Remote wireless control system for pipe automatic sling

Abstract

The application discloses a remote wireless control system of a pipe automatic lifting appliance, relates to the technical field of remote control of hoisting equipment, and comprises an appliance body, a remote control terminal and a sensing module, a control module and an execution module arranged on the appliance body. The sensing module comprises an appliance state sensor, a pipe posture sensor, an environment sensor and a digital wind speed and direction instrument, and is used for collecting the appliance state, the pipe posture, the environment obstacles and the wind speed and direction data. After the control module receives the data, a dynamic safety envelope is constructed, the collision risk is predicted through continuous collision detection, and a hierarchical obstacle avoidance instruction is generated. The execution module drives the lifting, walking and amplitude changing mechanism of the appliance to execute the obstacle avoidance action. The system is also integrated with a fault prediction and health management module and a digital twin visualization function. The application can realize real-time sensing of the dynamic posture of the pipe, avoid collision risks, realize equipment health prediction and operation transparency, and improve the safety, reliability and intelligent level of pipe hoisting.

Description

Technical Field

[0001] This invention relates to the field of remote control technology for hoisting equipment, specifically a remote wireless control system for automatic pipe hoisting tools. Background Technology

[0002] With the rapid development of modern industrial logistics and large-scale engineering construction, the automation and intelligence level of pipe hoisting operations, as an important basic material, has become a key factor affecting construction efficiency and safety. Traditional pipe hoisting relies heavily on manual operation or semi-automatic remote control equipment, which suffers from low operational precision, high safety risks, and high labor intensity. In recent years, remote wireless control technology has been gradually applied to the field of hoisting equipment, realizing the physical separation of operators from hoisting equipment and improving the working environment.

[0003] For example, some pipe-laying machines employ control systems based on remote controls and PLCs (Programmable Logic Controllers), enabling operators to remotely control the crane's basic movements. On the other hand, to further improve the efficiency of multi-device collaboration, existing technologies have developed solutions that use wireless communication modules to link multiple pipe-laying machines. These typically employ DSPs (Digital Signal Processors) as the core controller to integrate and control the hydraulic actuators, enabling functions such as anti-tipping alarms and fine-motion adjustments. These technologies have, to some extent, promoted the remote operation and initial collaboration of lifting operations.

[0004] However, existing remote wireless control systems still have significant technical blind spots when dealing with loads like pipes, which have large length-to-diameter ratios, are prone to swaying, and have unique shapes. On the one hand, the sensing systems of existing equipment mostly focus on the static detection of the lifting device's own state (such as position and angle) or environmental obstacles, lacking the ability to accurately perceive and model the real-time motion posture of the pipe itself (including swaying and bending deformation). On the other hand, their obstacle avoidance strategies are often reactive control based on fixed distance thresholds, that is, alarms or braking are only triggered when the lifting device or load approaches the preset safety distance boundary. This static protection method cannot effectively adapt to the irregular, large-scale, and time-varying spatial swaying of pipes during lifting due to factors such as wind and inertia. This makes it difficult for the system to predict the collision risk caused by the large swing of the pipe end or scraping against surrounding obstacles, constituting a potential hazard to safe production.

[0005] Therefore, a key technical problem that urgently needs to be solved in this field is: how to enable a remote wireless control system to perceive and predict the dynamic motion posture of the pipe in space in real time and accurately, and to construct a safety protection boundary that changes synchronously with the pipe's posture, thereby achieving proactive intervention and avoidance of irregular and time-varying collision risks. Existing technical solutions do not involve real-time modeling and collision prediction of the pipe's dynamic envelope, which inherently limits their safety assurance capabilities in complex dynamic environments.

[0006] In summary, although existing technologies have made some progress in remote control and basic safety monitoring of hoisting equipment, they are clearly insufficient in addressing the dynamic safety risks unique to pipe hoisting. Summary of the Invention

[0007] The purpose of this invention is to overcome the shortcomings of the prior art and provide a remote wireless control system for automatic pipe lifting devices, which can break through the limitations of static safety boundaries, thereby improving the ability to perceive, predict and actively avoid the risks of dynamic swaying of pipes.

[0008] To solve the above-mentioned technical problems, the present invention provides the following technical solution: a remote wireless control system for an automatic pipe lifting device, comprising a lifting device body, a remote control terminal, and a sensing module, a control module, and an execution module disposed on the lifting device body, wherein the remote control terminal is connected to the lifting device body via a wireless communication network, characterized in that:

[0009] The sensing module includes a lifting device status sensor, a pipe attitude sensor, and an environmental sensor. The lifting device status sensor is used to collect the status data of the lifting device, including the position, velocity, acceleration, and attitude angle of the lifting device. The pipe attitude sensor is used to collect the motion attitude data of the pipe, including point cloud data and image data of the pipe. The environmental sensor is used to collect environmental obstacle data and environmental wind speed data.

