Sectional dismantling and hoisting cooperative control method and system for net rack in complex environment

Abstract

The invention relates to the technical field of image processing, and discloses a cooperative control method and system for sectional dismantling and hoisting of a net rack in a complex environment. The method comprises the following steps: collecting net rack data through a multi-source sensor, and performing fusion processing to form a digital model; performing mechanical analysis and segmentation according to the digital model to obtain a segmentation scheme; connecting nodes are recognized according to the segmentation scheme, the dismantling sequence is optimized, and a dismantling path is formed; tracking segmented positions, and planning a hoisting track; robot tasks are coordinated, the working environment is monitored, and a cooperative control instruction is generated; and analyzing the execution data, recording deviation, and forming optimized data. According to the method, collaborative operation of multiple robots is achieved, an operation evaluation and optimization mechanism is established, a comprehensive integrated net rack segmented dismantling and hoisting collaborative control method is provided, and it is ensured that dismantling operation of a net rack structure can be safely and efficiently completed under the complex environment condition.

Description

Technical Field

[0001] The present application relates to the field of image processing technology, and in particular to a method and system for coordinated control of segmented dismantling and hoisting of a grid in a complex environment. Background Art

[0002] With the development of modern building structures, large-scale grid structures, due to their large spans and wide adaptability, have been widely used in large public buildings such as stadiums, exhibition centers, and airport terminals. The demolition and renovation of these grid structures is increasing in the process of urban renewal and building structural upgrades. Traditional grid dismantling methods rely primarily on manual labor, requiring workers to dismantle connection nodes and hoist components in high-altitude environments. With the advancement of computer vision and robotics technologies, some semi-automated dismantling methods have begun to be applied in practical projects, such as structural measurement technology based on 3D laser scanning, robot-assisted connection point dismantling equipment, and remote-controlled hoisting equipment. These technologies have significantly improved the efficiency and safety of grid dismantling and reduced the need for personnel to perform high-altitude operations.

[0003] However, existing technologies still have significant shortcomings in the coordinated control of grid segment dismantling and hoisting in complex environments. First, traditional methods lack the ability to accurately digitally model grid structures, making it difficult to fully capture the grid's geometric characteristics, connection information, and structural status, resulting in unscientific segment planning. Second, existing technologies often rely on empirical judgment to plan dismantling paths, lacking quantitative analysis based on structural stability, making it difficult to ensure the safety of the dismantling process. Third, traditional dismantling and hoisting operations are usually completed independently by different equipment, lacking an effective coordinated control mechanism, resulting in poor coordination between equipment and low efficiency. Finally, existing technologies lack a comprehensive evaluation of the operation process and a mechanism for accumulating experience, making it difficult to achieve continuous technical optimization and knowledge iteration, which is not conducive to the development and advancement of grid dismantling technology in complex environments. Summary of the Invention

[0004] The present application provides a method and system for collaborative control of segmented dismantling and hoisting of a grid structure in a complex environment, which is used to realize multi-robot collaborative operation and establish an operation evaluation and optimization mechanism, and provides a fully integrated method for collaborative control of segmented dismantling and hoisting of a grid structure to ensure that the demolition operation of the grid structure can be completed safely and efficiently under complex environmental conditions.

[0005] In the first aspect, the present application provides a method for coordinated control of grid segment dismantling and hoisting in a complex environment, the method comprising: collecting data of the grid structure through multi-source sensors, fusing the collected data, and obtaining a grid digital model; performing mechanical analysis and segmentation processing on the grid structure based on the grid digital model, and obtaining a grid segmentation scheme; identifying and analyzing the connection nodes based on the grid segmentation scheme, optimizing the dismantling sequence, and obtaining a grid dismantling path; based on the grid dismantling path, tracking the grid segment positions in real time, planning and generating the hoisting trajectory, and obtaining a grid segment hoisting trajectory; allocating and coordinating robot tasks based on the grid dismantling path and the grid segment hoisting trajectory, dynamically monitoring the working environment, and obtaining a demolition and hoisting coordinated control instruction; utilizing the execution data of the coordinated control instruction to compare and analyze the working process, record the deviation between the actual execution and the scheme, and obtain grid dismantling and hoisting optimization data.

[0006] In a second aspect, the present application provides a coordinated control system for segmented dismantling and hoisting of a grid frame in a complex environment, the coordinated control system for segmented dismantling and hoisting of a grid frame in a complex environment comprising: The acquisition module is used to collect data on the grid structure through multi-source sensors, fuse the collected data, and obtain a digital model of the grid; A segmentation module is used to perform mechanical analysis and segmentation processing on the grid structure according to the grid digital model to obtain a grid segmentation scheme; an identification module for identifying and analyzing connection nodes according to the grid segmentation scheme, optimizing and calculating the dismantling sequence, and obtaining a grid dismantling path; A tracking module is used to track the positions of the grid sections in real time based on the grid dismantling path, plan and generate the hoisting trajectory, and obtain the grid section hoisting trajectory; A coordination module is used to allocate and coordinate robot tasks according to the grid dismantling path and the grid segmented lifting trajectory, dynamically monitor the working environment, and obtain dismantling and lifting coordinated control instructions; The comparison module is used to use the execution data of the collaborative control instructions to compare and analyze the operation process, record the deviation between the actual execution and the plan, and obtain the optimization data of the grid dismantling and hoisting.

[0007] In a third aspect, a coordinated control device for segmented dismantling and hoisting of a grid frame in a complex environment is provided, comprising: a memory and at least one processor, wherein the memory stores instructions; the at least one processor calls the instructions in the memory to enable the coordinated control device for segmented dismantling and hoisting of a grid frame in a complex environment to execute the above-mentioned coordinated control method for segmented dismantling and hoisting of a grid frame in a complex environment.

[0008] In a fourth aspect, a computer-readable storage medium is provided, wherein instructions are stored in the computer-readable storage medium, which, when executed on a computer, enables the computer to execute the above-mentioned method for coordinated control of segmented dismantling and installation of a grid frame in a complex environment.

