AI-based intelligent fiber optic wiring system

Abstract

The present invention provides an AI-based intelligent fiber optic wiring system, which relates to the field of fiber optic communication management technology. The present invention includes a fiber optic scene perception unit, a fiber optic intelligent wiring unit and a fiber optic jumper execution unit; the fiber optic scene perception unit and the fiber optic jumper execution unit are deployed on an on-site remote control robot, and the fiber optic intelligent wiring unit is deployed in the cloud; the system uses AI technology to collect and process images of scenes in the fiber optic wiring room, automatically identifies the fiber optic port status, and generates jumper tasks; the fiber optic jumper execution unit automatically performs fiber optic plugging and unplugging and harness management operations according to the task; after the fiber optic jumper execution unit receives the jumper task, the robot arm will select the appropriate terminal device and adjust the position to perform specific operations. The system reduces manual operation errors, improves work efficiency and network stability, is particularly suitable for the maintenance and upgrade of data centers and communication networks, and solves the time-consuming and error-prone problems in high-frequency jumper operations.

Description

Technical Field

[0001] The present invention relates to the field of optical fiber communication management technology, and in particular to an AI-based intelligent optical fiber wiring system. Background Art

[0002] Optical fiber has the characteristics of fast transmission speed, low loss, large bandwidth, low delay, strong anti-interference ability and high security. It is widely used in the field of modern communications and is the basis for the rapid development of modern communications. These superior characteristics of optical fiber communication make it the preferred transmission medium in various network applications such as data centers, backbone networks, metropolitan area networks and access networks.

[0003] However, with the continuous expansion and upgrading of communication systems, the frequency and complexity of fiber patching work have also increased. Fiber patching refers to the process of switching and adjusting between different optical fiber lines by connecting optical fiber jumpers. This process is particularly important in the maintenance and adjustment of communication systems. However, the traditional fiber patching process usually relies on manual operation, which is time-consuming and prone to errors. Manual fiber patching not only increases the workload of operation and maintenance, but may also cause network interruption and deterioration of service quality, causing inconvenience to communication operators and users.

[0004] Therefore, an AI-based intelligent fiber optic wiring system is needed to solve the above technical problems. Summary of the invention

[0005] In order to solve the above technical problems, the present invention provides an AI-based intelligent fiber optic wiring system, comprising a fiber optic scene perception unit, a fiber optic intelligent wiring unit and a fiber optic jumper execution unit; wherein the fiber optic scene perception unit and the fiber optic jumper execution unit are both deployed on an on-site remote control robot, and the fiber optic intelligent wiring unit is deployed in the cloud and communicated with the fiber optic scene perception unit and the fiber optic jumper execution unit through a wireless network;

[0006] The optical fiber scene perception unit includes a scene image acquisition module, a scene image processing module, a near-field target observation module and a port image processing module; wherein the scene image acquisition module is used to acquire scene images in the optical fiber wiring room, the scene image processing module is used to perform AI target marking on image targets in the scene image, the near-field target observation module is used to perform close-range observation of image targets and acquire near-field target images, and the port image processing module is used to identify optical fiber ports through near-field target images, and perform AI state judgment on optical fiber ports to obtain optical fiber port states;

[0007] The optical fiber intelligent wiring unit includes an optical fiber port management module, a jumper task configuration module, a scene perception control module, an optical fiber jumper decision module and an optical fiber network management module; wherein the optical fiber port management module is used to manage and update the optical fiber port archive, the jumper task configuration module is used to generate corresponding optical fiber jumper tasks according to user settings, the scene perception control module is used to control the on-site remote control robot to perform optical fiber scene perception and control optical fiber jumper actions, the optical fiber jumper decision module parses optical fiber jumper tasks and decides optical fiber jumper actions according to the optical fiber port status, and the optical fiber network management module is used to manage port network configuration;

[0008] The fiber optic patch cord execution unit includes an action execution robot arm and an action execution arm head device; the action execution robot arm is used to select the corresponding action execution arm head device according to the needs and adjust the arm position; the action execution arm head device is used to execute the corresponding fiber optic patch cord action, including a fiber optic plug clamp, a fiber optic wire bundle pick, a fiber optic wire bundle binder and a fiber optic wire bundle remover; wherein, the fiber optic plug clamp is used to clamp the fiber optic plug, the fiber optic wire bundle pick is used to pick up the fiber optic wire bundle, the fiber optic wire bundle binder is used to bind the fiber optic wire bundle, and the fiber optic wire bundle remover is used to remove the wire.

