Concrete construction digital intelligent robot control system

Abstract

The invention relates to the technical field of robot intelligent control, in particular to a concrete construction digital intelligent robot control system. Comprising a concrete data acquisition module, a construction path planning module, a robot control module, a quality monitoring module and an interactive feedback module. The data acquisition module acquires information such as concrete height and humidity of a construction area through a sensor; the path planning module generates an optimal construction path by using an artificial bee colony algorithm; the robot control module accurately executes a path instruction through sliding mode control and a fuzzy adaptive algorithm; the quality monitoring module adopts a support vector machine to evaluate construction quality and generate a report; and the interaction feedback module realizes path adjustment and information interaction through a visual page. Through combination of digital intelligence and algorithm optimization, integration of intelligent planning of a construction path, precise control of the robot and quality monitoring and evaluation is realized, and the intelligent level and engineering quality of concrete construction are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the field of robot intelligent control technology, and in particular to a digital intelligent robot control system for concrete construction. Background Art

[0002] With the deep integration of intelligent technology and the construction industry, the field of concrete construction is gradually moving towards automation and digitalization. In recent years, the rapid development of robotics has provided new possibilities for solving these problems. The concrete digital robot control system combines technologies such as artificial intelligence, big data, and the Internet of Things, and can monitor and control key parameters in the construction process in real time, thereby significantly improving construction efficiency and quality. Although the existing technology has made certain progress in the automation of concrete construction, the following problems still exist: the existing technology relies on manual or simple algorithms and cannot dynamically adapt to changes in the construction environment and material properties, resulting in low efficiency and waste of resources; the existing technology has poor control accuracy for robots, and it is difficult to cope with complex construction conditions through a single control algorithm, and it is easy to have action deviations, affecting the construction effect; manual operation and traditional machinery may cause uneven concrete surface, affecting the final ground quality, and may even affect later use. Summary of the invention

[0003] To solve the above problems, the present invention provides a digital robot control system for concrete construction, which solves the problem of how to enable the concrete construction robot to dynamically adapt to construction conditions in a complex environment, improve the accuracy of path planning and robot control, and ensure the automation and efficiency of construction quality, thereby improving the intelligence level of concrete construction and engineering quality.

[0004] To achieve the above object, the technical solution adopted by the present invention is:

[0005] The digital intelligent robot control system for concrete construction includes a concrete data acquisition module, a construction path planning module, a robot control module, a quality monitoring module and an interactive feedback module which are connected in sequence by communication;

[0006] The concrete data acquisition module is used to collect concrete data of the construction area through a sensor network;

[0007] The construction path planning module is used to construct a construction path planning model based on the concrete data using an artificial bee colony algorithm, generate an optimal path in combination with dynamic constraints of construction equipment, concrete hardening time characteristics and dynamic changes in the construction environment, and output an optimal path point set;

[0008] The robot control module is used to convert the optimal path point set into the robot's action instructions, drive the robot to execute the instruction operation through the sliding mode variable structure control algorithm combined with the fuzzy adaptive control algorithm, and monitor its position, posture and execution status in real time;

[0009] The quality monitoring module is used to monitor the quality of concrete after the robot construction is completed, and classify and evaluate the construction quality in combination with the support vector machine algorithm. At the same time, a regional analysis algorithm based on sparse feature clustering is used to perform local optimization evaluation on the flatness and uniformity of the construction surface to generate a construction quality report;

[0010] The interactive feedback module is used to exchange information with operators through a visual page, receive feedback information from the construction site, and provide robot planning adjustment suggestions and construction quality information.

[0011] Furthermore, the concrete data includes the height, flatness, moisture and material distribution of the concrete surface.

[0012] Furthermore, the operation process of the construction path planning module includes the following steps:

[0013] The construction area is gridded based on the collected concrete data, and the surface height, moisture and material distribution characteristics of each grid unit are extracted;

[0014] Generate dynamic constraints for path planning by combining concrete hardening time, equipment kinematic limitations, and construction environment changes;

[0015] Build a path planning model based on artificial bee colony algorithm, generate candidate paths by combining multi-objective optimization strategy, and select the optimal path through smoothing and local adjustment;

[0016] The optimal path is decomposed into a series of path point sets, and the optimal path point instructions suitable for robot execution are generated and transmitted to the robot control module.