[0010] The control module includes a processor and a memory. The memory stores a computer program, and the processor executes the computer program to perform the following steps:

[0011] Receive sensing data from the sensing module in real time;

[0012] Based on the pipe's motion posture data, a dynamic safety envelope is constructed using point cloud processing and posture recognition algorithms. The dynamic safety envelope is a continuum in three-dimensional space, and its surface maintains a dynamic safety distance from the outer surface of the pipe. The dynamic safety distance is dynamically adjusted according to the real-time swing angular velocity of the pipe and the ambient wind speed.

[0013] The dynamic safety envelope is fused with the environmental obstacle data, and the collision risk between the dynamic safety envelope and the obstacle is predicted by a continuous collision detection algorithm, and the shortest collision time is calculated.

[0014] When the shortest collision time is less than or equal to a preset threshold, an obstacle avoidance control command is generated.

[0015] The execution module receives the obstacle avoidance control command and drives the hoisting mechanism, traveling mechanism and luffing mechanism of the spreader to perform obstacle avoidance actions;

[0016] The wireless communication network adopts a 5G private network to ensure low-latency transmission of control commands. The remote control terminal provides a visual interface to display the dynamic safety envelope, obstacles, and lifting gear status in real time.

[0017] Furthermore, the spreader status sensor includes a navigation satellite system receiver and an inertial measurement unit. The navigation satellite system receiver is used to acquire the position coordinates of the spreader, and the inertial measurement unit is used to measure the angular velocity and linear acceleration of the spreader. The data is fused using a Kalman filter algorithm to output the real-time three-dimensional position, velocity, acceleration, pitch angle, and roll angle of the spreader.

[0018] The pipe attitude sensor includes a multi-line lidar and a vision sensor. The multi-line lidar is installed below the hanger and scans the suspended pipe in a high-speed scanning manner to generate high-density point cloud data. The vision sensor includes a high-resolution binocular camera, which extracts the depth information of the pipe through a stereo vision algorithm and fuses it with the point cloud data of the multi-line lidar to improve the accuracy of pipe attitude recognition. The fusion algorithm adopts point cloud registration and feature matching methods.

[0019] The environmental sensors include a 3D lidar and a millimeter-wave radar. The 3D lidar is deployed around the boom and the site to build a real-time 3D point cloud map of the working environment and identify static and dynamic obstacles. The millimeter-wave radar is used to detect the speed and direction of the obstacles and supplement the lidar's perception capabilities in adverse weather conditions.

[0020] The sensing module also includes a digital anemometer to measure ambient wind speed and direction in real time.

[0021] Furthermore, the step of constructing the dynamic security envelope specifically includes:

[0022] Preprocessing of point cloud data acquired by pipe attitude sensors includes outlier removal, smoothing filtering, and voxel mesh downsampling to reduce noise and improve computational efficiency.

[0023] Pipe point cloud is extracted by a point cloud segmentation algorithm. The point cloud segmentation algorithm is based on region growing and clustering methods, which identifies pipe point cloud clusters and calculates their centroids and bounding boxes.

[0024] Based on the extracted pipe point cloud, the central axis of the pipe is calculated using principal component analysis algorithm, and a three-dimensional wireframe model of the pipe is fitted.

[0025] Based on the real-time angular velocity of the pipe and ambient wind speed Calculate dynamic safety distance The calculation formula is:

[0026]

[0027] in This is the basic safety distance. It is the oscillation angular velocity weighting coefficient. It is the wind speed weighting coefficient. , and Preset positive real numbers;

[0028] The dynamic safety distance The explanation is as follows:

[0029] This represents the minimum safety margin of the pipe when it is stationary. This represents the degree to which the angular velocity of the swing affects the safe distance. This represents the degree to which wind speed affects the safe distance;

[0030] The fitted 3D wireframe model of the pipe is expanded outward by a dynamic safety distance along its surface normal direction. Generate a dynamic safety envelope represented by a triangular mesh, the volume and shape of which vary with... and Updated in real time.

[0031] Furthermore, the step of predicting collision risk specifically includes:

[0032] The dynamic safety envelope is represented as a triangular mesh model, and environmental obstacle data is represented as a 3D point cloud or bounding box model.

[0033] By using a spatiotemporal synchronization algorithm, the dynamic safety envelope model and obstacle model are unified into the same coordinate system, and the motion trajectory of the dynamic safety envelope and obstacles in future time intervals is predicted.

