Spacer mounting robot communication control system

Abstract

The application provides a spacer mounting robot communication control system, which comprises the following steps: preloading a preset path of an execution mechanism and line environment parameters through a ground end communication module, combining satellite remote sensing data to predict the shielding risk of a work area to generate a corresponding scene risk map; collecting actual control instructions and running state data of the execution mechanism through a sky end communication module, synchronously generating a target control strategy adapted to the execution mechanism according to the actual control instructions and the running state data based on the scene risk map; issuing the target control strategy to the execution mechanism to enable the execution mechanism to complete corresponding execution actions; when the shielding risk of the work area reaches a preset risk threshold or the signal strength attenuation exceeds a preset attenuation threshold, triggering a dual-mode link mode in advance to build a stable connection between the sky end communication module and the ground end communication module. The application can realize stable communication of the robot and improve the control efficiency.

Description

Technical Field

[0001] This invention relates to the field of power transmission and transformation equipment technology, and in particular to a communication control system for a spacer bar installation robot. Background Technology

[0002] Spacer bars are tools used in high-voltage transmission lines to fix the spacing between split conductors and suppress conductor vibration and galloping. They are available in two-split, four-split, six-split, and eight-split configurations.

[0003] There are two ways to install spacers: one is by manual installation, and the other is by spacer installation robot. The spacer installation robot can be automatically controlled. In automatic control mode, the spacer installation robot calculates the mileage based on the data fed back by the IMU (Inertial Measurement Unit) and the motor encoder, and completes the automatic installation of the spacer after the robot reaches the designated position.

[0004] However, sometimes it is necessary to manually control the spacer installation robot. The spacer installation robot on the high-voltage transmission line is far away from the ground personnel, ranging from tens of meters to nearly 200 meters. Communication is prone to interruption, making it difficult to achieve communication between the spacer installation robot and the ground personnel. Summary of the Invention

[0005] Based on this, the purpose of the present invention is to provide a communication control system for a spacer bar installation robot, so as to solve the problem of communication interruption that easily occurs in the process of controlling the spacer bar installation robot in the prior art.

[0006] The first aspect of the present invention proposes: A communication control system for a spacer bar installation robot includes an actuator, a sky-end communication module, and a ground-end communication module. The system comprises: The ground-based communication module preloads the preset path and line environment parameters of the actuator, and combines them with satellite remote sensing data to predict the occlusion risk of the work area, so as to generate a corresponding scene risk map. The actual control commands and operating status data of the actuator are collected through the sky-end communication module. Simultaneously, based on the scenario risk map, a target control strategy adapted to the actuator is generated according to the actual control commands and operating status data. The target control strategy is sent to the execution mechanism so that the execution mechanism can complete the corresponding execution action; The aerial communication module monitors changes in the operating environment of the actuator. When the obstruction risk in the operating area reaches a preset risk threshold or the signal strength attenuation exceeds a preset attenuation threshold, the dual-mode link mode is triggered in advance to establish a stable connection between the aerial communication module and the ground communication module, and the corresponding execution actions are completed synchronously.

[0007] The beneficial effects of this invention are as follows: This technical solution preloads the preset path and line environment parameters of the actuator on the ground end, and accurately predicts the occlusion risk of the operation area by combining satellite remote sensing data and generates a scene risk map. This enables the sky end to generate an adaptive target control strategy based on actual control commands and operating status data, ensuring the accurate execution of the actuator's actions. At the same time, by triggering the dual-mode link mode in advance, it actively avoids communication interruption problems caused by occlusion or signal attenuation, builds a stable communication connection, and ensures that the actuator's operation is continuous and uninterrupted, effectively improving the operational reliability and communication stability of the spacer bar installation robot.

[0008] Furthermore, the step of combining satellite remote sensing data to predict the occlusion risk of the work area and generate a corresponding scene risk map includes: Collect satellite remote sensing images of the work area for multiple consecutive time periods, and combine the path coordinates of the preset path with the line environmental parameters to crop out a temporal image subset along the preset range of the path; A time-series trend fitting algorithm is used to extract dynamic features from the time-series image subset, and abnormal pixel regions in the satellite remote sensing images are identified simultaneously through standard deviation analysis to generate a corresponding dynamic feature dataset. Based on the dynamic feature dataset, each risk source in the preset path is parsed out. Simultaneously, a rasterized risk map framework is constructed based on the planar coordinates of the work area. The scene risk map is then generated according to the rasterized risk map framework and each of the risk sources.

[0009] Furthermore, the step of parsing out each risk source in the preset path based on the dynamic feature dataset includes: Based on the temporal variation features in the dynamic feature dataset and the spectral features of the abnormal pixel region, a corresponding inverse mapping model of dynamic features and physical attributes is constructed by combining the elevation data and texture data of the satellite remote sensing image. The physical parameters of the entities corresponding to each feature in the preset path are parsed through the reverse mapping model, and the candidate risk source attribute set is screened in conjunction with the line environment parameters. The Monte Carlo simulation algorithm is used to calculate the obstruction probability of each candidate risk source in the candidate risk source attribute set relative to the communication link, and each risk source is determined synchronously according to the magnitude of the obstruction probability.

[0010] Furthermore, the step of constructing a rasterized risk map framework based on the planar coordinates of the work area, and generating the scene risk map according to the rasterized risk map framework and each of the risk sources, includes: Based on the degree of influence of each risk source relative to the communication link, each risk source is divided into a core layer, an influence layer and an outer layer, and the risk radiation intensity corresponding to the risk source in the core layer is calculated simultaneously. Substitute the risk radiation intensity and the type and risk level of each risk source into the grid frame, and simultaneously calculate the comprehensive risk value of each grid cell to generate the corresponding initial risk map. The signal transmission thresholds and dual-mode link triggering conditions of the sky-end communication module and the ground-end communication module are fused into the internal structure of the initial risk map to generate the corresponding scenario risk map.

[0011] Furthermore, the step of generating a target control strategy adapted to the actuator based on the scenario risk map, the actual control commands, and the operating status data includes: Risk features from the scenario risk map, instruction features from the actual control instructions, and state features of the actuator are extracted and synchronously fused into corresponding fusion feature vectors. Using the adaptation requirements of the actuator as a constraint, a corresponding objective function is created by combining the fused feature vector, and an initial control strategy is generated simultaneously by solving the problem through a robust algorithm. The initial control strategy is subjected to anti-interference testing in order to generate the target control strategy accordingly.

