled. For the first The principal normal vector of each node to be assembled. For the first The reference tangential direction of each node to be assembled. It is the cross product of vectors.

[0068] Specifically, calculating the target homogeneous pose of the node to be assembled includes:

[0069]

[0070]

[0071]

[0072] in, For the first The target rotation matrix of the local coordinate system of each node to be assembled relative to the global coordinate system. For the first The target homogeneous pose of the nodes to be assembled. For the first The target translation vector of each node to be assembled. For from the first The known offset of the geometric center of each node to be assembled from the design assembly datum point. This is the transpose symbol.

[0073] Specifically, error compensation based on the error between the actual pose and the target homogeneous pose until it is less than the error threshold includes: calculating the error between the actual pose and the target homogeneous pose, specifically:

[0074] The actual position of the fastener is:

[0075]

[0076] in, For the first The actual pose of the node to be assembled. For the first The actual rotation matrix of the local coordinate system of each node to be assembled relative to the global coordinate system. For the first The actual translation vector of each node to be assembled;

[0077] Calculate the translation error between the actual pose and the target homogeneous pose:

[0078]

[0079] in, For the first Translation error between the actual pose of the node to be assembled and the homogeneous pose of the target;

[0080] Calculate the rotation error between the actual pose and the target homogeneous pose:

[0081]

[0082] in, For the first The rotation error between the actual pose of the node to be assembled and the homogeneous pose of the target. This is an operation to convert an antisymmetric matrix into an axis-angle vector.

[0083]

[0084] in, For the first The pose error between the actual pose of the node to be assembled and the homogeneous pose of the target node.

[0085] Step 103: Measure the actual pose of the fastener, and perform error compensation based on the error between the actual pose and the target homogeneous pose until it is less than the error threshold, thereby completing the intelligent positioning of the assembly node.

[0086] Specifically, according to the first The pose error between the actual pose and the target homogeneous pose of each node to be assembled Error compensation until it is less than the error threshold includes: outputting the displacement correction amount of the actuator in the form of a PD through a control law or compensation law, thereby reducing the pose error. Less than the error threshold.

[0087] Specifically, when pose error When the error is less than the threshold, the self-locking mechanism is triggered to complete the locking and the locking status is monitored in real time. If locking fails, the displacement correction amount of the actuator is adjusted by the control law or compensation law.

[0088] Preferably, this embodiment maximizes the similarity between the current pose of the fastener and the target homogeneous pose by means of the following method:

[0089]

[0090] in, For the first The target potential energy index of each node to be assembled. It is a proportional gain matrix. For time Self-learning weights at time Here is the damping gain matrix. For the first The rate of change of pose error between the actual pose and the target homogeneous pose of each node to be assembled. As the weight of the locking force, This is the actual measured locking force. For reference locking force.

[0091] When pose error Once the value is less than the error threshold, the displacement correction amount is adjusted to make the first... The target potential energy index of each node to be assembled Minimize, thus minimizing pose error To achieve the minimum, so as to maximize the convergence of the current pose of the fastener with the target homogeneous pose.

[0092] Preferably, in this embodiment, the self-learning weights... :

[0093] Define the gain adjustment function:

[0094]

[0095]

[0096] in, Based on the scaling gain matrix, It is a 6×6 identity matrix. This is the first gain adjustment intensity coefficient. For the first The current error eigenvector matrix of each node to be assembled. Based on the basic damping gain matrix, This is the second gain adjustment intensity coefficient.

[0097] No. The current error eigenvector matrix of each node to be assembled Specifically:

[0098]

[0099] Self-learning weights The update formula is:

[0100]

[0101] in, For time Self-learning weights at time This is the learning rate.

[0102] Preferably, in this embodiment The matrix, each row representing the direction along the path. Direction, along Direction, along Direction, around Axis, winding Axis, winding axis, each column with In this way, each error component can be processed individually when performing PD control and calculating the target potential energy index.

[0103] Example 2

[0104] like Figure 2 As shown, this embodiment proposes an intelligent positioning system for assembly nodes of a self-locking inclined buckle system. The system specifically includes:

[0105] The node feature extraction module is used to acquire the 3D point cloud data of the region to be assembled, and to identify the node features of the node to be assembled based on the 3D point cloud data. The node features include: geometric center and surface normal vector set.