[0034] The motion trajectory prediction adopts a physics-based dynamic model. For lifting equipment and pipes, rigid body kinematics equations are used, and for obstacles, a uniform motion model is used.

[0035] Continuous collision detection is performed to calculate the shortest collision time between the dynamic safety envelope and the obstacle. The continuous collision detection algorithm adopts the separating axis theorem algorithm to traverse all possible contact surfaces between the dynamic safety envelope and the obstacle.

[0036] The shortest collision time The calculation formula is:

[0037]

[0038] in It is the first dynamic safety envelope surface The triangular facet and the surface of the obstacle. The current distance between the three triangular faces It is the component of the relative velocity between the two along the line connecting them. This indicates taking the minimum value among all faces;

[0039] The shortest collision time The parameters are explained as follows:

[0040] Represents the spatial interval at the current moment. Represents the approach rate.

[0041] Furthermore, the step of generating obstacle avoidance control commands adopts a hierarchical autonomous obstacle avoidance decision-making mechanism, specifically including:

[0042] Set the first threshold Second threshold , Greater than , and It is a time constant;

[0043] When the shortest collision time Greater than At this time, the system is in normal operating condition, only sending visual warning signals to the remote control terminal, and does not interfere with the operation of the lifting device;

[0044] when Less than or equal to and greater than When the system initiates intervention-level obstacle avoidance, it automatically generates an obstacle avoidance path plan. The obstacle avoidance path plan uses the artificial potential field method to calculate a smooth path without collisions, and controls the execution module to make the spreader move along the path while reducing the spreader's running speed.

[0045] when Less than or equal to When the system triggers emergency braking obstacle avoidance, it immediately generates a stop command, cuts off the power source of the spreader, and activates the brakes to achieve an emergency stop.

[0046] The hierarchical decision-making mechanism also considers obstacle type; for dynamic obstacles, and The value is dynamically adjusted based on the obstacle's speed, and the adjustment formula is:

[0047]

[0048] in This is the adjusted first threshold. It is the speed of the obstacle. This is the maximum permissible speed of the spreading equipment;

[0049] The parameters are explained as follows:

[0050] and Time boundaries representing different risk levels Represents the speed at which the obstacle moves. This represents the upper limit of system performance.

[0051] Furthermore, the wireless communication network adopts 5G private network technology, and its network architecture includes a terminal access layer, an edge computing layer, and a cloud control layer.

[0052] The terminal access layer is deployed at the lifting site, and 5G industrial modules are used to realize data transmission between the sensing module and the control module;

[0053] The edge computing layer is deployed on the edge server of the site and is responsible for running dynamic safety envelope modeling and collision risk prediction algorithms to reduce cloud load. The edge server is connected to the terminal access layer through the 5G core network.

[0054] The cloud control layer provides remote control terminal access to enable collaborative management of multiple lifting devices. The cloud and edge computing layers are connected via dedicated fiber optic lines to ensure data synchronization.

[0055] The wireless communication network also employs a redundant transmission protocol. Important control commands, including emergency stop commands and path update commands, are transmitted with temporal and spatial redundancy. Temporal redundancy means that the command is sent three times, and spatial redundancy means that it is received through multi-antenna diversity reception.

[0056] Furthermore, the visual interface of the remote control terminal is built based on digital twin technology, specifically including:

[0057] Create high-fidelity 3D models of lifting equipment, pipes, and the working environment, and keep them synchronized with the physical entities;

[0058] The dynamic safety envelope is rendered in real time, highlighted in semi-transparent red, and overlaid with obstacle position and shortest collision time information;

[0059] Provides operator interaction functions, including one-click emergency stop, manual route planning, and system status monitoring;

[0060] The visual interface also integrates an alarm log system, which records all obstacle avoidance events and sensor anomalies, and supports historical data playback.

[0061] The remote control terminal uses an industrial-grade tablet computer equipped with a touch screen and physical buttons to meet outdoor operation requirements.

[0062] Furthermore, the system also includes a fault prediction and health management module, which is integrated into the control module and specifically includes:

[0063] Real-time monitoring of the motor current signal in the execution module and acquisition of three-phase current waveform data;

[0064] Perform a fast Fourier transform on the current waveform to extract characteristic frequency components, including the fundamental and harmonic amplitudes.

[0065] The feature components are input into a deep learning model, which is a convolutional neural network with an input layer, a convolutional layer, a pooling layer and a fully connected layer, and outputs a motor health score.