[0009] As a further solution, the fiber optic port file consists of a port status item, a device number item, a port number item, a plug number item, a device position item, a port position item, a harness configuration item and a port configuration item; wherein the port status item is used to set the start, disable and idle states of the fiber optic port, the device number item, the port number item and the plug number item are used to describe the number of the wiring device where the port is located, the number of the fiber optic port itself and the number of the fiber optic plug in the port respectively; the device position item is used to describe the device position of the wiring device in the fiber optic wiring room, the port position item is used to describe the port position of the fiber optic port on the wiring device, the harness configuration item is used to describe the harness set configuration of the fiber optic harness connected to the fiber optic plug, and the port configuration item is used to describe the port network configuration of the fiber optic port in the wiring device.

[0010] As a further solution, the port image processing module includes a fiber optic port identification module, a port image partitioning module, a port outsourcing identification module, a port inner groove identification module and a port contact identification module; wherein the fiber optic port identification module is used to identify the fiber optic port through a close-up target image and obtain a close-up image of the fiber optic port, and the port image partitioning module is used to partition and segment the close-up image of the fiber optic port to obtain a port contact image, a port inner groove image and a port outsourcing image; the port outsourcing identification module identifies the outsourcing covering state through the port outsourcing image, the port inner groove identification module identifies the inner groove embedding state through the port inner groove image, and the contact identification module is used to identify the contact position state through the port contact image.

[0011] As a further solution, the port image partitioning module is preset with a fixed segmentation template for each fiber port. When performing image partitioning: first, the fixed segmentation template corresponding to the current fiber port is obtained, and the near-field target image is subjected to a plane mapping transformation; then, a fixed anchor point is found in the near-field target image after the plane mapping, and it is determined whether the fixed anchor point and the fixed segmentation template match each other; finally, the near-field target image is partitioned by the fixed segmentation template, and each partition image is obtained by partition output.

[0012] As a further solution, the jumper task configuration module configures the fiber jumper task into a site survey task, a plug replacement task, a wire harness storage task and a network configuration task; wherein the fiber jumper task is a combination of one or more subtasks.

[0013] As a further solution, the jumper task configuration module configures the fiber jumper task through the following steps:

[0014] Step 1: The user accesses the visual interface of the optical fiber intelligent wiring unit and visualizes the optical fiber port profile;

[0015] Step 2: The user temporarily configures the fiber port profile locally in the visual interface according to the fiber wiring requirements, and outputs the difference data to obtain the fiber wiring requirement data;

[0016] Step 3: The jumper task configuration module generates corresponding fiber jumper tasks according to the fiber wiring requirement data;

[0017] Step 4: The fiber jumper decision module obtains the fiber jumper task and parses it to obtain the fiber jumper action and action execution conditions corresponding to each subtask;

[0018] Step 5: The fiber jumper decision module continuously monitors the fiber port status, and when the action execution conditions are met, sends a control instruction corresponding to the fiber jumper action to the scene perception control module;

[0019] Step 6: The scene perception control module generates corresponding robot arm control data and arm head device control data according to the fiber jumper action to be performed;

[0020] Step 7: The fiber jumper execution unit receives the robot arm control data and the arm head device control data, executes the robot arm control data through the action execution robot arm, selects the action execution arm head device and executes the arm head device control data;

[0021] Step 8: Wait for the fiber jumper action to be completed, determine the completion status through the completion condition, and update the fiber port file information.

[0022] As a further solution, the field survey tasks include fiber optic wiring room survey, wiring equipment survey, fiber optic port survey and fiber optic bundle survey; wherein, the fiber optic wiring room survey and wiring equipment survey adjust the survey position by moving the on-site remote control robot, and observe through the scene image acquisition module; the fiber optic port survey and fiber optic bundle survey adjust the survey position by controlling the action execution robot arm, and control the fiber optic plug clamp and fiber optic bundle pick to cooperate with the close-range target observation module for observation.