[0009] In the technical solution provided in the present application, data of the grid structure is collected and fused by multi-source sensors to obtain a grid digital model, thereby achieving high-precision digital expression of the grid structure, providing comprehensive and accurate basic data for subsequent analysis and planning, and overcoming the problem of incomplete grid structure information acquisition in traditional methods; based on the grid digital model, the grid structure is mechanically analyzed and segmented to obtain a grid segmentation scheme, making the segmentation planning more scientific, ensuring that each segment meets the structural stability requirements and the load-bearing capacity limitations of the lifting equipment, and avoiding the risk of structural instability caused by traditional empirical segmentation; based on the grid segmentation scheme, the connection nodes are identified and analyzed and the demolition sequence is optimized and calculated to obtain the grid demolition path, making the demolition process safer and more controllable, and significantly reducing structural collapse. risk, while improving the demolition efficiency; based on the grid dismantling path, the grid segment position is tracked in real time and the lifting trajectory is planned to generate the grid segment lifting trajectory, which realizes the precise control of the lifting process, reduces the operation error and improves the lifting accuracy; according to the grid dismantling path and the grid segment lifting trajectory, the robot tasks are allocated and coordinated and the working environment is dynamically monitored to obtain the demolition and lifting collaborative control instructions, which realizes the collaborative operation of the multi-robot system, improves the equipment utilization efficiency, and enhances the system's adaptability to environmental changes; the collaborative control instruction execution data is used to compare and analyze the operation process and record the deviation between the actual execution and the plan to obtain the grid dismantling and lifting optimization data, establishes the experience accumulation and optimization improvement mechanism, and promotes the continuous progress of technology. In particular, in this scheme, artificial intelligence algorithms play a key role in grid structure analysis, demolition path optimization and multi-robot collaborative control: the grid structure is intelligently segmented by combining mechanical analysis with image processing, achieving a balance between structural stability and segmentation suitability; the demolition sequence is optimized by using a reinforcement learning algorithm, which improves demolition efficiency while ensuring structural safety; the multi-agent collaborative decision-making model trained by deep reinforcement learning realizes flexible scheduling and adaptive control of multiple robots in complex environments, greatly improving the system's response to environmental changes and risk response capabilities; based on visual servo control technology, precise positioning and trajectory adjustment are achieved during the grid segment lifting process, bringing the lifting accuracy to the centimeter level. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0011] Figure 1 This is a schematic diagram of an embodiment of a method for coordinated control of segmented dismantling and installation of a grid frame in a complex environment according to an embodiment of the present application; Figure 2 This is a schematic diagram of an embodiment of a coordinated control system for segmented dismantling and installation of a grid frame in a complex environment in an embodiment of the present application; Figure 3 It is a schematic block diagram of the structure of the coordinated control device for segmented dismantling and hoisting of a grid in a complex environment in an embodiment of the present invention. DETAILED DESCRIPTION

[0012] The embodiments of the present application provide a method and system for coordinated control of segmented dismantling and hoisting of a grid in a complex environment. The terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of this application and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "including" or "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or apparatus that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or that are inherent to these processes, methods, products or apparatus.

[0013] For ease of understanding, the specific process of the embodiment of the present application is described below. Figure 1 In an embodiment of the present application, an embodiment of a method for coordinated control of segmented dismantling and installation of a grid frame in a complex environment includes: Step S101: collecting data on the grid structure through multi-source sensors, fusing the collected data, and obtaining a grid digital model; Step S102: Perform mechanical analysis and segmentation processing on the grid structure based on the grid digital model to obtain a grid segmentation scheme; Step S103: Identify and analyze the connection nodes according to the grid segmentation plan, optimize and calculate the demolition sequence, and obtain the grid demolition path; Step S104: Based on the grid dismantling path, the grid segment positions are tracked in real time, and the hoisting trajectory is planned and generated to obtain the grid segment hoisting trajectory; Step S105: Based on the grid dismantling path and the grid segmented hoisting trajectory, the robot tasks are allocated and coordinated, the working environment is dynamically monitored, and dismantling and hoisting coordinated control instructions are obtained; Step S106: Using the execution data of the collaborative control instructions, the operation process is compared and analyzed, and the deviation between the actual execution and the plan is recorded to obtain the optimization data for the dismantling and hoisting of the grid.

[0014] It is understandable that the execution subject of this application can be a coordinated control system for the segmented dismantling and installation of a grid in a complex environment, or a terminal or a server, which is not limited here. The embodiment of this application is described by taking the server as the execution subject as an example.

[0015] In an embodiment of the present application, the implementation process of the method for coordinated control of segmented dismantling and hoisting of a grid in a complex environment first collects data on the grid structure through multi-source sensors, fuses the collected data, and obtains a digital model of the grid. In actual operation, a high-precision laser scanner performs a three-dimensional scan on the surface of the grid structure to obtain the original data of the grid point cloud. These point cloud data contain the spatial geometric information of the grid structure. A high-resolution industrial camera collects images of the grid surface features from different angles to obtain a grid surface feature image set. These image sets record the visual features of the grid surface, such as texture and color changes. An infrared thermal imager performs thermal imaging scanning on the grid structure to identify the stress concentration area of ​​the grid structure and obtain a grid thermal stress distribution map. The thermal stress distribution map reflects the stress conditions of different parts of the grid structure. The raw grid point cloud data is filtered for noise and aligned with coordinates to generate a grid geometric feature point set. Feature extraction and matching are performed on the grid surface feature image set to identify the types and locations of grid structure connection points, generating a grid connection point data table. Threshold segmentation and region marking are performed on the grid thermal stress distribution map to identify weak areas of the grid structure and generate a grid structure risk marker map. Finally, the grid geometric feature point set, grid connection point data table, and grid structure risk marker map are spatially aligned and fused to construct a grid digital model that includes geometric features, connection information, and structural status. Based on the grid digital model, mechanical analysis and segmentation of the grid structure are performed to obtain a grid segmentation scheme. In specific implementation, static load analysis of the grid digital model is performed using finite element calculation methods to calculate the forces at each grid node and obtain grid structure stress distribution data. Image enhancement and edge detection are performed on the grid surface features in the digital model to extract surface texture and defect features, generating a grid surface status map. The grid structure stress distribution data was overlaid and analyzed with the grid surface status map to construct a grid structure health heat map. Based on the grid structure health heat map, the region growing method was used to mark the grid structure segments, determine the critical segment boundaries, and generate preliminary grid segment regions. The center of mass and weight of the preliminary segment regions were calculated, and the suitability of each segment for lifting was evaluated. The segment boundaries were adjusted to generate grid segment boundary data. The topological relationship analysis of the grid segment boundary data was performed, and the demolition sequence of each segment was determined based on the structural support dependencies, thus generating a grid segmentation plan.

[0016] Based on the truss segmentation scheme, connection nodes are identified and analyzed, and the dismantling sequence is optimized and calculated to determine the truss dismantling path. During implementation, the coordinate data of the connection points between truss segments is extracted based on the truss segmentation scheme, and a truss segment connection diagram is established. Nodes in the truss segment connection diagram are classified, and bolted, welded, and plug-in connections are identified to generate a connection point type labeling table. The dismantling difficulty of each connection point type is assessed, and the dismantling complexity of each connection point is calculated based on the connection point type, number, and location, forming a connection point dismantling difficulty index. Based on the connection point dismantling difficulty index, the truss segment dismantling sequence is refined, and the specific dismantling sequence for each connection point is determined, forming a connection point dismantling sequence table. The connection point dismantling sequence table is then verified for structural stability, assessing the impact of each dismantling step on the overall structural balance and generating a structural stability scoring curve. Based on the structural stability scoring curve, the connection point dismantling sequence table is optimized to form a truss dismantling path with a precise operation sequence.

[0017] Based on the truss dismantling path, the positions of truss segments are tracked in real time, and the lifting trajectory is planned and generated to obtain the truss segment lifting trajectory. In practical applications, the visual positioning system collects the spatial positions of truss segments in real time during the dismantling process, generating a truss segment position coordinate stream. This truss segment position coordinate stream is filtered and smoothed to eliminate position jitter interference, resulting in stable truss segment spatial positioning data. Based on the truss segment spatial positioning data, the center of gravity position and attitude angle of each truss segment are calculated to form a truss segment attitude parameter table. Combined with the three-dimensional spatial data of the operating environment, the kinematic characteristics of the lifting equipment are analyzed, the working range constraints of the lifting equipment are determined, and a lifting spatial constraint map is established. Based on the truss segment attitude parameter table and the lifting spatial constraint map, a path planning algorithm is used to generate a collision-free initial lifting trajectory. Feature point tracking technology is used to monitor the dynamic changes of the truss segments during the lifting process, generating dynamic correction data for the truss segments. This dynamic correction data is applied to the real-time adjustment of the initial lifting trajectory to form an adaptive truss segment lifting trajectory.