[0066] When the health score falls below a preset threshold, a predictive maintenance alert is generated, prompting the replacement of parts or adjustment of parameters;

[0067] The fault prediction and health management module also monitors the signal strength of the wireless communication network and automatically switches to the backup communication channel when the signal strength is lower than a set threshold.

[0068] Compared with existing technologies, the remote wireless control system of this automatic pipe lifting tool has the following advantages:

[0069] I. This invention collects pipe motion posture data by setting up pipe posture sensors, and constructs a dynamic safety envelope by combining point cloud processing algorithms and posture recognition algorithms. The safety distance of this dynamic safety envelope is dynamically adjusted according to the real-time swing angular velocity of the pipe and the ambient wind speed. At the same time, a continuous collision detection algorithm is used to predict the collision risk between the dynamic safety envelope and obstacles and calculate the shortest collision time. This allows for real-time and accurate perception of the dynamic motion posture of the pipe, and the construction of a dynamic safety protection boundary that changes synchronously with the pipe posture. This breaks through the limitations of existing static safety boundaries, effectively avoids the collision risk caused by the dynamic swing of the pipe, solves the problem that existing remote wireless control systems cannot cope with the dynamic swing of the pipe, resulting in insufficient safety protection capabilities, and improves the dynamic safety assurance level of pipe hoisting operations.

[0070] Second, this invention integrates a fault prediction and health management module into the control module, monitors the current signal of the motor in the execution module in real time, and outputs a motor health score based on a deep learning model. At the same time, it adopts a 5G private network redundant transmission protocol to ensure the reliable transmission of important control commands such as emergency stop commands and path update commands. The remote control terminal relies on digital twin technology to realize the visualization of the working environment and equipment status and the playback of historical data. It can realize predictive maintenance of equipment health status, stable transmission of control commands, and transparent management of the operation process, thereby reducing equipment downtime due to failure, improving the reliability of remote control and the monitorability of operation, and enhancing the overall operational stability and maintenance convenience of the system.

[0071] Other advantages, objectives and features of the invention will be set forth in part in the description which follows, and in part will be apparent to those skilled in the art from the following examination or study, or may be learned from the practice of the invention. Attached Figure Description

[0072] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without any creative effort.

[0073] Figure 1 This is a schematic diagram of the workflow of the present invention;

[0074] Figure 2 This is a schematic diagram of the system hardware architecture and data flow of the present invention;

[0075] Figure 3 This is a schematic diagram of the dynamic safety envelope construction and obstacle avoidance decision-making process of the present invention. Detailed Implementation

[0076] To further illustrate the technical means and effects of the present invention in achieving its intended purpose, the following detailed description of the specific implementation methods, structures, features, and effects of the present invention, in conjunction with the accompanying drawings and preferred embodiments, is provided below.

[0077] Example 1

[0078] like Figure 2 and Figure 3 As shown, this embodiment discloses a remote wireless control system for an automatic pipe lifting device, aiming to solve the technical problems of existing remote pipe lifting control systems being unable to perceive the dynamic attitude of the pipe in real time and the static safety boundary being unable to cope with complex risks. The system includes a lifting device body, a remote control terminal, and a sensing module, a control module, and an execution module mounted on the lifting device body; the remote control terminal and the lifting device body achieve low-latency communication via a 5G private network. The sensing module collects data on the lifting device status, pipe attitude, and environment; the control module constructs a safety envelope that dynamically adjusts with the swing angular velocity and wind speed based on the pipe attitude data, predicts collision risks through a continuous collision detection algorithm, and generates graded obstacle avoidance commands; the execution module responds to commands and drives the lifting device to move; simultaneously, it integrates digital twin visualization and fault prediction functions to achieve remote, intelligent, and highly safe pipe lifting.

[0079] System overall architecture setup:

[0080] The remote wireless control system for the automatic pipe lifting device in this embodiment comprises a lifting device body, a remote control terminal, and integrated sensing, control, and execution modules within the lifting device body. At the software level, it utilizes 5G private network communication protocols, digital twin modeling algorithms, and fault prediction models to achieve functional closed-loop operation. The lifting device body is a conventional mechanical structure for pipe lifting, possessing hoisting, traveling, and luffing functions, and its structural design meets industrial-grade lifting load requirements. The remote control terminal uses an industrial-grade tablet computer equipped with a high-definition touchscreen and a physical emergency stop button, adapting to harsh outdoor operating environments such as strong light and dust. The modules are mechanically and electrically connected through standardized interfaces, ensuring the stability and reliability of signal transmission.

