Special equipment intelligent navigation system based on AI identification

Abstract

The invention discloses an intelligent navigation system for special equipment based on AI identification. The intelligent navigation system comprises a scheme identification module, an equipment management module, an AI identification module, a cloud platform module, a data docking module and a mobile phone APP terminal, in the scheme compiling stage, the project, equipment, process and personnel information of the special equipment is imported through a scheme identification module; before operation, an operation plan is input through the equipment management module, and construction procedures are confirmed; in the operation process, pictures are uploaded through the mobile phone APP terminal; the data docking module judges the current working procedure through the collected data and broadcasts the attention key points of the current working procedure through voice; and the AI identification module performs intelligent operation on the collected data to obtain the safety state of the equipment, the safety protection measures of the personnel and the behaviors of the operating personnel in the dangerous area. The safety of the whole using process of the special equipment is improved through an AI intelligent means.

Description

Technical Field

[0001] The present invention relates to the field of safety monitoring of special equipment for bridge construction. More specifically, the present invention relates to an intelligent navigation system for special equipment based on AI recognition. Background Art

[0002] Bridge construction involves a lot of special equipment, including tower cranes, bridge cranes, hanging baskets, cable-mounted cranes, truck cranes, crawler cranes, gantry cranes, portal cranes, beam hoists, etc. Currently, the monitoring of these special equipment mostly only monitors the equipment's own status during operation and the safety behavior of personnel in specific dangerous areas.

[0003] From the on-site to the off-site phase, these special equipment undergo three processes: installation, operation, and disassembly. The operation phase is further divided into different steps according to the construction schedule, and some special equipment requires repeated installation and disassembly to relocate. Compared to the entire operational process, the installation and disassembly phase is currently dominated by on-site management, while the operational phase relies primarily on on-site management supplemented by sensor-based safety monitoring. This management approach relies heavily on the experience, commitment, and mental state of on-site managers and operators. Large-scale construction companies put a huge amount of special equipment into operation each year, and uneven on-site management can easily pose safety risks. To address this issue, a system is needed that uses AI to objectively determine whether on-site special equipment meets requirements and conducts acceptance inspections during installation, disassembly, and operation. It also proposes corrective measures for any deficiencies and identifies whether they need to be rectified. Furthermore, it assesses the professional level of the current operational team based on the number of rectifications and provides management advice.

[0004] In bridge construction, from the scheme preparation stage, the key points and construction procedures of the installation, disassembly and use stages of special equipment are imported into the special equipment intelligent navigation system based on AI recognition. The current special equipment construction stage and process are identified through photos, videos, sensors, locations, etc. uploaded from the construction site, and judgments and calculations are made according to the key point status set in the process, such as anchor status, bolt tightening degree, balance, etc. Finally, the statistical and alarm information is fed back to the company level, project level, and site level, and the corrective measures are sent to the on-site construction management personnel, and the loop is closed through photos, videos, and sensors. Summary of the Invention

[0005] The purpose of this invention is to use AI intelligent means to determine whether the key points and processes of on-site special equipment during installation, dismantling and use meet the requirements, solve the dependence on management personnel and operators during the use of special equipment, avoid subjective factors in safety management, and improve the safety of special equipment use.

[0006] The technical solution adopted by the present invention to solve this technical problem is: an intelligent navigation system for special equipment based on AI recognition, including: a solution recognition module, an equipment management module, an AI recognition module, a cloud platform module, a data docking module and a mobile phone APP terminal; During the plan preparation stage, the project, equipment, process and personnel information of the special equipment is imported through the plan identification module; Before operation, input the operation plan through the equipment management module to confirm the construction process; During the operation, photos are uploaded through the mobile phone APP; the data docking module determines the current process through the collected data, and the key points of the current process are announced by voice; the AI recognition module performs intelligent calculations on the collected data to obtain the safety status of the equipment, the safety protection measures of the personnel, and the behavior of the operators in the dangerous areas; the key information of the equipment is displayed on the monitoring board of the cloud platform module, and the alarm information is prompted through the human-computer interaction device, so as to evaluate the operation team during construction and timely determine the professional level of the operation team.

