A tower crane cluster cooperative control method and system for a smart construction site and a medium

Abstract

The application discloses a tower crane cluster cooperative control method and system for a smart construction site, and a medium, and belongs to the technical field of tower crane control, and is used for solving the technical problem that the cluster control among existing multiple tower cranes mainly relies on manual control, is prone to cause great safety hazards in complex tower crane operation, and is difficult to realize intelligent cooperative control of the work among multiple tower cranes. The method comprises the following steps: performing trajectory error prediction of a tower crane rope on the current dynamic environment characteristics in the current construction area to obtain wind resistance-swing angle coupling error data; performing trajectory intersection overlap judgment of the tower crane rope operation trajectory of a tower crane to be executed in multiple tower cranes in a future time period, predicting and determining an abnormal tower crane operation trajectory diagram; performing optimal path generation processing on the abnormal tower crane operation trajectory diagram in a federated learning framework to obtain an optimized tower crane operation trajectory diagram; and performing trajectory avoidance execution processing on a risk tower crane with trajectory intersection risk to obtain a tower crane avoidance strategy of the risk tower crane.

Description

TECHNICAL FIELD

[0001] The application relates to the field of tower crane control, in particular to a tower crane cluster cooperative control method and system for a smart construction site and a medium. BACKGROUND

[0002] With the rapid increase of super high-rise buildings and large infrastructure projects, it has become a common practice to densely deploy tower cranes (>=8) in single construction sites. Statistics from the Ministry of Housing and Urban-Rural Development show that 67% of tower crane collision accidents occur in multi-tower intersection operation areas, and high-altitude falling accidents caused by "dynamic blind area" account for as high as 81%.

[0003] In the traditional manual scheduling mode, the main operation process is to rely on intercom communication and experience prediction. The defects are: (1) the visual range of the human eye only covers the area around the tower crane; (2) when multiple tower crane drivers observe the same area at the same time, the visual angle overlap rate is relatively small; (3) when the sudden hoisting object swings, the manual response delay time is long.

[0004] Moreover, the real-time anti-collision system based on UWB, that is, using tower crane installation positioning base station, real-time measurement of tower crane spacing method, still has some limitations, and can only trigger an alarm 1-2 seconds before collision. In the static protection logic, it is impossible to predict the motion trajectory of the hoisting object (when the steel cable swing angle is >15°, the actual collision range is expanded by 2.3 times). At the same time, when multiple tower cranes enter the warning zone at the same time, the false alarm rate is relatively high, and it is difficult to truly achieve scientific tower crane cluster cooperative control. SUMMARY

[0005] The application embodiment provides a tower crane cluster cooperative control method and system for a smart construction site and a medium, which is used to solve the following technical problems: the existing cluster control between multiple tower cranes mainly relies on manual control, and is assisted by simple collision warning, which is easy to cause great safety hazards in complex tower crane operation, and it is difficult to realize intelligent cooperative control of the work between multiple tower cranes.

[0006] The application embodiment adopts the following technical scheme:

[0007] In one aspect, the embodiment of the present application provides a tower crane cluster cooperative control method for a smart construction site, comprising: predicting a tower crane rope trajectory error of a current dynamic environment feature in a current construction area by an environmental monitoring sensor, to obtain wind resistance-swing angle coupling error data; performing trajectory intersection overlap judgment of a tower crane rope operation trajectory of a tower crane to be executed in a future time period according to the wind resistance-swing angle coupling error data, to predict and determine an abnormal tower crane operation trajectory graph; performing optimal path generation processing of the abnormal tower crane operation trajectory graph based on a federated learning framework, to obtain an optimized tower crane operation trajectory graph; performing trajectory avoidance execution processing of a risk tower crane with trajectory intersection risk according to the optimized tower crane operation trajectory graph, to obtain a tower crane avoidance strategy of the risk tower crane; and performing correlation feedback of the tower crane avoidance strategy to multiple tower cranes by a preset cloud cooperative management platform, to generate a tower crane cluster cooperative control strategy in the future time period.

[0008] The embodiment of the present application can identify potential trajectory intersection and abnormal conditions in advance by predicting the trajectory error of the tower crane rope, thereby avoiding construction accidents and improving the safety of the construction site. The trajectory intersection overlap in the future time period can be predicted to plan the operation trajectory of the tower crane in advance, reduce waiting and adjustment time, and improve construction efficiency. Based on the optimized tower crane operation trajectory, unnecessary movement and energy consumption can be reduced, which helps to save energy and reduce emissions. Moreover, the federated learning framework is used for optimal path generation processing, which embodies the characteristics of intelligent management and helps to improve the intelligent level of construction management. At the same time, real-time information sharing and cooperative control between multiple tower cranes can be realized, which improves the cooperativeness and response speed of construction. Moreover, the automatic trajectory intersection overlap judgment and avoidance strategy generation can reduce the dependence on manual work and reduce the error rate of manual intervention, and can ensure the accuracy of tower crane operation, thereby improving the construction quality.

[0009] In a feasible implementation, before the trajectory error prediction of the tower crane rope in the current dynamic environment characteristics in the current construction area by the environmental monitoring sensor, the wind resistance-swing angle coupling error data is obtained, the method further comprises: deploying an ultrasonic anemometer at the tower crane arm end of the tower crane and collecting wind data; wherein the wind data comprises: wind speed data and wind direction data; deploying a MEMS tilt sensor at the hook connecting shaft of the tower crane and measuring dynamic angle data of the hoisted object in real time; wherein the dynamic angle data of the hoisted object comprises: pitch angle data and yaw angle data; deploying a cable tension sensor to the fixed end of the hoisting steel wire rope of the tower crane and collecting rope dynamic tension data; deploying a temperature and humidity sensor to the top of the tower crane cab of the tower crane and collecting air density parameter data; determining the wind data, the dynamic angle data of the hoisted object, the dynamic tension data of the rope and the air density parameter data as the tower crane external environment data.

