Intelligent path planning method and system for bridge pier climbing robot

Abstract

The invention relates to the technical field of pier-climbing robots, and discloses an intelligent path planning method and system for a bridge pier-climbing robot, and the method comprises the steps: firstly splitting bridge pier column members to build a database, constructing a pier column digital twin model with marked space coordinates in combination with bridge design data, and integrating static and dynamic data; collecting pier stud surface images, robot postures and environmental meteorological data, and building a collected data time sequence library according to timestamps; fusing data to establish an operation coordinate system, associating coordinates and planning an initial advancing track; then collecting data in real time, identifying difference points, correcting a path, and driving the robot to execute and feed back; and finally recording track and equipment operation data to establish an execution database, and extracting correction data to update a component library so as to iteratively optimize subsequent path planning. According to the method, the path planning precision of the bridge pier climbing robot is realized, the operation process is stabilized, and the adaptability and the operation efficiency of the robot to bridge pier column operation are greatly improved.

Description

Technical Field

[0001] This invention relates to the field of bridge pier climbing robot technology, and more specifically, to an intelligent path planning method and system for a bridge pier climbing robot. Background Technology

[0002] Bridge piers, as the core load-bearing components of bridges, are exposed to complex outdoor environments for extended periods, requiring regular inspection and maintenance to ensure the structural safety of the bridge. Pier-climbing robots, capable of replacing manual labor in high-altitude and high-risk operations, are increasingly being used in bridge pier maintenance. The rationality of path planning directly determines the safety, accuracy, and efficiency of the robot's operations.

[0003] Current path planning methods for bridge pier climbing robots have many shortcomings. Most rely on preset fixed paths, failing to adequately adapt to the differences in components of different piers. When facing complex structures such as pier reinforcements and embedded components, path jamming or deviation is prone to occur. There is a lack of dynamic adaptation to the real-time environment and robot posture. Meteorological factors such as wind speed and temperature, as well as robot posture deviations, often lead to a decrease in operational accuracy. At the same time, there is a lack of a complete feedback and iteration mechanism, making it impossible to correct trajectory deviations during the movement in a timely manner, and it is also difficult to optimize subsequent path planning through historical operation data, resulting in poor operational stability and a narrow range of adaptability.

[0004] Therefore, it is necessary to design an intelligent path planning method and system for bridge pier climbing robots to solve the problems of insufficient adaptability, poor real-time performance, lack of closed-loop optimization, and weak anti-interference ability in traditional path planning methods. Summary of the Invention

[0005] In view of this, the present invention proposes an intelligent path planning method and system for a bridge pier climbing robot, aiming to solve the problems of insufficient adaptability, poor real-time performance, lack of closed-loop optimization, and weak anti-interference ability in traditional path planning methods.

[0006] In one aspect, this invention proposes an intelligent path planning method for a bridge pier climbing robot, comprising: A component library is established by disassembling bridge pier components. Bridge design data is read and combined with the component library data to construct a bridge digital twin model that maps the three-dimensional spatial structure of the piers. Spatial coordinates of each part of the piers are marked on the bridge digital twin model, and static parameters and dynamic adaptation data are integrated. Collect image data of bridge pier surface, robot posture data and environmental meteorological data, and store each data synchronously according to the collection timestamp to establish a time-series database of collected data. By integrating the collected data time series library with the bridge digital twin model, a robot operation coordinate system with the center point of the bottom of the pier as the origin is established, the spatial coordinates of the robot and the pier components are associated, the associated data and attribute data of the components in the component library are retrieved, the initial path of the marked control points is planned, and a continuous initial travel trajectory is formed. The system collects obstacle and position data in real time during the robot's movement, compares the real-time data with the digital twin model data, identifies data discrepancies, corrects the initial path while retaining control points, calls the drive actuator to move along the corrected path, and synchronously provides feedback on the completion status of the action. Record trajectory data and equipment operation data during the path execution process, establish a path execution database by storing the execution timestamps and corresponding path nodes, extract correction data from the path execution database to update the component library, and iteratively optimize the basis for subsequent initial path planning.

[0007] Furthermore, when disassembling bridge pier components to establish a component library and construct a digital twin model of the bridge, the process includes: taking the pier body, pier reinforcement, pier embedded components, and pier connection nodes as disassembly objects, and extracting the geometric parameters, connection relationships, and surface features of each component; Data is stored and categorized according to component type, attribute data is recorded according to geometric parameters and material parameters, and the assembly relationship and spatial position relationship between components are recorded as association data to build a component library containing three types of data. Read bridge structural drawings, pier geometric dimensions, pier material parameters, and bridge completion acceptance data, and match them with corresponding component data from the component library; A digital twin model mapping the three-dimensional spatial structure of bridge piers is constructed, and the attribute data and related data of the component library are associated. The spatial coordinates of each part of the pier are marked, and the static parameters in the design data are integrated with the dynamic adaptation data of the component library.

[0008] Furthermore, when collecting relevant data on bridge piers and establishing a time-series database of collected data, the following steps are taken: collecting surface image data of piers through image acquisition equipment; collecting robot position coordinate data, angle deflection data, and posture change data through attitude sensors; and collecting environmental meteorological data such as wind speed, wind direction, temperature, and humidity through meteorological acquisition equipment. The collected data of various types are classified and preprocessed to remove invalid data caused by image blurring and sensor jitter. Valid data are synchronized and associated according to the acquisition timestamp so that image data, pose data and environmental data at the same time point correspond one-to-one. Establish a structured time-series database for collected data and divide the storage directories according to data types.

[0009] Furthermore, the data is integrated to establish an operational coordinate system and plan the initial path, including: extracting real-time parameters from the time series database of collected data, matching the corresponding coordinates and component attributes in the bridge digital twin model, and setting the three-dimensional coordinate axes to establish the robot's operational coordinate system with the center point of the bottom of the bridge pier as the origin; Accurately map robot posture data and environmental data to corresponding positions in the coordinate system, and associate robot position coordinates with the spatial coordinates of the pier component; Retrieve component association data and attribute data from the component library, and combine them with coordinate system spatial coordinate parameters to set the travel route according to the distribution order of pier components; The connection nodes and surface feature points of the pier components are selected as control points, and the path nodes are sequentially sorted according to spatial coordinates to form a continuous initial travel trajectory.

[0010] Furthermore, when correcting the path in real time and driving the robot to execute, the process includes: collecting data on protrusions, depressions, and external obstacles in the travel route through obstacle detection equipment, collecting robot position coordinate data in real time through positioning equipment, and updating the data synchronously according to the travel time to form a real-time data stream; Extract obstacle coordinates and robot position coordinates from the real-time data stream, match the corresponding component coordinates and spatial structure data in the digital twin model, and identify data discrepancies. Adjust the node coordinates of the initial path based on the differences, retain the original path control points, optimize the travel trajectory in the obstacle area, and reorder the travel order of the path nodes; The robot drives the actuators to move sequentially according to the node coordinates of the corrected path, passing through each control point and synchronously providing feedback on the completion status of the actions during the execution process, and recording the real-time status of the path execution.

[0011] Furthermore, recording path execution data and establishing a path execution database includes: collecting trajectory data during the path execution process, wherein the trajectory data includes the robot's actual travel coordinate sequence, node passage time, and trajectory deviation data; The equipment operation data is recorded synchronously, including robot drive mechanism operation parameters, sensor working status data, and actuator action feedback data. The trajectory data is associated and bound with the device operation data and corresponding path nodes according to the execution timestamp; Establish a structured path execution database and set up data indexing and query mechanisms.