[0007] As a further solution of the present invention, a method for identifying a construction plan of a characteristic device using a plan identification module includes: A1. Import the construction plan and qualification documents and check the plan; A2. Disassemble the construction plan and related accessories into modules; A3. Extract key data; A4. Extract the catalog and evaluate the completeness of the proposal. Evaluation method: Preset keywords and completeness ratios for each subheading and key data. The ratio is obtained when the word count and keyword presence requirements for each subheading are met. Key data must meet minimum quantity, type, data, or word count to achieve the ratio. The ratios for each sub-item are accumulated to obtain the proposal completeness. A5. Push key data to the device management module and notify solutions that do not meet the set completeness.

[0008] As a further solution of the present invention, the device management module includes an operation module for data interaction composed of a web page and a mobile phone, an entry module, an installation module, an operation module, a dismantling module, an exit module and a database.

[0009] As a further solution of the present invention, the operation process of the device management module is as follows: B1. Create an equipment usage plan: The project department's equipment management personnel create a special equipment usage plan and select a construction plan from the database pushed by the plan identification module; B2. The equipment management module automatically exports the equipment type from the database, assigns equipment codes based on the number of equipment, subdivides the process and key points, and the key parts inspection methods and key points; B3. The equipment manager uses the web and mobile operation modules to select the construction steps for each piece of equipment. B4. The equipment management module automatically generates a special equipment plan and detailed process and key points for attention based on the operation steps selected by the operator. After confirmation, the briefing materials and personnel information are entered; B5. The equipment management module publishes special equipment tasks and pushes them to the AI recognition module and cloud platform module.

[0010] As a further solution of the present invention, after the equipment management module releases the task, the construction sub-processes are gradually confirmed according to the construction progress on the mobile phone APP. The mobile phone APP broadcasts the key points of the current process, and takes photos of the key parts of the detailed process on site according to the construction progress and uploads them. The AI recognition module analyzes them until the current task is completed. After the task is completed, the AI recognition module verifies the process completion degree of the equipment in the current field of view.

[0011] As a further solution of the present invention, the data docking module collects the operating parameters, environmental parameters, position parameters and camera videos of the special equipment, and uploads them to the cloud platform module and the AI recognition module respectively.

[0012] As a further solution of the present invention, the AI recognition module includes: a process recognition algorithm, a visual recognition algorithm, an equipment dangerous working condition recognition algorithm and an AI database.

[0013] As a further solution of the present invention, the cloud platform module includes a monitoring page, a human-computer interaction system and an alarm page.

[0014] As a further solution of the present invention, it also includes a job evaluation module, and the evaluation method is as follows: Establish evaluation matrix R i =V a ×M a，b= 〔R1,R2,R3,.....,R a 〕; Evaluation coefficient A=R1+R2+R3+.....+R a， According to the size of A value, it is divided into unqualified, qualified and excellent; in, is the evaluation set matrix Evaluation set element value = (total score - unqualified deduction) / total score. Personnel quality and safety awareness are reported by on-site safety management personnel; safety protection capability, professional ability, and construction efficiency are read and calculated from the equipment management system to obtain the original evaluation value. V a It is weighted data, and the weight ratio is set by the enterprise equipment management personnel; After the equipment is removed from the site, the operation team evaluation is recorded in the database of the equipment management module along with the information of the operation project leader and project operators.

[0015] The present invention has at least the following beneficial effects: 1. This system automatically generates key points of work tasks based on the construction plan, and urges project equipment management personnel and construction personnel to fill in and confirm the task requirements. The AI intelligent recognition algorithm makes judgments and supervises management personnel and construction personnel to prevent construction personnel from reducing necessary processes in order to speed up the progress.

[0016] 2. This system realizes the closed-loop management of special equipment in plan preparation, actual operation and departure, avoiding the safety risks caused by missed inspections or insufficient inspections of key points during the installation and use of special equipment, as well as factors in the on-site construction management supervisor.