[0010] In a feasible implementation, the trajectory error prediction of the tower crane rope in the current dynamic environment characteristics in the current construction area by the environmental monitoring sensor, to obtain the wind resistance-swing angle coupling error data, specifically includes: through the PTP precise clock protocol, the time stamp alignment processing of the multiple types of collected data sets in the tower crane external environment data is carried out, and based on the Kalman filter denoising processing, the filtered tower crane external data is obtained; the wind speed vector decomposition processing of the wind data in the filtered tower crane external data is carried out, and the hoisted object coordinate system is obtained; based on the swing angle, wind resistance of the wind data in the filtered tower crane external data and the air density compensation of the air density parameter data, the hoisted object swing angle dynamics function is generated; through XGBoost, the wind speed, wind direction angle, swing angle, swing speed, cable length and hoisting weight mass in the filtered tower crane external data are determined as the model input features, and based on the hoisted object swing angle dynamics function, the actual measured swing angle and the physical model calculated swing angle are calculated, and the model target value is obtained; according to the model input features and the model target value, the wind resistance term and the swing angle term are residual compensated and trained, and the wind resistance-swing angle data driven fusion model is generated; inputting the current tower crane external environment data into the wind resistance-swing angle data driven fusion model, predicting and obtaining the final predicted swing angle based on the prediction data driven compensation; through the final predicted swing angle and the cable lifting trajectory, the horizontal offset and the vertical offset of the hoisted object are calculated, and the wind resistance-swing angle coupling error data based on the standard cable lifting trajectory is obtained.

[0011] In a feasible implementation, according to the wind resistance-swing angle coupling error data, the tower crane rope running track of the tower crane to be executed in the tower crane task in the plurality of tower cranes is subjected to track intersection judgment in a future time period, an abnormal tower crane running track graph is predicted and determined, and specifically includes: based on the BIM building information of the current construction area and the position information of each tower crane, a global coordinate system graph based on time synchronization processing is generated; wherein the global coordinate system graph contains a local coordinate system graph of each tower crane; according to the preset UWB positioning base station unit and the hook inertial navigation unit, and based on the future time period, the tower crane rope running track in the local coordinate system graph is subjected to non-disturbance generation processing, and the ideal tower crane rope running track graph of each tower crane at the same time is obtained; through the wind resistance-swing angle coupling error data, the coordinate system connecting line in the ideal tower crane rope running track graph of each tower crane is subjected to deviation compensation, and the deviation coordinate system connecting line predicted in the future time period is obtained; wherein the deviation coordinate system connecting line is the tower crane rope running prediction track of the local coordinate system graph of each tower crane; according to the global space-time grid corresponding to the global coordinate system graph, the deviation coordinate system connecting line of each tower crane local coordinate system graph is subjected to track voxel filling calculation about the safety margin of the hook influence sphere radius, and the current voxel of each deviation coordinate system connecting line is obtained; according to the time slice index, the current voxel of each deviation coordinate system connecting line is subjected to occupation coincidence judgment of the heat map generated by the conflict voxel clustering, and the abnormal deviation coordinate connecting line is determined; based on the abnormal deviation coordinate connecting line, the abnormal tower crane with hoisting risk and the corresponding abnormal tower crane running track graph are determined.

[0012] In a feasible implementation, the abnormal tower crane operation trajectory graph is subjected to optimal path generation processing based on a federal learning framework to obtain an optimized tower crane operation trajectory graph, specifically including: identifying the execution time stamp of the abnormal tower crane involved in the abnormal tower crane operation trajectory graph in relation to the to-be-executed tower crane task, determining the task start time node of each tower crane; and determining the latest start time node; extracting the conflict voxel clustering trajectory feature of the abnormal tower crane operation trajectory graph corresponding to the latest start time node, determining the abnormal trajectory feature and the corresponding abnormal spatio-temporal feature; wherein the abnormal spatio-temporal feature is the position coordinate feature of the conflict region in the conflict voxel clustering trajectory feature and the predicted operation time feature; guiding the decision through the federal model under the pre-trained federal learning framework, performing trajectory constraint processing on the abnormal trajectory feature under the target cost function to obtain a constraint execution action; wherein the target cost function includes: time cost, energy consumption cost and risk cost; the constraint execution action includes: time lag constraint, hook cargo type constraint and hoisting speed constraint; performing execution judgment based on the safety level of the constraint execution action to determine the optimal constraint execution action; according to the optimal constraint execution action, performing constraint processing on the deviation coordinate connection in the local coordinate system graph under the highest safety level to obtain the optimized deviation coordinate system connection in the future time period; based on the optimized deviation coordinate system connection, obtaining the optimized tower crane operation trajectory graph of the tower crane corresponding to the latest start time node.

[0013] In a feasible implementation, according to the optimized tower crane operation trajectory graph, the risk tower crane with trajectory intersection risk is subjected to trajectory avoidance execution processing to obtain the tower crane avoidance strategy of the risk tower crane, specifically including: determining the tower crane corresponding to the optimized tower crane operation trajectory graph as the risk tower crane with trajectory intersection risk; extracting the optimal constraint execution action in the optimized tower crane operation trajectory graph; modifying the control item in the to-be-executed tower crane task of the risk tower crane through the data control parameter in the optimal constraint execution action to generate a trajectory avoidance execution action; based on the trajectory avoidance execution action, generating the tower crane avoidance strategy of the risk tower crane.

[0014] In an implementable embodiment, the tower crane avoidance strategy is associated with feedback of multiple tower cranes through a preset cloud collaborative management platform, a tower crane cluster collaborative control strategy in the future time period is generated, and specifically includes: sending the tower crane avoidance strategy and corresponding risk tower crane information to the cloud collaborative management platform; according to the avoidance tower crane rope running track of the risk tower crane in the tower crane avoidance strategy in the first future time period, performing dynamic adjustment processing on the associated tower crane in the second future time period which has track intersection with the avoidance tower crane rope running track, to obtain an associated optimal constraint execution action in the optimal tower crane running track graph of the associated tower crane; wherein the second future time period is a post-time period of the first future time period; the associated hoisting execution strategy of the associated tower crane is generated through the associated optimal constraint execution action; and the tower crane cluster collaborative control strategy is generated based on the tower crane avoidance strategy and the associated hoisting execution strategy.

[0015] In an implementable embodiment, after the track intersection judgment of the tower crane rope running track of the tower crane to be executed in the multiple tower cranes in the future time period is performed according to the wind resistance-swing angle coupling error data, and the abnormal tower crane running track graph is predicted and determined, the method further includes: determining the abnormal tower crane number in the abnormal tower crane running track graph as a risk tower crane number; sending the risk tower crane number and the abnormal tower crane running track graph to the cloud collaborative management platform, and generating hoisting warning information; and sending the hoisting warning information to the associated staff terminal corresponding to the risk tower crane number, so that the tower crane pilots in mutual collaborative association obtain the hoisting warning information.