[0012] Furthermore, the component library is iteratively updated based on path execution data, including: extracting trajectory deviation data, equipment operation data, and obstacle data from the path execution database; Based on the extracted data, adjust the component attribute data in the component library, supplement the component surface feature records corresponding to the obstacle area, and update the spatial position relationship in the component association data; Optimize component classification criteria, refine attribute record dimensions, and add attribute fields related to path execution effects; The updated component library data will be used as the basis for subsequent initial path planning.

[0013] Furthermore, the determination of data difference point identification and path correction includes: preset multi-level deviation thresholds, including a first preset component coordinate deviation threshold, a second preset component coordinate deviation threshold, a first preset obstacle distance threshold, a second preset obstacle distance threshold, a first preset robot position deviation threshold, and a second preset robot position deviation threshold, and satisfying that the first preset component coordinate deviation threshold is less than the second preset component coordinate deviation threshold, the first preset obstacle distance threshold is less than the second preset obstacle distance threshold, and the first preset robot position deviation threshold is less than the second preset robot position deviation threshold; Calculate the differences in component coordinates, obstacle distances, and robot positions between real-time data and digital twin model data, respectively; The calculated differences are compared with the corresponding preset graded deviation thresholds to determine the path adjustment requirements: When the component coordinate difference is less than or equal to the upper limit of the first preset component coordinate deviation threshold, the obstacle distance difference is less than or equal to the upper limit of the first preset obstacle distance threshold, and the robot position difference is less than or equal to the upper limit of the first preset robot position deviation threshold, the path is determined to be unchanged and the current trajectory is maintained. When the component coordinate difference is greater than the upper limit of the first preset component coordinate deviation threshold but less than or equal to the upper limit of the second preset component coordinate deviation threshold, the obstacle distance difference is greater than the upper limit of the first preset obstacle distance threshold but less than or equal to the upper limit of the second preset obstacle distance threshold, or the robot position difference is greater than the upper limit of the first preset robot position deviation threshold but less than or equal to the upper limit of the second preset robot position deviation threshold, a data difference point is determined to exist, the path correction process is initiated, and the path node coordinates are adjusted in stages according to the interval where the difference exists. If the component coordinate difference is within the range of the upper limit of the first preset component coordinate deviation threshold and the lower limit of the second preset component coordinate deviation threshold, the obstacle distance difference is within the range of the upper limit of the first preset obstacle distance threshold and the lower limit of the second preset obstacle distance threshold, or the robot position difference is within the range of the upper limit of the first preset robot position deviation threshold and the lower limit of the second preset robot position deviation threshold, perform first-level path correction and adjust the path node coordinate offset to the first offset. If the component coordinate difference is within the second preset component coordinate deviation threshold range, the obstacle distance difference is within the second preset obstacle distance threshold range, or the robot position difference is within the second preset robot position deviation threshold range, perform secondary path correction and adjust the path node coordinate offset to the second offset. The first offset is less than the second offset, and the adjusted path node coordinates must be within the safe operating range of the pier column components.

[0014] Furthermore, the evaluation and secondary correction of the path execution effect includes: setting graded evaluation indicators and corresponding standard values ​​for the path execution effect. The evaluation indicators include node passing accuracy, trajectory deviation range, and equipment operation stability coefficient. The first preset node passing accuracy standard value, the second preset node passing accuracy standard value, the third preset node passing accuracy standard value, the first preset trajectory deviation range standard value, the second preset trajectory deviation range standard value, the third preset trajectory deviation range standard value, the first preset equipment operation stability coefficient standard value, the second preset equipment operation stability coefficient standard value, and the third preset equipment operation stability coefficient standard value are all set. The first preset node passing accuracy standard value is greater than the second preset node passing accuracy standard value, which is greater than the third preset node passing accuracy standard value; the first preset trajectory deviation range standard value is less than the second preset trajectory deviation range standard value, which is less than the third preset trajectory deviation range standard value; and the first preset equipment operation stability coefficient standard value is greater than the second preset equipment operation stability coefficient standard value, which is greater than the third preset equipment operation stability coefficient standard value. At the same time, the priority of the evaluation indicators is set, and the priority from high to low is as follows: the priority of equipment operation stability coefficient is higher than the priority of node passing accuracy, which is higher than the priority of trajectory deviation range. When the judgment results of multiple indicators conflict, the judgment result of the indicator with higher priority shall prevail. Extract relevant data from the path execution database and calculate the actual values ​​of node passing accuracy, trajectory deviation range, and equipment operation stability coefficient. The actual values ​​of each evaluation indicator are compared with the corresponding preset standard values ​​to determine the effectiveness of the path execution. When the actual value of the node passing accuracy is greater than or equal to the second preset node passing accuracy standard value, the actual value of the trajectory deviation range is less than or equal to the second preset trajectory deviation range standard value, and the actual value of the equipment operation stability coefficient is greater than or equal to the second preset equipment operation stability coefficient standard value, the path execution is deemed qualified and no secondary correction is required. When the actual value of node passing accuracy is less than the third preset node passing accuracy standard value, the actual value of trajectory deviation range is greater than the third preset trajectory deviation range standard value, or the actual value of equipment operation stability coefficient is less than the third preset equipment operation stability coefficient standard value, the path execution is determined to be unqualified. The cause of the deviation is analyzed and a secondary correction instruction is generated to adjust the path node spacing and execution parameters, and the corrected path is re-executed. When the actual value of the node passing accuracy is greater than or equal to the third preset node passing accuracy standard value and less than or equal to the second preset node passing accuracy standard value, the actual value of the trajectory deviation range is greater than the second preset trajectory deviation range standard value and less than or equal to the third preset trajectory deviation range standard value, or the actual value of the equipment operation stability coefficient is greater than the second preset equipment operation stability coefficient standard value and less than or equal to the third preset equipment operation stability coefficient standard value, the path execution is determined to be basically qualified, a first-level secondary correction instruction is generated, and the path node angle and drive mechanism operation parameters are finely adjusted. When the actual value of the node passing accuracy is greater than or equal to the first preset node passing accuracy standard value, the actual value of the trajectory deviation range is less than or equal to the first preset trajectory deviation range standard value, and the actual value of the equipment operation stability coefficient is greater than or equal to the first preset equipment operation stability coefficient standard value, the path execution is judged to be excellent, the current path parameters are recorded as the optimal reference case, and stored in the reference data directory of the component library.

[0015] Compared with the prior art, the beneficial effects of the present invention are as follows: 1. By disassembling the pier components to establish a component library and combining it with design data to construct a digital twin model, the three-dimensional spatial structure of the pier is accurately mapped. At the same time, the spatial coordinates of each part are marked, so that the path planning can call the component association data and attribute data, adapting to the complex structural differences of different piers such as the body, stiffeners, and embedded parts. This greatly expands the robot's adaptability to various types of piers and avoids the problems of getting stuck or deviating caused by fixed paths.

[0016] 2. By synchronously collecting images of the pier surface, robot posture, and environmental meteorological data, a time-series database is established. The data is then fused to construct a robot operation coordinate system. At the same time, obstacle and robot position data are collected in real time. Differences between real-time data and the digital twin model are quickly identified and the path is corrected, enabling dynamic adjustment of path planning and improving the real-time response speed and position accuracy of robot operations.

[0017] 3. By recording trajectory data and equipment operation data during the path execution process, a path execution database is established, and the component library is iteratively updated by extracting correction data, forming a complete closed loop of "planning-execution-feedback-optimization". This ensures that the basis for subsequent initial path planning is continuously improved, thereby continuously enhancing the rationality of path planning and operational stability.

[0018] 4. In the data acquisition stage, environmental and meteorological data are integrated, and in the path correction stage, external obstacles such as protrusions and depressions are addressed in a targeted manner. This ensures that path planning fully considers the impact of wind speed, temperature, and sudden conditions on the surface of the piers, effectively resisting interference from complex environments and sudden obstacles, and ensuring the safety and continuity of the robot's high-altitude operations.

