Cable tunnel robot inspection system and task scheduling method based on edge-end collaboration

Abstract

The present invention provides a cable tunnel robot inspection system and task scheduling method based on edge-end collaboration, including a cable tunnel inspection robot, a cable tunnel robot inspection system and a communication module. The cable tunnel robot inspection system in the present invention uses sensor perception information as the source of analysis data on the end side, and uses the analysis results of the end-side intelligent reasoning module and the edge intelligent reasoning service on the end side and the edge side as the basis for generating new tasks, respectively, so that the task types and decision-making basis of the robot inspection are richer. In addition, through the overall design of the cable tunnel robot inspection system, a set of inspection task scheduling methods based on edge-end collaboration is formed, which can perform effective edge-end interactive collaboration according to various tasks formed by the end side and the edge side, thereby improving the flexibility of dynamic adjustment and real-time scheduling of robot tasks.

Description

Technical Field

[0001] The present invention belongs to the technical field of cable tunnel inspection, and specifically relates to a cable tunnel robot inspection system and a task scheduling method based on edge-end collaboration. Background Art

[0002] In traditional cable tunnel inspections, inspections of cables and their ancillary facilities installed in the tunnels are mainly carried out manually. In addition to the low efficiency of manual inspections, there are certain personal risks in tunnel operations. With the development of intelligent software and hardware technologies, power robot inspection technology has also been rapidly improved, and robot-based cable tunnel inspections have become the main development direction.

[0003] In the application process of cable tunnel robots, different inspection tasks are generally set according to the different types of inspection targets, and tasks are scheduled according to the priority of the inspection target type. However, the cable tunnel robots that previously used track operation generally had fixed inspection routes, weak perception capabilities, and few task types, so their inspection task modes and task priority settings were relatively fixed. As the number and types of sensors that can be mounted on cable tunnel robots increase, the perception capabilities of cable tunnel robots are enhanced, the task execution model changes, and the types of executable tasks and priority judgment criteria have also increased accordingly. Summary of the invention

[0004] In view of this, the present invention aims to propose a cable tunnel robot inspection system and task scheduling method based on edge collaboration through the design of a cable tunnel robot inspection system, thereby improving the flexibility of dynamic adjustment and real-time scheduling of robot tasks through the optimization of edge perception interaction methods, and improving the environmental adaptability and intelligence level of robot inspection.

[0005] In order to solve the above technical problems, the technical solutions provided by the present invention are as follows:

[0006] In a first aspect, the present invention provides a cable tunnel robot inspection system based on edge-end collaboration, comprising: a cable tunnel inspection robot, a cable tunnel robot inspection system and a communication module;

[0007] The cable tunnel inspection robot and the cable tunnel inspection system communicate with each other via a communication module;

[0008] The cable tunnel inspection robot includes an end-side intelligent reasoning module;

[0009] The terminal-side intelligent reasoning module is used to perform real-time reasoning on the robot's perception data, so that the cable tunnel inspection robot can make decisions in real time based on the reasoning results, adjust task priorities, and schedule tasks.

[0010] The cable tunnel inspection robot is used to report task priority adjustment and task scheduling results to the cable tunnel robot inspection system; it is also used to receive inspection tasks or scheduling instructions from the cable tunnel robot inspection system and carry out specific inspection tasks;

[0011] The cable tunnel robotic inspection system includes edge intelligent inference services;

[0012] The edge intelligent reasoning service is used to reason about the perception data reported by the cable tunnel robot, so that the cable tunnel robot inspection system can perform intelligent analysis in real time based on the reasoning results, and formulate inspection tasks and plans based on the analysis results;

[0013] The cable tunnel robot inspection system is used to issue inspection tasks or dispatch instructions according to the established inspection tasks and plans; it is also used to receive and synchronize robot inspection task list information.