[0018] Based on the truss dismantling path and the truss segmented lifting trajectory, robot tasks are allocated and coordinated, and the operating environment is dynamically monitored to generate collaborative control instructions for dismantling and lifting. In actual operation, a unified spatiotemporal coordinate framework is established to convert the truss dismantling path and truss segmented lifting trajectory into standardized operation sequences, generating standardized task data. Task decomposition and resource analysis are performed on this standardized task data. Based on the robot equipment capability parameters, the operating boundaries of the dismantling robots and the lifting equipment are divided, generating an initial task allocation plan. This initial task allocation plan is then optimized for timing to eliminate task conflicts and resource competition, generating a collaborative robot schedule. Each operating robot is configured with independent visual perception parameters to monitor the operating environment from multiple angles, generating a real-time data stream of environmental status. Feature extraction and state recognition are performed on this real-time environmental status data stream to detect environmental changes and potential risks, generating environmental risk warning signals. Based on these environmental risk warning signals, the collaborative robot schedule is dynamically adjusted to generate collaborative control instructions for dismantling and lifting with risk mitigation capabilities.

[0019] Utilizing the execution data of collaborative control instructions, the operation process is compared and analyzed, and deviations between actual execution and the plan are recorded to generate optimized data for grid dismantling and hoisting. During the specific implementation process, the control parameters during the execution of collaborative control instructions are sampled in time series to generate instruction execution trajectory data. Image processing technology is used to extract key frames from the operation video recordings, mark important operation nodes, and generate key point data for the actual operation process. The instruction execution trajectory data is compared with the theoretical control instructions, and the control deviation value is calculated to form a control accuracy evaluation table. The key point data of the actual operation process is matched with the planned operation nodes in time and space, and the execution deviation is measured to generate operation quality assessment data. The control accuracy evaluation table and operation quality assessment data are analyzed to identify the main links and causes of deviations and construct a deviation cause correlation diagram. Based on the deviation cause correlation diagram, optimization suggestions are annotated for the grid segmentation plan, dismantling path, and hoisting trajectory to generate optimized data for grid dismantling and hoisting.

[0020] In a specific embodiment, the process of executing step S101 may specifically include the following steps: Use a high-precision laser scanner to perform three-dimensional scanning on the surface of the grid structure to obtain the original data of the grid point cloud; The surface features of the grid are captured from different angles using multiple high-resolution industrial cameras to obtain a set of grid surface feature images. Use an infrared thermal imager to perform thermal imaging scanning on the grid structure, identify the stress concentration area of ​​the grid structure, and obtain the grid thermal stress distribution map; Perform noise filtering and coordinate registration on the original grid point cloud data to generate a grid geometric feature point set; Extract and match the grid surface feature image set, identify the type and position of the grid structure connection points, and obtain the grid connection point data table; Perform threshold segmentation and region marking on the grid thermal stress distribution map to identify weak areas of the grid structure and generate a grid structure risk marking map; The grid geometric feature point set, grid connection point data table and grid structure risk marker map are spatially registered and fused to construct a grid digital model containing geometric features, connection information and structural status.

[0021] Specifically, a high-precision laser scanner performs a three-dimensional scan of the grid structure surface to obtain the grid point cloud raw data. By emitting a laser beam and receiving its reflected signal, the laser scanner measures the distance and angle between the laser and the target surface, thereby obtaining the spatial coordinate data of the grid structure surface. This data is presented in the form of a point cloud, with each point containing a three-dimensional spatial coordinate value and a reflection intensity value. At the same time, multiple high-resolution industrial cameras capture images of the grid surface features from different angles to obtain a set of grid surface feature images. The industrial cameras are arranged at different locations around the grid structure to ensure full coverage of the grid. The captured images contain visual feature information such as the color, texture, and rust condition of the grid surface. This information is an important basis for the subsequent identification of the grid structure status and connection point type.

[0022] In addition, an infrared thermal imager performs thermal imaging scans on the grid structure, identifying areas of stress concentration within the structure and generating a thermal stress distribution map. The infrared thermal imager captures the temperature distribution on the surface of the grid structure. Since stress concentration areas often exhibit different temperature characteristics from the surrounding areas, thermal imaging scans can visually display the stress distribution within the grid structure, providing an important reference for subsequent structural stability analysis. The acquired raw grid point cloud data undergoes noise filtering and coordinate registration to generate a grid geometric feature point set. Noise filtering uses statistical outlier analysis to remove outliers that significantly deviate from the main structure. The coordinate registration criterion uses an iterative nearest point algorithm to unify the point cloud data obtained from multiple scans into the same coordinate system, forming a complete grid geometric feature point set that accurately reflects the spatial geometry of the grid.

[0023] Feature extraction and matching are performed on the grid surface feature image set to identify the type and location of the grid structure connection points and generate a grid connection point data table. During image processing, an edge detection algorithm is applied to extract the grid structure's contour features. Template matching is then used to identify various connection nodes, including bolted, welded, and plug-in connections. Each connection point is classified and labeled, and its spatial coordinates are recorded to form a structured connection point data table. Threshold segmentation and region labeling are performed on the grid thermal stress distribution map to identify weak areas of the grid structure and generate a grid structure risk marker map. The threshold segmentation method extracts temperature anomalies in the thermal stress distribution map and labels and numbers these areas through connected region analysis. This identifies weak areas in the grid structure and generates an intuitive risk marker map, providing a risk assessment basis for subsequent segmentation planning.

[0024] Finally, the grid geometric feature point set, grid connection point data table, and grid structure risk marker map were spatially registered and fused to construct a grid digital model that incorporates geometric features, connection information, and structural status. The spatial registration process employed a feature point matching method to align the three types of data within the same coordinate system. Data fusion, through a multi-layered data structure design, integrated multidimensional information such as geometry, connection information, and structural status into a unified digital model, forming a complete grid digital model. This provided comprehensive data support for subsequent segment planning and demolition path optimization.

[0025] In a specific embodiment, the process of executing step S102 may specifically include the following steps: Use the finite element calculation method to perform static load analysis on the grid digital model, calculate the force value of each node of the grid, and obtain the stress distribution data of the grid structure; Perform image enhancement and edge detection on the grid surface features in the grid digital model, extract the grid surface texture and defect features, and form a grid surface state map; Superimpose and analyze the stress distribution data of the grid structure with the grid surface state map to construct a heat map of the grid structure health; Based on the heat map of grid structure health, the region growing method is used to mark the grid structure partitions, determine the critical boundary lines of the segments, and generate preliminary segmented regions of the grid. Calculate the centroid and weight of the preliminary segmented areas of the grid, evaluate the suitability of each segment for hoisting, adjust the segment boundaries, and form the grid segment boundary data; Perform topological relationship analysis on the boundary data of the grid segments, determine the order of dismantling each segment based on the structural support dependency, and generate a grid segmentation plan.