[0019] On the other hand, this application also provides an intelligent path planning system for a bridge pier climbing robot, characterized in that it includes: a component modeling module, a data acquisition module, a path planning module, a path execution module, and an iterative update module; The component modeling module is used to disassemble bridge pier components, establish a component library, read bridge design data, and construct a digital twin model of the bridge. The data acquisition module is used to acquire various types of data on images, attitude, environment, and obstacles through image acquisition devices, attitude sensors, meteorological acquisition devices, and obstacle detection devices, and to establish a time-series database of the acquired data. The path planning module is used to integrate the collected data with the digital twin model, establish a work coordinate system, plan the initial path, and correct the path based on real-time data. The path execution module includes a drive actuator and a positioning device. The path execution module is used to receive path correction instructions, drive the robot to move along the path, synchronously provide feedback on the completion status of the action, and record trajectory data and device operation data. The iterative update module is used to establish a path execution database, extract trajectory deviation data, equipment operation data, and obstacle data from the path execution database to update the component library, adjust the path planning basis, and perform iterative optimization of the initial path planning. The component modeling module, data acquisition module, path planning module, path execution module, and iterative update module are connected via a communication bus.

[0020] It is understandable that the above-mentioned intelligent path planning method and system for bridge pier climbing robots have the same beneficial effects, and will not be elaborated further here. Attached Figure Description

[0021] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings: Figure 1 A flowchart of an intelligent path planning method for a bridge pier climbing robot provided in an embodiment of the present invention; Figure 2 This is a functional block diagram of the intelligent path planning system for a bridge pier climbing robot provided in an embodiment of the present invention. Detailed Implementation

[0022] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the disclosure to those skilled in the art. It should be noted that, unless otherwise specified, embodiments and features in the embodiments of the present invention can be combined with each other. The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0023] Reference Figure 1 In some embodiments of this application, an intelligent path planning method for a bridge pier climbing robot includes the following steps: Step S100: Disassemble the bridge pier components to establish a component library, read the bridge design data and combine it with the component library data to construct a bridge digital twin model that maps the three-dimensional spatial structure of the piers, mark the spatial coordinates of each part of the piers on the bridge digital twin model, and integrate static parameters and dynamic adaptation data. Step S200: Collect surface image data of bridge piers, robot posture data and environmental meteorological data, store each data synchronously according to the collection timestamp, and establish a time-series database of collected data; Step S300: Integrate the time series data collection library with the bridge digital twin model, establish a robot operation coordinate system with the center point of the bottom of the pier as the origin, associate the spatial coordinates of the robot and the pier components, retrieve the associated data and attribute data of the components in the component library, plan the initial path of the marked control points, and form a continuous initial travel trajectory. Step S400: Collect obstacle data and position data during the robot's movement in real time, compare the real-time data with the digital twin model data, identify data differences, correct the initial path and retain control points, call the drive actuator to move along the corrected path, and synchronously provide feedback on the completion status of the action; Step S500: Record trajectory data and equipment operation data during the path execution process, establish a path execution database by storing the data according to the execution timestamp and path nodes, extract the correction data from the path execution database to update the component library, and iteratively optimize the basis for subsequent initial path planning.

[0024] Specifically, when disassembling bridge pier components to establish a component library and construct a digital twin model of the bridge, the process includes: taking the pier body, pier reinforcement, pier embedded components, and pier connection nodes as disassembly objects, and extracting the geometric parameters, connection relationships, and surface features of each component. Data is stored and categorized according to component type, attribute data is recorded according to geometric parameters and material parameters, and the assembly relationship and spatial position relationship between components are recorded as association data to build a component library containing three types of data. Read bridge structural drawings, pier geometric dimensions, pier material parameters, and bridge completion acceptance data, and match them with corresponding component data from the component library; A digital twin model mapping the three-dimensional spatial structure of bridge piers is constructed, and the attribute data and related data of the component library are associated. The spatial coordinates of each part of the pier are marked, and the static parameters in the design data are integrated with the dynamic adaptation data of the component library.

[0025] Specifically, when collecting data related to bridge piers and establishing a time-series database of the collected data, the following steps are included: Image data of the pier surface is collected by image acquisition equipment, robot position coordinate data, angle deflection data, and posture change data are collected by attitude sensor, and environmental meteorological data such as wind speed, wind direction, temperature, and humidity are collected by meteorological acquisition equipment. The collected data of various types are classified and preprocessed to remove invalid data caused by image blurring and sensor jitter. Valid data are synchronized and associated according to the acquisition timestamp so that image data, pose data and environmental data at the same time point correspond one-to-one. Establish a structured time-series database for collected data and divide the storage directories according to data types.

[0026] Specifically, the process of fusing data to establish a work coordinate system and plan the initial path includes: Extract real-time parameters from the time series database of collected data, match the corresponding coordinates and component attributes in the bridge digital twin model, and establish a robot operation coordinate system with the center point of the bottom of the bridge pier as the origin and set the three-dimensional coordinate axis. Accurately map robot posture data and environmental data to corresponding positions in the coordinate system, and associate robot position coordinates with the spatial coordinates of the pier component; Retrieve component association data and attribute data from the component library, and combine them with coordinate system spatial coordinate parameters to set the travel route according to the distribution order of pier components; The connection nodes and surface feature points of the pier components are selected as control points, and the path nodes are sequentially sorted according to spatial coordinates to form a continuous initial travel trajectory.

[0027] Specifically, real-time path correction and robot execution include: The robot collects data on protrusions, depressions, and external obstacles along its path using obstacle detection equipment, and collects the robot's position coordinates in real time using positioning equipment, which are then updated synchronously according to the travel time to form a real-time data stream. Extract obstacle coordinates and robot position coordinates from the real-time data stream, match the corresponding component coordinates and spatial structure data in the digital twin model, and identify data discrepancies. Adjust the node coordinates of the initial path based on the differences, retain the original path control points, optimize the travel trajectory in the obstacle area, and reorder the travel order of the path nodes; The robot drives the actuators to move sequentially according to the node coordinates of the corrected path, passing through each control point and synchronously providing feedback on the completion status of the actions during the execution process, and recording the real-time status of the path execution.

[0028] Specifically, recording path execution data and establishing a path execution database includes: The trajectory data collected during the path execution process includes the robot's actual travel coordinate sequence, node passage time, and trajectory deviation data. The equipment operation data is recorded synchronously, including the operating parameters of the robot drive mechanism, sensor working status data, and actuator action feedback data. The trajectory data is associated and bound with the device operation data and corresponding path nodes according to the execution timestamp; Establish a structured path execution database and set up data indexing and query mechanisms.

[0029] Specifically, the component library is updated iteratively based on path-based data execution, including: Extract trajectory deviation data, equipment operation data, and obstacle data from the path execution database; Based on the extracted data, adjust the component attribute data in the component library, supplement the component surface feature records corresponding to the obstacle area, and update the spatial position relationship in the component association data; Optimize component classification criteria, refine attribute record dimensions, and add attribute fields related to path execution effects; The updated component library data will be used as the basis for subsequent initial path planning.

[0030] Understandably, by breaking down the pier components to establish a component library and combining it with design data to construct a digital twin model, the three-dimensional spatial structure of the pier can be accurately mapped. At the same time, the spatial coordinates of each part can be marked, so that the path planning can call the component association data and attribute data, adapt to the complex structural differences of different piers such as the body, stiffeners, and embedded parts, and greatly expand the robot's adaptability to various types of piers, avoiding the problems of jamming and deviation caused by fixed paths.