[0012] Furthermore, the step of creating a corresponding objective function by combining the fused feature vector with the adaptation requirements of the actuator as a constraint, and simultaneously solving for and generating the initial control strategy through a robust algorithm, includes: The core components of the fused feature vector are extracted using a sparse autoencoder network, and the adaptation requirements of the actuator are simultaneously decomposed into corresponding elastic constraint intervals. The mapping relationship between the core components and the elastic constraint interval is established by using a kernel function to create a corresponding constraint-feature mapping table. Simultaneously, the objective function is created based on the constraint-feature mapping table. The whale optimization algorithm is selected, and the objective function is iteratively solved based on the fluctuation amplitude of the core component to extract the optimal solution. The optimal solution is then back-verified through the constraint-feature mapping table and combined and mapped to the initial control strategy.

[0013] Furthermore, the step of performing anti-interference testing on the initial control strategy to generate the target control strategy includes: Based on the risk characteristics of the scenario risk map and historical operation interference data, a corresponding interference prediction model is constructed through transfer learning algorithm to output the interference risk warning level for each operation stage. Based on the interference risk warning level and the fused feature vector, a multimodal anti-interference strategy library is generated through fuzzy decision matching; The initial control strategy is modified using the multimodal interference strategy library to generate the target control strategy accordingly.

[0014] Furthermore, the step of pre-triggering the dual-mode link mode to establish a stable connection between the sky-end communication module and the ground-end communication module, and synchronously completing the corresponding execution actions, includes: Based on the monitoring data collected by the sky-end communication module and combined with the scene risk map, the characteristics and movement trajectory of the obstruction are identified to generate corresponding link early warning signals and dual-mode link activation sequence. In response to the link warning signal, an independent redundant channel is allocated to the dual-mode link, and corresponding adjustment execution parameters are generated synchronously. The adjusted execution parameters are sent to the execution mechanism to synchronously complete the corresponding execution actions.

[0015] Furthermore, the step of allocating independent redundant channels to the dual-mode link and synchronously generating corresponding adjustment execution parameters includes: Based on the characteristics of the obstruction, the movement trajectory, and the scene risk map, the corresponding obstruction scene level is divided, and the corresponding independent redundant channel is allocated synchronously according to the obstruction scene level. The actual transmission characteristics of the independent redundant channels are collected, and initial execution parameters are generated by combining them with the job task type of the actuator. The initial execution parameters are validated to generate the corresponding adjusted execution parameters.

[0016] Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0017] Figure 1 This is a structural block diagram of a communication control system for a spacer bar installation robot provided in an embodiment of the present invention; Figure 2 This is a three-dimensional structural diagram of a spacer provided in an embodiment of the present invention; Figure 3A three-dimensional structural schematic diagram of a four-split spacer rod installation robot provided in an embodiment of the present invention; Figure 4 This is a three-dimensional structural diagram of the spacer bar clamping mechanism and the cable adjustment mechanism provided in an embodiment of the present invention; Figure 5 This is a three-dimensional structural diagram of the robotic arm of a spacer bar installation robot according to an embodiment of the present invention.

[0018] The following detailed description, in conjunction with the accompanying drawings, will further illustrate the present invention. Detailed Implementation

[0019] To facilitate understanding of the present invention, a more complete description will be given below with reference to the accompanying drawings. Several embodiments of the invention are illustrated in the drawings. However, the invention can be implemented in many different forms and is not limited to the embodiments described herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete.

[0020] It should be noted that when a component is said to be "fixed to" another component, it can be directly on the other component or there may be an intervening component. When a component is said to be "connected to" another component, it can be directly connected to the other component or there may be an intervening component. The terms "vertical," "horizontal," "left," "right," and similar expressions used in this document are for illustrative purposes only.

[0021] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. The terminology used herein in the description of the invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0022] Please see Figure 1 The image shows a communication control system for a spacer bar installation robot provided in an embodiment of the present invention. The communication control system for a spacer bar installation robot provided in this embodiment enables the robot to remain in a stable control state, thereby improving the robot's control efficiency.

[0023] Specifically, this embodiment provides: A communication control system for a spacer bar installation robot includes an actuator, a sky-end communication module, and a ground-end communication module. The system comprises: The ground-based communication module preloads the preset path and line environment parameters of the actuator, and combines them with satellite remote sensing data to predict the occlusion risk of the work area, so as to generate a corresponding scene risk map. It should be noted that, firstly, to address the issue of unknown communication risks before high-altitude operations, the ground-based communication module preloads the actuator's preset path (such as the direction of the transmission line conductor and the coordinates of the spacer installation points) and line environmental parameters (such as line altitude, surrounding topography, and vegetation distribution). Simultaneously, it combines satellite remote sensing data (such as high-resolution optical images and radar images) to predict the occlusion risks in the work area. Specifically, satellite remote sensing data can cover a wide area of ​​the work scene, accurately capturing the location and shape of potential obstructions (such as tall trees, mountains, and temporary buildings). By analyzing the relative positions of obstructions and the communication link (sky-to-ground), a scene risk map is generated. This map visually presents the distribution of communication risks in the work area, providing a risk constraint basis for the generation of subsequent control strategies and avoiding the shortcomings of traditional operations that involve "blind execution and passive response after communication interruption."

[0024] The actual control commands and operating status data of the actuator are collected through the sky-end communication module. Simultaneously, based on the scenario risk map, a target control strategy adapted to the actuator is generated according to the actual control commands and operating status data. It should be noted that, secondly, the aerial communication module is mounted on the robot itself or its accompanying drone, and collects in real time the actual control commands (such as gripping commands, tightening commands, and displacement commands) and operational status data (such as robot posture, gripping force, and motor speed) of the actuator. Based on the risk distribution in the scenario risk map, the module performs correlation analysis between command characteristics and status data (such as reducing command transmission frequency and improving status data sampling accuracy in high-risk occlusion areas) to generate a target control strategy adapted to the actuator. Specifically, the strategy adaptability design ensures that under different risk scenarios, the control commands can meet the operational accuracy requirements and adapt to the transmission capabilities of the communication link, avoiding operational errors caused by command loss or delay.