[0017] 3. Aiming at the key factors of safety risks, an AI intelligent recognition algorithm for key points of special equipment use was invented. Through data statistics and self-learning methods, it provides a practical and reliable scientific tool for information monitoring during the use of special equipment.

[0018] 4. An assessment method was designed to judge the professional level of the operating team based on factors such as operation time, number of risk warnings, and dangerous behaviors of personnel, in order to achieve management of the operating team.

[0019] Other advantages, objectives and features of the present invention will be reflected in part through the following description, and in part will be understood by those skilled in the art through study and practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 It is a framework diagram of the special equipment intelligent navigation system of the present invention. DETAILED DESCRIPTION

[0021] The present invention is described in detail and completely below with reference to the accompanying drawings. Those skilled in the art will be able to implement the present invention based on this description. Before describing the present invention with reference to the accompanying drawings, it should be noted that the technical solutions and technical features provided in various parts of the present invention, including those described below, may be combined with each other unless they conflict.

[0022] In addition, the embodiments of the present invention described below are generally only part of the embodiments of the present invention, rather than all of the embodiments. Therefore, based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without making any creative efforts should fall within the scope of protection of the present invention.

[0023] Special equipment is classified according to its type, including but not limited to: tower crane, truck crane, crawler crane, gantry crane, portal crane, beam lifting machine, bridge erection machine, beam transporter, elevator, hanging basket, etc. The following description applies to all special equipment.

[0024] The present invention will be further described in detail below with reference to the accompanying drawings and embodiments. The specific implementation process is as follows: like Figure 1 As shown, the present invention provides an intelligent navigation system for special equipment based on AI recognition, including: a solution recognition module, an equipment management module, an AI recognition module, a cloud platform module, a data docking module and a mobile phone APP terminal; During the plan preparation stage, the project, equipment, process and personnel information of the special equipment is imported through the plan identification module; Before operation, input the operation plan through the equipment management module to confirm the construction process; During the operation, photos are uploaded through the mobile phone APP; the data docking module determines the current process through the collected data, and the key points of the current process are announced by voice; the AI recognition module performs intelligent calculations on the collected data to obtain the safety status of the equipment, the safety protection measures of the personnel, and the behavior of the operators in the dangerous areas; the timely status of the equipment, alarm information, operation evaluation and other key information are displayed on the monitoring board of the cloud platform module, and the alarm information is prompted by sound, light, SMS, voice and other means through human-computer interaction equipment, so as to evaluate the operation team during construction and judge the professional level of the operation team in a timely manner.

[0025] In another technical solution, the solution identification module mainly includes: solution format checking, key data extraction, and solution completeness evaluation; the solution identified by the solution identification module sets a standard format template according to the type of special equipment: name, directory, various levels of title, table format, key data volume, but does not limit other content. The method of identifying the construction plan of characteristic equipment using the solution identification module includes: A1. Import the construction plan and qualification documents, check the title, table of contents, and headers of each table, and highlight any areas that do not meet format requirements until they are modified. A2. Disassemble the construction plan and related accessories into modules; A3. Extract key data: project name, equipment type, technical parameters, construction personnel information, installation and disassembly procedures, key parts and precautions for each procedure, construction procedures, key parts and precautions for each procedure; A4. Extract the catalog and evaluate the completeness of the proposal. Evaluation method: Preset keywords and completeness ratios for each subheading and key data. The ratio is obtained when the word count and keyword presence requirements for each subheading are met. Key data must meet minimum quantity, type, data, or word count to achieve the ratio. The ratios for each sub-item are accumulated to obtain the proposal completeness. A5. Push key data to the device management module and notify solutions that do not meet the set 80% completeness.

[0026] In another technical solution, the equipment management module is a system for enterprise equipment management, which includes an operation module for data interaction composed of a web side and a mobile side, an entry module, an installation module, an operation module, a dismantling module, an exit module and a database.

[0027] The operation module is a human-computer interaction page, where equipment managers enter data and select modules on the web or mobile phone.

[0028] The entry module is mainly used to select and enter information related to the entry of special equipment, including contracting team information, equipment entry time, transportation method, installation personnel information, etc.