[0031] Understandably, by simultaneously collecting images of the pier surface, robot posture, and environmental meteorological data to establish a time-series database, and fusing the data to construct a robot operation coordinate system, while simultaneously collecting obstacle and robot position data in real time, the system can quickly identify differences between real-time data and the digital twin model and correct the path, thereby achieving dynamic adjustment of path planning and improving the real-time response speed and position accuracy of robot operations.

[0032] Understandably, by recording trajectory data and equipment operation data during the path execution process, a path execution database is established, and the component library is iteratively updated by extracting correction data, forming a complete closed loop of "planning-execution-feedback-optimization". This allows the basis for subsequent initial path planning to be continuously improved, thereby continuously enhancing the rationality of path planning and operational stability.

[0033] Understandably, the data collection phase integrates environmental and meteorological data, while the path correction phase addresses external obstacles such as protrusions and depressions. This ensures that path planning fully considers the impact of wind speed, temperature, and sudden conditions on the pier surface, effectively resisting interference from complex environments and sudden obstacles, and guaranteeing the safety and continuity of the robot's high-altitude operations.

[0034] In a specific embodiment of this application, the above steps are implemented as follows: First, by disassembling the pier body, pier reinforcing bars, pier embedded components, pier connection nodes, and other components, the geometric parameters, connection relationships, and surface features of each component are extracted. Data is categorized and stored according to component type, and attribute data is recorded according to geometric and material parameters. The assembly relationships and spatial positional relationships between components are recorded as associated data to construct a component library. Then, bridge structural drawings, pier geometric dimensions, material parameters, and completion acceptance design data are read and matched with the corresponding data in the component library to construct a digital twin model mapping the pier's three-dimensional spatial structure. Spatial coordinates of each part of the pier are marked, and static parameters and dynamic adaptation data are integrated. Subsequently, image acquisition devices, attitude sensors, and meteorological acquisition devices are used to collect pier surface image data, robot position coordinates, angle deflection, attitude change data, and environmental meteorological data such as wind speed, wind direction, temperature, and humidity. After classifying and preprocessing the collected data to remove invalid data, valid data is synchronously associated according to the collection timestamp, establishing a structured time-series library of collected data categorized by data type. Then, the collected data is extracted... Real-time parameters in the data time series library are matched with the corresponding coordinates and component attributes in the digital twin model. A three-dimensional coordinate system is established with the center point of the bottom of the pier as the origin to establish the robot's operating coordinate system. The robot's posture data and environmental data are accurately mapped to the coordinate system. The spatial coordinates of the robot and the pier component are associated. The associated data and attribute data of the component library are retrieved. The travel route is set according to the distribution order of the pier components. The component connection nodes and surface feature points are selected as control points to sort the path nodes and form a continuous initial travel trajectory. During the robot's travel, obstacle detection and positioning devices collect data on protrusions, depressions, foreign obstacles and robot position coordinates in real time. The data is updated synchronously according to the travel time to form a real-time data stream. The obstacle coordinates and robot position coordinates in the data stream are extracted. The data differences are identified by matching the corresponding component coordinates and spatial structure data in the digital twin model. The initial path node coordinates are adjusted based on the differences and the control points are retained. The obstacle area trajectory is optimized and the path nodes are reordered. The drive actuator is called to travel along the corrected path. The completion status of the action is fed back synchronously and the real-time status of the path execution is recorded.Finally, trajectory data such as the robot's actual travel coordinate sequence, node passage time, and trajectory deviation data, as well as equipment operation data such as drive mechanism operating parameters and sensor working status data, are collected. These two types of data are associated and bound to corresponding path nodes according to the execution timestamp, establishing a structured path execution database with data indexing and query mechanisms. Trajectory deviation data, equipment operation data, and obstacle data are extracted from this database. Component attribute data in the component library is adjusted, component surface feature records are supplemented, and spatial positional relationships in component association data are updated. Simultaneously, component classification standards are optimized, attribute record dimensions are refined, and path execution effect-related attribute fields are added. The updated component library data serves as the basis for subsequent initial path planning, adapting to the differences in the complex structures of various piers and enabling dynamic adjustment of path planning. This forms a complete closed loop of "planning-execution-feedback-optimization," effectively resisting interference from complex environments and sudden obstacles, and ensuring the robot's adaptability, accuracy, stability, and safety in high-altitude operations on bridge piers.

[0035] The above scenarios are merely preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

[0036] Specifically, the identification of data discrepancies and the determination of path correction include: The system includes preset multi-level deviation thresholds, including a first preset component coordinate deviation threshold, a second preset component coordinate deviation threshold, a first preset obstacle distance threshold, a second preset obstacle distance threshold, a first preset robot position deviation threshold, and a second preset robot position deviation threshold, and satisfies that the first preset component coordinate deviation threshold is less than the second preset component coordinate deviation threshold, the first preset obstacle distance threshold is less than the second preset obstacle distance threshold, and the first preset robot position deviation threshold is less than the second preset robot position deviation threshold. Calculate the differences in component coordinates, obstacle distances, and robot positions between real-time data and digital twin model data, respectively; The calculated differences are compared with the corresponding preset graded deviation thresholds to determine the path adjustment requirements: When the component coordinate difference is less than or equal to the upper limit of the first preset component coordinate deviation threshold, the obstacle distance difference is less than or equal to the upper limit of the first preset obstacle distance threshold, and the robot position difference is less than or equal to the upper limit of the first preset robot position deviation threshold, the path is determined to be unchanged and the current trajectory is maintained. When the component coordinate difference is greater than the upper limit of the first preset component coordinate deviation threshold but less than or equal to the upper limit of the second preset component coordinate deviation threshold, the obstacle distance difference is greater than the upper limit of the first preset obstacle distance threshold but less than or equal to the upper limit of the second preset obstacle distance threshold, or the robot position difference is greater than the upper limit of the first preset robot position deviation threshold but less than or equal to the upper limit of the second preset robot position deviation threshold, a data difference point is determined to exist, the path correction process is initiated, and the path node coordinates are adjusted in stages according to the interval where the difference exists. If the component coordinate difference is within the range of the upper limit of the first preset component coordinate deviation threshold and the lower limit of the second preset component coordinate deviation threshold, the obstacle distance difference is within the range of the upper limit of the first preset obstacle distance threshold and the lower limit of the second preset obstacle distance threshold, or the robot position difference is within the range of the upper limit of the first preset robot position deviation threshold and the lower limit of the second preset robot position deviation threshold, perform first-level path correction and adjust the path node coordinate offset to the first offset. If the component coordinate difference is within the second preset component coordinate deviation threshold range, the obstacle distance difference is within the second preset obstacle distance threshold range, or the robot position difference is within the second preset robot position deviation threshold range, perform secondary path correction and adjust the path node coordinate offset to the second offset. The first offset is less than the second offset, and the adjusted path node coordinates must be within the safe operating range of the pier column components.

