Intelligent control method for collaborative operation of concrete leveling robots
By dynamically allocating tasks and building a data interaction network based on the terrain data model, the problem of collaborative operation of concrete leveling robots in complex terrain was solved, achieving efficient, accurate regional leveling and seamless connection.
Patent Information
- Application Number
- CN202511023725.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2025-07-24
- Publication Date
- 2026-11-17
- Estimated Expiration
- 2045-07-24
AI Technical Summary
Existing concrete leveling robots lack effective communication mechanisms and overall coordination when dealing with large-area and complex terrain operations, resulting in uneven division of work areas, repeated processing of some areas or neglect of others, which affects construction progress and quality.
By acquiring elevation and slope distribution information through a pre-established terrain data model, the initial work area is divided. Tasks are dynamically allocated based on the position status and workload of multiple leveling robots. A data interaction network is built to share location and progress information in real time, determine unified work rhythm control parameters, analyze elevation deviations between adjacent areas, and dynamically adjust area division and equipment paths to ensure seamless integration.
It enables precise control of multi-equipment collaborative operation in complex terrain, ensuring area flatness and seamless connection, and improving construction efficiency and quality.
Smart Images

Figure CN121363311B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of engineering construction robot control technology, and in particular to an intelligent control method for collaborative operation of concrete leveling robots. Background Technology
[0002] In the construction industry, concrete leveling technology is a crucial link in ensuring project quality and efficiency. Especially in complex terrain environments, intelligent leveling robots have irreplaceable value in improving construction accuracy and safety. With the construction industry's ever-growing demand for efficient and precise operations, the development of leveling robots capable of adapting to various terrains and achieving automated operation has become an urgent need for industry development.
[0003] However, current mainstream leveling robots or methods often exhibit insufficient coordination and a lack of overall planning capabilities when dealing with large-area, complex terrain operations. Many leveling robots lack effective communication mechanisms when collaborating in multi-robot operations, leading to uneven division of work areas. Some areas may be repeatedly processed, while others are neglected, thus affecting the overall construction progress and quality. This problem is not simply due to outdated technology, but rather to a lack of systematic design for the dynamic interaction and task coordination of multiple leveling robots.
[0004] A deeper analysis of the challenges in this field reveals that real-time information sharing among multiple leveling robots is the primary bottleneck. In large-area operations, if the leveling robots cannot promptly understand each other's positions and statuses, it becomes difficult to establish a unified work rhythm, leading to chaotic task allocation. Furthermore, due to the lack of precise handling of the work area boundaries, elevation discrepancies often occur between adjacent areas, affecting the overall leveling effect. Summary of the Invention
[0005] The purpose of this invention is to solve the above-mentioned problems by providing an intelligent control method for collaborative operation of concrete leveling robots.
[0006] The technical solution of this invention is implemented as follows: This invention provides an intelligent control method for collaborative operation of concrete leveling robots. By using a pre-established terrain data model, elevation and slope distribution information is obtained from the work area. To adapt to complex terrain, terrain features are analyzed and the initial work area is divided, and the boundary range and elevation benchmark data of each area are obtained. Based on the initial work area division results, the position status update data of multiple leveling robots are obtained. Combined with the workload and mobility of each leveling robot, the task is dynamically allocated to determine the area of responsibility and work sequence of each leveling robot. Based on the distribution of leveling robots after dynamic task allocation, real-time position and work progress information are obtained from the sensors of each leveling robot. Through a real-time information sharing mechanism, a data interaction network for multi-leveling robot collaboration is constructed to obtain unified work rhythm control parameters. By controlling the work rhythm parameters, the boundary data of the work area between adjacent leveling robots is obtained. Based on the boundary elevation connection requirements, the elevation deviation between adjacent areas is analyzed to determine the boundary processing priority and adjustment plan. Based on the boundary processing priority, the data on the difference in work progress between leveling robots is obtained. If the progress of a certain leveling robot is lagging behind, the task allocation strategy is dynamically updated by adjusting the data interaction frequency to obtain the optimized work path of the leveling robot.
[0007] In one embodiment, the control method further includes: For the optimized leveling robot's operation path, real-time feedback information from terrain data collection is obtained, terrain changes encountered during path execution are analyzed, it is determined whether the area division scheme needs to be adjusted, and updated area boundary data is obtained. Based on the updated regional boundary data, obtain the latest status information of multiple leveling robots working together, adjust the operation rhythm among the leveling robots to achieve the goal of seamless regional connection, and determine the final elevation consistency control parameters. By using the final elevation consistency control parameters, the flatness data of the area after each leveling robot has completed its operation is obtained, and the presence of deviations is analyzed. If the deviation exceeds the preset threshold, secondary processing of the local area is triggered to obtain the corrected flatness distribution information.
[0008] In one implementation method, the method for determining the boundary extent and elevation datum data of each region includes: By using a pre-set terrain data model, the elevation and slope distribution of the work area are obtained, resulting in elevation and slope distribution data. Digital elevation models are used to process elevation and slope distribution data, extract terrain features, and determine the set of terrain features. If the slope value in the terrain feature set is greater than a preset threshold, it is marked as a complex terrain region, and a set of complex terrain regions is obtained. The K-means clustering algorithm is used to divide the complex terrain region into initial work areas and determine the initial work area set. Based on the initial set of work areas, calculate the boundary range of each area to obtain the boundary range data; Interpolation algorithms are used to process boundary range data and elevation distribution data to generate elevation benchmark data; By rasterizing the elevation benchmark data, an elevation benchmark raster map of the work area is generated, and the final work area division result is determined.
[0009] In one implementation, the method for determining the area of responsibility and work sequence of each leveling robot includes: The system acquires the position status and update data of multiple leveling robots, and collects the position information and operating status of each leveling robot in real time through a pre-established monitoring system to determine the current working area and status change trend of the leveling robot. Based on the trend of status changes and combined with workload data, a preset load balancing strategy is used to analyze the current workload of each leveling robot and determine whether the leveling robot is in a high-load state. If the load exceeds the preset threshold, it is marked as a leveling robot that needs to be adjusted. For leveling robots marked as needing adjustment, their mobility data is obtained. By comparing the mobility with the distance and path information between the work area and the work area, a list of leveling robots that can be reassigned to the new work area is determined. Based on the list of leveling robots and the area division data, a dynamic programming algorithm is used to optimize the task allocation, resulting in a new area of responsibility and work sequence for each leveling robot. By updating data and status monitoring results, it is determined whether the newly assigned area of responsibility matches the current position status of the leveling robot. If the position status does not match the assigned area, the movement path is recalculated and the work sequence is adjusted. Based on the adjusted work sequence, obtain the task execution progress of each leveling robot, update the workload data through the real-time monitoring system, and determine the execution effect of the task allocation; By combining performance data with status monitoring information, the dynamic adjustment needs of the leveling robot within the work area are continuously tracked to obtain the final task allocation and area division results.