[0017] Furthermore, the formula of the path planning model is as follows:

[0018]

[0019] in, represents the comprehensive optimization value of the construction path; P represents the set of construction path points; d(P i , P i+1 ) represents the i-th path point P i and the i+1th path point P i+1 The Euclidean distance of i represents the speed of the robot on the i-th path; E(P i , P i+1 ) represents the energy consumption on the i-th path; θ(P i-1 , P i , P i+1) represents the angle between the i-1th, i and i+1th path points; π represents a straight path; N represents the total number of path points; w1, w2 and w3 represent weight coefficients.

[0020] Furthermore, the operation process of the robot control module includes the following steps:

[0021] Receive and parse the optimal path point set output by the construction path planning module, extract key posture parameters and motion characteristic data, and generate an action instruction set suitable for the robot through the mapping relationship between the path points and the robot kinematic model;

[0022] Based on the action instruction set, a sliding mode variable structure control algorithm and a fuzzy adaptive control algorithm are used to calculate the dynamic motion of the robot between the path points and generate a robot control strategy;

[0023] Based on the robot control strategy, the robot is driven to execute the path instruction, and its position, posture and task execution status are monitored in real time through the built-in sensors;

[0024] Compare the actual execution status with the expected path points, dynamically correct the robot's motion trajectory, and optimize subsequent path instructions;

[0025] Monitor the robot's operating status and construction execution results in real time, and feed back status data to the interactive feedback module; when an abnormal situation is detected, trigger the automatic response mechanism to perform action correction or emergency processing, and generate a detailed abnormality report.

[0026] Furthermore, the formula of the robot control strategy is as follows:

[0027]

[0028] Where u(t) represents the robot control input signal; k s ·sign(σ) represents the sliding mode switching control; σ represents the sliding mode surface function, which is used to describe the tracking error of the construction path planning point; η·σ represents the sliding mode surface compensation; represents the error change rate adjustment item; Δx represents the plane position error between the current robot and the path point; Δθ represents the angular deviation between the current robot posture and the target posture of the path point; Δv represents the difference between the current robot speed and the target speed between the path points; Δω represents the difference between the current angular velocity of the robot and the target angular velocity between the path points; k s , η and k d represents the control gain parameter; μ f , μ a , α and β represent fuzzy control parameters.

[0029] Furthermore, the operation process of the quality monitoring module includes the following steps:

[0030] The embedded sensor network is used to collect the concrete surface data after the robot construction is completed, and the dynamic filtering algorithm is used to complete the data denoising and standardization processing;

[0031] Processing the concrete surface data by partitioning based on a sparse feature clustering algorithm, extracting surface flatness and material uniformity features, and identifying local abnormal areas and potential quality problems;

[0032] Use the support vector machine algorithm to classify, divide the quality levels and generate quality labels for specific areas, and combine the set standards to generate the grading results and optimization tips of the construction area quality;

[0033] Generate a construction quality report based on the analysis results, including a surface feature distribution map and a comprehensive quality score.

[0034] Furthermore, the quality monitoring module generates a three-dimensional quality distribution map of the concrete construction area in real time based on global point cloud data analysis technology.

[0035] Furthermore, the interactive feedback module includes a touch display screen and a voice interaction device, which are used to display the construction path planning results, robot status information and construction quality assessment report in real time, and support voice or manual input of construction adjustment suggestions.

[0036] The beneficial effects of the present invention are:

[0037] The present invention uses a sensor network to collect concrete data of the construction area in real time through a concrete data acquisition module, which can fully grasp the construction environment and material properties, and provide reliable basic data support for subsequent construction path planning and robot control. The construction path planning module adopts an artificial bee colony algorithm, combined with the concrete hardening time characteristics, the dynamic constraints of construction equipment and the real-time changes of the construction environment, to generate an optimal path point set to ensure the efficiency and accuracy of the construction path, thereby reducing construction time and resource waste. The robot control module converts the optimal path point set into the robot's action instructions, and adopts a control method that combines a sliding mode variable structure control algorithm with a fuzzy adaptive control algorithm to improve the accuracy and stability of the robot's execution of actions. At the same time, the real-time monitoring function ensures the high reliability of the robot during the construction process. The quality monitoring module classifies and evaluates the concrete quality after the construction is completed through a support vector machine algorithm, and combines a regional analysis algorithm based on sparse feature clustering to optimize and evaluate the flatness and uniformity of the construction surface, generate a detailed construction quality report, and ensure that the construction quality meets the expected standards. The interactive feedback module realizes intuitive interaction with operators through a visual interface. It can receive feedback information from the construction site in real time, and provide robot planning adjustment suggestions and construction quality information based on the construction situation to further optimize the construction process. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 It is a module schematic diagram of the concrete construction digital intelligent robot control system of the present invention.