[0026] Specifically, a static load analysis of the grid digital model was performed using finite element methods. The forces acting on each node were calculated, yielding stress distribution data for the grid structure. Finite element analysis discretizes the grid structure into a finite number of elements, establishing a stiffness matrix for each element. This matrix is ​​then assembled into an overall stiffness matrix to determine the node displacements and internal force distributions under the structure's own weight and external loads. These calculations yield stress and deformation values ​​at each node, providing a visual representation of stress concentrations and weak points within the grid structure. Simultaneously, image enhancement and edge detection were performed on the grid surface features within the digital model to extract surface texture and defect features, creating a grid surface status map. Image enhancement utilizes histogram equalization to enhance image contrast, while edge detection utilizes the Canny algorithm to identify edge information within the image and highlight surface texture features. These image processing techniques effectively identify rusted areas, crack locations, and areas of material deterioration on the grid surface. This information is aggregated into a grid surface status map, visually displaying the surface health of the grid structure.

[0027] The stress distribution data of the grid structure is overlaid with the grid surface condition map to construct a grid structure health heatmap. During the overlay analysis, the stress distribution data and surface condition information are aligned using a spatial registration method. A weighted overlay algorithm is designed to comprehensively consider the impact of stress magnitude and surface condition on structural health, generating comprehensive assessment indicators. These indicators are presented as a heatmap, using a color gradient to represent the structural health of different regions. Red areas indicate weak areas, while green areas indicate areas in good condition. Based on the grid structure health heatmap, a region growing method is used to segment the grid structure, identify critical segment boundaries, and generate preliminary grid segmentation regions. The region growing method selects green areas in the health heatmap as seed points and expands to surrounding areas. The expansion stops when encountering areas with significantly reduced structural health. These stopped expansion boundaries form natural segmentation critical lines. This method divides the grid structure into multiple preliminary segmentation regions, each with relatively consistent structural health within it, separated by boundaries of lower health.

[0028] Centroid calculation and weight estimation are performed on the preliminary truss segment areas to assess the suitability of each segment for lifting, adjust the segment boundaries, and generate truss segment boundary data. Centroid calculation is based on the spatial coordinates and density information of all grid points within each segment to determine the segment's geometric center. Weight estimation is performed by multiplying the material density and volume of each component within the segment to obtain the total weight of the segment. Based on the load capacity of the lifting equipment and the operating radius constraints, the suitability of each segment for lifting is assessed. Boundaries of overweight or irregularly shaped segments are adjusted to ensure that each segment meets the operating requirements of the lifting equipment, thus generating optimized truss segment boundary data.

[0029] Finally, a topological analysis of the grid segment boundary data is performed. Based on the structural support dependencies, the order in which each segment should be dismantled is determined, generating a grid segmentation plan. This topological analysis constructs a support relationship diagram between grid segments, identifying the dependencies between segments and clarifying which segments, as supporting structures, need to be dismantled later and which can be dismantled first. Using a topological sorting algorithm from graph theory, a dismantling sequence chain that meets structural stability requirements is calculated, ensuring the grid structure remains stable throughout the dismantling process and avoiding the risk of structural collapse caused by improper dismantling sequence. Ultimately, a complete grid segmentation plan is formed, including segment boundary definitions and dismantling sequence arrangements.

[0030] In a specific embodiment, the process of executing step S103 may specifically include the following steps: Based on the grid segmentation scheme, the coordinate data of the connection points between the grid segments are extracted to establish a grid segment connection relationship diagram; Perform node classification processing on the grid segment connection relationship diagram, identify bolt connection points, welding connection points and plug-in connection points, and generate a connection point type marking table; Evaluate the difficulty of dismantling each type of connection point. Calculate the complexity of each connection point dismantling based on the type, number, and location of the connection point to form a connection point dismantling difficulty index. Based on the difficulty index of connection point dismantling, the order of dismantling the grid sections is refined, the specific time sequence of dismantling each connection point is determined, and a connection point dismantling time sequence table is formed; Verify the structural stability of the connection point removal schedule, evaluate the impact of each removal step on the overall structural balance, and generate a structural stability scoring curve; According to the structural stability scoring curve, the connection point dismantling schedule is optimized to form a grid dismantling path with a precise operation sequence.

[0031] Specifically, based on the truss segmentation scheme, the coordinate data of the connection points between truss segments is extracted to create a truss segment connection diagram. The connection point coordinate data is extracted from the truss digital model and contains the 3D spatial location of each connection point and the segment number to which it connects. This data is used to construct a truss segment connection diagram. This diagram represents truss segments with nodes and the connections between segments with edges. The edge weights represent the number of connection points, visually presenting the spatial connection topology between segments. Node classification is performed on the truss segment connection diagram to identify bolted, welded, and plug-in connections, generating a connection point type labeling table. Node classification is performed based on the geometric and image features of the connection points. Bolted connections typically exhibit regular rounded heads and hexagonal shapes, welded connections appear as continuous weld lines, and plug-in connections have specific slot and protrusion structures. A feature recognition algorithm is used to classify each connection point, generating a connection point type labeling table containing the connection point number, spatial coordinates, connection type, and segment information.

[0032] An assessment of the difficulty of disassembly is conducted on each type of connection point. Based on the type, number, and location of the connection point, the complexity of disassembly of each connection point is calculated to form an index of the difficulty of disassembly of the connection point. The assessment of the difficulty of disassembly takes into account a variety of factors. The difficulty of bolted connections is related to the specifications, number, and rust of the bolts; the difficulty of welded connections is related to the length, thickness, and quality of the weld; and the difficulty of plug-in connections is related to the tightness and degree of deformation of the plug-in structure. In addition, the spatial position of the connection point affects the difficulty of tool operation. Connection points at high altitudes or in narrow environments are more difficult to disassemble. Taking these factors into consideration, a disassembly complexity value is calculated for each connection point to form an index of the difficulty of disassembly of the connection point, which provides an important reference for the subsequent disassembly sequence planning.

[0033] Based on the connection point removal difficulty index, the truss segment dismantling sequence is refined, determining the specific time sequence for dismantling each connection point and creating a connection point dismantling schedule. This refinement breaks down the truss segment dismantling sequence into specific connection points, prioritizing those with the lowest removal difficulty while also considering the spatial distribution of connection points within the same segment to ensure efficient dismantling. The dismantling schedule also takes into account the time costs of tool changeovers and work platform movement, ultimately generating a dismantling schedule that includes the precise removal time nodes for each connection point.

[0034] The structural stability of the connection point removal schedule was verified, assessing the impact of each removal step on the overall structural balance and generating a structural stability score curve. This verification process employed incremental static analysis to simulate the changes in the mechanical state of the structure after each connection point was removed, calculating stress changes and displacement increments at key nodes to assess structural stability. A structural stability score was calculated for each removal step, generating a score curve that changes with the progress of the removal process. This curve visually illustrates the dynamic changes in structural stability during the removal process and identifies key removal steps where structural stability significantly decreases. Based on the structural stability score curve, the connection point removal schedule was optimized to generate a grid truss removal path with a precise operational sequence. During the optimization process, adjustments were made to the removal steps that showed a significant decrease in stability, either by adjusting the removal sequence or by adding temporary support measures to maintain structural stability. Through multiple iterations of optimization, structural stability remained within a safe range throughout the entire removal process. Ultimately, a complete grid truss removal path was generated, including the precise removal sequence for each connection point, the selection of removal tools, and the temporary support scheme, providing detailed guidance for on-site operations.