[0010] In one implementation, the method for obtaining the unified work rhythm control parameters includes: Real-time position and work progress data are obtained from the leveling robot's sensors, and a standardized data stream is obtained using a timestamp synchronization mechanism; Standardized data streams are transmitted through a data interaction network, and a message queue protocol is used to ensure data consistency among multiple leveling robots. If the data consistency meets the preset threshold, the distribution of leveling robots is clustered using the K-means algorithm based on real-time location and work progress to obtain the grouping results of leveling robots; Based on the grouping results of the leveling robots, the task allocation strategy is dynamically adjusted, and resource allocation is optimized through linear programming algorithm to generate a task allocation scheme. The workload of each leveling robot is extracted from the task allocation scheme, and the operation rhythm trend is predicted by time series analysis to obtain the rhythm prediction result; By broadcasting rhythm prediction results through a collaborative network, the control parameters of each leveling robot are adjusted to generate unified operation rhythm control parameters. If the deviation of the control parameters exceeds the preset threshold, the deviation information is fed back through the data interaction network, the task allocation scheme is iteratively optimized, and the updated control parameters are obtained.
[0011] In one implementation, the method for determining the boundary processing priority and adjustment scheme includes: By obtaining the boundary data of adjacent leveling robot operation areas from the leveling robot log using the operation rhythm parameters, a set of boundary coordinates is obtained. The boundary elevation values between adjacent areas are calculated using the boundary coordinate set, resulting in an elevation data table; If the elevation deviation between adjacent areas in the elevation data table exceeds a preset threshold, the deviation correction value is calculated using a linear interpolation algorithm to obtain the corrected elevation set. Based on the priority of boundary connection analysis of the modified elevation set, a priority ranking table is obtained; The adjustment scheme is assigned by prioritizing the sorting table, and the boundary elevation connection is optimized by least squares method to obtain the set of adjustment parameters; Obtain the set of adjustment parameters, update the boundary data of the leveling robot's working area, and obtain the updated boundary coordinate set; The elevation deviations of adjacent areas are recalculated based on the updated boundary coordinate set to obtain the final elevation consistency data.
[0012] In one implementation, the method for obtaining the optimized leveling robot's work path includes: By obtaining work progress data from the leveling robot's operation log, a preliminary comparison is made of the current task completion status of each leveling robot to determine the progress difference value. Based on the progress difference value obtained from the comparison, a preset threshold is used for screening. If the difference value of a certain leveling robot exceeds the threshold, it is marked as a lagging leveling robot, and a list of lagging leveling robots is obtained. For the leveling robots in the list of lagging leveling robots, obtain their historical data interaction records, analyze the correlation between interaction frequency and task completion efficiency, and determine the potential impact range of frequency adjustment. Based on the analysis results, the data interaction frequency of the lagging leveling robot is dynamically adjusted, and a temporary task allocation scheme is generated in combination with the current task load. Based on the temporary task allocation scheme, a genetic algorithm is used to optimize the working path of the leveling robot, recalculate the task execution order of each leveling robot, and determine the improved path planning. The improved path planning data is obtained, the task allocation strategy is updated in real time, and synchronized to the control modules of each leveling robot to complete the dynamic adjustment process. By continuously monitoring the leveling robot's operation progress and path execution, and cyclically comparing the differences, if a new lagging leveling robot is found, a new round of frequency adjustment and path optimization process is triggered. In one embodiment, the method for obtaining the updated region boundary data includes: Real-time feedback information is obtained from the terrain data acquisition module. The terrain change characteristics during the execution of the leveling robot's work path are initially extracted. The change characteristics are classified according to the preset classification criteria to obtain the classification results of terrain changes. Based on the classification results of terrain changes, the scope of influence during path execution is analyzed. Combined with the current state of the leveling robot's working path, the deviation of the path execution is determined by comparing it with the preset terrain adaptation rules. If the deviation of the path execution exceeds the preset threshold, an adaptive assessment is performed on the regional division scheme, the distribution data of the current regional boundary is obtained, the impact of terrain change characteristics on the boundary distribution is analyzed, and the adjustment requirements for regional division are obtained. Based on the adjustment requirements of the regional division, combined with the terrain change characteristics and the actual situation of the leveling robot's operation path, a genetic algorithm is used to recalculate the regional boundaries and generate a preliminary boundary adjustment plan. For the preliminary boundary adjustment scheme, the historical records of the leveling robot path optimization are obtained, the matching degree between the adjustment scheme and the historical records is analyzed, and the feasibility of the scheme is judged by comparing it with the preset optimization rules, so as to obtain the verified boundary adjustment data. Based on the verified boundary adjustment data, the area division scheme is updated in real time and synchronized to the control module of the leveling robot's operation path to complete the dynamic matching of the path and the boundary, and generate updated area boundary information. By continuously monitoring the execution status of the leveling robot's work path using updated area boundary information, and iteratively comparing the terrain change features in the real-time feedback information, a new adjustment process is triggered if new deviations are found, to determine the final path adaptation solution. In one implementation, the method for determining the final elevation consistency control parameters includes: By obtaining the latest regional boundary information and real-time status data of the leveling robot collaboration from the data storage module, a dynamic model of regional division is constructed to obtain preliminary boundary division results. Based on the boundary division results, the operation rhythm distribution of the leveling robot in each area is analyzed, and the rhythm is compared using a preset threshold to determine the adjustment direction required for rhythm optimization. Based on the optimization direction of the rhythm, and combined with real-time status data, the coordination deviation between leveling robots is judged. If the deviation exceeds the preset range, the optimized coordination scheme of leveling robots is obtained by adjusting the operation rhythm parameters. The optimized leveling robot collaborative scheme was obtained, the degree of seamless connection between areas was analyzed, the connection boundary was smoothed using the information processing module, and the preliminary control parameters for elevation consistency were determined. Based on the preliminary control parameters and combined with the status monitoring data, the operation effect of the leveling robot after adjustment is analyzed. If the effect does not meet the preset standard, the control parameters are adjusted through iterative calculation to obtain the final elevation consistency parameters. By updating the collaborative operation instructions of the leveling robots using the final elevation consistency parameters, the data update module sends the adjusted parameters to each leveling robot, thus completing the dynamic adjustment of rhythm optimization and area connection. Based on the results of dynamic adjustments, real-time status monitoring data is acquired. By comparing this data with historical data, the stability of the area boundary and the coordination of the leveling robot is assessed, and a continuous optimization scheme for system operation is determined. In one embodiment, the method for obtaining the flatness distribution information includes: Using elevation consistency parameters, initial regional flatness data are obtained from each leveling robot. Outliers are removed by data cleaning methods to obtain a pre-processed flatness dataset. Based on the pre-processed flatness dataset, a deviation analysis is performed on the regional flatness distribution. If the analysis results show that the deviation exceeds the preset threshold, the local areas with deviations are marked, and the range of areas requiring secondary processing is determined. The marked local area data is obtained, and these areas are finely segmented using the region division method to generate detailed grid information of the local area. By using detailed grid information of local areas, a secondary processing method is used to adjust the elevation of each grid cell to obtain the adjusted local flatness data. Based on the adjusted local flatness data, the flatness distribution of the entire area is recalculated to generate updated flatness distribution information; The updated flatness distribution information is verified again using the deviation analysis method. If there are still areas where the deviation exceeds the preset threshold, the secondary processing flow is triggered repeatedly to obtain the final corrected flatness distribution dataset. By integrating elevation consistency parameters and work completion data with the final corrected flatness distribution dataset, a complete regional flatness analysis result is generated.