[0029] The installation module, operation module, and dismantling module are databases that classify and store the data pushed by the scheme identification module according to equipment type-project-construction steps-subdivided processes and key points to note-key parts and key points to note; the database is not deleted with the completion of the project. The database is summarized according to the keywords of equipment type-equipment data-subdivided processes and key points to note-key parts and key points to note, and the standard requirements for special equipment are gradually improved with the intervention of special equipment professional and technical personnel.

[0030] The exit module mainly closes the loop of equipment use, including the selection and entry of exit-related information of special equipment, including the reason for exit, exit time, relevant responsible persons, etc.

[0031] In another technical solution, the operation process of the project-side device management module is as follows: B1. Create an equipment usage plan: The project department's equipment management personnel create a special equipment usage plan and select a construction plan from the database pushed by the plan identification module; B2. The equipment management module automatically exports the equipment type from the database, assigns equipment codes based on the number of equipment, subdivides the process and key points, and the key parts inspection methods and key points; B3. The equipment manager uses the web and mobile operating modules to select construction steps for each device. Construction steps include entry, installation, operation, dismantling, and exit. Entry and exit can only be selected once, while installation, operation, and dismantling can be selected repeatedly as needed. Only after selecting entry can installation, operation, and dismantling be selected. Construction steps can only be carried out according to the installation-operation-dismantling steps and cannot be reversed. After exiting, the equipment's usage plan can be sealed and modified.

[0032] B4. The equipment management module automatically generates a special equipment (installation and operation) plan and detailed process details and key points for attention, including key parts and key points, based on the operating steps selected by the operator. After confirmation by the equipment manager, the briefing materials and personnel information are entered. B5. The equipment management module publishes special equipment (installation, removal, operation) tasks and pushes them to the AI recognition module and cloud platform module.

[0033] In another technical solution, the mobile app primarily serves as a mobile interface for construction workers after tasks are issued by the equipment management system. After the equipment management module issues a task, construction workers gradually confirm the detailed construction steps based on the mobile app's progress. The mobile app then broadcasts key points for the current step and, based on the progress, allows on-site workers to take photos of key details and upload them to the AI recognition module for analysis until the current task is completed. Upon completion, the AI recognition module verifies the completion of the steps on the equipment in the current field of view.

[0034] Take the bridge erection machine through hole as an example: Step 1: Before starting the task, click Start Task on the mobile app, and the mobile app will announce the task requirements.

[0035] Step 2: Construction preparation process: Manually confirm the process, take photos of the on-site working environment and personnel, and confirm them. The AI recognition module pushes the on-site environment and personnel safety protection identification results.

[0036] Step 3: Main beam moving process: Manually confirm the process and announce the key points to note when moving the main beam of the bridge-building machine; before moving the main beam, upload photos of the main beam and the supporting legs being unanchored, and photos of the non-supporting supporting legs being unanchored from the structure, and the AI recognition module will push the recognition results.

[0037] Step 4: Outrigger movement process: Manually confirm the process and announce the key points to note when moving the outriggers. Before moving, upload photos of the outriggers, structures, and main beams being unanchored, and the AI recognition module will push the recognition results.

[0038] Step 5: Task completed: Manually confirm the process and announce the key points to note after the bridge-building machine has completed the hole-passing process; upload photos of the anchorage between each leg and the main beam, and the anchorage between the support legs and the structure, and the AI recognition module will push the recognition results.

[0039] Step 6: Mission completed.

[0040] In another technical solution, a data connection module connects the construction site's special equipment terminal acquisition system with a third-party cloud platform. This module collects the special equipment's operating parameters, environmental parameters, location parameters, and camera video, and uploads them to the cloud platform module and AI recognition module, respectively.

[0041] The data docking module adopts common communication protocols such as MQTT, FTP, MODBUS TCP / RTU transparent transmission, and has protocol conversion, data reception, data sending functions and standard communication protocol templates to connect with newly purchased equipment, leased equipment or local security monitoring platform needs.

[0042] The special equipment terminal acquisition system can transmit data and video to the data docking module through the Internet of Things or wired network.