[0037] Specifically, the evaluation and secondary correction of the path execution effect include: Set graded evaluation indicators and corresponding standard values ​​for path execution performance. Evaluation indicators include node passing accuracy, trajectory deviation range, and equipment operation stability coefficient. Pre-set first preset node passing accuracy standard value, second preset node passing accuracy standard value, third preset node passing accuracy standard value, first preset trajectory deviation range standard value, second preset trajectory deviation range standard value, third preset trajectory deviation range standard value, first preset equipment operation stability coefficient standard value, second preset equipment operation stability coefficient standard value, and third preset equipment operation stability coefficient standard value. The first preset node passing accuracy standard value is greater than the second preset node passing accuracy standard value, which is greater than the third preset node passing accuracy standard value; the first preset trajectory deviation range standard value is less than the second preset trajectory deviation range standard value, which is less than the third preset trajectory deviation range standard value; and the first preset equipment operation stability coefficient standard value is greater than the second preset equipment operation stability coefficient standard value, which is greater than the third preset equipment operation stability coefficient standard value. At the same time, the priority of the evaluation indicators is set, and the priority from high to low is as follows: the priority of equipment operation stability coefficient is higher than the priority of node passing accuracy, which is higher than the priority of trajectory deviation range. When the judgment results of multiple indicators conflict, the judgment result of the indicator with higher priority shall prevail. Extract relevant data from the path execution database and calculate the actual values ​​of node passing accuracy, trajectory deviation range, and equipment operation stability coefficient. The actual values ​​of each evaluation indicator are compared with the corresponding preset standard values ​​to determine the effectiveness of the path execution. When the actual value of the node passing accuracy is greater than or equal to the second preset node passing accuracy standard value, the actual value of the trajectory deviation range is less than or equal to the second preset trajectory deviation range standard value, and the actual value of the equipment operation stability coefficient is greater than or equal to the second preset equipment operation stability coefficient standard value, the path execution is deemed qualified and no secondary correction is required. When the actual value of node passing accuracy is less than the third preset node passing accuracy standard value, the actual value of trajectory deviation range is greater than the third preset trajectory deviation range standard value, or the actual value of equipment operation stability coefficient is less than the third preset equipment operation stability coefficient standard value, the path execution is determined to be unqualified. The cause of the deviation is analyzed and a secondary correction instruction is generated to adjust the path node spacing and execution parameters, and the corrected path is re-executed. When the actual value of the node passing accuracy is greater than or equal to the third preset node passing accuracy standard value and less than or equal to the second preset node passing accuracy standard value, the actual value of the trajectory deviation range is greater than the second preset trajectory deviation range standard value and less than or equal to the third preset trajectory deviation range standard value, or the actual value of the equipment operation stability coefficient is greater than the second preset equipment operation stability coefficient standard value and less than or equal to the third preset equipment operation stability coefficient standard value, the path execution is determined to be basically qualified, a first-level secondary correction instruction is generated, and the path node angle and drive mechanism operation parameters are finely adjusted. When the actual value of the node passing accuracy is greater than or equal to the first preset node passing accuracy standard value, the actual value of the trajectory deviation range is less than or equal to the first preset trajectory deviation range standard value, and the actual value of the equipment operation stability coefficient is greater than or equal to the first preset equipment operation stability coefficient standard value, the path execution is judged to be excellent, the current path parameters are recorded as the optimal reference case, and stored in the reference data directory of the component library.

[0038] Specifically, multiple preset graded deviation thresholds are established, taking into account the millimeter-level precision requirements of bridge pier operations and the measurement range of robot sensors. Specific threshold ranges are set (the thresholds can be dynamically fine-tuned according to different pier sizes and operating conditions): these include a first preset component coordinate deviation threshold (0-2mm), a second preset component coordinate deviation threshold (2-5mm), a first preset obstacle distance threshold (5-10cm), a second preset obstacle distance threshold (10-20cm), a first preset robot position deviation threshold (1-3mm), and a second preset robot position deviation threshold (3-6mm). Furthermore, the first preset threshold must be strictly less than the corresponding second preset threshold to accommodate pier component processing errors (typically ±2mm) and robot positioning. The industry standard accuracy (±1mm) is used; the differences in component coordinates between real-time data and digital twin model data are calculated separately (formula: component coordinate difference = |real-time acquired component 3D coordinates - model preset component 3D coordinates|), obstacle distance difference (formula: obstacle distance difference = |robot real-time position to obstacle surface straight-line distance - theoretical safe distance of corresponding position in the model|), and robot position difference (formula: robot position difference = |robot real-time positioning coordinates - model preset travel coordinates|). Two decimal places are retained during the calculation to ensure the accuracy of the differences; each calculated difference is compared one by one with the corresponding preset graded deviation threshold to accurately determine the path adjustment requirements: when the component coordinate difference ≤ the first preset component coordinate... When the upper limit of the deviation threshold (2mm), the obstacle distance difference is ≤ the upper limit of the first preset obstacle distance threshold (10cm), and the robot position difference is ≤ the upper limit of the first preset robot position deviation threshold (3mm), the path is determined to be unchanged and the current trajectory is maintained. At this time, the robot moves at the original speed. When the component coordinate difference is in the range of (2mm, 5mm), the obstacle distance difference is in the range of (10cm, 20cm), or the robot position difference is in the range of (3mm, 6mm), a data difference point is immediately determined, the path correction process is started, and the path node coordinates are adjusted in stages according to the range of the difference. During the correction process, the robot's direction of travel remains unchanged, and only the node coordinate offset is adjusted. Within the transition range between the upper limit of the first preset threshold and the lower limit of the second preset threshold (component coordinate difference: 2-3.5mm, obstacle distance difference: 10-15cm, robot position difference: 3-4.5mm), a first-level path correction is performed, adjusting the path node coordinate offset to the first offset (0.5-1mm). This offset is suitable for scenarios with slight deviations, avoiding over-correction that could cause trajectory jitter. If the differences are within the core range of the second preset threshold (component coordinate difference: 3.5-5mm, obstacle distance difference: 15-20cm, robot position difference: 4.5-6mm), a second-level path correction is performed, adjusting the path node coordinate offset to the second offset (1-2mm). The first offset (0.5-1mm) is used to adjust the path node coordinate offset.The offset must be less than the second offset (1-2mm), and all adjusted path node coordinates must be within the safe operating range of the pier component (the safe operating range is the area 5cm inward from the edge of the pier surface, avoiding the pier corners and edges of embedded components).

[0039] Specifically, a tiered evaluation index and corresponding quantitative standard values ​​are set for the path execution effect. The evaluation index focuses on operational accuracy and equipment safety, including node passing accuracy, trajectory deviation range, and equipment operation stability coefficient. Specific numerical standards are pre-set: first preset node passing accuracy standard value (≥99.5%), second preset node passing accuracy standard value (≥98%), third preset node passing accuracy standard value (≥95%), first preset trajectory deviation range standard value (≤1mm), second preset trajectory deviation range standard value (≤3mm), third preset trajectory deviation range standard value (≤5mm). The equipment operation stability coefficient standard value is set at three levels: a second preset standard value (≥0.98), a third preset standard value (≥0.95), and a fourth preset standard value (≥0.90). These values ​​strictly adhere to the principle of decreasing accuracy standard value, increasing deviation range standard value, and decreasing stability coefficient standard value, matching the high-precision requirements of bridge maintenance operations. Furthermore, the priority of evaluation indicators is clearly defined, with the priority order from highest to lowest: equipment operation stability coefficient > node passing accuracy > trajectory deviation range. This priority setting is based on the premise that equipment stability is the core prerequisite for operational safety, and node accuracy determines the quality of operation. The quantity and trajectory deviation can be corrected through fine-tuning. When multiple indicator judgment results conflict (such as node passing accuracy being qualified but equipment operation stability coefficient not meeting the standard), the judgment result of the indicator with higher priority shall be strictly taken as the standard. Extract trajectory data, equipment operation logs and other relevant data from the path execution database, and calculate the actual values ​​of the three evaluation indicators respectively: actual value of node passing accuracy = (number of control points actually passed accurately / total number of preset control points) × 100%, actual value of trajectory deviation range = the maximum difference between the actual coordinates and planned coordinates of each node during path execution, and actual value of equipment operation stability coefficient = (equipment fault-free operation time / path) Total execution time) × Equipment load stability rate (load stability rate = the reciprocal of the ratio of actual load fluctuation value to rated load); The actual values ​​of each evaluation indicator are compared with the corresponding preset standard values ​​to accurately determine the path execution effect and implement corresponding processing measures: When the actual value of node passing accuracy is ≥98%, the actual value of trajectory deviation range is ≤3mm, and the actual value of equipment operating stability coefficient is ≥0.95, the path execution is deemed qualified, no secondary correction is required, and the path parameters are recorded as a routine operation template; when the actual value of node passing accuracy is <95%, the actual value of trajectory deviation range is >5mm, or the actual value of equipment operating stability coefficient is <0.At 90, the path execution is deemed unqualified. The trajectory deviation data and equipment fault logs are reviewed through the path execution database to analyze the cause of the deviation (e.g., unidentified obstacles, sensor signal interference, etc.) and generate a secondary correction instruction. This adjusts the path node spacing (from 5cm to 3cm) and execution parameters (reducing the robot's travel speed by 20% and doubling the sensor sampling frequency). The corrected path is then re-executed. The path is re-executed when the actual node passing accuracy is within the [95%, 98%] range, the actual trajectory deviation is within the (3mm, 5mm] range, or the actual equipment operating stability coefficient is within a certain range. When the path is within the range of (0.90, 0.95), the path execution is deemed basically satisfactory, and a first-level secondary correction instruction is generated to fine-tune the path node angles (adjustment range ±0.5°) and the driving mechanism operating parameters (fine-tune the motor speed ±5%). When the actual value of node passing accuracy is ≥99.5%, the actual value of trajectory deviation range is ≤1mm, and the actual value of equipment operating stability coefficient is ≥0.98, the path execution is deemed excellent, and the current path parameters (including coordinate system parameters, threshold settings, driving parameters, etc.) are recorded as the optimal reference case and stored in the reference data directory of the component library for direct use in subsequent similar pier operations.