[0016] In a second aspect, the embodiments of the present application further provide a tower crane cluster collaborative control system for a smart construction site, which can perform the tower crane cluster collaborative control method for a smart construction site according to any of the above-mentioned embodiments.

[0017] In a third aspect, the embodiments of the present application further provide a non-volatile computer storage medium, which is a non-volatile computer readable storage medium, and stores at least one program, each of which includes instructions, which, when executed by a terminal, cause the terminal to perform the tower crane cluster collaborative control method for a smart construction site according to any of the above-mentioned embodiments.

[0018] The present application provides a tower crane cluster collaborative control method, system and medium for a smart construction site. Compared with the prior art, the embodiments of the present application have the following beneficial technical effects:

[0019] 1. Improved construction safety: By predicting the trajectory error of tower crane ropes, potential trajectory intersections and abnormal situations can be identified in advance, avoiding construction accidents and improving the safety of the construction site.

[0020] 2. Optimized tower crane operation efficiency: By predicting the trajectory intersection overlap in future time periods, the operation trajectory of the tower crane can be planned in advance, reducing waiting and adjustment time and improving construction efficiency.

[0021] 3. Reduced energy consumption: By optimizing the operation trajectory of the tower crane, unnecessary movement and energy consumption can be reduced, helping to save energy and reduce emissions.

[0022] 4. Improved intelligent level of construction management: Using the federated learning framework for optimal path generation processing, the intelligent management characteristics are reflected, which helps to improve the intelligent level of construction management.

[0023] 5. Real-time collaborative control: Through the cloud collaborative management platform, real-time information sharing and collaborative control between multiple tower cranes can be achieved, improving the collaboration and response speed of construction.

[0024] 6. Adapt to dynamic environment: Environmental monitoring sensors can monitor the dynamic environmental characteristics of the construction area in real time, enabling the tower crane cluster to adapt to environmental changes and ensure the continuity and stability of construction.

[0025] 7. Reduce manual intervention: Through automatic trajectory intersection overlap judgment and avoidance strategy generation, the dependence on manual intervention can be reduced, and the error rate of manual intervention can be reduced.

[0026] 8. Improve construction quality: Through precise trajectory control, the accuracy of tower crane operations can be ensured, thereby improving construction quality. BRIEF DESCRIPTION OF DRAWINGS

[0027] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiment or prior art description. Obviously, the drawings in the following description are only some embodiments described in the present application, and those skilled in the art can obtain other drawings according to these drawings without creative labor. In the drawings:

[0028] Figure 1 A flow chart of a tower crane cluster collaborative control method for a smart construction site is provided for the embodiments of the present application;

[0029] Figure 2 A structural schematic diagram of a tower crane cluster collaborative control system for a smart construction site is provided for the embodiments of the present application. DETAILED DESCRIPTION

[0030] In order for those skilled in the art to better understand the technical solutions in the present application, the technical solutions in the embodiments of the present application will be clearly and completely described in the following with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, not all. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without creative labor should fall within the scope of protection of the present application.

[0031] The embodiment of the present application provides a tower crane cluster cooperative control method for a smart construction site, as shown in Figure 1 The tower crane cluster cooperative control method for the smart construction site specifically includes steps S101-S105:

[0032] S101, through an environment monitoring sensor, trajectory error prediction of a tower crane rope is performed on a current dynamic environment feature in a current construction area, and wind resistance-swing angle coupling error data is obtained.

[0033] Specifically, first, various environment monitoring sensors need to be deployed into the tower crane equipment, that is: (1) an ultrasonic anemometer is deployed at the end of the tower crane arm of the tower crane, and wind data is collected. The wind data includes wind speed data and wind direction data. (2) A MEMS tilt sensor is deployed at the hook connecting shaft of the tower crane, and real-time measurement of the hoisting object dynamic angle data is performed. The hoisting object dynamic angle data includes pitch angle data and yaw angle data. (3) A cable tension sensor is deployed to the fixed end of the tower crane hoisting wire rope, and rope dynamic tension data is collected. (4) A temperature and humidity sensor is deployed to the top of the tower crane cab, and air density parameter data is collected. Finally, the wind data, hoisting object dynamic angle data, rope dynamic tension data and air density parameter data are determined as the tower crane external environment data.

[0034] Further, through the PTP precise time clock protocol, the multiple types of collected data sets in the tower crane external environment data are timestamped and aligned, and based on Kalman filter denoising processing, filtered tower crane external data is obtained. The wind speed vector decomposition processing is performed on the filtered tower crane external data and the wind data, and the hoisting object coordinate system is obtained.

[0035] In one embodiment, the tower crane external environment data needs to be synchronized and filtered for multiple sources of data. The code form that can be used is, for example:

[0036] def preprocess_data():

[0037] # Time stamp alignment (PTP precise time clock protocol)

[0038] synced_data = time_align(wind_data, angle_data, tension_data)

[0039] # Kalman filter denoising (take angle data as an example)

[0040] kf = KalmanFilter(F=1, H=1, Q=0.003, R=0.5)

[0041] clean_angle = kf.filter(synced_data['angle'])

[0042] # Wind speed vector decomposition (convert to hoist coordinate system)

[0043] wind_vector = transform_coord(wind_data, crane_orientation)

[0044] return { 'wind': wind_vector, 'angle': clean_angle, 'tension':tension_data}

[0045] Further, based on the wind force data in the filtered tower crane external data, the swing angle, the air resistance, and the air density compensation of the air density parameter data, a hoist swing angle dynamics function is generated. In one embodiment, the hoist swing angle dynamics function can be: where θ is the swing angle (radian); , and represents the air resistance; , and represents the air density compensation; m is the weight of the cargo in the hook, b is a mathematical constant, and g is the acceleration of gravity.

[0046] Further, by XGBoost, the wind speed, wind direction angle, swing angle, swing speed, cable length, and hoist mass in the filtered tower crane external data are determined as model input features, and based on the hoist swing angle dynamics function, the actual measured swing angle and the physical model calculated swing angle are calculated by difference, to obtain the model target value.

[0047] In one embodiment, XGBoost is used to learn the physical model residuals, where the model input features are [wind speed, wind direction angle, swing angle, swing speed, cable length, and suspended weight]; the model objective is the actual measured swing angle minus the swing angle calculated by the physical model; for example: model = xgb.XGBRegressor(objective='reg:squarederror'); model.fit(X_train, y_residual), which finally trains the residual compensation model.