[0035] In a specific embodiment, the process of executing step S104 may specifically include the following steps: The spatial position of the grid segments during the dismantling process is collected in real time through the visual positioning system to generate a grid segment position coordinate stream; Perform filtering and smoothing on the grid segment position coordinate stream to eliminate position jitter interference and obtain stable grid segment spatial positioning data; Based on the spatial positioning data of the grid segment, the center of gravity position and attitude angle of each grid segment are calculated to form a grid segment attitude parameter table; Combined with the three-dimensional spatial data of the working environment, the kinematic characteristics of the lifting equipment are analyzed, the working range constraints of the lifting equipment are determined, and the lifting space constraint diagram is established; Based on the grid segment posture parameter table and the lifting space constraint diagram, a path planning algorithm is used to generate a collision-free initial lifting trajectory line; The dynamic changes of the grid segments during the hoisting process are monitored through feature point tracking technology to generate dynamic correction data for the grid segments; The dynamic correction data of the grid segment is applied to the real-time adjustment of the initial hoisting trajectory to form an adaptive grid segment hoisting trajectory.

[0036] Specifically, a visual positioning system collects the spatial positions of truss segments during dismantling in real time, generating a stream of truss segment position coordinates. The visual positioning system, comprised of multiple high-speed cameras, marks feature points on the truss segments and uses multi-view triangulation principles to calculate the spatial coordinates of these feature points, thereby determining the positions of the truss segments. The acquisition frequency is typically 30 Hz, ensuring the real-time nature of the position data. The generated stream of position coordinates contains information on the continuous position changes of the truss segments in three-dimensional space. The stream of truss segment position coordinates is filtered and smoothed to eliminate position jitter interference and generate stable spatial positioning data for the truss segments. This filtering and smoothing process utilizes the Kalman filter algorithm, which combines position prediction and observation update processes to effectively suppress position jitter caused by random noise and measurement errors. Through filtering, the fluctuation amplitude of the position data is significantly reduced, resulting in smooth and continuous spatial positioning data, providing stable and reliable input for subsequent posture calculations.

[0037] Based on the spatial positioning data of the truss segments, the center of gravity position and attitude angle of each truss segment are calculated to form a truss segment attitude parameter table. The center of gravity position is calculated using a weighted average method based on the truss segment's geometry and mass distribution. The attitude angles are calculated by analyzing the spatial distribution of multiple feature points on the truss segment to determine the segment's roll, pitch, and yaw angles, fully describing the segment's attitude in three-dimensional space. These parameters are aggregated into a truss segment attitude parameter table, which records the spatial position and attitude state of each segment and provides a key basis for lifting trajectory planning. Combined with the three-dimensional spatial data of the operating environment, the kinematic characteristics of the lifting equipment are analyzed to determine the working range constraints of the lifting equipment and create a lifting spatial constraint diagram. The analysis considers technical parameters such as the lifting equipment's maximum lifting capacity, boom length, and slew angle range. Furthermore, the obstacle distribution at the work site is considered to calculate the working capacity and spatial accessibility of the lifting equipment at different locations. Through these analyses, the working range constraint boundaries of the lifting equipment are determined, forming a three-dimensional visual lifting space constraint diagram, which intuitively displays the safe operating space during the lifting process.

[0038] Based on the grid segment posture parameter table and the lifting space constraint diagram, a path planning algorithm was used to generate a collision-free initial lifting trajectory. This path planning employed a hybrid potential field-random tree algorithm, which models the lifting space as a potential field, with obstacle areas defined as high potential energy and safe areas as low potential energy. By finding the path with the fastest potential energy drop and expanding the search space with a random tree search method, the optimal path from the starting position to the target position was ultimately generated. The resulting initial lifting trajectory met the collision-free requirements and accounted for the grid segment posture changes and lifting equipment performance constraints. Feature point tracking technology was used to monitor the dynamic changes of the grid segments during the lifting process and generate dynamic correction data for the grid segments. Feature point tracking, based on the optical flow method in computer vision, calculates the displacement of feature points in successive image frames in real time, thereby detecting any slight sway, deformation, or rotation of the grid segments during the lifting process. These dynamic changes significantly impact lifting accuracy, especially in strong winds or high lifting speeds. Timely capture of these changes is crucial for ensuring lifting safety.

[0039] Finally, the dynamic correction data for each grid segment is applied to the real-time adjustment of the initial hoisting trajectory, forming an adaptive grid segment hoisting trajectory. This adjustment process employs a predictive-correction control strategy. Based on the currently monitored dynamic changes in the grid segments, future trends are predicted, and the hoisting trajectory is fine-tuned in advance to suppress segment swing and posture changes. Through this adaptive adjustment mechanism, the hoisting trajectory can respond to segment state changes in real time, ensuring the smoothness and accuracy of the entire hoisting process, and forming a complete grid segment hoisting trajectory that includes position, velocity, and acceleration information.

[0040] In a specific embodiment, the process of executing step S105 may specifically include the following steps: Establish a unified space-time coordinate framework, convert the grid dismantling path and grid segmented lifting trajectory into a standardized operation sequence, and form standardized data for the operation task; Perform task decomposition and resource analysis on standardized task data, divide the operation boundaries of demolition robots and lifting equipment according to the robot equipment capability parameters, and generate an initial task allocation plan; Optimize the timing of the initial task allocation plan to eliminate task conflicts and resource competition, and form a robot collaboration schedule; Configure independent visual perception parameters for each operating robot to monitor the operating environment from multiple angles and generate real-time data streams of environmental status; Perform feature extraction and status recognition on real-time environmental status data streams to detect environmental changes and potential risks, and generate environmental risk warning signals; According to the environmental risk warning signals, the robot collaborative timing table is dynamically adjusted to form demolition and lifting collaborative control instructions with risk response capabilities.

[0041] Specifically, a unified spatiotemporal coordinate framework is established to convert the grid dismantling path and the grid segmented lifting trajectory into standardized operation sequences, generating standardized task data. This unified spatiotemporal coordinate framework uses a global coordinate system to unify the positions and postures of all equipment and work objects within a common reference frame, eliminating differences between local coordinate systems. During the conversion process, connection point operations in the dismantling path and position control instructions in the lifting trajectory are time-synchronized and spatially mapped, forming a standardized data structure containing operation type, target position, execution time, and execution conditions. This allows different types of tasks to be described and scheduled within the same framework. Task decomposition and resource analysis are performed on the standardized task data. Based on the robot equipment capability parameters, the operational boundaries between the dismantling robot and the lifting equipment are delineated, generating an initial task allocation plan. The task decomposition process breaks down complex tasks into basic operational units, such as connection point removal, segmented securing, and lifting movement. Resource analysis assesses each robot's suitability for performing different tasks based on parameters such as its working range, load capacity, and operational accuracy. By matching task requirements with equipment capabilities, the most suitable robot equipment for each operational unit is identified, forming a preliminary task allocation plan that clearly defines the specific work content and responsibilities of each equipment.