[0043] In another technical solution, the AI recognition module includes: process recognition algorithm, visual recognition algorithm, equipment hazardous condition recognition algorithm and AI database.

[0044] The AI recognition module has built-in process recognition algorithms for various special equipment. This is achieved through real-time monitoring data uploaded by the special equipment terminal acquisition system. This data comes from various sensors installed on the special equipment, including displacement sensors, cylinder pressure sensors, force state sensors, and induction switches. The current process is determined by the sequential combination and range of this data. The equipment process recognition algorithm is mainly designed for the equipment movement and safety status during the equipment operation phase. The detailed process sensor logical relationship of each type of special equipment is pre-built in the AI algorithm. The visual recognition algorithm uses an improved SSD and Deep Sort-based recognition algorithm. It verifies and confirms workers' safety behaviors within the video field of view, including whether they are wearing safety helmets, whether they are equipped with safety belts when working at heights, and whether they are wearing reflective vests. In process identification, it also verifies the completion of special equipment construction processes. The image recognition algorithm uses a convolutional upscaling network algorithm based on an improved VGG-16 model. The AI recognition algorithm uploads the judgment results to the equipment management system.

[0045] The algorithm for identifying dangerous working conditions of equipment is mainly aimed at the overturning risk of special equipment. A mathematical model is established according to the type of equipment. Data such as the equipment's uniformly distributed load, wind load, hoisting weight, moving load, vibration load, fulcrum support reaction force and position are input into the mathematical model, and the anti-overturning coefficient of the current equipment, the range of motion of the lifting part, the hoisting weight ratio, and the theoretical / actual support reaction force ratio are output. The equipment construction size, weight, and relative position data related to the equipment's uniformly distributed load in the mathematical model are read from the equipment management system, and data such as wind speed, hoisting weight, timely position, fulcrum position, vibration acceleration, and inclination angle are read from the data docking module.

[0046] The AI database is mainly a visual recognition model library, which is classified according to the key points of device types. In the initial stage, photos will be collected for model training. During the application of this navigation system, as the APP side continuously uploads photos and videos, the model library will expand, and the AI model will be trained during the maintenance period to improve the recognition efficiency and accuracy of key points.

[0047] The call of the AI recognition module is carried out according to the special equipment construction steps: During the installation and removal phases, the image recognition algorithm is called. After the task starts on the mobile app, the trained model is called based on the key points of the current process, and the recognition results are fed back to the equipment management module. The call ends when the task is completed. When in the operation stage, the process recognition algorithm, equipment dangerous working condition recognition algorithm and visual recognition algorithm identify the unsafe behavior algorithm of personnel. It is triggered after the installation step is completed and confirmed on the mobile APP side until the dismantling step begins, and the result is fed back to the direct cloud platform module for processing in real time.

[0048] In another technical solution, the cloud platform module includes a monitoring page, a human-computer interaction system, and an alarm page. The cloud platform module has a historical tracing function, saving data according to the construction steps and process sequence. The saved data includes uploaded images, AI recognition results, sensor data, equipment hazardous condition identification algorithm results, task completion time, alarm items and processes, and work efficiency. The task completion time, alarm items and processes, and work efficiency are uploaded to the equipment management module's database after the task is completed.

[0049] The monitoring page has a digital twin function, which creates a 3D model and displays different pages according to different construction stages: 1) Installation phase: Displays the installed components of the current equipment, with components being installed flashing and components not yet installed hidden. Displays the installation progress percentage for installed components = (1 - number of days required to install remaining components / total number of days required to install the equipment) × 100%. Displays the installation efficiency percentage for installed components = (actual number of days installed / theoretical number of days required to install the component) × 100%. Displays the current operator information and AI recognition results. Displays images of key components uploaded by the current app and their recognition results in a timely manner based on the process sequence of the current task. Processes that fail to meet the recognition requirements are magnified, flashed, and recorded on the alarm page.