[0025] The target control strategy is sent to the execution mechanism so that the execution mechanism can complete the corresponding execution action; It should be noted that, next, the target control strategy is sent to the actuator through a stable communication link, driving the actuator to complete actions such as grasping, positioning, and installing the spacer, thereby achieving precise execution of the task.

[0026] The aerial communication module monitors changes in the operating environment of the actuator. When the obstruction risk in the operating area reaches a preset risk threshold or the signal strength attenuation exceeds a preset attenuation threshold, the dual-mode link mode is triggered in advance to establish a stable connection between the aerial communication module and the ground communication module, and the corresponding execution actions are completed synchronously.

[0027] It should be noted that, finally, regarding the dynamic changes in occlusion risk during high-altitude operations, the sky-side communication module monitors changes in the operational scenario in real time (such as trees swaying and causing increased occlusion, or changes in drone flight attitude causing signal attenuation). A dual trigger threshold is set: occlusion risk reaches a preset threshold (e.g., obstruction covering the communication link ≥ 50%) or signal strength attenuation exceeds a preset attenuation threshold (e.g., signal strength decrease ≥ 30dB). When either threshold is met, the dual-mode link is triggered in advance (e.g., the main link is 4G / 5G, and the backup link is microwave or laser communication), establishing a redundant and stable connection between the sky and ground ends. This ensures continuous transmission of control commands and status data, and synchronous execution of actions. Specifically, triggering the dual-mode link in advance avoids the lag of "switching after communication interruption," ensuring the continuity and safety of the operation.

[0028] Furthermore, the step of combining satellite remote sensing data to predict the occlusion risk of the work area and generate a corresponding scene risk map includes: Collect satellite remote sensing images of the work area for multiple consecutive time periods, and combine the path coordinates of the preset path with the line environmental parameters to crop out a temporal image subset along the preset range of the path; A time-series trend fitting algorithm is used to extract dynamic features from the time-series image subset, and abnormal pixel regions in the satellite remote sensing images are identified simultaneously through standard deviation analysis to generate a corresponding dynamic feature dataset. Based on the dynamic feature dataset, each risk source in the preset path is parsed out. Simultaneously, a rasterized risk map framework is constructed based on the planar coordinates of the work area. The scene risk map is then generated according to the rasterized risk map framework and each of the risk sources.

[0029] It should be noted that, firstly, in order to capture the dynamic changes in the risk of occlusion in the work area, satellite remote sensing images of the work area for multiple consecutive time periods are collected (such as daily images of the past month, covering different weather and vegetation growth status). Combined with the coordinate information of the preset path and the line environmental parameters, a time-series image subset along the preset range (such as 50 meters on both sides of the conductor) is cropped. Specifically, the time-series image subset focuses on the core area of ​​the work, removes irrelevant background information, and improves the efficiency and accuracy of subsequent feature extraction.

[0030] It should be noted that, secondly, time-series trend fitting algorithms (such as linear fitting and polynomial fitting) are used to extract dynamic features (such as changes in tree height and the construction and demolition of temporary buildings) from time-series image subsets. These dynamic features directly reflect the changing trends of occlusion risk. Simultaneously, abnormal pixel regions in satellite remote sensing images (such as regions where pixel gray values ​​deviate from the normal range, corresponding to tall trees, mountain shadows, and other occlusions) are identified through standard deviation analysis. The dynamic features and the features of abnormal pixel regions are integrated to generate a dynamic feature dataset. Specifically, this dataset contains both static location information of occlusions and dynamic change patterns, providing data support for accurate analysis of risk sources.

[0031] Finally, based on the dynamic feature dataset, each risk source in the preset path is analyzed (such as determining the type, location, height, and occlusion range of the obstruction). Using the planar coordinates of the work area as the basis, a gridded risk map framework is constructed (such as dividing the work area into 10m × 10m grid units). The information of each risk source is mapped to the corresponding grid unit, and the risk level of the grid is marked (such as high risk, medium risk, and low risk). Finally, a scene risk map is generated. Specifically, the gridded presentation makes the risk distribution intuitive and controllable, which facilitates the targeted formulation of subsequent control strategies.

[0032] Furthermore, the step of parsing out each risk source in the preset path based on the dynamic feature dataset includes: Based on the temporal variation features in the dynamic feature dataset and the spectral features of the abnormal pixel region, a corresponding inverse mapping model of dynamic features and physical attributes is constructed by combining the elevation data and texture data of the satellite remote sensing image. The physical parameters of the entities corresponding to each feature in the preset path are parsed through the reverse mapping model, and the candidate risk source attribute set is screened in conjunction with the line environment parameters. The Monte Carlo simulation algorithm is used to calculate the obstruction probability of each candidate risk source in the candidate risk source attribute set relative to the communication link, and each risk source is determined synchronously according to the magnitude of the obstruction probability.

[0033] It should be noted that, firstly, since there is a correspondence between the features of satellite remote sensing images (such as spectrum, texture, and elevation) and the physical properties of physical occluders (such as height, density, and material), a reverse mapping model of dynamic features and physical properties is constructed based on the temporal variation features and spectral features of abnormal pixel regions in the dynamic feature dataset, combined with the elevation data (such as digital elevation model DEM data) and texture data (such as the surface roughness of occluders) of satellite remote sensing images. Specifically, this model can transform abstract features in images into specific physical parameters, such as mapping the spectral features of a certain abnormal pixel region to the physical parameters of "tree height 15 meters, crown width 10 meters", thus solving the problem that "image features cannot directly correspond to physical risks".

[0034] Secondly, the physical parameters of the entities corresponding to each feature in the preset path are analyzed by the reverse mapping model. At the same time, the path environment parameters (such as the height of the conductor and the height of the robot operation) are combined to screen out the set of candidate risk source attributes. Specifically, the candidate risk source must meet the conditions such as "height ≥ vertical distance between the conductor and the ground and location on the communication link path". Low-risk objects that do not affect communication (such as low shrubs) are eliminated, and the focus is on the core occlusion risk.