[0023] As a further solution, the plug replacement task includes a plug removal task and a plug insertion task; wherein, the execution condition of the plug removal task is: the optical fiber port exploration has been performed on the designated port number, and the inner groove embedding state has been identified as embedded, and the execution completion condition is: the optical fiber plug of the corresponding port number has been pulled out, and the inner groove embedding state has been identified as not embedded; the execution condition of the plug insertion task is: the optical fiber port exploration has been performed on the designated port number, and the contact position state has been identified as not moved, and the execution completion condition is: the optical fiber plug of the corresponding port number has been inserted, and the outer covering state has been identified as covered.

[0024] As a further solution, the wiring harness storage task includes wiring harness binding and wiring harness removal; wherein, the wiring harness binding involves picking up the optical fiber wiring harness that needs to be adjusted onto the corresponding wiring harness set by means of an optical fiber wiring harness pick, and then binding the optical fiber wiring harness by means of an optical fiber wiring harness binder; the wiring harness removal involves removing the wiring harness of the wiring harness set by means of an optical fiber wiring harness picker, and then picking up and separating the optical fiber wiring harness that needs to be adjusted from the wiring harness set by means of an optical fiber wiring harness pick.

[0025] As a further solution, the network configuration task includes network pause, network configuration and network restart; wherein, the network is automatically paused before the plug replacement task is performed, the network configuration is updated after the plug replacement task is completed, and the network is restarted.

[0026] Compared with the related art, the AI-based intelligent optical fiber wiring system provided by the present invention has the following beneficial effects:

[0027] The present invention includes a fiber optic scene perception unit, a fiber optic intelligent wiring unit and a fiber optic jumper execution unit. The fiber optic scene perception unit and the fiber optic jumper execution unit are deployed on an on-site remote control robot, and the fiber optic intelligent wiring unit is deployed in the cloud. The system uses AI technology to collect and process images of scenes in the fiber optic wiring room, automatically identifies the status of the fiber optic port, and generates jumper tasks. The jumper execution unit automatically performs fiber optic plugging and unplugging and harness management according to the task, realizing accurate and efficient fiber optic jumper operations. The system reduces manual operation errors, improves work efficiency and network stability, and is particularly suitable for the maintenance and upgrade of data centers and communication networks. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 A schematic diagram of the structure of an AI-based intelligent optical fiber wiring system provided by the present invention;

[0029] Figure 2 A schematic diagram of the steps of an AI-based intelligent optical fiber distribution system provided by the present invention. DETAILED DESCRIPTION

[0030] The present invention will be further described below in conjunction with the accompanying drawings and implementation modes.

[0031] like Figure 1 As shown, the AI-based intelligent fiber optic wiring system includes a fiber optic scene perception unit, a fiber optic intelligent wiring unit and a fiber optic jumper execution unit; wherein the fiber optic scene perception unit and the fiber optic jumper execution unit are both deployed on an on-site remote control robot, and the fiber optic intelligent wiring unit is deployed in the cloud and communicates with the fiber optic scene perception unit and the fiber optic jumper execution unit through a wireless network;

[0032] The optical fiber scene perception unit includes a scene image acquisition module, a scene image processing module, a near-field target observation module and a port image processing module; wherein the scene image acquisition module is used to acquire scene images in the optical fiber wiring room, the scene image processing module is used to perform AI target marking on image targets in the scene image, the near-field target observation module is used to perform close-range observation of image targets and acquire near-field target images, and the port image processing module is used to identify optical fiber ports through near-field target images, and perform AI state judgment on optical fiber ports to obtain optical fiber port states;

[0033] The optical fiber intelligent wiring unit includes an optical fiber port management module, a jumper task configuration module, a scene perception control module, an optical fiber jumper decision module and an optical fiber network management module; wherein the optical fiber port management module is used to manage and update the optical fiber port archive, the jumper task configuration module is used to generate corresponding optical fiber jumper tasks according to user settings, the scene perception control module is used to control the on-site remote control robot to perform optical fiber scene perception and control optical fiber jumper actions, the optical fiber jumper decision module parses optical fiber jumper tasks and decides optical fiber jumper actions according to the optical fiber port status, and the optical fiber network management module is used to manage port network configuration;