[0040] Understandably, by pre-setting multiple graded deviation thresholds adapted to industry standards, combined with precise difference calculation and interval comparison, data difference points can be quickly and accurately identified and graded path corrections can be implemented. This avoids trajectory jitter caused by over-correction and ensures that path nodes are always within the safe operating range, effectively addressing issues such as pier component deviation, sudden obstacles, and robot position offset. Simultaneously, a graded evaluation system for path execution performance is constructed with clear quantitative standards and priority settings. By accurately calculating the actual values ​​of evaluation indicators and matching corresponding secondary correction strategies, it not only addresses issues such as insufficient accuracy and equipment instability in path execution but also accumulates excellent path parameter cases, further strengthening the closed-loop optimization capability of path planning. This significantly improves the accuracy, stability, and safety reliability of bridge pier climbing robot path execution, while providing efficient reference data for subsequent similar operations.

[0041] In a specific embodiment of this application, the above steps are implemented in the following ways: The intelligent path planning method is applicable to the operation and maintenance of concrete piers and steel-concrete composite piers of various bridges such as highways and railways. Before the operation, the bridge climbing robot is deployed on the working platform at the bottom of the pier. The robot is equipped with a high-definition industrial camera (image acquisition device), a six-axis gyroscope (attitude sensor), a miniature weather station (weather acquisition device), a lidar (obstacle detection device), and a Beidou positioning module (positioning device). Each device is interconnected with the robot's main control unit through a communication bus. The main control unit pre-stores the structural drawings, completion acceptance and other design data of the target bridge, as well as a blank component library. After the operation is started, the component decomposition and modeling process is executed first. The technicians use the main control unit to list the pier body, ring reinforcement, embedded steel plate, and pier-beam connection nodes as decomposition objects, extract the geometric parameters such as diameter, thickness, and spacing of each component, the material parameters such as reinforced concrete and steel, and the assembly relationships such as welding and bolt connections between components, and store them according to type to build a component library containing classification data, attribute data, and association data. Then, the pre-stored bridge design data is read and matched with the component library data. A 1:1 model of the pier three-dimensional spatial structure is built in the digital twin system of the main control unit, the three-dimensional spatial coordinates of each component are marked, and the static parameters in the design data are integrated with the dynamically adjustable adaptation data in the component library to provide basic model support for subsequent path planning. After the model is built, the robot starts various data acquisition devices. The high-definition industrial camera takes an image of the pier surface every 0.5 seconds, the six-axis gyroscope collects the robot's position coordinates, angle deflection and attitude change data in real time, and the micro weather station records wind speed, wind direction, temperature and humidity data simultaneously. The main control unit preprocesses the collected data, removes blurry images caused by camera shake and invalid data generated by instantaneous sensor interference, and synchronizes and associates the valid data according to the collection timestamp to establish a structured time-series database of collected data classified and stored according to images, attitude and meteorological data. Next, the main control unit integrates the time-series data collection library and the digital twin model. Taking the center point at the bottom of the pier as the origin, it establishes a three-dimensional operation coordinate system with the X-axis along the horizontal tangent of the pier, the Y-axis along the radial direction of the pier, and the Z-axis along the vertical direction of the pier. The robot's real-time posture data and environmental data are accurately mapped to the coordinate system. The robot's current position is associated with the coordinates of the pier components. Data such as the spacing of the reinforcing ribs and the position of the embedded components are retrieved from the component library. The route is set in the order of spiral movement around the pier from bottom to top. Feature points such as the connection between the reinforcing ribs and the body and the corners of the embedded steel plates are selected as control points. The initial movement trajectory is formed by sorting the spatial coordinates.After the robot starts moving along its initial trajectory, the lidar continuously scans the path within a 1-meter radius ahead, collecting real-time data on obstacles such as protruding cracks and fallen concrete blocks. The Beidou positioning module updates the robot's position coordinates every second, forming a real-time data stream. The main control unit compares this real-time data stream with the data from the digital twin model, calculating the differences in component coordinates, obstacle distances, and robot position using formulas (results are rounded to two decimal places). These results are then compared with preset graded deviation thresholds. When the component coordinate difference is ≤2mm, the obstacle distance difference is ≤10cm, and the robot position difference is ≤3mm, the robot maintains a constant speed of 0.2m / s. When the difference falls within the range of (2mm, 5mm), (10cm, 20cm), or (3mm, 6mm), path correction is immediately initiated. If the difference falls within the transition range of 2-3.5mm, 10-15cm, or 3-4.5mm, a first-level path correction is performed, shifting the path node coordinates by 0.5. -1mm; if the difference is within the core range of 3.5-5mm, 15-20cm, or 4.5-6mm, a secondary path correction is performed, adjusting the offset to 1-2mm. All adjusted node coordinates are limited to a safe operating range of 5cm inward from the edge of the pier surface to avoid contact with sharp edges and embedded components. During path execution, the main control unit synchronously records the robot's actual travel coordinate sequence, node passage time, trajectory deviation, and other trajectory data, as well as equipment operation data such as drive motor speed, sensor voltage, and actuator action feedback. This data is associated with path nodes according to the execution timestamp, establishing a path execution database with index query functionality for easy subsequent data retrieval. After the operation is completed, the main control unit extracts trajectory deviation data, equipment fault-free operating time, obstacle distribution, and other information from the database. It updates the surface feature records of the corresponding area in the component library, refines newly added fields such as "surface flatness" and "obstacle distribution density" in the component attributes, and optimizes the component classification standards.Meanwhile, the main control unit calculates the path execution effect according to preset evaluation indicators. Node passing accuracy is calculated as "actual number of accurately passed control points / preset total number of control points × 100%". The trajectory deviation range is taken as the maximum difference between the actual and planned coordinates of each node. The equipment operation stability coefficient is calculated as "fault-free operation time / total execution time × load stability rate". If the actual values ​​of the three indicators are ≥98%, ≤3mm, and ≥0.95 respectively, the result is deemed qualified and saved as a routine operation template. If the indicators are below 95%, >5mm, or <0.90, the logs are reviewed to analyze the cause of the deviation (such as sensor signal interference, sudden strong winds, etc.), and the node spacing is adjusted from 5... The cm value is adjusted to 3cm, the robot speed is reduced by 20%, and the sensor sampling frequency is increased by 100%, and the corrected path is re-executed. If the index is in the middle range, the path node angle is finely adjusted by ±0.5° and the motor speed by ±5%. If the index reaches the standards of ≥99.5%, ≤1mm, and ≥0.98, the path parameters such as the coordinate system parameters and threshold settings are listed as the optimal case and stored in the component library reference directory for direct use in subsequent similar pier operations. Through multiple operation iterations, the component library data is continuously improved, and the adaptability, accuracy, and stability of the robot path planning are gradually improved, realizing the intelligent and efficient operation of bridge pier maintenance.