[0035] Finally, considering the randomness of occlusion risk (such as the change in occlusion probability caused by trees swaying in the wind), the Monte Carlo simulation algorithm is used to calculate the occlusion probability of each candidate risk source relative to the communication link in the candidate risk source attribute set. Specifically, the algorithm simulates the dynamic state of the occluder (such as different swaying angles of trees) through a large number of random samples, counts the occlusion duration and frequency under different states, and quantifies the occlusion probability. Based on the magnitude of the occlusion probability (e.g., occlusion probability ≥60% is a high-risk source, 30%-60% is a medium-risk source, and <30% is a low-risk source), each risk source is determined, providing an accurate risk classification basis for the construction of the risk map.

[0036] Furthermore, the step of constructing a rasterized risk map framework based on the planar coordinates of the work area, and generating the scene risk map according to the rasterized risk map framework and each of the risk sources, includes: Based on the degree of influence of each risk source relative to the communication link, each risk source is divided into a core layer, an influence layer and an outer layer, and the risk radiation intensity corresponding to the risk source in the core layer is calculated simultaneously. Substitute the risk radiation intensity and the type and risk level of each risk source into the grid frame, and simultaneously calculate the comprehensive risk value of each grid cell to generate the corresponding initial risk map. The signal transmission thresholds and dual-mode link triggering conditions of the sky-end communication module and the ground-end communication module are fused into the internal structure of the initial risk map to generate the corresponding scenario risk map.

[0037] It should be noted that, firstly, based on the degree of impact of each risk source relative to the communication link, the risk sources are divided into three levels: core layer risk sources (such as tall trees or mountains that directly block the communication link, with a blocking probability ≥60%), influence layer risk sources (such as buildings near the communication link, with a blocking probability of 30%-60%), and peripheral layer risk sources (such as low vegetation far from the communication link, with a blocking probability <30%). Simultaneously, the risk radiation intensity of the core layer risk sources is calculated (e.g., radiation intensity = blocking probability × height of the obstruction, quantifying the scope of the risk's impact). Specifically, risk stratification can distinguish the priority of different risk sources, and risk radiation intensity can characterize the spread of the risk, providing accurate risk quantification indicators for subsequent link switching strategies.

[0038] Secondly, the risk radiation intensity and the types and risk levels of each risk source are substituted into the gridded framework, and a weighted summation algorithm is used to calculate the comprehensive risk value of each grid cell (e.g., comprehensive risk value = core layer risk radiation intensity × 0.6 + influence layer risk probability × 0.3 + outer layer risk probability × 0.1). The risk level of the grid cell is divided according to the magnitude of the comprehensive risk value to generate an initial risk map. Specifically, the weighted summation method can take into account the influence of risk sources at different levels and avoid the assessment bias caused by a single risk source.

[0039] Finally, to enable the risk map to directly guide the control of the communication link, the signal transmission thresholds (such as the minimum signal strength threshold and the minimum transmission rate threshold) of the communication modules at the sky and ground ends, as well as the dual-mode link triggering conditions (such as triggering the dual-mode link when the obstruction probability is ≥60%), are integrated into the initial risk map. The signal thresholds and triggering conditions are marked in the corresponding grid cells. The resulting scenario risk map not only contains the risk distribution but also integrates the key parameters of communication control, achieving an organic combination of "risk visualization" and "control strategy".

[0040] Furthermore, the step of generating a target control strategy adapted to the actuator based on the scenario risk map, the actual control commands, and the operating status data includes: Risk features from the scenario risk map, instruction features from the actual control instructions, and state features of the actuator are extracted and synchronously fused into corresponding fusion feature vectors. Using the adaptation requirements of the actuator as a constraint, a corresponding objective function is created by combining the fused feature vector, and an initial control strategy is generated simultaneously by solving the problem through a robust algorithm. The initial control strategy is subjected to anti-interference testing in order to generate the target control strategy accordingly.

[0041] It should be noted that, firstly, in order to comprehensively characterize the core factors affecting the control strategy, risk features (such as risk level, occlusion probability, and risk radiation range) from the scenario risk map, instruction features (such as instruction type, execution priority, and transmission frequency) from the actual control instructions, and state features of the actuators (such as attitude deviation, clamping force, and motor load rate) are extracted. A feature fusion algorithm (such as principal component analysis) is used to integrate the three types of features into a fused feature vector. Specifically, the fused feature vector can comprehensively cover the coupling relationship of "risk-instruction-state" and avoid the one-sidedness of the strategy caused by a single feature.

[0042] Secondly, using the adaptation requirements of the actuator as constraints (such as clamping force ≤ rated value, displacement accuracy ≤ ±1mm, communication delay ≤ 100ms), an objective function is created by combining the fused feature vectors. The optimization objective of the objective function is to "maximize the operation accuracy, minimize the communication delay, and optimize the actuator load rate". The objective function is solved by a robust algorithm (such as a robust optimization algorithm) to generate an initial control strategy. Specifically, the robust algorithm can ensure that the control strategy can still meet the execution requirements under uncertain risk scenarios, thereby improving the strategy's anti-interference ability.

[0043] Finally, to verify the effectiveness of the initial control strategy under complex interference scenarios, the initial control strategy was subjected to anti-interference testing: communication interruptions, signal attenuation, and other interferences under different risk scenarios were simulated in a simulation environment to test the execution effect of the strategy; based on the test results, the parameters in the strategy were modified (such as reducing the command transmission frequency and increasing the redundancy of state data in high interference scenarios), and finally the target control strategy was generated. Specifically, anti-interference testing can identify the defects of the strategy in advance and ensure the stability of the strategy in actual operation.

[0044] Furthermore, the step of creating a corresponding objective function by combining the fused feature vector with the adaptation requirements of the actuator as a constraint, and simultaneously solving for and generating the initial control strategy through a robust algorithm, includes: The core components of the fused feature vector are extracted using a sparse autoencoder network, and the adaptation requirements of the actuator are simultaneously decomposed into corresponding elastic constraint intervals. The mapping relationship between the core components and the elastic constraint interval is established by using a kernel function to create a corresponding constraint-feature mapping table. Simultaneously, the objective function is created based on the constraint-feature mapping table. The whale optimization algorithm is selected, and the objective function is iteratively solved based on the fluctuation amplitude of the core component to extract the optimal solution. The optimal solution is then back-verified through the constraint-feature mapping table and combined and mapped to the initial control strategy.