[0042] The initial task allocation plan is optimized for timing, eliminating task conflicts and resource contention, and forming a robot collaboration schedule. Timing optimization first identifies inter-task dependencies and constructs a task pre-configuration graph to ensure that the order of task execution satisfies logical dependencies. Resource usage conflicts are then analyzed. When multiple tasks require the same resources, resource contention is resolved by adjusting task start times or inserting wait operations. Finally, spatial interference is considered to prevent collisions when multiple robots operate simultaneously. Through these optimization measures, a robot collaboration schedule with detailed time nodes is generated, providing a precise scheduling basis for coordinated equipment operation. Each working robot is configured with independent visual perception parameters to monitor the working environment from multiple angles and generate a real-time data stream of environmental status. Visual perception parameter configuration, including camera resolution, frame rate, field of view, and sensitivity, is customized based on the working characteristics and monitoring requirements of each robot. Multiple visual sensors are strategically deployed to achieve comprehensive coverage of the working environment, capturing environmental information from multiple angles and generating a rich visual data stream. This data is initially filtered and compressed by edge processing nodes to reduce data transmission burden, preserve critical environmental information, and ultimately converge into a data stream reflecting the real-time status of the working environment.

[0043] Feature extraction and state recognition are performed on the real-time environmental state data stream to detect environmental changes and potential risks, generating environmental risk warning signals. Feature extraction utilizes computer vision technologies such as image segmentation, object detection, and motion analysis to identify key elements in the environment and their state changes. State recognition, based on a pre-defined risk model, assesses the degree of deviation between the current environmental state and safety standards. These analyses enable real-time detection of potential risk factors such as wind speed changes, rain and snow, environmental vibrations, and human intrusion. Warning signals are generated based on the risk level, providing environmental awareness support for collaborative control. Based on the environmental risk warning signals, the robot collaborative sequence is dynamically adjusted to generate risk-responsive collaborative control instructions for dismantling and lifting. This dynamic adjustment mechanism implements appropriate response strategies based on the level and type of the warning signal, such as reducing operating speed, suspending specific tasks, increasing safety margins, or initiating emergency procedures. The adjusted collaborative sequence ensures that robots can complete tasks in a coordinated and safe manner despite environmental changes. The resulting collaborative control instructions contain complete operational parameters and adjustment strategies, guiding the multi-robot system to safely and efficiently complete grid dismantling and lifting operations in complex environments.

[0044] In a specific embodiment, the process of executing step S106 may specifically include the following steps: The control parameters in the execution process of the collaborative control instruction are sampled in time series to form instruction execution trajectory data; Use image processing technology to extract key frames from operation video records, mark important operation nodes, and generate key point data of the actual operation process; Compare the instruction execution trajectory data with the theoretical control instructions, calculate the control deviation value, and form a control accuracy evaluation table; Perform spatiotemporal matching of key point data of the actual operation process with the planned operation nodes, measure execution deviations, and generate operation quality assessment data; Analyze the control accuracy assessment table and operation quality assessment data, identify the main links and causes of deviations, and construct a deviation cause correlation diagram; According to the deviation cause correlation diagram, optimization suggestions are marked for the grid segmentation plan, dismantling path and lifting trajectory to form grid dismantling and lifting optimization data.

[0045] Specifically, the final stage of the coordinated control method for truss segment dismantling and installation in complex environments requires evaluating the quality of the entire operation process and generating optimization data to provide a reference for subsequent similar projects. First, the control parameters during the execution of coordinated control instructions are sampled in time series to generate instruction execution trajectory data. During this process, the robot's position, velocity, torque, and other control parameters are recorded at regular intervals (typically 10-100 milliseconds) and saved as structured data with timestamps. This data details the actual motion trajectory and control state of each robot during the dismantling and installation tasks, providing ground truth for subsequent analysis. Image processing techniques are used to extract key frames from the operation video recordings, marking important operation nodes and generating key point data for the actual operation process. Key frame extraction employs scene change detection methods to identify frames in the video sequence where the content changes significantly. These frames typically correspond to transition points between important operation stages, such as the completion of connection point dismantling, the start of segment movement, and the lifting into place. Image analysis is performed on the extracted key frames to determine their corresponding operation types and execution time points, forming a key point dataset containing time coordinates and operation identifiers. These data intuitively reflect the temporal characteristics of the actual operation process.

[0046] The command execution trajectory data is compared with theoretical control instructions, and control deviations are calculated to form a control accuracy assessment table. During this comparative analysis, the actual control parameters are time-aligned with the original planned instructions. Indicators such as position error, time delay, and response sensitivity are calculated to assess the control system's execution accuracy. This deviation data is categorized and statistically analyzed by robot type, operation type, and time period to form a structured control accuracy assessment table. This table visually displays the control accuracy performance of different robots at different task stages, revealing the stability and reliability of the control system. Key point data from the actual operation process is temporally and spatially matched with the planned operation nodes, and execution deviations are measured to form operation quality assessment data. This temporal and spatial matching uses a nearest neighbor matching algorithm to pair the extracted key points with the key nodes set during the planning phase, calculating temporal and spatial position deviations. This deviation data reflects the overall execution quality of the operation process, including timing accuracy, position accuracy, and operational integrity. After statistical analysis, it forms the operation quality assessment data, providing a direct basis for operation process optimization.

[0047] Control accuracy assessment tables and work quality assessment data were analyzed to identify the main links and causes of deviations and construct a deviation cause correlation diagram. Correlation analysis and anomaly detection methods were used to identify operational links and patterns with significant deviations from the extensive deviation data. By tracing the contextual conditions and environmental factors at these deviation points, the main causes of the deviations, such as equipment performance limitations, environmental interference, and poor planning, were determined. These causal relationships were graphically presented to form a deviation cause correlation diagram, clearly demonstrating the correlations and root causes between different deviations and providing guidance for systematic improvements. Based on the deviation cause correlation diagram, optimization suggestions were identified for the grid truss segmentation plan, dismantling path, and hoisting trajectory, generating optimized data for grid truss dismantling and hoisting. These optimization suggestions address the main identified issues and propose targeted improvement measures, such as adjusting segment boundaries to avoid center of gravity instability, optimizing the sequence of connection point removal to reduce structural stress fluctuations, and modifying the hoisting trajectory to accommodate actual wind conditions. These suggestions were annotated and added to the original plan, indicating priority and expected improvement effects. This resulted in a complete grid truss dismantling and hoisting optimization data set, providing experience accumulation and technical guidance for subsequent similar projects, enabling continuous methodological improvement and knowledge iteration.

[0048] The above describes the method for coordinated control of grid section dismantling and hoisting in a complex environment in the embodiment of the present application. The following describes the coordinated control system of grid section dismantling and hoisting in a complex environment in the embodiment of the present application. Figure 2 In the embodiment of the present application, an embodiment of the coordinated control system for segmented dismantling and installation of a grid frame in a complex environment includes: The acquisition module is used to collect data on the grid structure through multi-source sensors, fuse the collected data, and obtain a digital model of the grid; The segmentation module is used to perform mechanical analysis and segmentation processing on the grid structure based on the grid digital model to obtain a grid segmentation plan; The identification module is used to identify and analyze the connection nodes according to the grid segmentation plan, optimize the demolition sequence, and obtain the grid demolition path; The tracking module is used to track the position of the grid segments in real time based on the grid dismantling path, plan and generate the lifting trajectory, and obtain the grid segment lifting trajectory; The coordination module is used to allocate and coordinate robot tasks based on the grid dismantling path and the grid segmented lifting trajectory, dynamically monitor the working environment, and obtain dismantling and lifting coordinated control instructions; The comparison module is used to use the execution data of the collaborative control instructions to compare and analyze the operation process, record the deviation between the actual execution and the plan, and obtain the optimization data for the dismantling and lifting of the grid.