[0014] Furthermore, the cable tunnel inspection robot also includes: a central control system;

[0015] The central control system is used to receive patrol tasks or dispatch instructions from the cable tunnel robot inspection system and drive the robot to carry out specific patrol tasks; it is also used to collect the robot's sensor perception information, adjust and dispatch task priorities according to the perception situation, and report to the cable tunnel robot inspection system through the communication module; it is also used to call the end-side intelligent reasoning module to process the perception information in real time, and make decisions in real time according to the processing results, adjust task priorities and schedule tasks; it is also used to feedback the battery power information fed back by the battery management module to the cable tunnel robot inspection system, and dispatch the robot to return to the charging point when the battery is low; it is also used to display the robot status and environmental conditions through the alarm indication module according to the feedback results of the sensor perception module and the battery management module, combined with the reasoning results of the end-side intelligent reasoning module.

[0016] Furthermore, the cable tunnel inspection robot further comprises: a motion control module;

[0017] The motion control module is used to receive control instructions from the central control system and control the robot to perform specified actions according to the information specified by the control instructions.

[0018] Furthermore, the cable tunnel robot inspection system also includes: a background management system;

[0019] The background management system is used to implement basic data management; it is also used to implement the formulation and issuance of robot inspection tasks; it is also used to implement real-time monitoring and display of the execution status of inspection tasks; it is also used to implement real-time monitoring and display of robot status, and it is also used to implement analysis and display of inspection alarm information; it is also used to implement analysis and display of inspection task result records; it is also used to interact with edge intelligent reasoning services, obtain intelligent reasoning results of visual image data, and automatically plan tasks based on the results; it is also used to interact with patrol task scheduling services, and adjust the priority and plan and schedule patrol tasks based on the patrol task planning results and the reasoning results provided by the intelligent reasoning service; it is also used to interact with the cable tunnel inspection robot side to implement the issuance of tasks and instructions, and to obtain patrol data and status data.

[0020] Furthermore, the cable tunnel robot inspection system also includes: inspection task scheduling service;

[0021] The patrol task scheduling service is used to receive patrol plans made by users through the background management system, set priorities, and synchronize them to the robot. It is also used to automatically plan patrol tasks, set priorities, and synchronize them to the robot based on the analysis results of the edge intelligent reasoning service. It is also used to receive and synchronize the robot patrol task list information based on the analysis results or intelligent reasoning results of the cable tunnel inspection robot on the sensor perception information and priority setting feedback.

[0022] Furthermore, the states of the cable tunnel robot include: running state, idle state, and fault state.

[0023] Furthermore, the cable tunnel robot can execute tasks only when the cable tunnel robot is in the running state or the idle state. The tasks it needs to execute include: manual control tasks, calibration tasks, charging tasks, real-time tasks, planned tasks, and panoramic tasks.

[0024] Furthermore, the tasks that the cable tunnel robot needs to perform are prioritized according to the main priority code and the sub-priority code, wherein the main priority code is arranged from A to F from high to low, and the sub-priority code is arranged from 1 to 1024 from high to low.

[0025] Furthermore, when in motion, the cable tunnel robot first determines the primary priority code, and similar tasks under the same primary priority are determined according to the secondary priority code.

[0026] In a second aspect, the present invention provides a cable tunnel robot inspection task scheduling method based on edge-end collaboration, comprising the following steps:

[0027] Generate a patrol task list and determine whether the task queue is empty. If not, proceed to the next step;

[0028] Take out the highest priority task;

[0029] Determine whether there is a task being executed;

[0030] If not, the highest priority task is removed from the task queue, and the highest priority task is executed and ends;

[0031] If so, compare whether the highest priority task is lower than the priority of the task being executed;

[0032] If so, insert the highest priority task back into the task queue and end;

[0033] If not, the task being executed is interrupted;

[0034] Determine whether the task being executed needs to be restored;

[0035] If yes, insert the executing task back into the task queue and continue to execute the subsequent steps;

[0036] If not, proceed directly to the next step;

[0037] The highest priority task will be removed from the task queue, and the highest priority task will be executed and ended.