[0045] It should be noted that, firstly, since there are redundant features in the fused feature vector (such as features that are weakly related to the adaptability of the actuator), a sparse autoencoder network is used to extract the core components in the fused feature vector. Specifically, the sparse autoencoder network can use unsupervised learning to filter out the features that have the greatest impact on the control strategy (such as risk level, instruction priority, motor load rate), eliminate redundant features, and reduce the complexity of subsequent optimization calculations.

[0046] Secondly, the adaptation requirements of the actuator are broken down into corresponding elastic constraint ranges (e.g., the elastic constraint range of clamping force is [500N, 800N], and the elastic constraint range of displacement accuracy is [±0.5mm, ±1mm]). The elastic constraint range can take into account both operational accuracy and the fault tolerance of the actuator. The mapping relationship between the core components and the elastic constraint range is established by using a kernel function (e.g., Gaussian kernel function), and a constraint-feature mapping table is constructed. This table clarifies the constraint range range corresponding to different core feature values. Based on the constraint-feature mapping table, an objective function is created. The constraint conditions of the objective function are directly related to the elastic constraint range of the actuator, and the optimization objective is linked to the value of the core feature.

[0047] Finally, the whale optimization algorithm is selected to solve the objective function. This algorithm has the characteristics of fast convergence speed and strong global optimization ability, and is suitable for complex multi-constraint optimization problems. The iteration parameters of the algorithm are adjusted based on the fluctuation amplitude of the core components (e.g., increasing the exploration step size when the core component fluctuation is large, and increasing the development step size when the fluctuation is small). The optimal solution of the objective function is solved through multiple rounds of iteration. The optimal solution is back-verified through the constraint-feature mapping table to ensure that the parameters corresponding to the optimal solution meet the elastic constraint requirements of the actuator. The optimal solution is then combined and mapped to the initial control strategy. Specifically, the back-verification can prevent the optimal solution from exceeding the physical limit of the actuator and ensure the feasibility of the strategy.

[0048] Furthermore, the step of performing anti-interference testing on the initial control strategy to generate the target control strategy includes: Based on the risk characteristics of the scenario risk map and historical operation interference data, a corresponding interference prediction model is constructed through transfer learning algorithm to output the interference risk warning level for each operation stage. Based on the interference risk warning level and the fused feature vector, a multimodal anti-interference strategy library is generated through fuzzy decision matching; The initial control strategy is modified using the multimodal interference strategy library to generate the target control strategy accordingly.

[0049] It should be noted that, firstly, based on the risk characteristics of the scenario risk map and historical operation interference data (such as historical communication interruption scenarios, interference types, and durations), a transfer learning algorithm is used to construct an interference prediction model. Specifically, transfer learning can transfer the interference patterns of historical operation scenarios to the current operation scenario, solving the problem of insufficient interference data in the new scenario. The model can predict the interference risk warning level (such as high risk, medium risk, and low risk) for each operation stage based on the current risk characteristics, providing a basis for matching anti-interference strategies.

[0050] Secondly, based on the interference risk warning level and the fused feature vector, a fuzzy decision algorithm is used to match the multimodal anti-interference strategy library. Specifically, the strategy library stores anti-interference strategies corresponding to different interference levels. For example, the high-risk level corresponds to "reducing the command transmission frequency + increasing data redundancy + triggering the dual-mode link in advance", and the medium-risk level corresponds to "adjusting command priority + enhancing signal filtering". Fuzzy decision can handle the uncertainty of interference risk and achieve accurate matching between risk level and anti-interference strategy.

[0051] Finally, the initial control strategy is modified using strategies from the multimodal anti-interference strategy library. For example, the transmission method of instructions is modified during high-interference risk stages, and the clamping force parameters are modified when the load rate of the actuator is too high. The modified control strategy retains the operational accuracy requirements of the initial strategy and has the ability to cope with complex interference, thus generating the target control strategy.

[0052] Furthermore, the step of pre-triggering the dual-mode link mode to establish a stable connection between the sky-end communication module and the ground-end communication module, and synchronously completing the corresponding execution actions, includes: Based on the monitoring data collected by the sky-end communication module and combined with the scene risk map, the characteristics and movement trajectory of the obstruction are identified to generate corresponding link early warning signals and dual-mode link activation sequence. In response to the link warning signal, an independent redundant channel is allocated to the dual-mode link, and corresponding adjustment execution parameters are generated synchronously. The adjusted execution parameters are sent to the execution mechanism to synchronously complete the corresponding execution actions.

[0053] It should be noted that, firstly, monitoring data collected by the sky-end communication module (such as signal strength and real-time images of obstructions) is combined with the risk distribution in the scene risk map. Target detection algorithms (such as the YOLO algorithm) are used to identify the characteristics (such as type, height, and obstruction range) and movement trajectories (such as the swaying trajectory of trees and the flight trajectory of drones). Based on the characteristics and trajectories of the obstructions, the attenuation trend of the communication link is predicted, generating link warning signals (such as "signal strength will drop below the threshold in 30 seconds") and dual-mode link activation timing (such as activating the backup link 10 seconds in advance). Specifically, early warning and timing control can avoid the lag in link switching and ensure the continuity of communication.

[0054] Secondly, in response to link warning signals, independent redundant channels are allocated to the dual-mode links. Specifically, the main link and the backup link use different transmission frequency bands and transmission protocols (e.g., the main link is 5G, and the backup link is microwave communication). Independent redundant channels can avoid interference between links and improve communication reliability. At the same time, corresponding adjustment execution parameters are generated according to the timing requirements of link switching (e.g., reducing the movement speed of the actuator and pausing high-precision positioning actions during the link switching phase). Specifically, adjusting execution parameters can ensure the stability of the actuator's actions during link switching and avoid operational deviations caused by command delays.