[0106] Specifically, identifying the node features of the node to be assembled based on the 3D point cloud data includes: denoising and downsampling the 3D point cloud data to obtain downsampled 3D point cloud data;

[0107] The average value of the coordinates of all points in the downsampled 3D point cloud data is used as the coordinates of the geometric center of the node to be assembled.

[0108] Principal component analysis was used to obtain the set of surface normal vectors from the downsampled 3D point cloud data;

[0109] The average direction of all surface normal vectors in the set of surface normal vectors is taken as the principal normal vector.

[0110] The pose acquisition module is used to construct a local coordinate system of the node based on the node features, calculate the target homogeneous pose of the node to be assembled, and control the actuator to translate and adjust the attitude of the fastener so that the current pose of the fastener is similar to the target homogeneous pose.

[0111] Specifically, constructing a local coordinate system for a node based on its characteristics includes:

[0112]

[0113] in, For the first The x-axis of the local coordinate system of the nodes to be assembled. For the first The y-axis of the local coordinate system of the nodes to be assembled. For the first The z-axis of the local coordinate system of the nodes to be assembled. For the first The principal normal vector of each node to be assembled. For the first The reference tangential direction of each node to be assembled. It is the cross product of vectors.

[0114] Specifically, calculating the target homogeneous pose of the node to be assembled includes:

[0115]

[0116]

[0117]

[0118] in, For the first The target rotation matrix of the local coordinate system of each node to be assembled relative to the global coordinate system. For the first The target homogeneous pose of the nodes to be assembled. For the first The target translation vector of each node to be assembled. For from the first The known offset of the geometric center of each node to be assembled from the design assembly datum point. This is the transpose symbol.

[0119] Specifically, error compensation based on the error between the actual pose and the target homogeneous pose until it is less than the error threshold includes: calculating the error between the actual pose and the target homogeneous pose, specifically:

[0120] The actual position of the fastener is:

[0121]

[0122] in, For the first The actual pose of the node to be assembled. For the first The actual rotation matrix of the local coordinate system of each node to be assembled relative to the global coordinate system. For the first The actual translation vector of each node to be assembled;

[0123] Calculate the translation error between the actual pose and the target homogeneous pose:

[0124]

[0125] in, For the first Translation error between the actual pose of the node to be assembled and the homogeneous pose of the target;

[0126] Calculate the rotation error between the actual pose and the target homogeneous pose:

[0127]

[0128] Among them, For the first The rotation error between the actual pose of the node to be assembled and the homogeneous pose of the target. This is an operation to convert an antisymmetric matrix into an axis-angle vector.

[0129]

[0130] in, For the first The pose error between the actual pose of the node to be assembled and the homogeneous pose of the target node.

[0131] The assembly module is used to measure the actual pose of the fastener and compensate for the error between the actual pose and the target homogeneous pose until the error is less than the error threshold, thereby completing the intelligent positioning of the assembly node.

[0132] Specifically, according to the first The pose error between the actual pose and the target homogeneous pose of each node to be assembled Error compensation until it is less than the error threshold includes: outputting the displacement correction amount of the actuator in the form of a PD through a control law or compensation law, thereby reducing the pose error. Less than the error threshold.

[0133] Specifically, when pose error When the error is less than the threshold, the self-locking mechanism is triggered to complete the locking and the locking status is monitored in real time. If locking fails, the displacement correction amount of the actuator is adjusted by the control law or compensation law.

[0134] Preferably, this embodiment maximizes the similarity between the current pose of the fastener and the target homogeneous pose by means of the following method:

[0135]

[0136] in, For the first The target potential energy index of each node to be assembled. It is a proportional gain matrix. For time Self-learning weights at time Here is the damping gain matrix. For the first The rate of change of pose error between the actual pose and the target homogeneous pose of each node to be assembled. As the weight of the locking force, This is the actual measured locking force. For reference locking force.

[0137] When pose error Once the value is less than the error threshold, the displacement correction amount is adjusted to make the first... The target potential energy index of each node to be assembled Minimize, thus minimizing pose error To achieve the minimum, so as to maximize the convergence of the current pose of the fastener with the target homogeneous pose.