[0038] In summary, the present invention provides a cable tunnel robot inspection system and task scheduling method based on edge-end collaboration, including a cable tunnel inspection robot, a cable tunnel robot inspection system and a communication module. The cable tunnel robot inspection system in the present invention uses sensor perception information as the source of analysis data on the end side, and uses the analysis results of the end-side intelligent reasoning module and the edge intelligent reasoning service on the end side and the edge side as the basis for generating new tasks, respectively, so that the task types and decision-making basis of the robot inspection are richer. In addition, through the overall design of the cable tunnel robot inspection system, a set of inspection task scheduling methods based on edge-end collaboration is formed, which can perform effective edge-end interactive collaboration according to various tasks formed by the end side and the edge side, thereby improving the flexibility of dynamic adjustment and real-time scheduling of robot tasks. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0040] Figure 1 A structural diagram of a cable tunnel robot inspection system based on edge-end collaboration provided by an embodiment of the present invention;

[0041] Figure 2 A flow chart of a cable tunnel robot inspection task scheduling method based on edge-end collaboration provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0042] In order to make the purpose, features and advantages of the present invention more obvious and easy to understand, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the embodiments described below are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0043] See also Figure 1 , an embodiment of the present invention provides a cable tunnel robot inspection system based on edge-end collaboration, including: a cable tunnel inspection robot, a cable tunnel robot inspection system and a communication module;

[0044] The cable tunnel inspection robot and the cable tunnel inspection system communicate with each other via a communication module;

[0045] The cable tunnel inspection robot includes an end-side intelligent reasoning module;

[0046] The terminal-side intelligent reasoning module is used to perform real-time reasoning on the robot's perception data, so that the cable tunnel inspection robot can make decisions in real time based on the reasoning results, adjust task priorities, and schedule tasks.

[0047] The cable tunnel inspection robot is used to report task priority adjustment and task scheduling results to the cable tunnel robot inspection system; it is also used to receive inspection tasks or scheduling instructions from the cable tunnel robot inspection system and carry out specific inspection tasks;

[0048] The cable tunnel robotic inspection system includes edge intelligent inference services;

[0049] The edge intelligent reasoning service is used to reason about the perception data reported by the cable tunnel robot, so that the cable tunnel robot inspection system can perform intelligent analysis in real time based on the reasoning results, and formulate inspection tasks and plans based on the analysis results;

[0050] The cable tunnel robot inspection system is used to issue inspection tasks or dispatch instructions according to the established inspection tasks and plans; it is also used to receive and synchronize robot inspection task list information.

[0051] The cable tunnel robot inspection system based on edge-end collaboration provided in this embodiment uses sensor perception information as the source of analysis data on the end side, and uses the analysis results of the end-side intelligent reasoning module and the edge intelligent reasoning service on the end side and edge side as the basis for generating new tasks, respectively, so that the task types and decision-making basis of the robot inspection are more abundant. In addition, through the overall design of the cable tunnel robot inspection system, a set of inspection task scheduling methods based on edge-end collaboration is formed, which can effectively perform edge-end interactive collaboration according to various tasks formed by the end side and the edge side, and improve the flexibility of dynamic adjustment and real-time scheduling of robot tasks.

[0052] For the cable tunnel robot, since it can mount more sensors with richer types, higher requirements are put forward for the data processing it perceives and the interaction with other systems such as the cable tunnel robot inspection system. Therefore, in another embodiment of the present invention, the cable tunnel inspection robot also includes a central control system. The central control system is the central system of the cable tunnel inspection robot body, responsible for: (1) receiving inspection tasks or scheduling instructions from the cable tunnel robot inspection system, and calling the motion control module to drive the robot to carry out specific inspection tasks; (2) collecting the robot's sensor perception information, adjusting and scheduling the task priority according to the perception situation, and reporting to the cable tunnel robot inspection system through the communication module; (3) calling the end-side intelligent reasoning module to process the visible light and infrared images in the sensor perception information in real time, and make decisions in real time according to the processing results, and adjust the task priority and schedule the task. (4) According to the battery power information fed back by the battery management module, it is fed back to the cable tunnel robot inspection system, and the robot is scheduled to return to the charging point when the battery is low. (5) Based on the feedback results of the sensor perception, battery management and other modules, and combined with the reasoning results of the terminal-side intelligent reasoning module, the robot status and environmental conditions are displayed through the alarm indication module.