[0039] Figure 2 It is a flowchart of the operation process of the construction path planning module provided by an embodiment of the present invention.

[0040] Figure 3 It is a flowchart of the operation process of the robot control module provided by one embodiment of the present invention. DETAILED DESCRIPTION

[0041] See also Figure 1-3 As shown, the present invention relates to a digital intelligent robot control system for concrete construction.

[0042] Example

[0043] The digital intelligent robot control system for concrete construction includes a concrete data acquisition module, a construction path planning module, a robot control module, a quality monitoring module and an interactive feedback module which are connected in sequence by communication;

[0044] The concrete data acquisition module is used to collect concrete data in the construction area through a sensor network; the concrete data includes the height, flatness, humidity and material distribution of the concrete surface.

[0045] It should be noted that the concrete data acquisition module mainly uses distributed sensor networks, edge computing devices and data preprocessing algorithms to achieve multi-dimensional and high-precision data acquisition in the construction area, covering key information such as the height, flatness, humidity and material distribution of the concrete surface. The module hardware components are as follows:

[0046] Height sensor: A combination of laser ranging sensors (such as LiDAR) and ultrasonic sensors can measure the height difference of the concrete surface with millimeter-level accuracy (accurate to 0.1 mm). The height sensors are arranged in an array to cover the construction area and dynamically record surface changes.

[0047] Humidity sensor: integrated capacitive and conductive humidity sensors are placed on the construction robot or fixed bracket to collect humidity data on the concrete surface. The sensor supports multi-point measurement and real-time monitoring of humidity uniformity.

[0048] Optical scanner: Based on laser scanning or structured light technology (such as blue light scanning), it captures 3D point cloud data of the construction surface. Through the linkage of high-definition cameras and scanning systems, material distribution can be further analyzed.

[0049] Material Analyzer: Use a portable X-ray fluorescence spectrometer (XRF) to perform non-destructive testing of concrete components and identify the distribution ratio of sand, stone and cement.

[0050] Edge computing devices: deployed near the data collection terminal, perform preliminary processing on the data collected by the sensor in real time, including denoising, filtering, formatting and other operations, to reduce the burden on the central processing system.

[0051] In one embodiment, the robot device is controlled by an STM32F4 series single-chip microcomputer; the motor drive uses the CANOPEN protocol; the laser reception uses the CAN protocol; the body drive uses a single 1500W motor rear drive, and the speed uses software-defined 8-level speed change; the steering power uses a servo motor to accurately control the direction.

[0052] Specifically, after the system is started, the sensor network performs self-calibration, such as the height sensor calibrating its measurement benchmark through laser positioning. The optical scanner establishes an initial three-dimensional model of the construction area through pre-scanning. The sensor obtains the height difference of the concrete surface through laser scanning, generates two-dimensional grid data, and transmits it to the edge computing device in real time. The optical scanner combines three-dimensional point cloud technology to extract the flatness characteristics of the concrete surface and generate a thermal map for subsequent analysis. The humidity sensor records the surface humidity changes at fixed time intervals (such as 1 second) to form a humidity distribution curve. The material analyzer regularly samples the concrete composition of different areas and converts the sampled data into a material distribution matrix. Through the multi-sensor data fusion algorithm, combined with height, flatness, humidity and material distribution, a unified multi-dimensional construction area data model is generated. Wavelet denoising technology is used to eliminate measurement noise, and the Kalman filter algorithm is used to smooth dynamically changing data to ensure data quality. The processed data is stored in a unified format (such as JSON or HDF5) for the construction path planning module to call.