[0049] Through the coordinated cooperation of the above-mentioned components, data of the grid structure is collected and fused by multi-source sensors to obtain a grid digital model, which realizes high-precision digital expression of the grid structure, provides comprehensive and accurate basic data for subsequent analysis and planning, and overcomes the problem of incomplete grid structure information acquisition in traditional methods; according to the grid digital model, the grid structure is mechanically analyzed and segmented to obtain a grid segmentation scheme, making the segmentation planning more scientific, ensuring that each segment meets the structural stability requirements and the load-bearing capacity limitations of the lifting equipment, and avoiding the risk of structural instability caused by traditional empirical segmentation; according to the grid segmentation scheme, the connection nodes are identified and analyzed and the demolition sequence is optimized and calculated to obtain the grid demolition path, making the demolition process safer and more controllable, and significantly reducing the structural The system can reduce the risk of collapse and improve the demolition efficiency; based on the grid dismantling path, the grid segment position is tracked in real time and the lifting trajectory is planned to generate the grid segment lifting trajectory, which realizes the precise control of the lifting process, reduces the operation error and improves the lifting accuracy; according to the grid dismantling path and the grid segment lifting trajectory, the robot tasks are allocated and coordinated and the working environment is dynamically monitored to obtain the demolition and lifting collaborative control instructions, which realizes the collaborative operation of the multi-robot system, improves the equipment utilization efficiency, and enhances the system's adaptability to environmental changes; the collaborative control instruction execution data is used to compare and analyze the operation process and record the deviation between the actual execution and the plan to obtain the grid dismantling and lifting optimization data, establish an experience accumulation and optimization improvement mechanism, and promote the continuous progress of technology. In particular, in this scheme, artificial intelligence algorithms play a key role in grid structure analysis, demolition path optimization and multi-robot collaborative control: the grid structure is intelligently segmented by combining mechanical analysis with image processing, achieving a balance between structural stability and segmentation suitability; the demolition sequence is optimized by using a reinforcement learning algorithm, which improves demolition efficiency while ensuring structural safety; the multi-agent collaborative decision-making model trained by deep reinforcement learning realizes flexible scheduling and adaptive control of multiple robots in complex environments, greatly improving the system's response to environmental changes and risk response capabilities; based on visual servo control technology, precise positioning and trajectory adjustment are achieved during the grid segment lifting process, bringing the lifting accuracy to the centimeter level.

[0050] above Figure 2 From the perspective of modular functional entities, the grid section dismantling and hoisting collaborative control system in a complex environment in an embodiment of the present invention is described in detail. Below, from the perspective of hardware processing, the grid section dismantling and hoisting collaborative control equipment in a complex environment in an embodiment of the present invention is described in detail.

[0051] Figure 3This is a schematic diagram of the structure of a coordinated control device for grid segment dismantling and installation in a complex environment, provided by an embodiment of the present invention. The coordinated control device 300 for grid segment dismantling and installation in a complex environment can vary significantly due to different configurations or performance. The device may include one or more central processing units (CPUs) 310 (e.g., one or more processors), a memory 320, and one or more storage media 330 (e.g., one or more mass storage devices) storing application programs 333 or data 332. The memory 320 and storage media 330 may be either transient or persistent storage. The program stored in the storage medium 330 may include one or more modules (not shown), each of which may include a series of instructions and operations within the coordinated control device 300 for grid segment dismantling and installation in a complex environment. Furthermore, the processor 310 may be configured to communicate with the storage medium 330, allowing the coordinated control device 300 to execute the series of instructions and operations stored in the storage medium 330, thereby implementing the steps of the coordinated control method for grid segment dismantling and installation in a complex environment.

[0052] The grid segment dismantling and hoisting coordinated control device 300 under complex environment may also include one or more power supplies 340, one or more wired or wireless network interfaces 350, one or more input and output interfaces 360, and / or one or more operating systems 331, such as Windows Server, Mac OS X, Unix, Linux, FreeBSD, etc. It will be understood by those skilled in the art that Figure 3 The structure of the coordinated control equipment for segmented dismantling and hoisting of the grid in a complex environment shown does not constitute a limitation of the coordinated control equipment for segmented dismantling and hoisting of the grid in a complex environment provided by the present invention, and may include more or fewer components than shown in the figure, or a combination of certain components, or a different arrangement of components.

[0053] The present invention also provides a computer-readable storage medium, which may be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium. The computer-readable storage medium stores instructions. When the instructions are executed on a computer, the computer executes the steps of the method for coordinated control of segmented dismantling and hoisting of a grid frame in a complex environment.

[0054] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described systems, systems and units can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0055] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the existing technology, or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a grid segment dismantling and hoisting collaborative control device in a complex environment (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk, and other media that can store program code.

[0056] As described above, the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that the technical solutions described in the above embodiments can still be modified, or some of the technical features thereof can be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for coordinated control of truss segment dismantling and hoisting in a complex environment, characterized in that: The method comprises: The grid structure is collected through multi-source sensors, and the collected data is fused and processed to obtain a grid digital model; According to the grid digital model, mechanical analysis and segmentation processing are performed on the grid structure to obtain a grid segmentation scheme; According to the grid segmentation scheme, the connection nodes are identified and analyzed, and the demolition sequence is optimized and calculated to obtain the grid demolition path; Based on the grid dismantling path, the grid segment positions are tracked in real time, and the hoisting trajectory is planned and generated to obtain the grid segment hoisting trajectory; According to the grid dismantling path and the grid segmented hoisting trajectory, the robot tasks are allocated and coordinated, the working environment is dynamically monitored, and the dismantling and hoisting coordinated control instructions are obtained; The execution data of the collaborative control instructions is used to compare and analyze the operation process, record the deviation between the actual execution and the plan, and obtain the optimization data for the dismantling and lifting of the grid frame.

2. The method for coordinated control of segmented dismantling and installation of a grid frame in a complex environment according to claim 1 is characterized in that: The method of collecting data of the grid structure by using multi-source sensors and fusing the collected data to obtain a grid digital model includes: Use a high-precision laser scanner to perform three-dimensional scanning on the surface of the grid structure to obtain the original data of the grid point cloud; The surface features of the grid are captured from different angles using multiple high-resolution industrial cameras to obtain a set of grid surface feature images. Use an infrared thermal imager to perform thermal imaging scanning on the grid structure, identify the stress concentration area of ​​the grid structure, and obtain the grid thermal stress distribution map; Performing noise filtering and coordinate registration on the raw grid point cloud data to generate a grid geometric feature point set; Extracting and matching features of the grid surface feature image set, identifying the types and positions of grid structure connection points, and obtaining a grid connection point data table; Performing threshold segmentation and region marking on the grid thermal stress distribution map to determine weak areas of the grid structure and generate a grid structure risk marking map; The grid geometric feature point set, the grid connection point data table and the grid structure risk marker map are spatially aligned and data fused to construct a grid digital model containing geometric features, connection information and structural status.