[0034] The fiber optic patch cord execution unit includes an action execution robot arm and an action execution arm head device; the action execution robot arm is used to select the corresponding action execution arm head device according to the needs and adjust the arm position; the action execution arm head device is used to execute the corresponding fiber optic patch cord action, including a fiber optic plug clamp, a fiber optic wire bundle pick, a fiber optic wire bundle binder and a fiber optic wire bundle remover; wherein, the fiber optic plug clamp is used to clamp the fiber optic plug, the fiber optic wire bundle pick is used to pick up the fiber optic wire bundle, the fiber optic wire bundle binder is used to bind the fiber optic wire bundle, and the fiber optic wire bundle remover is used to remove the wire.

[0035] It should be noted that: in order to solve the many problems caused by traditional manual wiring operations, this embodiment adopts a wiring system that combines "cloud + on-site", the fiber optic scene perception unit and the fiber optic jumper execution unit are deployed on the on-site remote control robot, and the fiber optic intelligent wiring unit is deployed in the cloud; the system uses AI technology to collect and process images of scenes in the fiber optic wiring room, automatically identifies the status of fiber optic ports, and generates jumper tasks; the fiber optic jumper execution unit automatically performs fiber optic plugging and unplugging and harness management operations according to the task; after receiving the jumper task, the fiber optic jumper execution unit, the robotic arm will select the appropriate terminal device and adjust the position to perform specific operations; this embodiment reduces manual operation errors and improves work efficiency and network stability.

[0036] As a further solution, the fiber optic port file consists of a port status item, a device number item, a port number item, a plug number item, a device position item, a port position item, a harness configuration item and a port configuration item; wherein the port status item is used to set the start, disable and idle states of the fiber optic port, the device number item, the port number item and the plug number item are used to describe the number of the wiring device where the port is located, the number of the fiber optic port itself and the number of the fiber optic plug in the port respectively; the device position item is used to describe the device position of the wiring device in the fiber optic wiring room, the port position item is used to describe the port position of the fiber optic port on the wiring device, the harness configuration item is used to describe the harness set configuration of the fiber optic harness connected to the fiber optic plug, and the port configuration item is used to describe the port network configuration of the fiber optic port in the wiring device.

[0037] As a further solution, the port image processing module includes a fiber optic port identification module, a port image partitioning module, a port outsourcing identification module, a port inner groove identification module and a port contact identification module; wherein the fiber optic port identification module is used to identify the fiber optic port through a close-up target image and obtain a close-up image of the fiber optic port, and the port image partitioning module is used to partition and segment the close-up image of the fiber optic port to obtain a port contact image, a port inner groove image and a port outsourcing image; the port outsourcing identification module identifies the outsourcing covering state through the port outsourcing image, the port inner groove identification module identifies the inner groove embedding state through the port inner groove image, and the contact identification module is used to identify the contact position state through the port contact image.

[0038] It should be noted that: this embodiment is the basis for the subsequent rapid plug replacement task. By identifying the status of different areas of the optical fiber port, different plug operations can be coordinated, which is fast and accurate.

[0039] As a further solution, the port image partitioning module is preset with a fixed segmentation template for each fiber port. When performing image partitioning: first, the fixed segmentation template corresponding to the current fiber port is obtained, and the near-field target image is subjected to a plane mapping transformation; then, a fixed anchor point is found in the near-field target image after the plane mapping, and it is determined whether the fixed anchor point and the fixed segmentation template match each other; finally, the near-field target image is partitioned by the fixed segmentation template, and each partition image is obtained by partition output.

[0040] It should be noted that the fixed segmentation template can be understood as a paper-cut template. We only need to select the correct paper-cut template and align the template (find a fixed anchor point) to quickly and accurately partition the foreground target image after plane mapping. The image of each area is used for further status judgment.

[0041] As a further solution, the jumper task configuration module configures the fiber jumper task into a site survey task, a plug replacement task, a wire harness storage task and a network configuration task; wherein the fiber jumper task is a combination of one or more subtasks.