[0053] By constructing a central control system of the cable tunnel robot, the cable tunnel robot of the present invention can more effectively cope with the processing of enhanced perceptible data and changes in executable tasks compared to traditional cable robots that perform relatively single tasks.

[0054] Based on the central control system provided for the cable tunnel robot, in a further embodiment of the present invention, the central control system is also connected to a sensor perception module, a battery management module and an alarm indication module.

[0055] The sensor perception module is responsible for acquiring information from various sensors, including robot gimbal visible light sensors, robot gimbal infrared sensors, RFID sensors, various gas sensors, temperature and humidity sensors, ultrasonic radars, etc.

[0056] On this basis, the terminal-side intelligent reasoning module is responsible for the real-time reasoning of the robot's perception data, mainly performing intelligent reasoning based on deep learning on the visible light photos and infrared photos obtained by the robot's gimbal visible light sensor and the robot's gimbal infrared sensor, and feeding back the reasoning results to the central control system, which is then fed back to the cable tunnel robot inspection system.

[0057] The battery management module is responsible for monitoring the battery power information and feeding back the battery power information and alarm information to the central control system. The central control system feeds back the battery status to the background management and control system, and performs tasks scheduling and alarm indication for possible alarm situations.

[0058] The alarm indication module is responsible for analyzing and judging the relevant information of the battery management, sensor perception or motion control modules by the central controller, and prompting the robot body and environmental conditions by controlling devices such as alarm indicators or buzzers.

[0059] Increasing the number and types of sensors carried by cable robots can not only effectively utilize the processing power of their central control systems, but also further improve their perception capabilities and enhance their ability to perform tasks in complex environments.

[0060] In another embodiment of the present invention, the cable tunnel inspection robot also includes a motion control module, which is responsible for receiving control instructions from the central control system, controlling the robot's motion mechanism to move or stop, and controlling the robot's pan-tilt to adjust its posture, etc. according to the specific position, pan-tilt angle and other information specified by the control instructions.

[0061] In order to enhance the perception and data processing capabilities of the cable tunnel robot, the cable tunnel robot inspection system can improve the efficiency of edge-end collaboration to adapt to the enhanced capabilities of the robot. In another embodiment of the present invention, the cable tunnel robot inspection system also includes a background management system. The backend management system is the core module of the cable tunnel robot inspection system, which is responsible for: (1) basic data management, including the entry and management of equipment ledgers, inspection points, user information, etc.; (2) formulation and issuance of robot inspection tasks, including the formulation and issuance of different periodic tasks such as daily inspections and special inspections; (3) real-time monitoring and display of the execution status of inspection tasks, including visible light video streams, infrared video streams, inspection images, etc.; (4) real-time monitoring and display of robot status, including battery power, fault status, etc.; (5) analysis and display of inspection alarm information; (6) analysis and display of inspection task result records, etc.; (7) interaction with edge intelligent reasoning service to obtain intelligent reasoning results of visual image data and automatically plan tasks based on the results; (8) interaction with patrol task scheduling service to adjust the priority and plan the patrol tasks according to the patrol task planning results and the reasoning results provided by the intelligent reasoning service; (9) interaction with the cable tunnel inspection robot to implement the issuance of tasks and instructions and the acquisition of patrol data and status data.