3. The method for coordinated control of truss segmented dismantling and hoisting in complex environments according to claim 1 is characterized in that: The grid structure is subjected to mechanical analysis and segmentation processing based on the grid digital model to obtain a grid segmentation scheme, including: Using the finite element calculation method to perform static load analysis on the grid digital model, calculate the force value of each node of the grid, and obtain the stress distribution data of the grid structure; Performing image enhancement and edge detection on the grid surface features in the grid digital model, extracting the grid surface texture and defect features, and forming a grid surface state map; Overlaying and analyzing the grid structure stress distribution data with the grid surface state diagram to construct a grid structure health heat map; Based on the grid structure health heat map, the grid structure is partitioned and marked using a region growing method to determine the critical boundary lines of the segments and generate preliminary segmented regions of the grid; Calculating the centroid and weight of the preliminary segmented areas of the grid, evaluating the suitability of each segment for hoisting, adjusting the segment boundaries, and forming grid segment boundary data; A topological relationship analysis is performed on the boundary data of the grid segment, and the order of dismantling each segment is determined according to the structural support dependency relationship to generate a grid segmentation plan.

4. The method for coordinated control of truss segmented dismantling and hoisting in complex environments according to claim 1 is characterized in that: The method of identifying and analyzing the connection nodes according to the grid segmentation scheme and optimizing the demolition sequence to obtain the grid demolition path includes: Based on the grid segmentation scheme, extracting the coordinate data of the connection points between the grid segments and establishing a grid segment connection relationship diagram; Performing node classification processing on the grid segment connection relationship diagram, identifying bolt connection points, welding connection points and plug connection points, and generating a connection point type marking table; Evaluate the difficulty of dismantling each type of connection point. Calculate the complexity of each connection point dismantling based on the type, number, and location of the connection point to form a connection point dismantling difficulty index. Based on the connection point removal difficulty index, the grid frame segment removal sequence is refined, the specific time sequence for removing each connection point is determined, and a connection point removal time sequence table is formed; Verifying the structural stability of the connection point removal schedule, evaluating the impact of each removal step on the overall structural balance, and generating a structural stability scoring curve; According to the structural stability scoring curve, the connection point dismantling time sequence table is optimized to form a grid dismantling path including a precise operation sequence.

5. The method for coordinated control of truss segmented dismantling and hoisting in complex environments according to claim 1 is characterized in that: The method of tracking the grid segment positions in real time based on the grid dismantling path and planning and generating the hoisting trajectory to obtain the grid segment hoisting trajectory includes: The spatial position of the grid segments during the dismantling process is collected in real time through the visual positioning system to generate a grid segment position coordinate stream; Performing filtering and smoothing processing on the grid segment position coordinate stream to eliminate position jitter interference and obtain stable grid segment spatial positioning data; Based on the spatial positioning data of the grid segments, the center of gravity position and attitude angle of each grid segment are calculated to form a grid segment attitude parameter table; Combined with the three-dimensional spatial data of the working environment, the kinematic characteristics of the lifting equipment are analyzed, the working range constraints of the lifting equipment are determined, and the lifting space constraint diagram is established; Based on the grid segment posture parameter table and the hoisting space constraint diagram, a path planning algorithm is used to generate a collision-free initial hoisting trajectory line; The dynamic changes of the grid segments during the hoisting process are monitored through feature point tracking technology to generate dynamic correction data for the grid segments; The grid segment dynamic correction data is applied to the real-time adjustment of the initial hoisting trajectory to form an adaptive grid segment hoisting trajectory.

6. The method for coordinated control of truss segmented dismantling and hoisting in complex environments according to claim 1 is characterized in that: The method allocates and coordinates robot tasks according to the grid dismantling path and the grid segmented hoisting trajectory, dynamically monitors the working environment, and obtains dismantling and hoisting coordinated control instructions, including: Establishing a unified space-time coordinate framework, converting the grid dismantling path and the grid segmented hoisting trajectory into a standardized operation sequence to form standardized operation task data; Performing task decomposition and resource analysis on the standardized data of the operation tasks, dividing the operation boundaries of the demolition robot and the lifting equipment according to the robot equipment capability parameters, and generating an initial task allocation plan; Optimizing the timing of the initial task allocation plan to eliminate task conflicts and resource competition, and forming a robot collaboration timing table; Configure independent visual perception parameters for each operating robot to monitor the operating environment from multiple angles and generate real-time data streams of environmental status; Perform feature extraction and status recognition on the real-time data stream of the environmental status, detect environmental changes and potential risks, and generate environmental risk warning signals; According to the environmental risk warning signal, the robot collaborative timing table is dynamically adjusted to form a demolition and lifting collaborative control instruction with risk response capability.

7. The method for coordinated control of segmented dismantling and installation of a grid frame in a complex environment according to claim 1 is characterized in that: The collaborative control instruction execution data is used to compare and analyze the operation process, record the deviation between the actual execution and the plan, and obtain the grid frame dismantling and hoisting optimization data, including: Sampling the control parameters in the execution process of the collaborative control instruction in a time series to form instruction execution trajectory data; Use image processing technology to extract key frames from operation video records, mark important operation nodes, and generate key point data of the actual operation process; Comparing the instruction execution trajectory data with the theoretical control instructions, calculating the control deviation value, and forming a control accuracy evaluation table; Performing spatiotemporal matching of the key point data of the actual operation process with the planned operation nodes, measuring execution deviations, and forming operation quality assessment data; Analyze the control accuracy evaluation table and the operation quality evaluation data, identify the main links and causes of deviations, and construct a deviation cause correlation diagram; According to the deviation cause association diagram, optimization suggestions are marked for the grid segmentation plan, dismantling path and lifting trajectory to form grid dismantling and lifting optimization data.

8. A coordinated control system for the segmented dismantling and installation of a grid in a complex environment, characterized by: The method for realizing coordinated control of segmented dismantling and hoisting of a grid frame in a complex environment according to any one of claims 1 to 7, wherein the coordinated control system for segmented dismantling and hoisting of a grid frame in a complex environment comprises: The acquisition module is used to collect data on the grid structure through multi-source sensors, fuse the collected data, and obtain a digital model of the grid; A segmentation module is used to perform mechanical analysis and segmentation processing on the grid structure according to the grid digital model to obtain a grid segmentation scheme; an identification module for identifying and analyzing connection nodes according to the grid segmentation scheme, optimizing and calculating the dismantling sequence, and obtaining a grid dismantling path; A tracking module is used to track the positions of the grid sections in real time based on the grid dismantling path, plan and generate the hoisting trajectory, and obtain the grid section hoisting trajectory; A coordination module is used to allocate and coordinate robot tasks according to the grid dismantling path and the grid segmented lifting trajectory, dynamically monitor the working environment, and obtain dismantling and lifting coordinated control instructions; The comparison module is used to use the execution data of the collaborative control instructions to compare and analyze the operation process, record the deviation between the actual execution and the plan, and obtain the optimization data of the grid dismantling and hoisting.

9. A coordinated control device for the segmented dismantling and hoisting of a grid in a complex environment, characterized in that: It includes a memory and a processor, the memory stores a computer program that can be run on the processor, and when the processor executes the computer program, it implements the coordinated control method of segmented dismantling and lifting of a grid frame in a complex environment under any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the processor executes the method for coordinated control of segmented dismantling and installation of a grid in a complex environment according to any one of claims 1 to 7.

Images