[0048] Furthermore, based on the model input features and the model target value, residual compensation training is performed on the wind resistance and sway angle terms to generate a wind resistance-sway angle data-driven fusion model. The current external environment data of the tower crane is then input into the wind resistance-sway angle data-driven fusion model to predict and derive the final predicted sway angle based on the predicted data-driven compensation.

[0049] Furthermore, by predicting the swing angle and the cable lifting trajectory, trajectory compensation calculations are performed on the horizontal and vertical offsets of the hoisted object to obtain wind resistance-swing angle coupling error data based on the standard cable lifting trajectory.

[0050] In one embodiment, a fusion model driven by wind resistance and sway angle data is needed to predict dynamic errors. Inputs include: current environmental parameters (wind speed v_wind, wind direction φ, air density ρ) and hoisting status (cable length L, hoisting weight m, current sway angle θ). Then, physical baseline values ​​are calculated, and compensation is performed based on the predicted data; specifically, trajectory compensation calculations are performed for the horizontal and vertical offsets of the hoisted object. Possible code examples include:

[0051] F_wind = 0.5 * ρ * C_d * A * v_wind**2

[0052] θ_physical = solve_pendulum_eq(F_wind, L, m) # Solve the differential equation

[0053] residual = model.predict([[v_wind, φ, θ, θ_velocity, L, m]])

[0054] θ_pred = θ_physical + residual # Final predicted pendulum angle

[0055] Finally, the trajectory error mapping process is performed, that is, the horizontal and vertical offset amounts of the hoisted object are calculated for trajectory compensation, so as to obtain the horizontal offset amount Δx and the vertical offset amount Δy, and then according to the final predicted swing angle and the sequential prediction change of the cable lifting trajectory, the error of the standard cable lifting trajectory is calculated, so as to obtain the wind resistance-swing angle coupling error data.

[0056] S102, according to the wind resistance-swing angle coupling error data, the tower crane rope running trajectory of the tower crane to be executed in the tower crane task is judged for trajectory intersection in the future time period, and an abnormal tower crane running trajectory graph is predicted and determined.

[0057] Specifically, it is also necessary to generate a global coordinate system graph based on time synchronization processing based on the BIM building information of the current construction area and the position information of each tower crane. The global coordinate system graph contains the local coordinate system graph of each tower crane.

[0058] Further, according to the preset UWB positioning base station unit and the hook inertial navigation unit, and based on the future time period, the tower crane rope running trajectory in the local coordinate system graph is generated without disturbance, and the ideal tower crane rope running trajectory graph of each tower crane at the same time is obtained.

[0059] It should be noted that the UWB positioning base station is a wireless device based on ultra-wideband (UWB) technology, mainly used for indoor and outdoor high-precision positioning. By receiving positioning tag signals and calculating position information, real-time tracking with centimeter-level accuracy is achieved. The UWB positioning base station transmits nanosecond-level narrow pulse signals, uses time difference (TDOA) or angle of arrival (PDOA) algorithms, accurately measures the distance or angle from the tag, and finally determines the target position through triangulation.

[0060] In one embodiment, the tower crane rope running trajectory needs to be processed in space-time trajectory, that is, coordinate system conversion and time synchronization are needed. First, the global coordinate system is established: taking the construction reference control point as the origin (RTK positioning accuracy ±1cm), then calculating all tower crane pose conversion functions according to the tower crane swing angle and tower crane base coordinates, and generating a global coordinate system graph under time synchronization processing. Then, according to the preset UWB positioning base station unit and the hook inertial navigation unit, and in the same future time period, for example, within 20 seconds, the tower crane rope running trajectory in the local coordinate system graph is generated without disturbance, and the ideal tower crane rope running trajectory graph of each tower crane at the same time is obtained.

[0061] In one embodiment, the code for generating a single tower crane trajectory prediction graph can be adopted, for example:

[0062] # Input: current state + operation instruction + environment data

[0063] def predict_trajectory(crane_state, command, env_data):

[0064] # 1. Basic kinematic prediction

[0065] base_path = kinematic_model(command, crane_state)

[0066] # 2. Inject wind sway-angle error

[0067] wind_sway = wind_sway_predictor(env_data)# Call the aforementioned coupling model

[0068] compensated_path = []

[0069] for t in range(0, 20):# Predict for the next 20 seconds (step 0.1 seconds)

[0070] point = base_path[t]

[0071] # Calculate the wind-induced offset vector (magnitude + direction)

[0072] offset = calc_wind_offset(wind_sway[t], point.height)

[0073] compensated_path.append(point + offset)

[0074] # 3. Add inertia delay compensation

[0075] final_path = inertia_compensate(compensated_path)

[0076] return final_path

[0077] Finally, predict and output the ideal crane rope running trajectory graph for each crane at the same time (the next 20 seconds).

[0078] Furthermore, by using the wind resistance-swing angle coupling error data, the coordinate system connection lines in the ideal tower crane rope running trajectory diagram of each tower crane are offset to obtain the predicted offset coordinate system connection lines in the future time period. These offset coordinate system connection lines represent the predicted tower crane rope running trajectory in the local coordinate system diagram of each tower crane.

[0079] Furthermore, based on the global spatiotemporal grid corresponding to the global coordinate system diagram, the trajectory voxel filling calculation of the deviation coordinate system connection line of each tower crane local coordinate system diagram is performed to determine the safety margin under the influence of the hook ball radius, thus obtaining the current voxel of each deviation coordinate system connection line.

[0080] In one embodiment, the global coordinate map needs to be first divided into spatiotemporal grids, configured with a spatial resolution of 0.5m × 0.5m × 0.5m (X / Y / Z) and a temporal resolution of 0.2 seconds (covering 100 time slices for a 20-second prediction). Then, using a trajectory voxel filling algorithm, the deviation coordinate systems of each tower crane's local coordinate system are connected to calculate the trajectory voxel filling for the safety margin under the influence of the hook's sphere radius. That is:

[0081] def voxelize_trajectory(trajectory):

[0082] voxel_grid = np.zeros((100,200,200)) # Time slice × spatial grid

[0083] for t, point in enumerate(trajectory):

[0084] # Calculate the influence of the hook on the sphere radius (size of the object being lifted + safety margin)

[0085] radius = load_radius + 1.5# 1.5m safety margin

[0086] # Marking the placebo voxel

[0087] x_idx, y_idx, z_idx = coord_to_voxel(point)

[0088] mark_sphere(voxel_grid[t], x_idx, y_idx, z_idx, radius)

[0089] return voxel_grid

[0090] Finally, the current voxel of each deviation coordinate system connection is obtained.