[0055] Finally, the adjusted execution parameters are sent to the actuator, which continues to work according to the adjusted parameters, and simultaneously completes the execution action of installing the spacer. Specifically, the link switching and the adjustment of execution parameters are carried out simultaneously, achieving a dual guarantee of "stable communication" and "continuous operation".

[0056] Furthermore, the step of allocating independent redundant channels to the dual-mode link and synchronously generating corresponding adjustment execution parameters includes: Based on the characteristics of the obstruction, the movement trajectory, and the scene risk map, the corresponding obstruction scene level is divided, and the corresponding independent redundant channel is allocated synchronously according to the obstruction scene level. The actual transmission characteristics of the independent redundant channels are collected, and initial execution parameters are generated by combining them with the job task type of the actuator. The initial execution parameters are validated to generate the corresponding adjusted execution parameters.

[0057] It should be noted that, firstly, based on the characteristics, movement trajectory, and scene risk map of the obstructing object, the obstruction scene is classified into three levels: Level 1 obstruction scene (such as continuous obstruction by a fixed building, with an obstruction probability ≥80%), Level 2 obstruction scene (such as intermittent obstruction by swaying trees, with an obstruction probability of 40%-80%), and Level 3 obstruction scene (such as brief obstruction by a temporary obstruction, with an obstruction probability <40%). Corresponding independent redundant channels are allocated according to the obstruction scene level. For example, Level 1 scenes are allocated with dual backup links (microwave + laser), Level 2 scenes with a single backup link (microwave), and Level 3 scenes do not require a backup link. Specifically, scene classification enables on-demand allocation of redundant channels, avoiding resource waste and improving the efficiency of link switching.

[0058] Secondly, the actual transmission characteristics (such as transmission rate, latency, and bit error rate) of the independent redundant channels are collected, and the initial execution parameters are generated by combining them with the operation task type of the actuator (such as clamping task, tightening task, and positioning task). Specifically, different operation tasks have different communication requirements. For example, positioning tasks require low latency and high transmission rate, while clamping tasks have a higher tolerance for latency. The initial execution parameters need to match the task type and channel transmission characteristics.

[0059] Finally, the initial execution parameters are verified: the execution effect of the parameters is simulated by the dynamic simulation model of the actuator (such as the ADAMS simulation model) to test whether the parameters meet the requirements of operational accuracy and stability; parameters that exceed the physical limits of the actuator (such as the movement speed exceeding the rated value) are eliminated, and unreasonable parts of the parameters are corrected; finally, the adjusted execution parameters are generated to ensure that the parameters are both adapted to the transmission characteristics of the dual-mode link and meet the operational requirements of the actuator.

[0060] Additionally, in this embodiment, it should be noted that, please refer to... Figures 2 to 5 Spacer bars are tools used in high-voltage transmission lines to fix the spacing between split conductors and suppress conductor vibration and galloping, such as... Figure 2 As shown, the spacer 5 includes a body 50 and latches 51 located at the four corners of the body 50. The body 50 is formed as a rectangular frame, consisting of an upper frame 501, a lower frame 502, a left frame 503, and a right frame 504 connected end to end. The lower frame 502 has a positioning hole 52 for inserting a positioning post to position the spacer; the upper frame 501, the left frame 503, and the right frame 504 have through holes 53 for engaging grippers, thereby enabling the spacer 5 to be gripped. The latches 51 include a palm rest 511 and a grip 512. The palm rest 511 is fixed, while the grip 512 can be opened and closed, thereby opening and closing the latches 51.

[0061] The four-split spacer installation robot is a component used to install spacers, such as... Figure 3As shown, the four-split spacer installation robot includes a drone 6, a fixed frame 7, a spacer clamping mechanism 8, a robotic arm 9, and a cable adjustment mechanism 10. The drone 6 and the fixed frame 7 are connected for transporting the spacer installation robot. The spacer clamping mechanism 8 is used to clamp and rotate the spacer 5. The cable adjustment mechanism 10 is used to grasp the cable and place it in the two clips 51 below the spacer 5. The robotic arm 9 is used to lock the clips 51 of the spacer 5.

[0062] The mounting bracket 7 includes a mounting frame 71 and four support legs 72. A guide rod 73 for guiding the wires is provided between the front and rear support legs 72. The guide rod 73 includes a first guide rod 731, a second guide rod 732, a third guide rod 733, and a fourth guide rod 734. The first guide rod 731 and the fourth guide rod 734 are parallel to each other and extend vertically. The second guide rod 732 extends downwards at an angle, and the third guide rod 733 extends upwards at an angle. One end of the first guide rod 731 is fixedly connected to the mounting frame 71. The other end of the first guide rod 731 is fixedly connected to one end of the second guide rod 732. The other end of the second guide rod 732 is fixedly connected to one end of the third guide rod 733. The other end of the third guide rod 733 is fixedly connected to one end of the fourth guide rod 734. The other end of the fourth guide rod 734 is fixedly connected to the mounting frame 71.

[0063] Because the second guide rod 732 and the third guide rod 733 have an inclined angle, even if the UAV positioning accuracy is inaccurate and the wire does not fall into the appropriate fixed position, the wire can still slide upward along the second guide rod 732 and the third guide rod 733 under the guidance of the second guide rod 732 and the third guide rod 733, and slide vertically upward along the first guide rod 731 and the fourth guide rod 734 to fall into the appropriate fixed position.

[0064] like Figure 4 As shown, the spacer bar gripping mechanism 8 includes a gripper 81, which is used to engage with the through hole 53 of the spacer bar 5, thereby gripping the spacer bar 5. The gripper 81 is driven by a first hydraulic cylinder 82, and a first proximity switch is provided on the gripper 81. The first proximity switch is used to detect whether the spacer bar 5 is close to the gripper 81 or moves away from the gripper 81. The spacer bar gripping mechanism 8 is rotatably connected to the fixed frame 71 and is driven by a rotary motor 83. When the spacer bar gripping mechanism 8 starts working to grip the spacer bar 5, it makes the plane where the spacer bar 5 is located parallel to the wire to avoid interference between the spacer bar 5 and the wire, facilitating wire placement. When the wire is placed, when the photoelectric gate on the support leg 72 senses the wire, the rotary motor 83 rotates, driving the spacer bar gripping mechanism 8 to rotate, making the plane where the spacer bar is located perpendicular to the direction of the wire, facilitating the wire to be inserted into the latch 51 of the spacer bar 5.