[0013] The advantages or beneficial effects of the above technical solutions include at least the following: This invention discloses an intelligent control method for multi-device collaborative operations in complex terrain. It obtains elevation and slope distribution information through a pre-established terrain data model, divides the initial work area, and dynamically allocates tasks. The invention constructs a data interaction network for multi-device collaboration, sharing location and progress information in real time to obtain unified operation rhythm control parameters. To address the need for boundary elevation connection, it analyzes elevation deviations between adjacent areas and determines processing solutions. Furthermore, the invention can dynamically adjust area division and equipment paths based on real-time feedback, and ensure seamless area connection through elevation consistency control parameters. Finally, by analyzing flatness data to trigger local secondary processing, this invention effectively solves the problem of precise control for multi-device collaborative operations in complex terrain, achieving efficient flattening and seamless connection of the work area. Attached Figure Description
[0014] The accompanying drawings illustrate exemplary embodiments of the present invention and, together with the description thereof, serve to explain the principles of the invention. These drawings are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification.
[0015] Figure 1 A flowchart of the control method according to an embodiment of the present invention is shown; Figure 2 A schematic diagram of the leveling robot according to an embodiment of the present invention is shown. Detailed Implementation
[0016] Embodiments of the present invention will now be described in more detail with reference to the accompanying drawings. While some embodiments of the invention are shown in the drawings, it should be understood that the invention can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided to provide a more thorough and complete understanding of the invention. It should be understood that the accompanying drawings and embodiments are for illustrative purposes only and are not intended to limit the scope of protection of the invention.
[0017] It should be noted that, unless otherwise specified, the embodiments and features described in this invention can be combined with each other. The invention will now be described in detail with reference to the accompanying drawings and embodiments.
[0018] It should be understood that the term "comprising" and its variations as used herein are open-ended, meaning "including but not limited to". The term "based on" means "at least partially based on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". Definitions of other terms will be given in the following description. It should be noted that the concepts of "first", "second", etc., mentioned in this invention are used only to distinguish different devices, modules, or units, and are not intended to limit the order of functions performed by these devices, modules, or units or their interdependencies.
[0019] It should be noted that the terms "a" and "a plurality of" used in this invention are illustrative rather than restrictive. Those skilled in the art should understand that, unless otherwise expressly indicated in the context, they should be understood as "one or more".
[0020] The names of the messages or information exchanged between the multiple devices in the embodiments of the present invention are for illustrative purposes only and are not intended to limit the scope of these messages or information.
[0021] like Figure 1-2 This embodiment of an intelligent control method for collaborative operation of concrete leveling robots may specifically include: S101. Using a pre-established terrain data model, obtain elevation and slope distribution information from the work area. To adapt to complex terrain, analyze terrain features and divide the initial work area, obtaining the boundary range and elevation benchmark data of each area.
[0022] By using a pre-defined terrain data model, the elevation and slope distribution of the work area are obtained, resulting in elevation and slope distribution data. A digital elevation model is then used to process the elevation and slope distribution data, extracting terrain features and determining a set of terrain features. If the slope value in the terrain feature set exceeds a preset threshold, it is marked as a complex terrain area, resulting in a set of complex terrain areas. A K-means clustering algorithm is used to initially divide the complex terrain area set into work areas, determining the initial work area set. Based on the initial work area set, the boundary range of each area is calculated, obtaining boundary range data. An interpolation algorithm is used to process the boundary range data and elevation distribution data to generate elevation benchmark data. The elevation benchmark data is then rasterized to generate an elevation benchmark raster map of the work area, determining the final work area division result.
[0023] Specifically, based on a pre-established terrain data model, assuming a Digital Elevation Model (DEM) format with a spatial resolution of 1 meter, covering a 1000m × 1000m work area, including elevation values ranging from 0 to 500 meters, the elevation distribution information is extracted first. Through raster data processing, the DEM file is read using Python's GDAL library, the elevation value of each raster cell is calculated, and an elevation distribution histogram is generated. The area percentage of elevation intervals (e.g., 0-100 meters, 100-200 meters, etc.) is statistically analyzed. Assuming the results show 0-100 meters accounts for 30%, 100-200 meters for 50%, and 200-500 meters for 20%, the slope distribution is calculated. Using ArcGIS's slope analysis tool, the slope of each raster is calculated based on the DEM using the formula: atan(√(dz / dx)). 2 + dz / dy 2 ))×57.29578, where dz / dx and dz / dy are the elevation gradients in the x and y directions, respectively, with a slope range of 0-45°. Statistical results show that 0-5° accounts for 60%, 5-15° accounts for 30%, and 15-45° accounts for 10%.
[0024] To adapt to complex terrain, a K-means clustering algorithm (K=3) was used to analyze terrain features and divide the region into flat areas (elevation 0-100 meters, slope 0-5°), gentle slope areas (elevation 100-200 meters, slope 5-15°), and steep slope areas (elevation 200-500 meters, slope 15-45°). After clustering, the areas of each region were 600 hectares, 300 hectares, and 100 hectares, respectively. The initial working area was delineated, and GIS spatial analysis tools were used to generate vector boundaries for each region. The boundaries were stored in GeoJSON format, containing a sequence of coordinate points and an elevation benchmark (the average elevation within the region, e.g., 50 meters for flat areas). Finally, the boundary ranges of each region (e.g., the boundary of the flat area is [(0,0), (0,500), (500,500), (500,0)]) and elevation benchmark data (e.g., benchmark elevation of 50 meters for flat areas, 150 meters for gentle slope areas, and 300 meters for steep slope areas) are output and stored as a database table to ensure that subsequent operation planning can be optimized based on this data. The entire process is implemented through automated scripts, with rigorous logic and efficient data processing, providing precise support for adapting to complex terrain.
[0025] S102. Based on the initial work area division results, obtain the position status update data of multiple leveling robots, and combine the workload and mobility of each leveling robot to perform dynamic task allocation and determine the responsible area and work sequence of each leveling robot.
[0026] Among them, such as Figure 2As shown, the leveling robot includes a body 10, on which an RTK high-precision positioning system 12 is installed for locating the robot's position. Wheels 11 are installed under the body 10, and spring suspension is provided between the wheels 11 and the body 10 to effectively reduce the impact of terrain on the robot's movement. A leveling mechanism 20 is also installed at the front end of the body 10. The leveling mechanism includes a leveling plate 22 connected to the body 10, on which a vibration motor 23 is installed for leveling concrete. A laser receiver 21 is also installed to receive laser signals from a laser transmitter and feed them back to the control system. The laser transmitter is installed at a location with good visibility outside the construction site to determine the robot's horizontal elevation based on the site elevation lines.
[0027] The system acquires the position status and update data of multiple leveling robots. A pre-established monitoring system collects the position information and operational status of each leveling robot in real time to determine its current work area and status change trend. Based on the status change trend and workload data, a preset load balancing strategy is used to analyze the current workload of each leveling robot to determine if it is under high load. If the load exceeds a preset threshold, the robot is marked as needing adjustment. For the marked robots, their mobility data is acquired. By comparing the mobility capabilities with the distance and path information between the work area and the new work area, a list of leveling robots that can be reassigned to new areas is determined. Based on the robot list and area division data, a dynamic programming algorithm is used to optimize task allocation, resulting in new work areas and work sequences for each robot. By updating the data and comparing the status monitoring results, it is determined whether the newly assigned work area matches the current position status of the leveling robot. If the position status does not match the assigned area, the movement path is recalculated and the work sequence is adjusted. Based on the adjusted work sequence, the task execution progress of each leveling robot is acquired. The workload data is updated in real time through the monitoring system to determine the execution effect of the task allocation. By combining performance data with status monitoring information, the dynamic adjustment needs of the leveling robot within the work area are continuously tracked to obtain the final task allocation and area division results.