[0053] The construction path planning module is used to construct a construction path planning model based on the concrete data using an artificial bee colony algorithm, generate an optimal path in combination with dynamic constraints of construction equipment, concrete hardening time characteristics and dynamic changes in the construction environment, and output an optimal path point set;

[0054] The operation process of the construction path planning module includes the following steps:

[0055] The construction area is gridded based on the collected concrete data, and the surface height, moisture and material distribution characteristics of each grid unit are extracted;

[0056] Specifically, the construction area is divided into grids according to fixed sizes (such as 0.1 m × 0.1 m) to generate a two-dimensional or three-dimensional grid model. The size of the grid unit is dynamically adjusted according to the construction accuracy requirements. The unique ID of each grid unit is used to identify its spatial position.

[0057] The features are extracted as follows:

[0058] Surface height: Based on the laser ranging data, the average height and height difference of each grid cell are calculated to generate a height distribution map.

[0059] Humidity: Using the data from the humidity sensor, calculate the average humidity of each grid and mark areas with abnormal humidity (such as too dry or too wet).

[0060] Material distribution: Combine optical scanning and material analyzer to extract the proportion of sand, gravel and cement in the grid and generate a material distribution matrix.

[0061] The feature data of each grid cell is formatted and stored (such as JSON or matrix form) for subsequent path planning model calls.

[0062] Generate dynamic constraints for path planning by combining concrete hardening time, equipment kinematic limitations, and construction environment changes;

[0063] Specifically, based on real-time humidity and temperature data, a concrete hardening time model (such as the Arrhenius model) is used to predict the hardening speed of the construction area. The hardening time is divided into different stages (such as the initial setting period and the final setting period), and different construction priorities are assigned to each stage. The kinematic parameters of the construction robot (such as the maximum speed, the minimum turning radius, and the motion stability limit) are extracted to generate a dynamic kinematic model of the equipment. The robot's turning angle and linear motion distance are limited in path planning to avoid unreasonable trajectory planning. Environmental parameters such as wind speed, temperature, and humidity are monitored in real time, and path planning conditions are dynamically adjusted (such as avoiding high wind speed areas). Environmental changes are introduced into the dynamic weight model to provide a basis for path optimization.

[0064] Build a path planning model based on artificial bee colony algorithm, generate candidate paths by combining multi-objective optimization strategy, and select the optimal path through smoothing and local adjustment;

[0065] Specifically, the grid cells are used as the search space and the swarm is initialized into multiple random paths, each of which consists of a series of grid cells.

[0066] Fitness function design: path smoothness (minimizing path turning angles and path length); construction priority (preferentially covering areas with short hardening time); environmental adaptability (avoiding areas with poor environmental parameters). A comprehensive weight function is used to evaluate the fitness of the path.

[0067] Use the Pareto optimal solution to generate multiple candidate paths to meet the optimization requirements under different weights. In each iteration, a combination of random search and neighborhood search is used to update the path to improve fitness. Smooth the candidate path and use B-spline curve to fit the path to ensure the continuity and smoothness of the path. Correct the unreasonable local path points to avoid path deviation due to environmental changes or dynamic constraints.

[0068] The optimal path is decomposed into a series of path point sets, and the optimal path point instructions suitable for robot execution are generated and transmitted to the robot control module.

[0069] Specifically, the optimal path is decomposed into a series of discrete path points, each of which contains position information (such as X, Y, Z coordinates), timestamp and execution status. Ensure that the intervals between path points meet the motion accuracy requirements of the construction robot (such as 1-3 mm). Based on the inverse kinematics algorithm, the robot joint angles and dynamic parameters corresponding to each path point are calculated. Combined with the sliding mode variable structure control algorithm, specific execution instructions (such as moving speed, dwell time, etc.) are generated for the path points. The path point set and instructions are transmitted to the robot control module through the wireless communication module in a unified protocol (such as ROS or a custom protocol). At the same time, the path point priority and dynamic adjustment identifier are attached to facilitate the robot to optimize the execution strategy in real time.

[0070] Furthermore, the formula of the path planning model is as follows:

[0071]

[0072] in, represents the comprehensive optimization value of the construction path; P represents the set of construction path points; d(P i , P i+1 ) represents the i-th path point P i and the i+1th path point P i+1 The Euclidean distance is used to measure the path length; v i represents the movement speed of the robot on the i-th path, which is affected by the concrete hardening time characteristics, the dynamic constraints of the construction equipment and the dynamic changes of the construction environment; E(P i , P i+1 ) represents the energy consumption on the i-th path; θ(P i-1 , P i , P i+1 ) represents the angle between the i-1th, i and i+1th path points, which is used to evaluate the path smoothness; π represents a straight path; N represents the total number of path points; w1, w2 and w3 represent weight coefficients, which respectively represent the influence of path time, energy consumption and smoothness on the objective function.