[0065] The cable adjustment mechanism 10 includes a left jaw 101 and a right jaw 102. The left jaw 101 and right jaw 102 are used to place the two lower cables into the two latches 51 below the spacer bar 5. The left jaw 101 and right jaw 102 are controlled by a second hydraulic cylinder and a third hydraulic cylinder to extend and retract in the left and right directions, respectively; controlled by a fourth hydraulic cylinder and a fifth hydraulic cylinder to close and open, respectively; and controlled by a sixth hydraulic cylinder to move up and down in the vertical direction. Three displacement sensors are provided on the left jaw 101 and right jaw 102: a first displacement sensor, a second displacement sensor, and a third displacement sensor. The first displacement sensor detects the vertical displacement of the left and right jaws; the second displacement sensor detects the extension and retraction of the left jaw 101 in the left and right directions; and the third displacement sensor detects the extension and retraction of the right jaw 102 in the left and right directions. A second proximity switch and a third proximity switch are also provided on the left jaw 101 and the right jaw 102. The second proximity switch is used to determine whether the left jaw 101 and the right jaw 102 are close to the cable, and the third proximity switch is used to determine whether the left jaw 101 and the right jaw 102 are clamping the cable.

[0066] like Figure 5 As shown, the robotic arm 9 includes a motion joint mechanism 90, a first clamping plate 91, and a second clamping plate 92. The first clamping plate 91 can move back and forth and is used to hold the grip 512 of the buckle 51 of the spacer bar 5. The second clamping plate 92 is fixed and is used to hold the palm support 511 of the buckle 51 of the spacer bar 5. When it is necessary to lock the buckle 51 of the spacer bar 5, the motion joint mechanism 90 is driven to move, so that the second clamping plate 92 holds the palm support 511 of the spacer bar 5, and the first clamping plate 91 moves towards the grip 512 of the spacer bar 5 and holds the grip 512 of the spacer bar 5, thereby locking the buckle 51 of the spacer bar 5.

[0067] The motion joint mechanism 90 includes a first joint 901, a second joint 902, a third joint 903, and a fourth joint 904. The first joint 901 includes a lead screw stepper motor 9011 and a first joint rod 9012 connected to the lead screw stepper motor 9011. The lead screw stepper motor 9011 drives the first joint rod 9012 to move up and down in the Z-axis direction, thereby realizing the up-and-down movement of the first joint 901 in the Z-axis direction. The second joint 902 includes a first joint motor 9021 and a first joint arm 9022 pivotally connected to the first joint motor 9021. The rotation of the first joint motor 9021 drives the first joint arm 9022 to rotate in the X and Y planes, thereby realizing the rotation of the second joint 902 in the X and Y planes. The third joint 903 includes a second joint motor 9031 and a second joint arm 9032 pivotally connected to the second joint motor 9031. Rotation of the second joint motor 9031 drives rotation of the second joint arm 9032 in the X and Y planes, thereby achieving rotation of the third joint 903 in the X and Y planes. The fourth joint 904 includes a third joint motor 9041 and a first joint body 9042 pivotally connected to the third joint motor 9041. Rotation of the third joint motor 9041 drives rotation of the first joint body 9042 in the X and Y planes, thereby achieving rotation of the fourth joint 904 in the X and Y planes. The second joint 902, third joint 903, and fourth joint 904 complete movement in the same plane. The other end of the first joint rod 9012 is fixedly connected to the first joint motor 9021, the other end of the first joint arm 9022 is fixedly connected to the second joint motor 9031, and the other end of the second joint arm 9032 is fixedly connected to the third joint motor 9041. The first clamping plate 91 and the second clamping plate 92 are mounted on the first joint body 9042. The first joint body 9042 is equipped with a fourth joint motor (not shown). The rotation of the fourth joint motor drives the first clamping plate 91 to move and abut against the grip 512 of the spacer bar 5. The second joint 902, the third joint 903, and the fourth joint 904 rotate in the X and Y planes, while the first joint 901 moves up and down in the Z-axis direction, thereby driving the first clamping plate 91 and the second clamping plate 92 to their respective latching positions. At this time, the fourth joint motor is activated to drive the first clamping plate 91 to move and abut against the grip 512 of the spacer bar 5, thereby locking the various latches 51 of the spacer bar 5.

[0068] In summary, the communication control system for spacer rod installation robots provided in the above embodiments of the present invention enables the robot to remain in a stable control state, thereby improving the robot's control efficiency.

[0069] It should be noted that the above modules can be functional modules or program modules, and can be implemented through software or hardware. For modules implemented through hardware, the above modules can reside in the same processor; or the above modules can be located in different processors in any combination.

[0070] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-including system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device.

[0071] More specific examples of computer-readable media (a non-exhaustive list) include: electrical connections (electronic devices) having one or more wires, portable computer disk drives (magnetic devices), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Furthermore, computer-readable media can even be paper or other suitable media on which the program can be printed, because the program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in computer memory.

[0072] It should be understood that various parts of the present invention can be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0073] In the description of this specification, references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0074] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the appended claims.

Claims

1. A communication control system for a spacer bar installation robot, characterized in that, The system includes an actuator, a sky-to-ground communication module, and a ground-to-ground communication module. The ground-based communication module preloads the preset path and line environment parameters of the actuator, and combines them with satellite remote sensing data to predict the occlusion risk of the work area, so as to generate a corresponding scene risk map. The actual control commands and operating status data of the actuator are collected through the sky-end communication module. Simultaneously, based on the scenario risk map, a target control strategy adapted to the actuator is generated according to the actual control commands and operating status data. The target control strategy is sent to the execution mechanism so that the execution mechanism can complete the corresponding execution action; The aerial communication module monitors changes in the operating environment of the actuator. When the obstruction risk in the operating area reaches a preset risk threshold or the signal strength attenuation exceeds a preset attenuation threshold, the dual-mode link mode is triggered in advance to establish a stable connection between the aerial communication module and the ground communication module, and the corresponding execution actions are completed synchronously.