[0062] On this basis, the edge intelligent reasoning service is responsible for reasoning the perception data reported by the cable tunnel robot, mainly for intelligent analysis based on deep learning of the visible light photos / videos and infrared photos / videos obtained by the robot's pan-tilt visible light sensor and the robot's pan-tilt infrared sensor, and the analysis results are fed back to the background management system, which visualizes the analysis results and calls the patrol task scheduling service to develop patrol tasks and plans based on the analysis results. It can be seen that the ability of the inspection system can be improved through the background management system, thereby improving the reasoning ability of the edge intelligent reasoning service for perception data.

[0063] In the edge-end collaborative cable tunnel inspection system, due to the improvement of the robot's perception ability and the improvement of the inspection system's reasoning ability, the inspection tasks are complex and changeable. Therefore, in another embodiment of the present invention, the cable tunnel robot inspection system also includes a patrol task scheduling service, which is responsible for setting and synchronizing patrol tasks and priorities: (1) receiving the patrol plan formulated by the user through the background management system and setting the priority, and synchronizing it to the robot central control system; (2) automatically planning patrol tasks, setting priorities, and synchronizing them to the robot central control system according to the analysis results of the edge intelligent reasoning service; (3) receiving and synchronizing the robot patrol task list information according to the analysis results or intelligent reasoning results of the sensor perception information and the priority setting feedback of the cable tunnel inspection robot. By setting the patrol tasks and priorities through the patrol task scheduling service, the task execution efficiency of the cable tunnel robot in the multi-task state can be guaranteed.

[0064] Due to the collaborative operation of the edge, the types and number of inspection tasks of the cable tunnel robot have increased significantly, which poses a challenge to the allocation and scheduling of tasks. The effective execution of the inspection task depends not only on the reasoning generation of the inspection system, but also on the current state of the robot. Therefore, in other embodiments of the present invention, a task scheduling method is proposed according to the different states of the robot and the different task types. Based on the above-mentioned cable tunnel robot inspection system, the state of the cable tunnel robot can be divided into three main states: running state, idle state, and fault state.

[0065] The status codes of the cable tunnel robot inspection system are: operating state (O), idle state (I), and fault state (B).

[0066] The tasks that the cable tunnel robot needs to perform can be divided into six main types, including: manual control tasks (H), calibration tasks (C), charging tasks (R), real-time tasks (A), planning tasks (P), and panoramic tasks (F). The robot can only perform the above tasks when it is in a normal state.

[0067] Manual control tasks are tasks in which the tunnel inspection robot receives manual control instructions issued by the operation and maintenance personnel through the background management and control system and performs specified inspection actions.

[0068] The calibration task is a task that must be completed before the tunnel inspection robot can perform other types of inspection tasks except manual inspection tasks. It is mainly to calibrate the track with the help of the robot positioning sensor to determine the specific task of track length, slope and sensing node related information. In a typical implementation plan, RFID and visual sensors can be used as positioning sensors.

[0069] A charging task is a type of task generated by the robot's central control system and synchronized to the cable tunnel robot inspection system's inspection task scheduling service when the tunnel inspection robot's battery level drops to a pre-specified threshold or the remaining battery level is close to the amount of power required to return to the charging point. At the same time, it drives the robot to the charging pile for charging.

[0070] Real-time tasks are a type of specific inspection tasks that are generated by the patrol task scheduling service based on the terminal intelligent reasoning module and synchronized to the background management and control system by the robot central control system, or generated based on the analysis results of the edge intelligent reasoning service and sent to the robot central control system.

[0071] Planned tasks are pre-set by operation and maintenance personnel based on the tunnel equipment and environmental inspection cycle requirements, and are started by the robot at a specified time to go to a designated inspection point to collect equipment and environmental data.

[0072] Panoramic tasks are a type of task performed to improve the user-friendliness of the cable tunnel robot inspection system, provide maintenance personnel with intuitive views of the equipment and environment, and establish a panoramic image of the cable tunnel robot at a specified inspection point. In a typical implementation, the panoramic task will take a specified number of photos and angles around each inspection point to stitch together a panoramic image of the inspection point.