[0028] Specifically, based on the initial work area division, the system first divides the work area into 100×100 meter grids using a grid partitioning algorithm. Each grid is assigned a unique identifier, and its geographic coordinates and task requirements are recorded (e.g., 50MB of data to be processed per grid). The system obtains real-time position status updates from multiple leveling robots via an API interface. Assuming there are three leveling robots A, B, and C, with current coordinates (20,30), (50,60), and (80,20) respectively, movement speeds of 5m / s, 4m / s, and 3m / s respectively, and processing capacities of 10MB / s, 8MB / s, and 6MB / s respectively, the position data is updated every second via a GPS module and stored in a cloud database. The system uses the WebSocket protocol to ensure real-time performance. Based on the workload of the leveling robots, the system calculates the current task queue of each robot. For example, leveling robot A has 20MB of tasks to process, leveling robot B has 30MB, and leveling robot C has 15MB. The availability of the leveling robots is evaluated using a load balancing algorithm (weighted round-robin method, where weights are based on the inverse ratio of processing capacity to remaining tasks). Dynamic task allocation employs an improved genetic algorithm, initializing the population with 100 allocation schemes. Each scheme includes a mapping between leveling robots and grids, and the fitness function is a weighted sum of total travel distance and processing time (weights 0.4 and 0.6). Through 50 generations of crossover (probability 0.8) and mutation (probability 0.1), the optimal allocation scheme is selected. For example, leveling robot A is responsible for grids (20,30) to (40,50), leveling robot B is responsible for (50,60) to (70,80), and leveling robot C is responsible for (80,20) to (100,40). The task order is based on the priority of tasks within the grid (determined by the amount of data and urgency, with urgency scores of 1-5). The Dijkstra algorithm is used to calculate the shortest path for the leveling robot within its assigned area, ensuring that the travel distance is minimized.
[0029] For example, leveling robot A's path is (20,30)→(30,40)→(40,50), with a total distance of approximately 28.28 meters and an estimated time of 5.66 seconds. The analysis process records the load changes of each leveling robot. After processing, leveling robot A's load drops to 10MB. The system compares the initial and final loads through logs, verifying a 20% improvement in allocation efficiency. If the grid task volume suddenly increases (e.g., grid (50,60) increases to 100MB), the system re-runs the genetic algorithm to adjust the allocation, ensuring load balancing. This process is implemented through a distributed computing framework (such as Spark), automatically handling data flow and task scheduling to form a closed-loop optimization.
[0030] S103. Based on the distribution of leveling robots after dynamic task allocation, real-time position and work progress information are obtained from the sensors of each leveling robot. Through a real-time information sharing mechanism, a data interaction network for multi-leveling robot collaboration is constructed to obtain unified work rhythm control parameters.
[0031] Real-time position and work progress data are acquired from the leveling robot's sensors, and a timestamp synchronization mechanism is used to obtain a standardized data stream. This standardized data stream is transmitted via a data interaction network, using a message queue protocol to ensure data consistency among multiple leveling robots. If data consistency meets a preset threshold, the distribution of leveling robots is clustered using the K-means algorithm based on real-time position and work progress, resulting in robot grouping results. Based on the robot grouping results, the task allocation strategy is dynamically adjusted, and resource allocation is optimized using a linear programming algorithm to generate a task allocation scheme. The workload of each leveling robot is extracted from the task allocation scheme, and time series analysis is used to predict the work rhythm trend, obtaining a rhythm prediction result. The rhythm prediction result is broadcast through a collaborative network, and the control parameters of each leveling robot are adjusted to generate unified work rhythm control parameters. If the control parameter deviation exceeds a preset threshold, deviation information is fed back through the data interaction network, and the task allocation scheme is iteratively optimized to obtain updated control parameters.
[0032] Specifically, in constructing a system for collaborative operation of multiple leveling robots, position data is first collected in real time using GPS sensors and inertial measurement units (IMUs) on each leveling robot. Assuming leveling robot A is at coordinates (X:100.5, Y:200.3) and leveling robot B is at (X:105.2, Y:198.7), the data collection frequency is 5 times per second. The data is transmitted to the cloud server in JSON format via a 5G network, with transmission latency controlled within 50 milliseconds to ensure real-time performance. Next, progress information is obtained using work progress sensors (e.g., leveling robot A has completed 35% and leveling robot B has completed 42%). The overall progress is calculated using a weighted average algorithm: Total Progress = (Leveling Robot A Progress × Weight 0.6 + Leveling Robot B Progress × Weight 0.4), resulting in a total progress of 37.8%. The progress difference (leveling robot B is 7% ahead) is analyzed to provide a basis for subsequent pace adjustments. Furthermore, a real-time information sharing mechanism is established, employing the MQTT protocol to construct a publish-subscribe model. All leveling robots subscribe to the cloud topic "TaskProgress," updating data every 10 seconds to ensure information synchronization errors are less than 1%. Based on this, a collaborative data interaction network for multiple leveling robots is constructed. Graph theory algorithms are used to calculate the communication topology between leveling robots, establishing connections when the distance between them is less than 50 meters, generating an adjacency matrix, and optimizing data transmission paths using a shortest path algorithm to reduce latency to below 30 milliseconds. Finally, based on the above data, a PID control algorithm is used to calculate unified work rhythm parameters. The target progress is set at 40%, and the current deviation is 2.2%. Using a proportional coefficient Kp=0.5, an integral coefficient Ki=0.1, and a derivative coefficient Kd=0.05, an adjustment amount of 1.3% is calculated, outputting the adjusted work speed command for each leveling robot (e.g., leveling robot A accelerates to 1.2 meters / second), ensuring overall coordination. This process is automated in the cloud, combined with analysis of distance and progress differences between leveling robots, forming a closed-loop control logic to ensure work efficiency and consistency.
[0033] S104. By using the operation rhythm control parameters, obtain the boundary data of the operation area between adjacent leveling robots. Based on the boundary elevation connection requirements, analyze the elevation deviation between adjacent areas and determine the boundary processing priority and adjustment plan.
[0034] Boundary data of adjacent leveling robot work areas are obtained from the leveling robot logs using work rhythm parameters, resulting in a boundary coordinate set. The boundary elevation values between adjacent areas are calculated using this coordinate set, resulting in an elevation data table. If the elevation deviation between adjacent areas in the elevation data table exceeds a preset threshold, a deviation correction value is calculated using a linear interpolation algorithm, resulting in a corrected elevation set. The priority of boundary connection is analyzed based on the corrected elevation set, resulting in a priority ranking table. Adjustment schemes are allocated using the priority ranking table, and the least squares method is used to optimize boundary elevation connection, resulting in an adjustment parameter set. The adjustment parameter set is then used to update the boundary data of the leveling robot work areas, resulting in an updated boundary coordinate set. The elevation deviation between adjacent areas is recalculated based on the updated boundary coordinate set, resulting in final elevation consistency data.