2. The communication control system for the spacer bar installation robot according to claim 1, characterized in that, The step of combining satellite remote sensing data to predict the occlusion risk of the work area and generate a corresponding scene risk map includes: Collect satellite remote sensing images of the work area for multiple consecutive time periods, and combine the path coordinates of the preset path with the line environmental parameters to crop out a temporal image subset along the preset range of the path; A time-series trend fitting algorithm is used to extract dynamic features from the time-series image subset, and abnormal pixel regions in the satellite remote sensing images are identified simultaneously through standard deviation analysis to generate a corresponding dynamic feature dataset. Based on the dynamic feature dataset, each risk source in the preset path is parsed out. Simultaneously, a rasterized risk map framework is constructed based on the planar coordinates of the work area. The scene risk map is then generated according to the rasterized risk map framework and each of the risk sources.

3. The communication control system for the spacer bar installation robot according to claim 2, characterized in that, The step of parsing each risk source in the preset path based on the dynamic feature dataset includes: Based on the temporal variation features in the dynamic feature dataset and the spectral features of the abnormal pixel region, a corresponding inverse mapping model of dynamic features and physical attributes is constructed by combining the elevation data and texture data of the satellite remote sensing image. The physical parameters of the entities corresponding to each feature in the preset path are parsed through the reverse mapping model, and the candidate risk source attribute set is screened in conjunction with the line environment parameters. The Monte Carlo simulation algorithm is used to calculate the obstruction probability of each candidate risk source in the candidate risk source attribute set relative to the communication link, and each risk source is determined synchronously according to the magnitude of the obstruction probability.

4. The communication control system for the spacer bar installation robot according to claim 2, characterized in that, The step of constructing a rasterized risk map framework based on the planar coordinates of the work area, and generating the scene risk map according to the rasterized risk map framework and each of the risk sources, includes: Based on the degree of influence of each risk source relative to the communication link, each risk source is divided into a core layer, an influence layer and an outer layer, and the risk radiation intensity corresponding to the risk source in the core layer is calculated simultaneously. Substitute the risk radiation intensity and the type and risk level of each risk source into the grid frame, and simultaneously calculate the comprehensive risk value of each grid cell to generate the corresponding initial risk map. The signal transmission thresholds and dual-mode link triggering conditions of the sky-end communication module and the ground-end communication module are fused into the internal structure of the initial risk map to generate the corresponding scenario risk map.

5. The communication control system for the spacer bar installation robot according to claim 1, characterized in that, The step of generating a target control strategy adapted to the actuator based on the scenario risk map, the actual control commands, and the operational status data includes: Risk features from the scenario risk map, instruction features from the actual control instructions, and state features of the actuator are extracted and synchronously fused into corresponding fusion feature vectors. Using the adaptation requirements of the actuator as a constraint, a corresponding objective function is created by combining the fused feature vector, and an initial control strategy is generated simultaneously by solving the problem through a robust algorithm. The initial control strategy is subjected to anti-interference testing in order to generate the target control strategy accordingly.

6. The communication control system for the spacer bar installation robot according to claim 5, characterized in that, The steps of creating a corresponding objective function by combining the fused feature vector with the adaptation requirements of the actuator as a constraint, and simultaneously generating the initial control strategy by solving the problem using a robust algorithm, include: The core components of the fused feature vector are extracted using a sparse autoencoder network, and the adaptation requirements of the actuator are simultaneously decomposed into corresponding elastic constraint intervals. The mapping relationship between the core components and the elastic constraint interval is established by using a kernel function to create a corresponding constraint-feature mapping table. Simultaneously, the objective function is created based on the constraint-feature mapping table. The whale optimization algorithm is selected, and the objective function is iteratively solved based on the fluctuation amplitude of the core component to extract the optimal solution. The optimal solution is then back-verified through the constraint-feature mapping table and combined and mapped to the initial control strategy.

7. The communication control system for the spacer bar installation robot according to claim 5, characterized in that, The step of performing anti-interference testing on the initial control strategy to generate the target control strategy includes: Based on the risk characteristics of the scenario risk map and historical operation interference data, a corresponding interference prediction model is constructed through transfer learning algorithm to output the interference risk warning level for each operation stage. Based on the interference risk warning level and the fused feature vector, a multimodal anti-interference strategy library is generated through fuzzy decision matching; The initial control strategy is modified using the multimodal interference strategy library to generate the target control strategy accordingly.

8. The communication control system for the spacer bar installation robot according to claim 1, characterized in that, The steps of pre-triggering the dual-mode link mode to establish a stable connection between the sky-end communication module and the ground-end communication module, and synchronously completing the corresponding execution actions, include: Based on the monitoring data collected by the sky-end communication module and combined with the scene risk map, the characteristics and movement trajectory of the obstruction are identified to generate corresponding link early warning signals and dual-mode link activation sequence. In response to the link warning signal, an independent redundant channel is allocated to the dual-mode link, and corresponding adjustment execution parameters are generated synchronously. The adjusted execution parameters are sent to the execution mechanism to synchronously complete the corresponding execution actions.

9. The communication control system for the spacer bar installation robot according to claim 8, characterized in that, The steps of allocating independent redundant channels to the dual-mode link and synchronously generating corresponding adjustment execution parameters include: Based on the characteristics of the obstruction, the movement trajectory, and the scene risk map, the corresponding obstruction scene level is divided, and the corresponding independent redundant channel is allocated synchronously according to the obstruction scene level. The actual transmission characteristics of the independent redundant channels are collected, and initial execution parameters are generated by combining them with the job task type of the actuator. The initial execution parameters are validated to generate the corresponding adjusted execution parameters.

10. The communication control system for the spacer bar installation robot according to claim 1, characterized in that, The actuator is used to grip the spacer bar (5). The spacer bar (5) includes a body (50) and buckles (51) located at the four corners of the body (50). The body (50) is formed as a rectangular frame, which is surrounded by an upper frame (501), a lower frame (502), a left frame (503), and a right frame (504) connected end to end. The lower frame (502) is provided with a positioning hole (52), which is used for the insertion of a positioning post to position the spacer bar. The upper frame (501), the left frame (503), and the right frame (504) are provided with through holes (53), which are used to engage the gripper, thereby realizing the gripping of the spacer bar (5).

Images