[0050] 2) Operation Phase: The operation phase is divided into the normal operation phase and the equipment relocation phase. The main page displays the risk factor of the current equipment, including anti-overturning coefficient, wind speed, inclination angle, support reaction ratio, lifting weight, hook speed, and other items related to the equipment type and other items specified in the regulations.

[0051] The digital twin model on the main page reproduces the current operating status of the equipment, displaying the lifting weight and fulcrum reaction force at the hook and fulcrum positions respectively, and is equipped with a bar status display box. According to the actual / rated ratio, the display is displayed from small to large in the order of green (40%), yellow (60%), and red (80%). When it exceeds 90%, the alarm area flashes and notifies the on-site human-computer interaction system.

[0052] 3) During the equipment relocation phase, such as bridge erection machine passing through a hole, tower crane lifting, synchronous basket advancement, and mobile lifting equipment outrigger pads, the digital twin model confirms the process progress based on the task plan of the equipment management system and the mobile app. It zooms in on key points preset in the current process, such as anchor pins, leveling processes, and pre-stressing. It also displays the current equipment relocation progress percentage = (1-remaining relocation process hours / total equipment relocation hours) × 100%. The percentage will flash until the sensor data or AI recognition result is correct. If the construction personnel do not meet the requirements and proceed to the next process, the part of the main digital twin screen will flash rapidly to attract the attention of the manager and notify the on-site human-computer interaction system.

[0053] 4) Demolition phase: Displays the unremoved parts of the current equipment, with parts being removed flashing and parts that have been removed hidden. Displays the removal progress percentage based on the removed parts = (1 - number of days required to remove the remaining parts / total number of days for equipment removal) × 100%. Displays the installation efficiency percentage based on the removed parts = (actual number of days removed / theoretical number of days for removed parts × 100%). Displays the current operator information and AI recognition results. Displays the images of key parts uploaded by the current app and the recognition results in a timely manner according to the process sequence of the current task. Processes that do not meet the recognition requirements are magnified, flashed, and recorded on the alarm page.

[0054] Human-computer interaction is provided by a display screen, voice prompter, and red and green two-color lights set on site. The display screen can view the current equipment operating status data, risk factor, and alarm information; The voice prompter is controlled by the cloud platform and announces the alarm information that is currently approaching the threshold.

[0055] The red and green dual-color light flashes green when there is no alarm information, and flashes red when there is unprocessed alarm information, which is linked to the error flashing of the digital twin.

[0056] Another technical solution also includes a task evaluation module, which is a module for evaluating the quality of the operators. The personnel quality evaluation updates the evaluation data after each task is completed. The evaluation method is as follows: Establish evaluation matrix R i =V a ×M a，b=〔R1,R2,R3,.....,R a 〕; Evaluation coefficient A=R1+R2+R3+.....+R a， According to the size of A value, it is divided into unqualified, qualified and excellent; in, is the evaluation set matrix Evaluation set element value = (total score - unqualified deduction) / total score. Personnel quality and safety awareness are reported by on-site safety management personnel; safety protection capability, professional ability, and construction efficiency are read and calculated from the equipment management system to obtain the original evaluation value. V a It is weighted data, and the weight ratio is set by the enterprise equipment management personnel; After the equipment is removed from the site, the operation team evaluation is recorded in the database of the equipment management module along with the information of the operation project leader and project operators.

[0057] Although the embodiments of the present invention have been disclosed above, they are not limited to the applications listed in the description and implementation methods. They can be fully applied to various fields suitable for the present invention. For those familiar with the art, additional modifications can be easily implemented. Therefore, without departing from the general concept defined by the claims and the scope of equivalents, the present invention is not limited to the specific details and embodiments shown and described herein.

Claims

1. An intelligent navigation system for special equipment based on AI recognition, characterized in that: include: Solution identification module, device management module, AI identification module, cloud platform module, data docking module and mobile APP; During the plan preparation stage, the project, equipment, process and personnel information of the special equipment is imported through the plan identification module; Before operation, input the operation plan through the equipment management module to confirm the construction process; During the operation, photos are uploaded through the mobile phone APP; the data docking module determines the current process through the collected data, and the key points of the current process are announced by voice; the AI recognition module performs intelligent calculations on the collected data to obtain the safety status of the equipment, the safety protection measures of the personnel, and the behavior of the operators in the dangerous areas; the key information of the equipment is displayed on the monitoring board of the cloud platform module, and the alarm information is prompted through the human-computer interaction device, so as to evaluate the operation team during construction and timely determine the professional level of the operation team.