[0073] The robot control module is used to convert the optimal path point set into the robot's action instructions, drive the robot to execute the instruction operation through the sliding mode variable structure control algorithm combined with the fuzzy adaptive control algorithm, and monitor its position, posture and execution status in real time;

[0074] The operation process of the robot control module includes the following steps:

[0075] Receive and parse the optimal path point set output by the construction path planning module, extract key posture parameters and motion characteristic data, and generate an action instruction set suitable for the robot through the mapping relationship between the path points and the robot kinematic model;

[0076] Specifically, the optimal path point set output by the path planning module is received, including the coordinates (X, Y, Z), timestamp, attitude angle (Pitch, Roll, Yaw) and other task characteristics (such as speed requirements) of each path point. The path point data is formatted using a parser to ensure that the data is compatible with the robot motion model.

[0077] The key parameters for extraction are as follows:

[0078] Pose parameters: Extract the position information (X, Y, Z) and posture information (Pitch, Roll, Yaw) of the path points.

[0079] Motion characteristics: Extract velocity, acceleration, and dynamic task requirements (such as maintaining a fixed humidity) at waypoints.

[0080] Based on the robot kinematic model, the path points are mapped to robot joint motion instructions (such as joint angle and rotation speed). The motion instruction set includes the robot's position target, moving speed, dwell time and specific construction operations (such as pouring or leveling) at each path point.

[0081] Based on the action instruction set, a sliding mode variable structure control algorithm and a fuzzy adaptive control algorithm are used to calculate the dynamic motion of the robot between the path points and generate a robot control strategy;

[0082] Specifically, according to the position and posture of the target path point and the current state of the robot, the error function of the sliding mode control is defined to ensure that the error converges on the sliding surface. When the robot path deviates or is disturbed by the environment, the trajectory is adjusted by switching the sliding mode control law to maintain high-precision execution. The sliding mode control is highly robust to uncertainties (such as vibration in the construction environment) and can ensure stable operation of the robot.

[0083] Design fuzzy control rules to dynamically adjust control parameters based on input errors (such as path deviation, attitude angle error) and error change rate. Adjust the sliding mode control gain through fuzzy reasoning to optimize the controller response speed and accuracy to adapt to different construction scenarios. The dynamic adjustment of sliding mode parameters by fuzzy adaptive control compensates for the inherent chattering problem of sliding mode control and improves control smoothness.

[0084] Sliding mode control and fuzzy adaptive control are integrated to generate the dynamic motion strategy of the robot between path points, including target position, joint movements, speed and posture changes.

[0085] Based on the robot control strategy, the robot is driven to execute the path instruction, and its position, posture and task execution status are monitored in real time through the built-in sensors;

[0086] Compare the actual execution status with the expected path points, dynamically correct the robot's motion trajectory, and optimize subsequent path instructions;

[0087] Monitor the robot's operating status and construction execution results in real time, and feed back status data to the interactive feedback module; when an abnormal situation is detected, trigger the automatic response mechanism to perform action correction or emergency processing, and generate a detailed abnormality report.

[0088] Furthermore, the formula of the robot control strategy is as follows:

[0089]

[0090] Where u(t) represents the robot control input signal; k s ·sign(σ) represents the sliding mode switching control, which is used to quickly adjust the robot's movements that deviate from the path points and adapt to dynamic construction constraints; σ represents the sliding mode surface function, which is used to describe the tracking error of the construction path planning points; η·σ represents the sliding mode surface compensation, which is used to correct the deviation between the actual path of the robot and the planned path points to improve the path tracking accuracy; represents the error change rate adjustment term, combining the concrete hardening time characteristics and the equipment kinematic limitations; Δx represents the plane position error between the current robot and the path point; Δθ represents the angular deviation between the current robot posture and the target posture of the path point; Δv represents the difference between the current robot speed and the target speed between the path points; Δω represents the difference between the current angular velocity of the robot and the target angular velocity between the path points; k s , η and k d represents the control gain parameter; μ f , μ a , α and β represent fuzzy control parameters.