[0073] In a typical implementation scheme, the priority settings of the above 6 types of tasks in the running state are shown in the following table:

[0074]

[0075] When the robot is in operation, the movement of the cable tunnel robot is determined according to the following rules: (1) First, the main priority code is determined, and the priorities are arranged from high to low from A to F. (2) For tasks of the same type under the same main priority, the secondary priority code is used for determination, and the priorities are arranged from high to low from 1 to 1024.

[0076] The patrol task scheduling service of the cable tunnel robot inspection system maintains a task priority queue and sets task priorities and schedules tasks according to actual conditions.

[0077] The above is a detailed introduction to an embodiment of a cable tunnel robot inspection system based on edge-end collaboration of the present invention. The following is a detailed introduction to an embodiment of a cable tunnel robot inspection task scheduling method based on edge-end collaboration of the present invention.

[0078] See also Figure 2 This embodiment provides a cable tunnel robot inspection task scheduling method based on edge-end collaboration, comprising the following steps:

[0079] Generate a patrol task list and determine whether the task queue is empty. If not, proceed to the next step;

[0080] Take out the highest priority task;

[0081] Determine whether there is a task being executed;

[0082] If not, the highest priority task is removed from the task queue, and the highest priority task is executed and ends;

[0083] If so, compare whether the highest priority task is lower than the priority of the task being executed;

[0084] If so, insert the highest priority task back into the task queue and end;

[0085] If not, the task being executed is interrupted;

[0086] Determine whether the task being executed needs to be restored;

[0087] If yes, insert the executing task back into the task queue and continue to execute the subsequent steps;

[0088] If not, proceed directly to the next step;

[0089] The highest priority task will be removed from the task queue, and the highest priority task will be executed and ended.

[0090] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. The cable tunnel robot inspection system based on edge-end collaboration is characterized by: include: Cable tunnel inspection robot, cable tunnel inspection robot inspection system and communication module; The cable tunnel inspection robot and the cable tunnel inspection robot inspection system communicate with each other through the communication module; The cable tunnel inspection robot includes an end-side intelligent reasoning module; The terminal-side intelligent reasoning module is used to perform real-time reasoning on the robot's perception data, so that the cable tunnel inspection robot can make decisions in real time according to the reasoning results, adjust task priorities and schedule tasks; The cable tunnel inspection robot is used to report task priority adjustment and task scheduling results to the cable tunnel inspection robot inspection system; It is also used to receive the inspection tasks or dispatch instructions of the cable tunnel inspection robot inspection system and carry out specific inspection tasks; The cable tunnel inspection robot inspection system includes edge intelligent reasoning services; The edge intelligent reasoning service is used to reason about the perception data reported by the cable tunnel inspection robot, so that the inspection system of the cable tunnel inspection robot can perform intelligent analysis on the inspection task in real time according to the reasoning result, and use the analysis result and the task priority adjustment and task scheduling result reported by the cable tunnel inspection robot as the basis for generating new tasks, carry out the formulation of inspection tasks and plans, so as to issue inspection tasks or scheduling instructions to the cable tunnel robot, and realize dynamic adjustment of tasks; The cable tunnel inspection robot inspection system is used to issue the inspection tasks or dispatch instructions according to the formulated inspection tasks and plans; It is also used to receive and synchronize robot patrol task list information; The cable tunnel inspection robot inspection system also includes: a background management system and an inspection task scheduling service; The background management system is used to interact with the patrol task scheduling service to realize the formulation and issuance of robot patrol tasks, so as to adjust the priority and plan the patrol tasks according to the patrol task planning results and the reasoning results provided by the intelligent reasoning service; the patrol task scheduling service is used to receive the patrol plan formulated by the user through the background management system and set the priority, and synchronize it to the robot; or automatically plan the patrol task, set the priority, and synchronize it to the robot according to the analysis results of the edge intelligent reasoning service; or receive and synchronize the robot patrol task list information according to the analysis results of the sensor perception information or the intelligent reasoning results and priority setting feedback of the cable tunnel patrol robot.