[0035] Specifically, the following implementation method is used to obtain the boundary data of the working areas between adjacent leveling robots by using the operation rhythm control parameters, and to analyze the elevation deviation based on the boundary elevation connection requirements, and to determine the boundary processing priority and adjustment plan: First, using the operation rhythm control parameters, such as setting the leveling robot's operating speed to 5 kilometers per hour and the operation interval to 10 minutes, the boundary of the working areas of adjacent leveling robots A and B is calculated using the system's built-in geographic information algorithm. Assuming that the boundary coordinates of the working area of leveling robot A are (100.5, 200.5) and the boundary coordinates of leveling robot B are (100.8, 200.5), and the distance between the two points is calculated to be 0.3 meters, the data within a boundary area width of 0.5 meters is determined as the analysis object. Next, to address the boundary elevation connection requirements, elevation data within the boundary areas is extracted. Assuming the boundary elevation of area A for leveling robot is 50.2 meters and the boundary elevation of area B for leveling robot is 50.8 meters, the elevation deviation is calculated using an interpolation algorithm to be 0.6 meters. This is compared to a preset deviation threshold of 0.4 meters, and it is determined that this deviation exceeds the threshold and requires priority processing. Subsequently, the impact of the elevation deviation between adjacent areas is analyzed. Considering the slope requirements related to the business (assuming a maximum slope of 2%), the slope caused by the current deviation is calculated to be 0.6 / 0.3 = 2%, reaching the upper limit of the slope. Immediate adjustment is required to ensure smooth operation. Finally, the system determines the boundary processing priority and adjustment plan. Based on the deviation value and elevation adjustment rules (prioritizing areas with higher elevations), the system automatically generates an adjustment plan, smoothly adjusting the elevation of area B of the leveling robot from 50.8 meters to 50.5 meters, an adjustment increment of 0.3 meters, ensuring the deviation is reduced to 0.3 meters, below the threshold. Simultaneously, a smooth interpolation algorithm (such as linear interpolation) is used to calculate the elevation adjustment values of each point within the boundary, creating a smooth transition. The adjustment plan is automatically pushed to the leveling robot control system for execution. Through this method, a complete logical chain is formed from data acquisition to plan generation, ensuring that the boundary elevation connection meets operational requirements.
[0036] S105. Based on the boundary processing priority, obtain the work progress difference data between leveling robots. If the progress of a certain leveling robot is lagging behind, the task allocation strategy is dynamically updated by adjusting the data interaction frequency to obtain the optimized leveling robot work path.
[0037] By obtaining work progress data from the leveling robot's operation logs, a preliminary comparison of the current task completion status of each leveling robot is performed to determine the progress difference value. Based on the obtained progress difference value, a preset threshold is used for filtering. If the difference value of a leveling robot exceeds the threshold, it is marked as a lagging leveling robot, resulting in a list of lagging leveling robots. For the leveling robots in the lagging leveling robot list, their historical data interaction records are obtained, and the correlation between interaction frequency and task completion efficiency is analyzed to determine the potential impact range of frequency adjustment. Based on the analysis results, the data interaction frequency of lagging leveling robots is dynamically adjusted, and a temporary task allocation scheme is generated in combination with the current task load. Based on the temporary task allocation scheme, a genetic algorithm is used to optimize the leveling robot's work path, recalculate the task execution order of each leveling robot, and determine the improved path planning. The improved path planning data is obtained, the task allocation strategy is updated in real time, and synchronized to the control module of each leveling robot to complete the dynamic adjustment process. By continuously monitoring the leveling robot's work progress and path execution status, and cyclically comparing the difference value, if a new lagging leveling robot is found, a new round of frequency adjustment and path optimization process is triggered.
[0038] Specifically, when handling the differences in work progress and optimizing paths among leveling robots, the work data of multiple leveling robots are first analyzed using a boundary processing priority algorithm. Assuming there are three leveling robots A, B, and C, with work progress of 75%, 82%, and 60% respectively, the progress difference is calculated (e.g., the difference between A and C is 15%, and the difference between B and C is 22%). It is determined that leveling robot C is significantly lagging behind and has the highest priority, requiring task allocation adjustment. Next, for the lagging leveling robot, a data interaction frequency adjustment mechanism is used. The real-time monitoring system collects leveling robot status data every 5 minutes. Analyzing the workload of leveling robot C reveals that it processes 50 tasks per hour, while A and B process 80 and 75 tasks per hour respectively. This calculates that C's efficiency is 37.5% lower than A's. Therefore, the interaction frequency is dynamically adjusted to once every 3 minutes to increase data feedback density and more accurately grasp its status. Subsequently, based on the difference data and interaction frequency, a weighted task allocation algorithm was adopted to reduce the workload of leveling robot C by 20%, from 100 units to 80 units. Simultaneously, the reduced workload was proportionally allocated to robots A and B (A increased by 12 units, B increased by 8 units). Simulation analysis showed that the estimated completion times for the three leveling robots after the new allocation scheme were 10 hours, 10.2 hours, and 10.5 hours, respectively, with the difference narrowing to within 0.5 hours, indicating improved balance. Finally, using a path optimization algorithm (such as Dijkstra's algorithm), combined with the leveling robot position coordinates (assuming A is at (0,0), B at (5,5), and C at (10,0)) and task point distribution, the shortest operation path was recalculated. The results showed that the path length for leveling robot A was shortened from 15 units to 13 units, B from 18 units to 16 units, and C from 20 units to 17 units, achieving a total path optimization rate of 12%, thereby improving overall operation efficiency. The above process is completed through automated system calculations and real-time data interaction, ensuring rigorous logic and close correlation between task allocation and path optimization.
[0039] S106. For the optimized leveling robot operation path, obtain real-time feedback information of terrain data collection, analyze the terrain changes encountered during the path execution, determine whether the area division scheme needs to be adjusted, and obtain updated area boundary data.
[0040] Real-time feedback information is obtained from the terrain data acquisition module. Preliminary extraction of terrain change characteristics during the execution of the leveling robot's work path is performed, and these characteristics are categorized using a preset classification standard to obtain the terrain change classification results. Based on the classification results, the impact range during path execution is analyzed. Combined with the current state of the leveling robot's work path, and compared with preset terrain adaptation rules, it is determined whether the deviation exceeds the adaptation range, thus determining the degree of deviation in path execution. If the deviation exceeds a preset threshold, an adaptability assessment of the region division scheme is conducted. The distribution data of the current region boundaries is obtained, and the impact of terrain change characteristics on the boundary distribution is analyzed to determine the adjustment requirements for region division. Based on the adjustment requirements for region division, combined with the terrain change characteristics and the actual situation of the leveling robot's work path, a genetic algorithm is used to recalculate the region boundaries, generating a preliminary boundary adjustment scheme. For the preliminary boundary adjustment scheme, the historical records of leveling robot path optimization are obtained, and the matching degree between the adjustment scheme and the historical records is analyzed. By comparing with preset optimization rules, the feasibility of the scheme is determined, and validated boundary adjustment data is obtained. Based on the verified boundary adjustment data, the area division scheme is updated in real time and synchronized to the control module of the leveling robot's work path to complete the dynamic matching of the path and the boundary, generating updated area boundary information. Using the updated area boundary information, the execution status of the leveling robot's work path is continuously monitored, and the terrain change features in the real-time feedback information are iteratively compared. If a new deviation is found, a new round of adjustment is triggered to determine the final path adaptation scheme.