[0091] The quality monitoring module is used to monitor the quality of concrete after the robot construction is completed, and classify and evaluate the construction quality in combination with the support vector machine algorithm. At the same time, a regional analysis algorithm based on sparse feature clustering is used to perform local optimization evaluation on the flatness and uniformity of the construction surface to generate a construction quality report; the quality monitoring module is also specifically based on the global point cloud data analysis technology to generate a three-dimensional quality distribution map of the concrete construction area in real time.

[0092] The operation process of the quality monitoring module includes the following steps:

[0093] The embedded sensor network is used to collect the concrete surface data after the robot construction is completed, and the dynamic filtering algorithm is used to complete the data denoising and standardization processing;

[0094] Processing the concrete surface data by partitioning based on a sparse feature clustering algorithm, extracting surface flatness and material uniformity features, and identifying local abnormal areas and potential quality problems;

[0095] Specifically, the collected point cloud data or mesh data is divided into multiple regions, and each region is assigned a feature label according to surface characteristics (such as height, material distribution). Sparse feature clustering identifies areas with significant surface changes by constraining the adjacency of feature points and local feature differences (such as height gradients). The standard deviation, gradient direction, and local curvature of the height data in each partition are calculated to generate a flatness heat map. Statistical analysis is performed on the material distribution data to extract the distribution mean and variance of components such as sand, gravel, and cement to evaluate the uniformity of the material. The clustering results are compared with the preset standards to identify potential quality problem areas (such as excessive height and uneven material distribution). The location and characteristics of abnormal areas are marked as the focus of subsequent optimization.

[0096] Use the support vector machine algorithm to classify, divide the quality levels and generate quality labels for specific areas, and combine the set standards to generate the grading results and optimization tips of the construction area quality;

[0097] Specifically, a large amount of annotated construction quality data (such as A, B, and C grades) is used to train the support vector machine model offline. The model input includes multi-dimensional features such as flatness, uniformity, and material distribution, and the output is a quality grade label. The feature data of the current partition is input into the support vector machine model to predict the quality grade in real time. Generate a specific quality label for each partition (such as grade A: high quality, grade B: qualified, grade C: needs to be optimized). Combined with the global construction standards (such as the flatness or material uniformity requirements in the national standard GB50010-2010), calculate the overall quality classification of the construction area. According to the partition quality grade, generate targeted optimization suggestions (such as increasing wetting or re-leveling for the C-level area). Associate the optimization tips with the partition position for subsequent construction adjustments.

[0098] Generate a construction quality report based on the analysis results, including a surface feature distribution map and a comprehensive quality score.

[0099] It should be noted that the report content is generated as follows:

[0100] Surface feature distribution map:

[0101] Flatness distribution map: Use color gradient (such as green-excellent, red-poor) to show the flatness of the construction area.

[0102] Material Uniformity Distribution Diagram: Displays material distribution characteristics through histograms or heat maps.

[0103] Quality grading information: Lists the quality grade (A / B / C) of each partition and summarizes the overall construction quality grading. Additional optimization tips and improvement suggestions are provided for low-quality areas.

[0104] Comprehensive quality score: Based on the zoning level and area weight, a comprehensive score (e.g. 100 points) is calculated. The score result is compared with the industry standard to intuitively evaluate whether the construction effect meets the requirements.

[0105] The interactive feedback module is used to interact with the operator through a visual page, receive feedback information from the construction site, and provide robot planning adjustment suggestions and construction quality information; the interactive feedback module includes a touch display screen and a voice interaction device, which is used to display the construction path planning results, robot status information and construction quality assessment reports in real time, and supports voice or manual input of construction adjustment suggestions.

[0106] Specifically, multi-level and dynamic information display is achieved through the touch screen, as follows:

[0107] Path point collection: The optimal path is displayed in the form of a three-dimensional model, and the path points are marked with dynamic graphics to clearly show the starting point, end point and the location of each key point.

[0108] Real-time adjustment: Operators can modify the path directly through the touch screen, such as adding new waypoints or adjusting the positions of existing waypoints. The system will re-plan and display the updated path in real time.

[0109] Basic status: Real-time display of robot position, posture, speed, battery level and other information.

[0110] Execution status: Displays the completion status of the current construction task in the form of a progress bar, and marks abnormal status (such as deviation from the path, task delay, etc.).