[0091] Further, according to the time slice index, the current voxel of each deviation coordinate system connecting line is subjected to a hot map occupation coincidence judgment about the conflict voxel cluster generation, and an abnormal deviation coordinate connecting line is determined. Finally, based on the abnormal deviation coordinate connecting line, an abnormal tower crane with a lifting risk and a corresponding abnormal tower crane running trajectory graph are determined.

[0092] In one embodiment, the conflict label based on the time slice index can adopt a code, for example:

[0093] __global__ void detect_collision(int *grid, int *conflict_map) {

[0094] int t = blockIdx.x; / / time slice index

[0095] int x = threadIdx.x;

[0096] int y = threadIdx.y;

[0097] int z = threadIdx.z;

[0098] / / If the current voxel is occupied by ≥2 tower cranes

[0099] if (grid[t][x][y][z] >= 2) {

[0100] atomicAdd(&conflict_map[t], 1); / / Mark conflict

[0101] }

[0102] }

[0103] In one embodiment, it is also necessary to utilize a dynamic heat map to perform a hot map occupation coincidence judgment about the conflict voxel cluster generation for the current voxel of each deviation coordinate system connecting line, that is, to determine whether there is a conflict between two or more coordinate connecting lines, for example:

[0104] # Generate a heat map based on conflict voxel cluster

[0105] heatmap = np.zeros(space_grid)

[0106] for t in range(100):

[0107] cluster_labels = DBSCAN(voxel_conflicts[t], eps=3, min_samples=5)

[0108] for cluster in unique_clusters:

[0109] # Calculate the centroid of the conflict region

[0110] centroid = calc_centroid(cluster)

[0111] # Diffusion heat value centered on the center of mass

[0112] heatmap = gaussian_filter(heatmap, centroid, sigma=2.0)

[0113] Finally, the abnormal deviation coordinate lines were selected, and based on these lines, the abnormal tower cranes with lifting risks and their corresponding operating trajectories were identified.

[0114] As a feasible implementation method, the abnormal tower crane number in the abnormal tower crane operation trajectory map is identified as the risk tower crane number. The risk tower crane number and the abnormal tower crane operation trajectory map are then sent to the cloud-based collaborative management platform, generating a lifting warning message. Finally, the lifting warning message is sent to the terminal of the associated staff corresponding to the risk tower crane number, so that tower crane operators in the mutually collaborative association can receive the lifting warning information.

[0115] S103. Perform optimal path generation processing on the abnormal tower crane operation trajectory map based on the federated learning framework to obtain the optimized tower crane operation trajectory map.

[0116] Specifically, the execution timestamps of the tasks to be performed on the abnormal tower cranes involved in the abnormal tower crane operation trajectory map are first identified to determine the task start time node for each tower crane. The latest start time node is then determined.

[0117] Furthermore, the abnormal tower crane operation trajectory map corresponding to the latest start time node is processed by extracting relevant conflict voxel clustering trajectory features to determine the abnormal trajectory features and corresponding abnormal spatiotemporal features. Among them, the abnormal spatiotemporal features are the position coordinate features of the conflict area and the predicted operation time features in the conflict voxel clustering trajectory features.

[0118] In one embodiment, step 1: Identification of abnormal tower crane task timestamps:

[0119] 1.1. Collect real-time operating data of all tower cranes in the current construction area.

[0120] 1.2. For each abnormal tower crane, extract the timestamp of its task execution.

[0121] 1.3. Analyze the timestamp data to determine the task start time node for each abnormal tower crane.

[0122] 1.4. Mark the latest time node of task start in all abnormal tower cranes.

[0123] Step 2: Conflict voxel clustering trajectory feature extraction:

[0124] 2.1. For the abnormal tower crane with the latest start time node determined in step 1, extract the conflict voxel data of its running trajectory.

[0125] 2.2. Use spatial analysis algorithms to perform clustering analysis on conflict voxels to identify conflict areas.

[0126] 2.3. Extract the position coordinate features of the conflict area, including latitude, longitude, height, etc.

[0127] 2.4. Predict the trajectory of the abnormal tower crane in the future running time and extract the predicted running time features.

[0128] Step 3: Abnormal spatiotemporal feature determination:

[0129] 3.1. Combine the position coordinate features and predicted running time features extracted in step 2 to form an abnormal spatiotemporal feature set.

[0130] 3.2. Analyze the abnormal spatiotemporal feature set to determine the abnormal trajectory features, including trajectory stability, speed variation, etc.

[0131] 3.3. Determine the relationship between abnormal spatiotemporal features and tower crane running safety and efficiency.

[0132] Further, through the federal model under the pre-trained federal learning framework to guide the decision, the abnormal trajectory features are processed under the trajectory constraint of the target cost function, and the constraint execution action is obtained. The target cost function includes: time cost, energy cost and risk cost. The constraint execution action includes: time lag constraint, hook cargo type constraint and lifting speed constraint.

[0133] In one embodiment, it is necessary to design a communication protocol for a federal learning framework based on a transport layer, a security layer and a federal layer. Then perform distributed model training, that is, obtain a pre-trained federal learning framework under a target cost function for abnormal trajectory features, for example:

[0134] def federated_aggregation():

[0135] while not convergence:

[0136] # 1. Select 10% edge nodes

[0137] selected_nodes = random_select(edges, 0.1)

[0138] # 2. Deploy the global model

[0139] broadcast(global_model)

[0140] # 3. Local training (executed on the edge)

[0141] # Execute locally on edge devices:

[0142] for epoch in range(5): # Local 5 iterations

[0143] local_grad = compute_grad(local_data, global_model)

[0144] local_model = sgd_update(local_model, local_grad)

[0145] # 4. Secure Aggregation

[0146] encrypted_grads = []

[0147] for node in selected_nodes:

[0148] grad = receive_encrypted_grad(node)

[0149] encrypted_grads.append(paillier_add(grad)) # Homomorphic encrypted accumulation

[0150] # 5. Model Update

[0151] global_grad = paillier_decrypt(encrypted_grads)

[0152] global_model = update_model(global_model, global_grad)

[0153] Finally, the constraint execution action is obtained based on the trajectory constraint processing.

[0154] As a feasible implementation, the constraint execution action can be: postponing the execution start plan time of the tower crane; changing the type, size, and weight of the hook cargo; reducing the lifting speed, etc., and performing a power score calculation to obtain a constraint condition based on a target cost function.