[0041] Specifically, during the process of collecting terrain data and adjusting the area division for the optimized leveling robot's working path, the system first automatically collects terrain data every 10 minutes through the sensor network deployed on the leveling robot, including information such as slope, obstacle density, and soil hardness. Suppose that a leveling robot D finds that the slope suddenly increases from 5 degrees to 15 degrees and the obstacle density increases from 0.2 per square meter to 0.8 per square meter during path execution, the system uploads this data to the cloud analysis platform and uses the terrain adaptability assessment algorithm to calculate that the current path's traversal difficulty index has increased from the original 3.5 to 7.8, exceeding the safety threshold of 6.0, and determines that there is a potential risk in path execution. Next, the system automatically triggered the terrain change analysis module. By comparing historical data, it was found that the slope change rate in the area reached 200% in the past 24 hours, which is an abnormal fluctuation, requiring a reassessment of the area division scheme. Therefore, the system invoked a boundary dynamic adjustment algorithm to reduce the original boundary radius from 500 meters to 300 meters to avoid high-risk terrain. Simultaneously, the system analyzed the flatness of the surrounding terrain, finding that the slope in the eastern area was only 3 degrees and the obstacle density was 0.1 obstacles per square meter, making it suitable for re-division. Finally, the system generated updated area boundary data, shifting the working range of the leveling robot D 200 meters eastward, forming new boundary coordinates from (X: 450, Y: 320) to (X: 750, Y: 320). Simulation calculations confirmed that the new area's accessibility index had decreased to 2.9, meeting safety requirements. The entire process was completed automatically by the system, ensuring a close connection between terrain changes and area division logic, while also providing data support for subsequent path planning.
[0042] S107. Based on the updated regional boundary data, obtain the latest status information of multiple leveling robots working together, adjust the operation rhythm among the leveling robots for the goal of seamless regional connection, and determine the final elevation consistency control parameters.
[0043] By acquiring the latest regional boundary information and real-time status data of the leveling robots from the data storage module, a dynamic model for regional division is constructed, yielding preliminary boundary division results. Based on the boundary division results, the distribution of the leveling robots' work rhythm in each region is analyzed, and a preset threshold is used for rhythm comparison to determine the adjustment direction required for rhythm optimization. For the rhythm optimization direction, combined with real-time status data, the coordination deviation between leveling robots is judged. If the deviation exceeds a preset range, the work rhythm parameters are adjusted to obtain an optimized leveling robot coordination scheme. The optimized leveling robot coordination scheme is obtained, and the seamless connection between regions is analyzed. An information processing module is used to smooth the connection boundaries, determining preliminary control parameters for elevation consistency. Based on the preliminary control parameters and status monitoring data, the operational effect of the leveling robots after adjustment is analyzed. If the effect does not meet the preset standard, the control parameters are iteratively calculated and adjusted to obtain the final elevation consistency parameters. Using the final elevation consistency parameters, the operational instructions for leveling robot coordination are updated, and the adjusted parameters are sent to each leveling robot using a data update module, completing the dynamic adjustment of rhythm optimization and regional connection. Based on the results of dynamic adjustments, real-time status monitoring data is obtained. By comparing the data with historical data, the stability of the coordination between the area boundary and the leveling robot is determined, and a continuous optimization scheme for system operation is identified.
[0044] Specifically, based on the updated regional boundary data, the system first automatically loads the latest boundary vector data through a Geographic Information System (GIS). Assuming the boundary data covers the area from 116.3°E to 116.5°E and from 39.8°N to 40.0°N, the system uses spatial analysis algorithms to calculate the operational coverage of each leveling robot within the region, generating a coverage grid with a resolution of 0.5m × 0.5m to ensure accurate matching between the boundary data and the leveling robot positions. Next, the system acquires the latest status information of the collaborative work of multiple leveling robots. This data is collected in real-time via an IoT interface. For example, leveling robot A operates at a speed of 2.5m / s, while leveling robot B operates at 2.2m / s, with operational height errors of ±0.03m and ±0.05m respectively. Combining the status data with analysis of the collaborative efficiency between the leveling robots, the system calculates that the operational overlap rate between leveling robots A and B is 15%. A timestamp comparison reveals that leveling robot A leads leveling robot B by approximately 3 seconds, requiring an adjustment of their pace to reduce overlap. Subsequently, to achieve seamless regional connectivity, the system employed a dynamic programming algorithm to adjust the operating rhythm of the leveling robots. The speed of leveling robot A was reduced to 2.3 m / s, while that of leveling robot B was increased to 2.4 m / s. The adjusted overlap rate was calculated to be reduced to 5%, and simulation analysis confirmed that the error at the boundary connection was controlled within ±0.02 meters. Finally, elevation consistency control parameters were determined. Based on the elevation data collected by the leveling robots (average elevation of 50.25 meters for leveling robot A and 50.28 meters for leveling robot B), the system used the least squares method to fit elevation curves, deriving a consistency control parameter of an elevation deviation threshold of ±0.015 meters. The operating height commands of the leveling robots were adjusted to ensure that the elevation consistency error within the region was less than 0.01 meters. Simultaneously, corroborating business data such as soil moisture distribution (humidity range 20% to 30%) was used for auxiliary verification to ensure that the parameters adapted to different terrain conditions. Through this automated process, the system achieved end-to-end processing from data updates to parameter optimization, with rigorous logic and quantifiable results.
[0045] S108. By using the final elevation consistency control parameters, obtain the flatness data of the area after each leveling robot has completed its operation, analyze whether there is a deviation, and if the deviation exceeds the preset threshold, trigger secondary processing of the local area to obtain the corrected flatness distribution information.
[0046] Initial flatness data for the region is obtained from each leveling robot using elevation consistency parameters. Outliers are removed using data cleaning methods, resulting in a pre-processed flatness dataset. Based on this dataset, deviation analysis is performed on the regional flatness distribution. If the analysis shows deviations exceeding a preset threshold, local areas with deviations are marked, determining the areas requiring secondary processing. Data for these marked local areas is acquired, and these areas are finely segmented using a region division method, generating detailed grid information for each local area. Using this detailed grid information, the elevation of each grid cell is adjusted using a secondary processing method, resulting in adjusted local flatness data. Based on this adjusted local flatness data, the overall regional flatness distribution is recalculated, generating updated flatness distribution information. This updated distribution is then validated again using deviation analysis. If areas still exceed the preset threshold, the secondary processing flow is repeatedly triggered, resulting in a final corrected flatness distribution dataset. Finally, the final corrected flatness distribution dataset is integrated with elevation consistency parameters and work completion data to generate a complete regional flatness analysis result.