2. The special equipment intelligent navigation system based on AI recognition according to claim 1, characterized in that: The method for identifying the construction plan of characteristic equipment using the plan identification module includes: A1. Import the construction plan and qualification documents and check the plan; A2. Disassemble the construction plan and related accessories into modules; A3. Extract key data; A4. Extract the catalog and evaluate the completeness of the proposal. Evaluation method: Preset keywords and completeness ratios for each subheading and key data. The ratio is obtained when the word count and keyword presence requirements for each subheading are met. Key data must meet minimum quantity, type, data, or word count to achieve the ratio. The ratios for each sub-item are accumulated to obtain the proposal completeness. A5. Push key data to the device management module and notify solutions that do not meet the set completeness.

3. The special equipment intelligent navigation system based on AI recognition according to claim 1, characterized in that: The equipment management module includes an operation module for data interaction composed of a web page and a mobile phone, an entry module, an installation module, an operation module, a dismantling module, an exit module and a database.

4. The special equipment intelligent navigation system based on AI recognition according to claim 3, characterized in that: The operation process of the device management module is as follows: B1. Create an equipment usage plan: The project department's equipment management personnel create a special equipment usage plan and select a construction plan from the database pushed by the plan identification module; B2. The equipment management module automatically exports the equipment type from the database, assigns equipment codes based on the number of equipment, subdivides the process and key points, and the key parts inspection methods and key points; B3. The equipment manager uses the web and mobile operation modules to select the construction steps for each piece of equipment. B4. The equipment management module automatically generates a special equipment plan and detailed process and key points for attention based on the operation steps selected by the operator. After confirmation, the briefing materials and personnel information are entered; B5. The equipment management module publishes special equipment tasks and pushes them to the AI recognition module and cloud platform module.

5. The AI-based intelligent navigation system for special equipment according to claim 4, characterized in that: After the equipment management module releases the task, it gradually confirms the detailed construction processes according to the construction progress on the mobile APP. The mobile APP broadcasts the key points of the current process and takes photos of the key parts of the detailed processes on site according to the construction progress. The photos are uploaded to the AI recognition module for analysis until the current task is completed. After the task is completed, the AI recognition module verifies the process completion degree of the equipment in the current field of view.

6. The special equipment intelligent navigation system based on AI recognition according to claim 1, characterized in that: The data docking module collects the operating parameters, environmental parameters, location parameters, and camera videos of special equipment, and uploads them to the cloud platform module and AI recognition module respectively.

7. The special equipment intelligent navigation system based on AI recognition according to claim 1, characterized in that: The AI recognition module includes: process recognition algorithm, visual recognition algorithm, equipment hazardous condition recognition algorithm and AI database.

8. The special equipment intelligent navigation system based on AI recognition according to claim 1, characterized in that: The cloud platform module includes monitoring page, human-computer interaction system and alarm page.

9. The special equipment intelligent navigation system based on AI recognition according to claim 1, characterized in that: It also includes an assignment assessment module, and the evaluation methods are as follows: Establish evaluation matrix R i =V a ×M a，b= 〔R1,R2,R3,.....,R a 〕; Evaluation coefficient A=R1+R2+R3+.....+R a， According to the size of A value, it is divided into unqualified, qualified and excellent; in, is the evaluation set matrix Evaluation set element value = (total score - unqualified deduction) / total score. Personnel quality and safety awareness are reported by on-site safety management personnel; safety protection capability, professional ability, and construction efficiency are read and calculated from the equipment management system to obtain the original evaluation value. V a It is weighted data, and the weight ratio is set by the enterprise equipment management personnel; After the equipment is removed from the site, the operation team evaluation is recorded in the database of the equipment management module along with the information of the operation project leader and project operators.

Images