[0111] Construction quality assessment report: Displays the concrete construction quality grade (such as A, B, C) in the form of a chart, and provides the score details for each area (such as flatness, moisture uniformity, etc.). Displays the construction area quality distribution map generated based on the 3D point cloud in real time, and uses color gradients to show the quality differences between each area (such as green for qualified and red for need for optimization).

[0112] The operator can adjust the robot's operation through voice commands (such as "pause construction" and "replan the path"), and the system will feedback confirmation information through the voice assistant. Adjustment suggestions can be directly input through the touch screen, such as marking problem areas in the construction area or drawing new path planning. Handwriting and stylus operations are supported. According to the real-time monitored quality data and robot status, the module automatically generates construction adjustment suggestions, such as "increase humidity compensation" and "optimize the path to reduce deviation". Optimization suggestions are pushed to the screen in the form of pop-up windows, and voice broadcast reminders are supported. Operators can assign construction tasks to different robots through the screen and track their completion in real time. Support task priority setting to ensure that important tasks are executed first.

[0113] In summary, the present invention can collect multi-dimensional data such as the height, flatness, humidity and material distribution of the concrete surface in real time through a distributed sensor network (laser rangefinder, humidity sensor, optical scanner, material analyzer, etc.) and edge computing equipment, and perform efficient data preprocessing to ensure the high accuracy and reliability of construction data, and provide an accurate basis for subsequent construction planning and quality monitoring. The artificial bee colony algorithm is combined with a multi-objective optimization strategy to generate the optimal path planning through dynamic constraints (such as concrete hardening time, equipment kinematic limitations, environmental changes, etc.) to achieve path smoothness, priority coverage and energy consumption minimization. The path can be dynamically adjusted to adapt to the real-time changing construction environment, significantly improving the efficiency and accuracy of robot construction.

[0114] The present invention combines the sliding mode variable structure control algorithm with the fuzzy adaptive control algorithm, and the system realizes high-precision and dynamic adjustment of the robot path tracking, which can effectively deal with the uncertainty in the construction environment (such as vibration, path deviation, etc.), and ensure the stability and accuracy of the robot's action. Using the sparse feature clustering algorithm and the support vector machine algorithm, the system realizes the comprehensive monitoring and graded evaluation of the concrete construction quality, and can generate distribution feature maps such as flatness and material uniformity, and identify potential problem areas. Targeted optimization suggestions are provided through the quality assessment report to ensure that the construction effect meets industry standards. The interactive feedback module displays the construction path, robot status and construction quality assessment results in real time through the touch display and voice interaction device. It supports operators to make real-time adjustments through touch or voice, enhances the operability and flexibility of the system, and significantly improves the user experience through the visual interface.

[0115] The above implementation modes are merely descriptions of the preferred implementation modes of the present invention, and are not intended to limit the scope of the present invention. Without departing from the design spirit of the present invention, various modifications and improvements made to the technical solutions of the present invention by ordinary engineering and technical personnel in the field shall fall within the protection scope determined by the claims of the present invention.

Claims

1. The digital robot control system for concrete construction is characterized by: It includes a concrete data acquisition module, a construction path planning module, a robot control module, a quality monitoring module and an interactive feedback module which are sequentially connected in communication; The concrete data acquisition module is used to collect concrete data of the construction area through a sensor network; The construction path planning module is used to construct a construction path planning model based on the concrete data using an artificial bee colony algorithm, generate an optimal path in combination with dynamic constraints of construction equipment, concrete hardening time characteristics and dynamic changes in the construction environment, and output an optimal path point set; The robot control module is used to convert the optimal path point set into the robot's action instructions, drive the robot to execute the instruction operation through the sliding mode variable structure control algorithm combined with the fuzzy adaptive control algorithm, and monitor its position, posture and execution status in real time; The quality monitoring module is used to monitor the quality of concrete after the robot construction is completed, and classify and evaluate the construction quality in combination with the support vector machine algorithm. At the same time, a regional analysis algorithm based on sparse feature clustering is used to perform local optimization evaluation on the flatness and uniformity of the construction surface to generate a construction quality report; The interactive feedback module is used to exchange information with operators through a visual page, receive feedback information from the construction site, and provide robot planning adjustment suggestions and construction quality information.