[0047] Specifically, regional flatness data is obtained through elevation consistency control parameters. First, a LiDAR scanner is used to scan the leveling robot and collect 3D point cloud data of the work area. Assuming the area is 1000 square meters and the point cloud density is 100 points per square meter, coordinate and elevation data of 100,000 points are generated. Control parameters include an elevation reference value (e.g., 0.00 meters) and an allowable deviation threshold of ±0.05 meters. The least squares method is used to fit the point cloud data, and the difference between the actual elevation and the reference elevation of each point is calculated to generate a flatness distribution map.
[0048] When analyzing deviations, the point cloud data is traversed, and the mean and standard deviation of the elevation differences are calculated. The mean is assumed to be 0.03 meters and the standard deviation 0.02 meters. The elevation differences of each point are compared to a threshold of ±0.05 meters. If more than 30% of the points in a certain area (e.g., a 100-square-meter sub-area) have elevation differences exceeding the threshold, the analysis is performed.
[0049] For example, if 20 out of 50 points have an elevation difference greater than 0.05 meters, then the area is determined to have exceeded the deviation limit. When secondary processing is triggered, an automatic leveling robot is invoked to generate an optimized path based on the deviation distribution. The area with the largest elevation difference is processed first, and the processed elevation is recalculated using an interpolation algorithm (such as Kriging interpolation) to update the point cloud data.
[0050] Finally, the flatness distribution is recalculated to obtain the corrected mean elevation (e.g., 0.01 meters) and standard deviation (e.g., 0.015 meters), generating a new flatness distribution map and ensuring that the elevation difference in all areas is controlled within ±0.05 meters. If areas still exceed the standard, a secondary processing step is performed iteratively until the threshold requirement is met. The entire process is integrated through an automated system. Data processing uses Python scripts, calling the NumPy library for matrix operations, and Matplotlib to generate visual distribution maps, ensuring rigorous logic and requiring no manual intervention.
[0051] In the description of this invention, it should be noted that the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limiting this invention.
[0052] Those skilled in the art should understand that the above embodiments are merely for illustrating the present invention and are not intended to limit the scope of the invention. Those skilled in the art can make other changes or modifications based on the above disclosure, and these changes or modifications still fall within the scope of the present invention.
Claims
1. An intelligent control method for collaborative operation of concrete leveling robots, characterized in that: By using a pre-established terrain data model, elevation and slope distribution information is obtained from the work area. To adapt to complex terrain, terrain features are analyzed and the initial work area is divided, and the boundary range and elevation benchmark data of each area are obtained. Based on the initial work area division results, the position status update data of multiple leveling robots are obtained. Combined with the workload and mobility of each leveling robot, the task is dynamically allocated to determine the area of responsibility and work sequence of each leveling robot. Based on the distribution of leveling robots after dynamic task allocation, real-time position and work progress information are obtained from the sensors of each leveling robot. Through a real-time information sharing mechanism, a data interaction network for multi-leveling robot collaboration is constructed to obtain unified work rhythm control parameters. By controlling the work rhythm parameters, the boundary data of the work area between adjacent leveling robots is obtained. Based on the boundary elevation connection requirements, the elevation deviation between adjacent areas is analyzed to determine the boundary processing priority and adjustment plan. Based on the boundary processing priority, the data on the difference in work progress between leveling robots is obtained. If the progress of a certain leveling robot is lagging behind, the task allocation strategy is dynamically updated by adjusting the data interaction frequency to obtain the optimized work path of the leveling robot.
2. The intelligent control method for collaborative operation of concrete leveling robots according to claim 1, characterized in that: The control method further includes: For the optimized leveling robot's operation path, real-time feedback information from terrain data collection is obtained, terrain changes encountered during path execution are analyzed, it is determined whether the area division scheme needs to be adjusted, and updated area boundary data is obtained. Based on the updated regional boundary data, obtain the latest status information of multiple leveling robots working together, adjust the operation rhythm among the leveling robots to achieve the goal of seamless regional connection, and determine the final elevation consistency control parameters. By using the final elevation consistency control parameters, the flatness data of the area after each leveling robot has completed its operation is obtained, and the presence of deviations is analyzed. If the deviation exceeds the preset threshold, secondary processing of the local area is triggered to obtain the corrected flatness distribution information.
3. The intelligent control method for collaborative operation of concrete leveling robots according to claim 1, characterized in that: The methods for determining the boundary extent and elevation datum data of each region include: By using a pre-set terrain data model, the elevation and slope distribution of the work area are obtained, resulting in elevation and slope distribution data. Digital elevation models are used to process elevation and slope distribution data, extract terrain features, and determine the set of terrain features. If the slope value in the terrain feature set is greater than a preset threshold, it is marked as a complex terrain region, and a set of complex terrain regions is obtained. The K-means clustering algorithm is used to divide the complex terrain region into initial work areas and determine the initial work area set. Based on the initial set of work areas, calculate the boundary range of each area to obtain the boundary range data; Interpolation algorithms are used to process boundary range data and elevation distribution data to generate elevation benchmark data; By rasterizing the elevation benchmark data, an elevation benchmark raster map of the work area is generated, and the final work area division result is determined.
4. The intelligent control method for collaborative operation of concrete leveling robots according to claim 1, characterized in that: in, The methods for determining the area of responsibility and work sequence for each leveling robot include: The system acquires the position status and update data of multiple leveling robots, and collects the position information and operating status of each leveling robot in real time through a pre-established monitoring system to determine the current working area and status change trend of the leveling robot. Based on the trend of status changes and combined with workload data, a preset load balancing strategy is used to analyze the current workload of each leveling robot and determine whether the leveling robot is in a high-load state. If the load exceeds the preset threshold, it is marked as a leveling robot that needs to be adjusted. For leveling robots marked as needing adjustment, their mobility data is obtained. By comparing the mobility with the distance and path information between the work area and the work area, a list of leveling robots that can be reassigned to the new work area is determined. Based on the list of leveling robots and the area division data, a dynamic programming algorithm is used to optimize the task allocation, resulting in a new area of responsibility and work sequence for each leveling robot. By updating data and status monitoring results, it is determined whether the newly assigned area of responsibility matches the current position status of the leveling robot. If the position status does not match the assigned area, the movement path is recalculated and the work sequence is adjusted. Based on the adjusted work sequence, obtain the task execution progress of each leveling robot, update the workload data through the real-time monitoring system, and determine the execution effect of the task allocation; By combining performance data with status monitoring information, the dynamic adjustment needs of the leveling robot within the work area are continuously tracked to obtain the final task allocation and area division results.
5. The intelligent control method for collaborative operation of concrete leveling robots according to claim 1, characterized in that: The method for obtaining the unified work rhythm control parameters includes: Real-time position and work progress data are obtained from the leveling robot's sensors, and a standardized data stream is obtained using a timestamp synchronization mechanism; Standardized data streams are transmitted through a data interaction network, and a message queue protocol is used to ensure data consistency among multiple leveling robots. If the data consistency meets the preset threshold, the distribution of leveling robots is clustered using the K-means algorithm based on real-time location and work progress to obtain the grouping results of leveling robots; Based on the grouping results of the leveling robots, the task allocation strategy is dynamically adjusted, and resource allocation is optimized through linear programming algorithm to generate a task allocation scheme. The workload of each leveling robot is extracted from the task allocation scheme, and the operation rhythm trend is predicted by time series analysis to obtain the rhythm prediction result; By broadcasting rhythm prediction results through a collaborative network, the control parameters of each leveling robot are adjusted to generate unified operation rhythm control parameters. If the deviation of the control parameters exceeds the preset threshold, the deviation information is fed back through the data interaction network, the task allocation scheme is iteratively optimized, and the updated control parameters are obtained.