[0042] like Figure 2 As shown, as a further solution, the jumper task configuration module configures the fiber jumper task through the following steps:

[0043] Step 1: The user accesses the visual interface of the optical fiber intelligent wiring unit and visualizes the optical fiber port profile;

[0044] Step 2: The user temporarily configures the fiber port profile locally in the visual interface according to the fiber wiring requirements, and outputs the difference data to obtain the fiber wiring requirement data;

[0045] Step 3: The jumper task configuration module generates corresponding fiber jumper tasks according to the fiber wiring requirement data;

[0046] Step 4: The fiber jumper decision module obtains the fiber jumper task and parses it to obtain the fiber jumper action and action execution conditions corresponding to each subtask;

[0047] Step 5: The fiber jumper decision module continuously monitors the fiber port status, and when the action execution conditions are met, sends a control instruction corresponding to the fiber jumper action to the scene perception control module;

[0048] Step 6: The scene perception control module generates corresponding robot arm control data and arm head device control data according to the fiber jumper action to be performed;

[0049] Step 7: The fiber jumper execution unit receives the robot arm control data and the arm head device control data, executes the robot arm control data through the action execution robot arm, selects the action execution arm head device and executes the arm head device control data;

[0050] Step 8: Wait for the fiber jumper action to be completed, determine the completion status through the completion condition, and update the fiber port file information.

[0051] As a further solution, the field survey tasks include fiber optic wiring room survey, wiring equipment survey, fiber optic port survey and fiber optic bundle survey; wherein, the fiber optic wiring room survey and wiring equipment survey adjust the survey position by moving the on-site remote control robot, and observe through the scene image acquisition module; the fiber optic port survey and fiber optic bundle survey adjust the survey position by controlling the action execution robot arm, and control the fiber optic plug clamp and fiber optic bundle pick to cooperate with the close-range target observation module for observation.

[0052] As a further solution, the plug replacement task includes a plug removal task and a plug insertion task; wherein, the execution condition of the plug removal task is: the optical fiber port exploration has been performed on the designated port number, and the inner groove embedding state has been identified as embedded, and the execution completion condition is: the optical fiber plug of the corresponding port number has been pulled out, and the inner groove embedding state has been identified as not embedded; the execution condition of the plug insertion task is: the optical fiber port exploration has been performed on the designated port number, and the contact position state has been identified as not moved, and the execution completion condition is: the optical fiber plug of the corresponding port number has been inserted, and the outer covering state has been identified as covered.

[0053] It should be noted that: this embodiment aims to quickly determine the execution of the task through the port status without complex intelligent image recognition; the inner slot embedding state is whether there is anything in the port, the contact position state is whether the contact is exposed or deformed and moved, and the outer package covering state is whether the plug is inserted in place. If it is inserted in place, the outer package cannot be seen, otherwise the outer package can be found at a close distance (such as using color to show the outer package area); therefore, the prerequisite for the plug removal task is that there is a plug inside, and it may be in a semi-inserted state at this time, so it is only necessary to judge the inner slot embedding state as embedded (something is there), and there is no need to judge the outer package covering state (exposed in the semi-inserted state); after the plug is pulled out, the inner slot embedding state is directly judged as not embedded (nothing), which means that the plug is taken out; before the plug insertion task is performed, the contact position state is directly obtained, which can not only determine whether there is nothing in the slot, but also determine whether the contacts are shifted to avoid the plug being unable to align; after the plug is inserted, the outer package covering state is directly obtained, which can not only determine whether it is inserted but also determine whether the plug is fully inserted.

[0054] As a further solution, the wiring harness storage task includes wiring harness binding and wiring harness removal; wherein, the wiring harness binding involves picking up the optical fiber wiring harness that needs to be adjusted onto the corresponding wiring harness set by means of an optical fiber wiring harness pick, and then binding the optical fiber wiring harness by means of an optical fiber wiring harness binder; the wiring harness removal involves removing the wiring harness of the wiring harness set by means of an optical fiber wiring harness picker, and then picking up and separating the optical fiber wiring harness that needs to be adjusted from the wiring harness set by means of an optical fiber wiring harness pick.