[0042] The above scenarios are merely preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

[0043] Reference Figure 2 In some embodiments of this application, an intelligent path planning system for a bridge pier climbing robot is characterized by comprising: a component modeling module, a data acquisition module, a path planning module, a path execution module, and an iterative update module.

[0044] Specifically, the component modeling module is used to break down bridge pier components, establish a component library, read bridge design data, and build a digital twin model of the bridge. The data acquisition module is used to collect various types of data on images, attitude, environment, and obstacles through image acquisition devices, attitude sensors, meteorological acquisition devices, and obstacle detection devices, and to establish a time-series database of the acquired data; The path planning module is used to integrate the collected data with the digital twin model, establish the operation coordinate system, plan the initial path, and correct the path based on real-time data. The path execution module includes a drive actuator and a positioning device. The path execution module is used to receive path correction instructions, drive the robot to move along the path, synchronously provide feedback on the completion of actions, and record trajectory data and equipment operation data. The iterative update module is used to establish a path execution database, extract trajectory deviation data, equipment operation data, and obstacle data from the path execution database to update the component library, adjust the path planning basis, and perform iterative optimization of the initial path planning. The component modeling module, data acquisition module, path planning module, path execution module, and iterative update module are connected via a communication bus.

[0045] It is understandable that the above-mentioned intelligent path planning method and system for bridge pier climbing robots have the same beneficial effects, and will not be elaborated further here.

[0046] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program goods. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program goods embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0047] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program goods according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0048] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0049] These computer program instructions can also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0050] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. An intelligent path planning method for a bridge pier climbing robot, characterized in that, include: A component library is established by disassembling bridge pier components. Bridge design data is read and combined with the component library data to construct a bridge digital twin model that maps the three-dimensional spatial structure of the piers. Spatial coordinates of each part of the piers are marked on the bridge digital twin model, and static parameters and dynamic adaptation data are integrated. Collect image data of bridge pier surface, robot posture data and environmental meteorological data, and store each data synchronously according to the collection timestamp to establish a time-series database of collected data. By integrating the collected data time series library with the bridge digital twin model, a robot operation coordinate system with the center point of the bottom of the pier as the origin is established, the spatial coordinates of the robot and the pier components are associated, the associated data and attribute data of the components in the component library are retrieved, the initial path of the marked control points is planned, and a continuous initial travel trajectory is formed. The system collects obstacle and position data in real time during the robot's movement, compares the real-time data with the digital twin model data, identifies data discrepancies, corrects the initial path while retaining control points, calls the drive actuator to move along the corrected path, and synchronously provides feedback on the completion status of the action. Record trajectory data and equipment operation data during the path execution process, establish a path execution database by storing the execution timestamps and corresponding path nodes, extract correction data from the path execution database to update the component library, and iteratively optimize the basis for subsequent initial path planning.

2. The intelligent path planning method for the bridge pier climbing robot according to claim 1, characterized in that, When disassembling bridge pier components to establish a component library and constructing a digital twin model of the bridge, the following steps are included: The pier body, pier reinforcing bars, pier embedded components, and pier connection nodes are treated as decomposition objects, and the geometric parameters, connection relationships, and surface features of each component are extracted. Data is stored and categorized according to component type, attribute data is recorded according to geometric parameters and material parameters, and the assembly relationship and spatial position relationship between components are recorded as association data to build a component library containing three types of data. Read bridge structural drawings, pier geometric dimensions, pier material parameters, and bridge completion acceptance data, and match them with corresponding component data from the component library; A digital twin model mapping the three-dimensional spatial structure of bridge piers is constructed, and the attribute data and related data of the component library are associated. The spatial coordinates of each part of the pier are marked, and the static parameters in the design data are integrated with the dynamic adaptation data of the component library.

3. The intelligent path planning method for the bridge pier climbing robot according to claim 2, characterized in that, When collecting data related to bridge piers and establishing a time-series database of the collected data, the following is included: Image data of the pier surface is collected by image acquisition equipment, robot position coordinate data, angle deflection data, and posture change data are collected by attitude sensor, and environmental meteorological data such as wind speed, wind direction, temperature, and humidity are collected by meteorological acquisition equipment. The collected data of various types are classified and preprocessed to remove invalid data caused by image blurring and sensor jitter. Valid data are synchronized and associated according to the acquisition timestamp so that image data, pose data and environmental data at the same time point correspond one-to-one. Establish a structured time-series database for collected data and divide the storage directories according to data types.

4. The intelligent path planning method for the bridge pier climbing robot according to claim 3, characterized in that, The data is integrated to establish a work coordinate system and plan the initial path, including: Extract real-time parameters from the time series database of collected data, match the corresponding coordinates and component attributes in the bridge digital twin model, and establish a robot operation coordinate system with the center point of the bottom of the bridge pier as the origin and set the three-dimensional coordinate axis. Accurately map robot posture data and environmental data to corresponding positions in the coordinate system, and associate robot position coordinates with the spatial coordinates of the pier component; Retrieve component association data and attribute data from the component library, and combine them with coordinate system spatial coordinate parameters to set the travel route according to the distribution order of pier components; The connection nodes and surface feature points of the pier components are selected as control points, and the path nodes are sequentially sorted according to spatial coordinates to form a continuous initial travel trajectory.

5. The intelligent path planning method for the bridge pier climbing robot according to claim 4, characterized in that, When correcting the path in real time and driving the robot to execute, this includes: The robot collects data on protrusions, depressions, and external obstacles along its path using obstacle detection equipment, and collects the robot's position coordinates in real time using positioning equipment, which are then updated synchronously according to the travel time to form a real-time data stream. Extract obstacle coordinates and robot position coordinates from the real-time data stream, match the corresponding component coordinates and spatial structure data in the digital twin model, and identify data discrepancies. Adjust the node coordinates of the initial path based on the differences, retain the original path control points, optimize the travel trajectory in the obstacle area, and reorder the travel order of the path nodes; The robot drives the actuators to move sequentially according to the node coordinates of the corrected path, passing through each control point and synchronously providing feedback on the completion status of the actions during the execution process, and recording the real-time status of the path execution.

6. The intelligent path planning method for the bridge pier climbing robot according to claim 5, characterized in that, Record path execution data and establish a path execution database, including: The trajectory data collected during the path execution process includes the robot's actual travel coordinate sequence, node passage time, and trajectory deviation data. The equipment operation data is recorded synchronously, including robot drive mechanism operation parameters, sensor working status data, and actuator action feedback data. The trajectory data is associated and bound with the device operation data and corresponding path nodes according to the execution timestamp; Establish a structured path execution database and set up data indexing and query mechanisms.