[0155] Further, the constraint execution action is judged based on the safety level to determine the optimal constraint execution action. That is, the constraint condition needs to be evaluated to determine the optimal constraint execution action that meets the highest safety level and has the optimal cost. Then, according to the optimal constraint execution action, the deviation coordinate connection in the local coordinate system diagram is processed with respect to the constraint under the highest safety level to obtain an optimized deviation coordinate system connection in the future time period. That is, the deviation coordinate system connection is fine-tuned based on the data adjustment parameter corresponding to the optimal constraint execution action, thereby generating an optimized deviation coordinate system connection.

[0156] Further, based on the optimized deviation coordinate system connection, an optimized tower crane running trajectory diagram of the tower crane corresponding to the latest start time node is obtained.

[0157] S104, according to the optimized tower crane running trajectory diagram, the risk tower crane with trajectory intersection risk is processed with trajectory avoidance execution, and the tower crane avoidance strategy of the risk tower crane is obtained.

[0158] Specifically, the tower crane corresponding to the optimized tower crane running trajectory diagram is determined as the risk tower crane with trajectory intersection risk.

[0159] Further, the optimal constraint execution action in the optimized tower crane running trajectory diagram is extracted.

[0160] Further, the control item in the to-be-executed tower crane task of the risk tower crane is modified by the data control parameter in the optimal constraint execution action to generate a trajectory avoidance execution action. Finally, the tower crane avoidance strategy of the risk tower crane is generated based on the trajectory avoidance execution action.

[0161] S105, through the preset cloud collaborative management platform, the tower crane avoidance strategy is associated and fed back to multiple tower cranes to generate a tower crane cluster collaborative control strategy in the future time period.

[0162] Specifically, the tower crane avoidance strategy and the corresponding risk tower crane information also need to be sent to the cloud collaborative management platform.

[0163] Further, according to the avoidance crane rope running track of the risk crane in the crane avoidance strategy in the first future time period, a dynamic adjustment process is performed on the associated crane in the second future time period, which has track intersection overlap with the avoidance crane rope running track, to obtain the associated optimal constraint execution action in the optimized crane running track diagram of the associated crane. The second future time period is a subsequent time period of the first future time period.

[0164] Further, the associated lifting execution strategy of the associated crane is generated through the associated optimal constraint execution action. Finally, the crane cluster collaborative control strategy is generated based on the crane avoidance strategy and the associated lifting execution strategy.

[0165] In an embodiment, after the crane avoidance strategy is determined, the avoidance strategy and the corresponding risk crane information are uploaded to the cloud collaborative management platform through a data interface. It is ensured that the uploaded data includes the identification information of the risk crane, the specific details of the avoidance strategy and the execution time. Then, according to the avoidance strategy received by the cloud collaborative management platform, the cranes in the first future time period start to perform the avoidance action. The avoidance crane rope running track of the risk crane is monitored, and the real-time data is recorded. Then, the running track of the avoidance crane rope in the first future time period is analyzed to predict the possible track intersection overlap in the second future time period. The associated cranes having track intersection overlap with the avoidance crane rope running track are identified. The track paths of these associated cranes are dynamically adjusted to avoid the intersection overlap.

[0166] In an embodiment, according to the dynamically adjusted track path, the optimized crane running track diagram of the associated crane is generated. Then, the associated optimal constraint execution action in the optimized track diagram is determined. Then, based on the associated optimal constraint execution action in the above embodiment, the associated lifting execution strategy of the associated crane is generated. The strategy should include the adjustment of the lifting height, speed, time and other parameters. Then, the final crane cluster collaborative control strategy is generated by comprehensively considering the crane avoidance strategy and the associated lifting execution strategy. The strategy should ensure the safe and efficient operation of all cranes in the construction process. Finally, the generated collaborative control strategy is issued to all related cranes. The running state of the crane cluster is monitored in real time, including the position, speed, track and the like of the crane. Thus, according to the real-time monitoring data, the strategy is adjusted as necessary to adapt to the changes in the construction site.

[0167] In addition, the embodiment of the present application also provides a crane cluster collaborative control system for a smart construction site, as shown in Figure 2 The crane cluster collaborative control system 200 specifically includes:

[0168] The environment monitoring module 210 is configured to perform trajectory error prediction of the tower crane rope in the current dynamic environment in the current construction area by using the environment monitoring sensor, and obtain wind resistance-swing angle coupling error data.

[0169] The abnormal trajectory identification module 220 is configured to perform trajectory intersection judgment in a future time period on the tower crane rope operation trajectory of the tower crane to be executed in the multiple tower cranes according to the wind resistance-swing angle coupling error data, and predict and determine an abnormal tower crane operation trajectory diagram.

[0170] The optimal path generation module 230 is configured to perform optimal path generation processing on the abnormal tower crane operation trajectory diagram based on a federated learning framework, and obtain an optimized tower crane operation trajectory diagram.

[0171] The avoidance calculation module 240 is configured to perform trajectory avoidance execution processing on the risk tower crane with trajectory intersection risk according to the optimized tower crane operation trajectory diagram, and obtain a tower crane avoidance strategy of the risk tower crane.

[0172] The cooperative control module 250 is configured to perform associated feedback on the tower crane avoidance strategy by using a preset cloud cooperative management platform, and generate a tower crane cluster cooperative control strategy in a future time period.

[0173] The embodiments of the present application can predict the trajectory error of the tower crane rope, identify potential trajectory intersection and abnormal conditions in advance, avoid construction accidents, and improve the safety of the construction site. The trajectory intersection overlap in the future time period can be predicted in advance to plan the operation trajectory of the tower crane, reduce waiting and adjustment time, and improve construction efficiency. Based on the optimized tower crane operation trajectory, unnecessary movement and energy consumption can be reduced, which is helpful for energy saving and emission reduction. The federated learning framework is used for optimal path generation processing, which reflects the characteristics of intelligent management and helps to improve the intelligent level of construction management. At the same time, real-time information sharing and cooperative control between multiple tower cranes can be realized, which improves the cooperativeness and response speed of the construction. In addition, the automatic trajectory intersection overlap judgment and avoidance strategy generation can reduce the dependence on manual intervention, reduce the error rate of manual intervention, ensure the accuracy of tower crane operation, and improve the construction quality.