[0055] As a further solution, the network configuration task includes network pause, network configuration and network restart; wherein, the network is automatically paused before the plug replacement task is performed, the network configuration is updated after the plug replacement task is completed, and the network is restarted.

[0056] The above are only embodiments of the present invention, and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. AI-based intelligent fiber optic wiring system, characterized by: It includes an optical fiber scene perception unit, an optical fiber intelligent wiring unit and an optical fiber jumper execution unit; wherein the optical fiber scene perception unit and the optical fiber jumper execution unit are both deployed on an on-site remote control robot, and the optical fiber intelligent wiring unit is deployed in the cloud and communicates with the optical fiber scene perception unit and the optical fiber jumper execution unit through a wireless network; The optical fiber scene perception unit includes a scene image acquisition module, a scene image processing module, a near-field target observation module and a port image processing module; wherein the scene image acquisition module is used to acquire scene images in the optical fiber wiring room, the scene image processing module is used to perform AI target marking on image targets in the scene image, the near-field target observation module is used to perform close-range observation of image targets and acquire near-field target images, and the port image processing module is used to identify optical fiber ports through near-field target images, and perform AI state judgment on optical fiber ports to obtain optical fiber port states; The optical fiber intelligent wiring unit includes an optical fiber port management module, a jumper task configuration module, a scene perception control module, an optical fiber jumper decision module and an optical fiber network management module; wherein the optical fiber port management module is used to manage and update the optical fiber port archive, the jumper task configuration module is used to generate corresponding optical fiber jumper tasks according to user settings, the scene perception control module is used to control the on-site remote control robot to perform optical fiber scene perception and control optical fiber jumper actions, the optical fiber jumper decision module parses optical fiber jumper tasks and decides optical fiber jumper actions according to the optical fiber port status, and the optical fiber network management module is used to manage port network configuration; The fiber optic patch cord execution unit includes an action execution mechanical arm and an action execution arm head device; the action execution mechanical arm is used to select the corresponding action execution arm head device according to the demand and adjust the arm position; the action execution arm head device is used to execute the corresponding fiber optic patch cord action, including a fiber optic plug clamp, a fiber optic wire bundle pick, a fiber optic wire bundle binder and a fiber optic wire bundle remover; wherein, the fiber optic plug clamp is used to clamp the fiber optic plug, the fiber optic wire bundle pick is used to pick the fiber optic wire bundle, the fiber optic wire bundle binder is used to bind the fiber optic wire bundle, and the fiber optic wire bundle remover is used to remove the wire bundle; The jumper task configuration module configures the fiber jumper task through the following steps: Step 1: The user accesses the visual interface of the optical fiber intelligent wiring unit and visualizes the optical fiber port profile; Step 2: The user temporarily configures the optical fiber port profile locally in the visual interface according to the optical fiber wiring requirements, and outputs the difference data to obtain the optical fiber wiring requirement data; Step 3: The jumper task configuration module generates corresponding fiber jumper tasks according to the fiber wiring requirement data; Step 4: The fiber jumper decision module obtains the fiber jumper task and parses it to obtain the fiber jumper action and action execution conditions corresponding to each subtask; Step 5: The fiber jumper decision module continuously monitors the fiber port status, and when the action execution conditions are met, sends a control instruction corresponding to the fiber jumper action to the scene perception control module; Step 6: The scene perception control module generates corresponding robot arm control data and arm head device control data according to the fiber jumper action to be performed; Step 7: The fiber jumper execution unit receives the robot arm control data and the arm head device control data, executes the robot arm control data through the action execution robot arm, selects the action execution arm head device and executes the arm head device control data; Step 8: Wait for the fiber jumper action to be completed, determine the completion status through the completion condition, and update the fiber port file information.