2. The concrete construction digital intelligent robot control system according to claim 1 is characterized in that: The concrete data includes the height, flatness, moisture and material distribution of the concrete surface.

3. The concrete construction digital intelligent robot control system according to claim 1 is characterized in that: The operation process of the construction path planning module includes the following steps: The construction area is gridded based on the collected concrete data, and the surface height, moisture and material distribution characteristics of each grid unit are extracted; Generate dynamic constraints for path planning by combining concrete hardening time, equipment kinematic limitations, and construction environment changes; Build a path planning model based on artificial bee colony algorithm, generate candidate paths by combining multi-objective optimization strategy, and select the optimal path through smoothing and local adjustment; The optimal path is decomposed into a series of path point sets, and the optimal path point instructions suitable for robot execution are generated and transmitted to the robot control module.

4. The concrete construction digital intelligent robot control system according to claim 3 is characterized in that: The formula of the path planning model is as follows: in, represents the comprehensive optimization value of the construction path; P represents the set of construction path points; d(P i , P i+1 ) represents the i-th path point P i and the i+1th path point P i+1 The Euclidean distance of i represents the speed of the robot on the i-th path; E(P i , P i+1 ) represents the energy consumption on the i-th path; θ(P i-1 , P i , P i+1 ) represents the angle between the i-1th, i and i+1th path points; π represents a straight path; N represents the total number of path points; w1, w2 and w3 represent weight coefficients.

5. The concrete construction digital intelligent robot control system according to claim 1 is characterized in that: The operation process of the robot control module includes the following steps: Receive and parse the optimal path point set output by the construction path planning module, extract key posture parameters and motion characteristic data, and generate an action instruction set suitable for the robot through the mapping relationship between the path points and the robot kinematic model; Based on the action instruction set, a sliding mode variable structure control algorithm and a fuzzy adaptive control algorithm are used to calculate the dynamic motion of the robot between the path points and generate a robot control strategy; Based on the robot control strategy, the robot is driven to execute the path instruction, and its position, posture and task execution status are monitored in real time through the built-in sensors; Compare the actual execution status with the expected path points, dynamically correct the robot's motion trajectory, and optimize subsequent path instructions; Monitor the robot's operating status and construction execution results in real time, and feed back status data to the interactive feedback module; when an abnormal situation is detected, trigger the automatic response mechanism to perform action correction or emergency processing, and generate a detailed abnormality report.

6. The concrete construction digital intelligent robot control system according to claim 5 is characterized in that: The formula of the robot control strategy is as follows: Where u(t) represents the robot control input signal; k s ·sign(σ) represents the sliding mode switching control; σ represents the sliding mode surface function, which is used to describe the tracking error of the construction path planning point; η·σ represents the sliding mode surface compensation; represents the error change rate adjustment item; Δx represents the plane position error between the current robot and the path point; Δθ represents the angular deviation between the current robot posture and the target posture of the path point; Δv represents the difference between the current robot speed and the target speed between the path points; Δω represents the difference between the current angular velocity of the robot and the target angular velocity between the path points; k s , η and k d represents the control gain parameter; μ f , μ a , α and β represent fuzzy control parameters.

7. The concrete construction digital intelligent robot control system according to claim 1 is characterized in that: The operation process of the quality monitoring module includes the following steps: The embedded sensor network is used to collect the concrete surface data after the robot construction is completed, and the dynamic filtering algorithm is used to complete the data denoising and standardization processing; Processing the concrete surface data by partitioning based on a sparse feature clustering algorithm, extracting surface flatness and material uniformity features, and identifying local abnormal areas and potential quality problems; Use the support vector machine algorithm to classify, divide the quality levels and generate quality labels for specific areas, and combine the set standards to generate the grading results and optimization tips of the construction area quality; Generate a construction quality report based on the analysis results, including a surface feature distribution map and a comprehensive quality score.

8. The concrete construction digital intelligent robot control system according to claim 1 is characterized in that: The quality monitoring module is specifically based on global point cloud data analysis technology to generate a three-dimensional quality distribution map of the concrete construction area in real time.

9. The concrete construction digital intelligent robot control system according to claim 1 is characterized in that: The interactive feedback module includes a touch display screen and a voice interaction device, which is used to display the construction path planning results, robot status information and construction quality assessment report in real time, and supports voice or manual input of construction adjustment suggestions.

Images