7. The intelligent path planning method for a bridge pier climbing robot according to claim 6, characterized in that, The component library is updated iteratively based on path-based data execution, including: Extract trajectory deviation data, equipment operation data, and obstacle data from the path execution database; Based on the extracted data, adjust the component attribute data in the component library, supplement the component surface feature records corresponding to the obstacle area, and update the spatial position relationship in the component association data; Optimize component classification criteria, refine attribute record dimensions, and add attribute fields related to path execution effects; The updated component library data will be used as the basis for subsequent initial path planning.

8. The intelligent path planning method for the bridge pier climbing robot according to claim 7, characterized in that, The determination of data discrepancies and path correction includes: The system includes preset multi-level deviation thresholds, including a first preset component coordinate deviation threshold, a second preset component coordinate deviation threshold, a first preset obstacle distance threshold, a second preset obstacle distance threshold, a first preset robot position deviation threshold, and a second preset robot position deviation threshold, and satisfies that the first preset component coordinate deviation threshold is less than the second preset component coordinate deviation threshold, the first preset obstacle distance threshold is less than the second preset obstacle distance threshold, and the first preset robot position deviation threshold is less than the second preset robot position deviation threshold. Calculate the differences in component coordinates, obstacle distances, and robot positions between real-time data and digital twin model data, respectively; The calculated differences are compared with the corresponding preset graded deviation thresholds to determine the path adjustment requirements: When the component coordinate difference is less than or equal to the upper limit of the first preset component coordinate deviation threshold, the obstacle distance difference is less than or equal to the upper limit of the first preset obstacle distance threshold, and the robot position difference is less than or equal to the upper limit of the first preset robot position deviation threshold, the path is determined to be unchanged and the current trajectory is maintained. When the component coordinate difference is greater than the upper limit of the first preset component coordinate deviation threshold but less than or equal to the upper limit of the second preset component coordinate deviation threshold, the obstacle distance difference is greater than the upper limit of the first preset obstacle distance threshold but less than or equal to the upper limit of the second preset obstacle distance threshold, or the robot position difference is greater than the upper limit of the first preset robot position deviation threshold but less than or equal to the upper limit of the second preset robot position deviation threshold, a data difference point is determined to exist, the path correction process is initiated, and the path node coordinates are adjusted in stages according to the interval where the difference exists. If the component coordinate difference is within the range of the upper limit of the first preset component coordinate deviation threshold and the lower limit of the second preset component coordinate deviation threshold, the obstacle distance difference is within the range of the upper limit of the first preset obstacle distance threshold and the lower limit of the second preset obstacle distance threshold, or the robot position difference is within the range of the upper limit of the first preset robot position deviation threshold and the lower limit of the second preset robot position deviation threshold, perform first-level path correction and adjust the path node coordinate offset to the first offset. If the component coordinate difference is within the second preset component coordinate deviation threshold range, the obstacle distance difference is within the second preset obstacle distance threshold range, or the robot position difference is within the second preset robot position deviation threshold range, perform secondary path correction and adjust the path node coordinate offset to the second offset. The first offset is less than the second offset, and the adjusted path node coordinates must be within the safe operating range of the pier column components.

9. The intelligent path planning method for a bridge pier climbing robot according to claim 8, characterized in that, The evaluation and secondary correction of the path execution effect include: Set graded evaluation indicators and corresponding standard values ​​for path execution performance. Evaluation indicators include node passing accuracy, trajectory deviation range, and equipment operation stability coefficient. Pre-set first preset node passing accuracy standard value, second preset node passing accuracy standard value, third preset node passing accuracy standard value, first preset trajectory deviation range standard value, second preset trajectory deviation range standard value, third preset trajectory deviation range standard value, first preset equipment operation stability coefficient standard value, second preset equipment operation stability coefficient standard value, and third preset equipment operation stability coefficient standard value. The first preset node passing accuracy standard value is greater than the second preset node passing accuracy standard value, which is greater than the third preset node passing accuracy standard value; the first preset trajectory deviation range standard value is less than the second preset trajectory deviation range standard value, which is less than the third preset trajectory deviation range standard value; and the first preset equipment operation stability coefficient standard value is greater than the second preset equipment operation stability coefficient standard value, which is greater than the third preset equipment operation stability coefficient standard value. At the same time, the priority of the evaluation indicators is set, and the priority from high to low is as follows: the priority of equipment operation stability coefficient is higher than the priority of node passing accuracy, which is higher than the priority of trajectory deviation range. When the judgment results of multiple indicators conflict, the judgment result of the indicator with higher priority shall prevail. Extract relevant data from the path execution database and calculate the actual values ​​of node passing accuracy, trajectory deviation range, and equipment operation stability coefficient. The actual values ​​of each evaluation indicator are compared with the corresponding preset standard values ​​to determine the effectiveness of the path execution. When the actual value of the node passing accuracy is greater than or equal to the second preset node passing accuracy standard value, the actual value of the trajectory deviation range is less than or equal to the second preset trajectory deviation range standard value, and the actual value of the equipment operation stability coefficient is greater than or equal to the second preset equipment operation stability coefficient standard value, the path execution is deemed qualified and no secondary correction is required. When the actual value of node passing accuracy is less than the third preset node passing accuracy standard value, the actual value of trajectory deviation range is greater than the third preset trajectory deviation range standard value, or the actual value of equipment operation stability coefficient is less than the third preset equipment operation stability coefficient standard value, the path execution is determined to be unqualified. The cause of the deviation is analyzed and a secondary correction instruction is generated to adjust the path node spacing and execution parameters, and the corrected path is re-executed. When the actual value of the node passing accuracy is greater than or equal to the third preset node passing accuracy standard value and less than or equal to the second preset node passing accuracy standard value, the actual value of the trajectory deviation range is greater than the second preset trajectory deviation range standard value and less than or equal to the third preset trajectory deviation range standard value, or the actual value of the equipment operation stability coefficient is greater than the second preset equipment operation stability coefficient standard value and less than or equal to the third preset equipment operation stability coefficient standard value, the path execution is determined to be basically qualified, a first-level secondary correction instruction is generated, and the path node angle and drive mechanism operation parameters are finely adjusted. When the actual value of the node passing accuracy is greater than or equal to the first preset node passing accuracy standard value, the actual value of the trajectory deviation range is less than or equal to the first preset trajectory deviation range standard value, and the actual value of the equipment operation stability coefficient is greater than or equal to the first preset equipment operation stability coefficient standard value, the path execution is judged to be excellent, the current path parameters are recorded as the optimal reference case, and stored in the reference data directory of the component library.

10. An intelligent path planning system for a bridge pier climbing robot, applied to the intelligent path planning method for a bridge pier climbing robot as described in any one of claims 1-9, characterized in that, include: The module includes a component modeling module, a data acquisition module, a path planning module, a path execution module, and an iterative update module. The component modeling module is used to disassemble bridge pier components, establish a component library, read bridge design data, and construct a digital twin model of the bridge. The data acquisition module is used to acquire various types of data on images, attitude, environment, and obstacles through image acquisition devices, attitude sensors, meteorological acquisition devices, and obstacle detection devices, and to establish a time-series database of the acquired data. The path planning module is used to integrate the collected data with the digital twin model, establish a work coordinate system, plan the initial path, and correct the path based on real-time data. The path execution module includes a drive actuator and a positioning device. The path execution module is used to receive path correction instructions, drive the robot to move along the path, synchronously provide feedback on the completion status of the action, and record trajectory data and device operation data. The iterative update module is used to establish a path execution database, extract trajectory deviation data, equipment operation data, and obstacle data from the path execution database to update the component library, adjust the path planning basis, and perform iterative optimization of the initial path planning. The component modeling module, data acquisition module, path planning module, path execution module, and iterative update module are connected via a communication bus.

Images