2. The AI-based intelligent optical fiber wiring system according to claim 1, characterized in that: The optical fiber port file is composed of a port status item, a device number item, a port number item, a plug number item, a device position item, a port position item, a harness configuration item and a port configuration item; wherein the port status item is used to set the start, stop and idle states of the optical fiber port, the device number item, the port number item and the plug number item are used to describe the number of the wiring device where the port is located, the number of the optical fiber port itself and the number of the optical fiber plug in the port respectively; the device position item is used to describe the device position of the wiring device in the optical fiber wiring room, the port position item is used to describe the port position of the optical fiber port on the wiring device, the harness configuration item is used to describe the harness set configuration of the optical fiber harness connected to the optical fiber plug, and the port configuration item is used to describe the port network configuration of the optical fiber port in the wiring device.

3. The AI-based intelligent optical fiber wiring system according to claim 1, characterized in that: The port image processing module includes a fiber optic port identification module, a port image partitioning module, a port outsourcing identification module, a port inner groove identification module and a port contact identification module; wherein the fiber optic port identification module is used to identify the fiber optic port through a close-up target image and obtain a close-up image of the fiber optic port, and the port image partitioning module is used to partition and segment the close-up image of the fiber optic port to obtain a port contact image, a port inner groove image and a port outsourcing image; the port outsourcing identification module identifies the outsourcing covering state through the port outsourcing image, the port inner groove identification module identifies the inner groove embedding state through the port inner groove image, and the contact identification module is used to identify the contact position state through the port contact image.

4. The AI-based intelligent optical fiber wiring system according to claim 3, characterized in that: The port image partitioning module is preset with fixed segmentation templates for each optical fiber port. When performing image partitioning: first, the fixed segmentation template corresponding to the current optical fiber port is obtained, and the near-field target image is subjected to plane mapping transformation; then, a fixed anchor point is found in the near-field target image after plane mapping, and it is determined whether the fixed anchor point and the fixed segmentation template match each other; finally, the near-field target image is partitioned by the fixed segmentation template, and each partition image is obtained by partition output.

5. The AI-based intelligent optical fiber wiring system according to claim 1, characterized in that: The jumper task configuration module configures the fiber jumper task into a site survey task, a plug replacement task, a wire harness storage task, and a network configuration task; wherein the fiber jumper task is a combination of one or more subtasks.

6. The AI-based intelligent optical fiber wiring system according to claim 5, characterized in that: The field survey tasks include fiber optic wiring room survey, wiring equipment survey, fiber optic port survey and fiber optic bundle survey; wherein, the fiber optic wiring room survey and wiring equipment survey adjust the survey position by moving the on-site remote control robot, and observe through the scene image acquisition module; the fiber optic port survey and fiber optic bundle survey adjust the survey position by controlling the action execution robot arm, and control the fiber optic plug clamp and fiber optic bundle pick to cooperate with the close-range target observation module for observation.

7. The AI-based intelligent optical fiber wiring system according to any one of claims 5 and 6, characterized in that: The plug replacement task includes a plug removal task and a plug insertion task; wherein, the execution condition of the plug removal task is: the optical fiber port exploration has been performed on the designated port number, and the inner groove embedding state is identified as embedded, and the execution completion condition is: the optical fiber plug of the corresponding port number has been pulled out, and the inner groove embedding state is identified as not embedded; the execution condition of the plug insertion task is: the optical fiber port exploration has been performed on the designated port number, and the contact position state is identified as not moved, and the execution completion condition is: the optical fiber plug of the corresponding port number is inserted, and the outer covering state is identified as covered.

8. The AI-based intelligent optical fiber wiring system according to claim 5, characterized in that: The wiring harness storage tasks include wiring harness binding and wiring harness removal; wherein, the wiring harness binding involves picking up the optical fiber wiring harness that needs to be adjusted onto the corresponding wiring harness set by using an optical fiber wiring harness picker, and then binding the optical fiber wiring harness by using an optical fiber wiring harness binder; the wiring harness removal involves removing the wiring harness of the wiring harness set by using an optical fiber wiring harness remover, and then picking up and separating the optical fiber wiring harness that needs to be adjusted from the wiring harness set by using an optical fiber wiring harness picker.

9. The AI-based intelligent optical fiber wiring system according to claim 5, characterized in that: The network configuration tasks include network pause, network configuration and network restart; wherein, the network is automatically paused before the plug replacement task is performed, and the network configuration is updated and the network is restarted after the plug replacement task is completed.

Images