2. The cable tunnel robot inspection system based on edge-end collaboration according to claim 1 is characterized in that: The cable tunnel inspection robot further comprises: a central control system; The central control system is used to receive patrol tasks or scheduling instructions from the patrol system of the cable tunnel patrol robot, and drive the robot to carry out specific patrol tasks; it is also used to collect the robot's sensor perception information, adjust and schedule task priorities according to the perception situation, and report to the cable tunnel patrol robot patrol system through the communication module; it is also used to call the end-side intelligent reasoning module to process the perception information in real time, and make decisions in real time according to the processing results, adjust task priorities and schedule tasks; it is also used to feedback the battery power information fed back by the battery management module to the cable tunnel patrol robot patrol system, and schedule the robot to return to the charging point when the power is low; it is also used to display the robot status and environmental conditions through the alarm indication module according to the feedback results of the sensor perception module and the battery management module, and combined with the reasoning results of the end-side intelligent reasoning module.

3. The cable tunnel robot inspection system based on edge-end collaboration according to claim 2 is characterized in that: The cable tunnel inspection robot further comprises: a motion control module; The motion control module is used to receive control instructions from the central control system and control the robot to perform specified actions according to the information specified by the control instructions.

4. The cable tunnel robot inspection system based on edge-end collaboration according to claim 1 is characterized in that: The background management system is used to implement basic data management; it is also used to implement the formulation and issuance of robot inspection tasks; it is also used to implement real-time monitoring and display of the execution status of inspection tasks; it is also used to implement real-time monitoring and display of robot status, and it is also used to implement the analysis and display of inspection alarm information; it is also used to implement the analysis and display of inspection task result records; it is also used to implement interaction with the edge intelligent reasoning service, obtain the intelligent reasoning results of visual image data, and automatically plan tasks based on the results; It is also used to interact with the cable tunnel inspection robot to issue tasks and instructions and obtain inspection data and status data.

5. The cable tunnel robot inspection system based on edge-end collaboration according to claim 1 is characterized in that: The states of the cable tunnel robot include: running state, idle state, and fault state.

6. The cable tunnel robot inspection system based on edge-end collaboration according to claim 5 is characterized in that: The cable tunnel robot can execute tasks only when the state of the cable tunnel robot is the running state or the idle state. The tasks that need to be executed include: manual control tasks, calibration tasks, charging tasks, real-time tasks, planned tasks, and panoramic tasks.

7. The cable tunnel robot inspection system based on edge-end collaboration according to claim 6 is characterized in that: The tasks that the cable tunnel robot needs to perform are prioritized according to the main priority code and the secondary priority code, wherein the main priority code is arranged from high to low priority A to F, and the secondary priority code is arranged from high to low priority 1 to 1024.

8. The cable tunnel robot inspection system based on edge-end collaboration according to claim 6 is characterized in that: When in the running state, the cable tunnel robot first determines the main priority code, and for similar tasks under the same main priority, determines according to the secondary priority code.

9. A cable tunnel robot inspection task scheduling method based on edge-end collaboration is characterized in that: The cable tunnel robot inspection task scheduling method is applied to a cable tunnel inspection robot in a cable tunnel robot inspection system based on edge-end collaboration as claimed in any one of claims 1 to 8, comprising the following steps: Generate a patrol task list and determine whether the task queue is empty. If not, proceed to the next step; Take out the highest priority task; Determine whether there is a task being executed; If not, the highest priority task is deleted from the task queue, and the highest priority task is executed and terminated; If so, comparing whether the highest priority task is lower than the priority of the task being executed; If yes, insert the highest priority task back into the task queue and end; If not, interrupt the task being executed; Determine whether the task being executed needs to be restored; If yes, insert the executing task back into the task queue and continue to execute the subsequent steps; If not, proceed directly to the next step; The highest priority task is deleted from the task queue, and the highest priority task is executed and ended.

Images