[0174] Each of the embodiments in the present application is described in a progressive manner, and the same or similar parts of each embodiment can be referred to each other. Each embodiment mainly describes the difference from other embodiments. Especially, the device and medium embodiments are basically similar to the method embodiments, so the description is relatively simple, and the related parts can be referred to the part of the method embodiment.

[0175] The device and medium provided by the embodiments of the present application are one-to-one corresponding, and therefore the device and medium also have similar beneficial technical effects to the corresponding method. Since the beneficial technical effects of the method have been described in detail above, the beneficial technical effects of the device and medium will not be described here again.

[0176] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. In addition, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROMs, optical storage, etc.) containing computer-usable program code.

[0177] The present application is described with reference to flowcharts and / or block diagrams of the method, device (system), and computer program product according to the embodiments of the present application. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and the combination of flows and / or blocks in the flowcharts and / or block diagrams can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing apparatus to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing apparatus generate a means for implementing the functions specified in the flowcharts and / or block diagrams. Figure 1 one or more flows and / or blocks Figure 1 an apparatus that carries out the functions specified in one or more flows and / or blocks.

[0178] These computer program instructions can also be stored in a computer-readable memory that can direct the computer or other programmable data processing apparatus to work in a specific manner, so that the instructions stored in the computer-readable memory produce a manufactured product including instruction apparatus, which implements the functions specified in the flowcharts and / or block diagrams. Figure 1 one or more flows and / or blocks Figure 1 an apparatus that carries out the functions specified in one or more flows and / or blocks.

[0179] These computer program instructions can also be loaded onto a computer or other programmable data processing apparatus, so that a series of operation steps are performed on the computer or other programmable data processing apparatus to produce a computer-implemented process, so that the instructions executed on the computer or other programmable data processing apparatus provide a means for implementing the functions specified in the flowcharts and / or block diagrams. Figure 1 one or more flows and / or blocks Figure 1 an apparatus that carries out the functions specified in one or more flows and / or blocks.

[0180] In a typical configuration, a computing device includes one or more processors (CPUs), input / output interfaces, network interfaces, and memories.

[0181] Memory can include non-persistent memory and / or volatile memory, such as random access memory (RAM) and / or cache memory, etc. in the form of a computer-readable medium, read only memory (ROM), or flash memory, etc. Memory is an example of computer-readable media.

[0182] Computer-readable media includes permanent and non-permanent, removable and non-removable media implemented in any method or technology for storage of information such as computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read only memory (ROM), electrically erasable programmable read only memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), digital versatile disc (DVD), or other optical storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information accessible to a computing device. According to the definition herein, computer-readable media does not include transitory media, such as modulated data signals and carrier waves.

[0183] It should also be noted that the terms "comprising", "containing", or any other variant thereof, are intended to encompass a non-exclusive inclusion, such that a process, method, article or apparatus that comprises a list of elements does not include only those elements in the list, but can also include other elements not expressly listed or inherent to such process, method, article or apparatus. Without more limitations, the element defined by the statement "comprising a" does not exclude the presence of additional identical elements in the process, method, article or apparatus that includes the element.

[0184] The above description is merely illustrative of the application, and not restrictive. Various modifications and changes can be made by those skilled in the art which fall within the spirit and scope of the application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principles of the application should be included within the scope of the description.

Claims

1. A tower crane cluster cooperative control method for a smart construction site, characterized in that, The method comprises: Perform trajectory error prediction on the current dynamic environment characteristics in the current construction area by the environmental monitoring sensor to obtain wind resistance-swing angle coupling error data; According to the wind resistance-swing angle coupling error data, perform trajectory intersection overlap judgment on the tower crane rope operation trajectory of the tower crane to be executed in the future time period, and predict and determine the abnormal tower crane operation trajectory diagram; The optimal path generation processing based on the federal learning framework is performed on the abnormal tower crane operation trajectory diagram to obtain an optimized tower crane operation trajectory diagram; According to the optimized tower crane operation trajectory diagram, perform trajectory avoidance execution processing on the risk tower crane with trajectory intersection risk to obtain a tower crane avoidance strategy of the risk tower crane; Through the preset cloud collaborative management platform, the tower crane avoidance strategy is associated and fed back to multiple tower cranes to generate a tower crane cluster collaborative control strategy in the future time period. 2.The tower crane cluster cooperative control method for a smart construction site of claim 1, wherein, Before the trajectory error prediction on the current dynamic environment characteristics in the current construction area by the environmental monitoring sensor to obtain wind resistance-swing angle coupling error data, the method further comprises: Deploy an ultrasonic anemometer at the tower crane arm end of the tower crane and collect wind data; wherein the wind data includes wind speed data and wind direction data; Deploy a MEMS tilt sensor at the hook connecting shaft of the tower crane and measure real-time hoisting object dynamic angle data; wherein the hoisting object dynamic angle data includes pitch angle data and yaw angle data; Deploy a cable tension sensor to the hoisting steel wire rope fixed end of the tower crane and collect rope dynamic tension data; Deploy a temperature and humidity sensor to the top of the tower crane cab of the tower crane and collect air density parameter data; The wind data, the hoisting object dynamic angle data, the rope dynamic tension data, and the air density parameter data are determined as tower crane external environment data. 3.The tower crane cluster cooperative control method for smart construction site of claim 2, wherein, The trajectory error prediction on the current dynamic environment characteristics in the current construction area by the environmental monitoring sensor to obtain wind resistance-swing angle coupling error data specifically comprises: Through the PTP precise time clock protocol, perform timestamp alignment processing on the multiple types of collected data sets in the tower crane external environment data, and based on Kalman filter noise reduction processing, obtain filtered tower crane external data; Perform wind speed vector decomposition processing on the wind data in the filtered tower crane external data to obtain a hoisting object coordinate system; Based on the swing angle, wind resistance of the wind data in the filtered tower crane external data, and the air density compensation of the air density parameter data, generate a hoisting object swing angle dynamics function; Through XGBoost, determine the wind speed, wind direction angle, swing angle, swing speed, cable length, and hoisting weight mass in the filtered tower crane external data as model input features, and based on the hoisting object swing angle dynamics function, perform difference calculation on the actual measured swing angle and the physical model calculated swing angle to obtain a model target value; According to the model input features and the model target value, perform residual compensation training on the wind resistance term and the swing angle term to generate a wind resistance-swing angle data driven fusion model; The current tower crane external environment data is input into the wind resistance-swing angle data-driven fusion model, and the final predicted swing angle based on the predicted data-driven compensation is predicted and obtained; Through the final predicted swing angle and the cable lifting trajectory, the horizontal and vertical offset amounts of the hoisted object are calculated for trajectory compensation, and the wind resistance-swing angle coupling error data based on the standard cable lifting trajectory is obtained.