[0138] Preferably, in this embodiment, the self-learning weights... :

[0139] Define the gain adjustment function:

[0140]

[0141]

[0142] in, Based on the scaling gain matrix, It is a 6×6 identity matrix. This is the first gain adjustment intensity coefficient. For the first The current error eigenvector matrix of each node to be assembled. Based on the basic damping gain matrix, This is the second gain adjustment intensity coefficient.

[0143] No. The current error eigenvector matrix of each node to be assembled Specifically:

[0144]

[0145] Self-learning weights The update formula is:

[0146]

[0147] in, For time Self-learning weights at time This is the learning rate.

[0148] Preferably, in this embodiment The matrix, each row representing the direction along the path. Direction, along Direction, along Direction, around Axis, winding Axis, winding axis, each column with In this way, each error component can be processed individually when performing PD control and calculating the target potential energy index.

[0149] Example 3

[0150] This invention also proposes a storage medium storing multiple instructions for implementing the intelligent positioning method for assembly nodes of a self-locking inclined buckle system.

[0151] Optionally, in this embodiment, the storage medium may be located in any computer terminal in a group of computer terminals in a computer network, or in any mobile terminal in a group of mobile terminals.

[0152] Optionally, in this embodiment, the storage medium is configured to store program code for performing the method steps of Embodiment 1.

[0153] Example 4

[0154] This invention also proposes an electronic device, including a processor and a storage medium connected to the processor. The storage medium stores multiple instructions, which can be loaded and executed by the processor to enable the processor to execute the intelligent positioning method for assembly nodes of a self-locking inclined fastening system.

[0155] Specifically, the electronic device in this embodiment can be a computer terminal, which may include one or more processors and a storage medium.

[0156] The storage medium can be used to store software programs and modules, such as the intelligent positioning method for assembly nodes of a self-locking inclined fastening system in this embodiment of the invention. The processor executes various functional applications and data processing by running the software programs and modules stored in the storage medium, thereby realizing the aforementioned intelligent positioning method for assembly nodes of a self-locking inclined fastening system. The storage medium may include high-speed random access storage media, and may also include non-volatile storage media, such as one or more magnetic storage systems, flash memory, or other non-volatile solid-state storage media. In some instances, the storage medium may further include storage media remotely configured relative to the processor, which can be connected to the terminal via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0157] The processor can execute the method steps of Embodiment 1 by calling the information and application stored in the storage medium through the transmission system.

[0158] In the above embodiments of the present invention, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0159] In the several embodiments provided by this invention, it should be understood that the disclosed technical content can be implemented in other ways. The system embodiments described above are merely illustrative; for example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces, indirect coupling or communication connection between units or modules, and may be electrical or other forms.

[0160] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0161] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0162] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes: USB flash drives, read-only memory (ROM), random access memory (RAM), portable hard drives, magnetic disks, optical disks, and other media capable of storing program code.

[0163] Obviously, the above embodiments are merely illustrative examples for clear explanation and are not intended to limit the implementation. Those skilled in the art will recognize that other variations or modifications can be made based on the above description. It is neither necessary nor possible to exhaustively list all possible implementations here. However, obvious variations or modifications derived therefrom are still within the scope of protection of this invention.

Claims

1. An intelligent positioning method for assembly nodes of a self-locking inclined buckle system, characterized in that, include: Acquire 3D point cloud data of the region to be assembled, and identify the node features of the node to be assembled based on the 3D point cloud data. The node features include: geometric center and set of surface normal vectors. Identifying node features of nodes to be assembled based on 3D point cloud data includes: denoising and downsampling the 3D point cloud data to obtain downsampled 3D point cloud data; The average value of the coordinates of all points in the downsampled 3D point cloud data is used as the coordinates of the geometric center of the node to be assembled. Principal component analysis was used to obtain the set of surface normal vectors from the downsampled 3D point cloud data; The average direction of all surface normal vectors in the set of surface normal vectors is taken as the principal normal vector; Based on the node characteristics, a local coordinate system for the node is constructed, and the target homogeneous pose of the node to be assembled is calculated. The actuator is controlled to translate and adjust the attitude of the fastener so that the current pose of the fastener is similar to the target homogeneous pose. Constructing a local coordinate system for a node based on its characteristics includes: in, For the first The x-axis of the local coordinate system of the nodes to be assembled. For the first The y-axis of the local coordinate system of the nodes to be assembled. For the first The z-axis of the local coordinate system of the nodes to be assembled. For the first The principal normal vector of each node to be assembled. For the first The reference tangential direction of each node to be assembled. For vector cross product; Calculating the target homogeneous pose of the node to be assembled includes: ， ， ， in, For the first The target rotation matrix of the local coordinate system of each node to be assembled relative to the global coordinate system. For the first The target homogeneous pose of the nodes to be assembled. For the first The target translation vector of each node to be assembled. For from the first The known offset of the geometric center of each node to be assembled from the design assembly datum point. It is the transpose symbol; The actual pose of the fastener is measured, and error compensation is performed based on the error between the actual pose and the target homogeneous pose until it is less than the error threshold, thereby completing the intelligent positioning of the assembly node.