6. The intelligent control method for collaborative operation of concrete leveling robots according to claim 1, characterized in that: The method for determining the boundary processing priority and adjustment scheme includes: By obtaining the boundary data of adjacent leveling robot operation areas from the leveling robot log using the operation rhythm parameters, a set of boundary coordinates is obtained. The boundary elevation values between adjacent areas are calculated using the boundary coordinate set, resulting in an elevation data table; If the elevation deviation between adjacent areas in the elevation data table exceeds a preset threshold, the deviation correction value is calculated using a linear interpolation algorithm to obtain the corrected elevation set. Based on the priority of boundary connection analysis of the modified elevation set, a priority ranking table is obtained; The adjustment scheme is assigned by prioritizing the sorting table, and the boundary elevation connection is optimized by least squares method to obtain the set of adjustment parameters; Obtain the set of adjustment parameters, update the boundary data of the leveling robot's working area, and obtain the updated boundary coordinate set; The elevation deviations of adjacent areas are recalculated based on the updated boundary coordinate set to obtain the final elevation consistency data.
7. The intelligent control method for collaborative operation of concrete leveling robots according to claim 1, characterized in that: in, The optimized methods for obtaining the leveling robot's work path include: By obtaining work progress data from the leveling robot's operation log, a preliminary comparison is made of the current task completion status of each leveling robot to determine the progress difference value. Based on the progress difference value obtained from the comparison, a preset threshold is used for screening. If the difference value of a certain leveling robot exceeds the threshold, it is marked as a lagging leveling robot, and a list of lagging leveling robots is obtained. For the leveling robots in the list of lagging leveling robots, obtain their historical data interaction records, analyze the correlation between interaction frequency and task completion efficiency, and determine the potential impact range of frequency adjustment. Based on the analysis results, the data interaction frequency of the lagging leveling robot is dynamically adjusted, and a temporary task allocation scheme is generated in combination with the current task load. Based on the temporary task allocation scheme, a genetic algorithm is used to optimize the working path of the leveling robot, recalculate the task execution order of each leveling robot, and determine the improved path planning. The improved path planning data is obtained, the task allocation strategy is updated in real time, and synchronized to the control modules of each leveling robot to complete the dynamic adjustment process. By continuously monitoring the leveling robot's operation progress and path execution, and cyclically comparing the differences, if a new lagging leveling robot is found, a new round of frequency adjustment and path optimization process is triggered.
8. The intelligent control method for collaborative operation of concrete leveling robots according to claim 1, characterized in that: The method for obtaining the updated region boundary data includes: Real-time feedback information is obtained from the terrain data acquisition module. The terrain change characteristics during the execution of the leveling robot's work path are initially extracted. The change characteristics are classified according to the preset classification criteria to obtain the classification results of terrain changes. Based on the classification results of terrain changes, the scope of influence during path execution is analyzed. Combined with the current state of the leveling robot's working path, the deviation of the path execution is determined by comparing it with the preset terrain adaptation rules. If the deviation of the path execution exceeds the preset threshold, an adaptive assessment is performed on the regional division scheme, the distribution data of the current regional boundary is obtained, the impact of terrain change characteristics on the boundary distribution is analyzed, and the adjustment requirements for regional division are obtained. Based on the adjustment requirements of the regional division, combined with the terrain change characteristics and the actual situation of the leveling robot's operation path, a genetic algorithm is used to recalculate the regional boundaries and generate a preliminary boundary adjustment plan. For the preliminary boundary adjustment scheme, the historical records of the leveling robot path optimization are obtained, the matching degree between the adjustment scheme and the historical records is analyzed, and the feasibility of the scheme is judged by comparing it with the preset optimization rules, so as to obtain the verified boundary adjustment data. Based on the verified boundary adjustment data, the area division scheme is updated in real time and synchronized to the control module of the leveling robot's operation path to complete the dynamic matching of the path and the boundary, and generate updated area boundary information. By updating the area boundary information, the execution status of the leveling robot's work path is continuously monitored. The terrain change features in the real-time feedback information are compared cyclically. If a new deviation is found, a new round of adjustment process is triggered to determine the final path adaptation solution.
9. The intelligent control method for collaborative operation of concrete leveling robots according to claim 2, characterized in that: The method for determining the final elevation consistency control parameters includes: By obtaining the latest regional boundary information and real-time status data of the leveling robot collaboration from the data storage module, a dynamic model of regional division is constructed to obtain preliminary boundary division results. Based on the boundary division results, the operation rhythm distribution of the leveling robot in each area is analyzed, and the rhythm is compared using a preset threshold to determine the adjustment direction required for rhythm optimization. Based on the rhythm optimization direction and real-time status data, the coordination deviation between leveling robots is judged. If the deviation exceeds the preset range, the optimized coordination scheme of leveling robots is obtained by adjusting the operation rhythm parameters. The optimized leveling robot collaborative scheme was obtained, the degree of seamless connection between areas was analyzed, the connection boundary was smoothed using the information processing module, and the preliminary control parameters for elevation consistency were determined. Based on the preliminary control parameters and combined with the status monitoring data, the operation effect of the leveling robot after adjustment is analyzed. If the effect does not meet the preset standard, the control parameters are adjusted through iterative calculation to obtain the final elevation consistency parameters. By updating the collaborative operation instructions of the leveling robots based on the final elevation consistency parameters, the data update module sends the adjusted parameters to each leveling robot to complete the dynamic adjustment of rhythm optimization and area connection. Based on the results of dynamic adjustments, real-time status monitoring data is obtained. By comparing the data with historical data, the stability of the coordination between the area boundary and the leveling robot is determined, and a continuous optimization scheme for system operation is identified.
10. The intelligent control method for collaborative operation of concrete leveling robots according to claim 9, characterized in that: The method for obtaining the flatness distribution information includes: Using elevation consistency parameters, initial regional flatness data are obtained from each leveling robot. Outliers are removed by data cleaning methods to obtain a pre-processed flatness dataset. Based on the pre-processed flatness dataset, a deviation analysis is performed on the regional flatness distribution. If the analysis results show that the deviation exceeds the preset threshold, the local areas with deviations are marked, and the range of areas requiring secondary processing is determined. The marked local area data is obtained, and these areas are finely segmented using the region division method to generate detailed grid information of the local area. By using detailed grid information of local areas, a secondary processing method is used to adjust the elevation of each grid cell to obtain the adjusted local flatness data. Based on the adjusted local flatness data, the flatness distribution of the entire area is recalculated to generate updated flatness distribution information; The updated flatness distribution information is verified again using the deviation analysis method. If there are still areas where the deviation exceeds the preset threshold, the secondary processing flow is triggered repeatedly to obtain the final corrected flatness distribution dataset. By integrating elevation consistency parameters and work completion data with the final corrected flatness distribution dataset, a complete regional flatness analysis result is generated.
Citation Information
Patent Citations
Multi-scene multi-task multi-robot task allocation and collaborative operation method and system
CN117140524A
Robot grinding path optimization method based on industrial internet of things and MES linkage
CN120276401A