4. The tower crane cluster cooperative control method for a smart construction site according to claim 1, characterized in that, According to the wind resistance-swing angle coupling error data, the tower crane rope running trajectories of the tower cranes to be executed in the future time period are judged for trajectory intersection and overlap, and an abnormal tower crane running trajectory graph is predicted and determined, specifically including: Based on the BIM building information of the current construction area and the position information of each tower crane, a global coordinate system graph based on time synchronization processing is generated; wherein the global coordinate system graph contains the local coordinate system graph of each tower crane; According to the preset UWB positioning base station unit and the hook inertial navigation unit, and based on the future time period, the tower crane rope running trajectory in the local coordinate system graph is generated without disturbance, and an ideal tower crane rope running trajectory graph of each tower crane at the same time is obtained; Through the wind resistance-swing angle coupling error data, the coordinate system connecting line in the ideal tower crane rope running trajectory graph of each tower crane is compensated for deviation, and a deviation coordinate system connecting line predicted in the future time period is obtained; wherein the deviation coordinate system connecting line is the tower crane rope running prediction trajectory of the local coordinate system graph of each tower crane; According to the global space-time grid corresponding to the global coordinate system graph, the deviation coordinate system connecting line of each tower crane local coordinate system graph is calculated for trajectory voxel filling related to the safety margin of the hook influence sphere radius, and the current voxel of each deviation coordinate system connecting line is obtained. According to the time slice index, the current voxel of each deviation coordinate system connecting line is occupied and overlapped to determine an abnormal deviation coordinate connecting line. Based on the abnormal deviation coordinate connecting line, an abnormal tower crane with lifting risk and the corresponding abnormal tower crane running trajectory graph are determined. 5.The tower crane cluster cooperative control method for smart construction site of claim 1, wherein, The abnormal tower crane running trajectory graph is processed for optimal path generation based on the federated learning framework, and an optimized tower crane running trajectory graph is obtained, specifically including: The abnormal tower crane involved in the abnormal tower crane running trajectory graph is identified for the execution time stamp of the to-be-executed tower crane task, and the task start time node of each tower crane is determined; and the latest start time node is determined; The abnormal tower crane running trajectory graph corresponding to the latest start time node is extracted for the conflict voxel clustering trajectory feature, and an abnormal trajectory feature and a corresponding abnormal space-time feature are determined; wherein the abnormal space-time feature is the position coordinate feature and the predicted running time feature of the conflict region in the conflict voxel clustering trajectory feature; The abnormal trajectory feature is processed under a trajectory constraint of a target cost function through a federated model under a pre-trained federated learning framework to guide decision-making, and a constraint execution action is obtained; wherein the target cost function includes a time cost, an energy consumption cost, and a risk cost; the constraint execution action includes a time lag constraint, a hook cargo type constraint, and a hoisting speed constraint; The constraint execution action is judged based on a safety level to determine an optimal constraint execution action; According to the optimal constraint execution action, the deviation coordinate connection in the local coordinate system diagram is processed under the constraint of the highest safety level to obtain an optimized deviation coordinate system connection in the future time period; Based on the optimized deviation coordinate system connection, an optimized tower crane running trajectory diagram of the tower crane corresponding to the latest starting time node is obtained. 6.The tower crane cluster cooperative control method for smart construction site of claim 1, wherein, According to the optimized tower crane running trajectory diagram, a risk tower crane with a trajectory intersection risk is processed to obtain a tower crane avoidance strategy of the risk tower crane, specifically including: The tower crane corresponding to the optimized tower crane running trajectory diagram is determined as the risk tower crane with a trajectory intersection risk; The optimal constraint execution action in the optimized tower crane running trajectory diagram is extracted; The control item in the to-be-executed tower crane task of the risk tower crane is modified through the data control parameter in the optimal constraint execution action to generate a trajectory avoidance execution action; Based on the trajectory avoidance execution action, the tower crane avoidance strategy of the risk tower crane is generated. 7.The tower crane cluster cooperative control method for smart construction site of claim 1, wherein, Through a preset cloud collaborative management platform, the tower crane avoidance strategy is associated and fed back to multiple tower cranes to generate a tower crane cluster collaborative control strategy in the future time period, specifically including: The tower crane avoidance strategy and the corresponding risk tower crane information are sent to the cloud collaborative management platform; According to the avoidance tower crane rope running trajectory of the risk tower crane in the tower crane avoidance strategy in the first future time period, the associated tower crane with trajectory intersection overlap with the avoidance tower crane rope running trajectory in the second future time period is dynamically adjusted to obtain an associated optimal constraint execution action in the optimized tower crane running trajectory diagram of the associated tower crane; wherein the second future time period is a post-time period of the first future time period; Through the associated optimal constraint execution action, an associated hoisting execution strategy of the associated tower crane is generated; Based on the tower crane avoidance strategy and the associated hoisting execution strategy, the tower crane cluster collaborative control strategy is generated. 8.The tower crane cluster cooperative control method for smart construction site of claim 1, wherein, After predicting and determining the abnormal tower crane running trajectory diagram by judging the trajectory intersection overlap of the tower crane rope running trajectory of the to-be-executed tower crane task of multiple tower cranes in the future time period according to the wind resistance-swing angle coupling error data, the method further includes: The abnormal tower crane number in the abnormal tower crane running trajectory diagram is determined as a risk tower crane number; The risk tower crane number and the abnormal tower crane running trajectory diagram are sent to the cloud collaborative management platform, and a hoisting warning information is generated. The hoisting early warning information is sent to an associated staff terminal corresponding to the risk tower crane number, so that a tower crane driver in mutual coordination association obtains the hoisting early warning information.

9. A tower crane cluster cooperative control system for a smart construction site, characterized in that, The system can perform the tower crane cluster cooperative control method for a smart construction site according to any one of claims 1-8.

10. A non-transitory computer storage medium, comprising, The storage medium is a non-volatile computer readable storage medium, and the non-volatile computer readable storage medium stores at least one program, each program includes instructions, and the instructions make the terminal execute the tower crane cluster cooperative control method for a smart construction site according to any one of claims 1-8 when executed by the terminal.

Images