2. The intelligent positioning method for assembly nodes of a self-locking inclined buckle system as described in claim 1, characterized in that, Error compensation is performed based on the error between the actual pose and the target homogeneous pose until it is less than the error threshold. This includes calculating the error between the actual pose and the target homogeneous pose, specifically: The actual position of the fastener is: ， in, For the first The actual pose of the node to be assembled. For the first The actual rotation matrix of the local coordinate system of each node to be assembled relative to the global coordinate system. For the first The actual translation vector of each node to be assembled; Calculate the translation error between the actual pose and the target homogeneous pose: ， in, For the first Translation error between the actual pose of the node to be assembled and the homogeneous pose of the target; Calculate the rotation error between the actual pose and the target homogeneous pose: ， in, For the first The rotation error between the actual pose of the node to be assembled and the homogeneous pose of the target. This is an operation to convert an antisymmetric matrix into an axis-angle vector. ， in, For the first The pose error between the actual pose of the node to be assembled and the homogeneous pose of the target node.

3. The intelligent positioning method for assembly nodes of a self-locking inclined buckle system as described in claim 2, characterized in that, According to the The pose error between the actual pose and the target homogeneous pose of each node to be assembled Error compensation until it is less than the error threshold includes: outputting the displacement correction amount of the actuator in the form of a PD through a control law or compensation law, thereby reducing the pose error. Less than the error threshold.

4. The intelligent positioning method for assembly nodes of a self-locking inclined buckle system as described in claim 3, characterized in that, When pose error When the error is less than the threshold, the self-locking mechanism is triggered to complete the locking and the locking status is monitored in real time. If locking fails, the displacement correction amount of the actuator is adjusted by the control law or compensation law.

5. An intelligent positioning system for assembly nodes of a self-locking inclined buckle system, characterized in that, include: The node feature extraction module is used to acquire the 3D point cloud data of the region to be assembled, and to identify the node features of the node to be assembled based on the 3D point cloud data. The node features include: geometric center and surface normal vector set. Identifying node features of nodes to be assembled based on 3D point cloud data includes: denoising and downsampling the 3D point cloud data to obtain downsampled 3D point cloud data; The average value of the coordinates of all points in the downsampled 3D point cloud data is used as the coordinates of the geometric center of the node to be assembled. Principal component analysis was used to obtain the set of surface normal vectors from the downsampled 3D point cloud data; The average direction of all surface normal vectors in the set of surface normal vectors is taken as the principal normal vector; The pose acquisition module is used to construct a local coordinate system of the node based on the node features, calculate the target homogeneous pose of the node to be assembled, and control the actuator to translate and adjust the attitude of the fastener so that the current pose of the fastener is similar to the target homogeneous pose. Constructing a local coordinate system for a node based on its characteristics includes: ， in, For the first The x-axis of the local coordinate system of the nodes to be assembled. For the first The y-axis of the local coordinate system of the nodes to be assembled. For the first The z-axis of the local coordinate system of the nodes to be assembled. For the first The principal normal vector of each node to be assembled. For the first The reference tangential direction of each node to be assembled. For vector cross product; Calculating the target homogeneous pose of the node to be assembled includes: ， ， ， in, For the first The target rotation matrix of the local coordinate system of each node to be assembled relative to the global coordinate system. For the first The target homogeneous pose of the nodes to be assembled. For the first The target translation vector of each node to be assembled. For from the first The known offset of the geometric center of each node to be assembled from the design assembly datum point. It is the transpose symbol; The assembly module is used to measure the actual pose of the fastener and compensate for the error between the actual pose and the target homogeneous pose until the error is less than the error threshold, thereby completing the intelligent positioning of the assembly node.

Images