[0081] Data acquisition and processing of the sensing module:

[0082] The sensing module is the core of the system for sensing the environment and equipment status. It includes a spreader status sensor, a pipe attitude sensor, an environmental sensor, and a digital anemometer. All sensors work together to obtain comprehensive and accurate monitoring data.

[0083] The spreader status sensor consists of a navigation satellite system receiver and an inertial measurement unit (IMU). The navigation satellite system receiver acquires the real-time position coordinates of the spreader by receiving satellite signals, and its positioning accuracy meets the position awareness requirements of industrial lifting. The IMU has a built-in three-axis gyroscope and a three-axis accelerometer, which can measure the angular velocity and linear acceleration of the spreader in real time. To reduce the measurement error of a single sensor, the data collected by both are fused using a Kalman filter algorithm. The Kalman filter algorithm dynamically corrects the position data from the navigation satellite system receiver and the motion parameters from the IMU through a prediction-update iterative process, and finally outputs the real-time three-dimensional position, velocity, acceleration, and attitude angles of the spreader, providing basic state data of the spreader body for subsequent control decisions.

[0084] Pipe attitude sensor: Includes a multi-line LiDAR and a high-resolution binocular camera. The multi-line LiDAR is installed below the hanger and performs high-speed scanning of the suspended pipe in both circumferential and axial directions, generating high-density point cloud data. This point cloud data can completely reflect the pipe's shape and spatial position. The binocular camera extracts the pipe's depth information through the parallax of the left and right lenses, combined with stereo vision algorithms, compensating for the LiDAR's shortcomings in perceiving details on the pipe's surface. To improve the accuracy of pipe attitude recognition, the two data are fused using point cloud registration and feature matching algorithms: the point cloud registration algorithm converts the depth information acquired by the binocular camera into point cloud data and aligns it spatially with the LiDAR point cloud; the feature matching algorithm extracts geometric features of the pipe surface (such as end edges and surface protrusions) to achieve accurate correlation between the two types of point clouds, ultimately outputting point cloud data and image data that accurately reflect the pipe's motion attitude.

[0085] Environmental sensors and digital anemometers: Environmental sensors include 3D LiDAR and millimeter-wave radar. The 3D LiDAR is deployed at fixed locations on the top of the boom and around the site, continuously scanning the work area to build a real-time 3D point cloud map. This map clearly identifies the spatial location of both static and dynamic obstacles. The millimeter-wave radar is resistant to rain, fog, and dust interference, and can detect the speed and direction of dynamic obstacles in real time, supplementing the LiDAR's blind spots in adverse weather conditions. Digital anemometers are installed at an unobstructed high position on the crane body, measuring wind speed and direction data in real time to provide environmental parameters for calculating dynamic safety distances.

[0086] Algorithm execution and decision generation in the control module:

[0087] The control module includes a processor and a memory. The memory stores the computer program that implements the system functions. When the processor executes the program, it sequentially completes data reception, dynamic safety envelope construction, collision risk prediction, and obstacle avoidance command generation. The specific process is as follows:

[0088] Real-time data reception: The processor receives real-time data on the status of the lifting device, the posture of the pipe, environmental obstacles, and wind speed and direction transmitted by the sensor module through the terminal access layer of the 5G private network. The receiving frequency is synchronized with the sampling frequency of the sensor to ensure the real-time and continuous nature of the data. At the same time, the processor verifies the validity of the received data and removes abnormal data caused by transmission interference to ensure the accuracy of subsequent algorithm processing.

[0089] Construction of the dynamic safety envelope: The dynamic safety envelope is the core innovation of the system in addressing the risk of dynamic swaying of the pipe. Its construction process includes four steps: point cloud preprocessing, pipe model fitting, dynamic safety distance calculation, and envelope generation.

[0090] Point cloud preprocessing: The point cloud data collected by the pipe attitude sensor is sequentially processed by outlier removal, smoothing filtering, and voxel mesh downsampling. Outlier removal uses a statistical filtering algorithm, which calculates the distance deviation between each point and its neighbors to remove noise points that exceed a reasonable deviation range. Smoothing filtering uses a Gaussian filtering algorithm, which reduces the fluctuation noise of the point cloud data by weighted averaging of the neighbors of each point. Voxel mesh downsampling divides the point cloud space into uniform voxels, with each voxel retaining a representative point, reducing the amount of data and improving the efficiency of subsequent calculations while ensuring the accuracy of the pipe contour.

[0091] Pipe Model Fitting: A point cloud segmentation algorithm based on region growing and clustering is used to identify and extract point cloud clusters belonging to the pipe material from the preprocessed point cloud data. The centroid of the pipe material is determined by calculating the mean spatial coordinates of these point cloud clusters, and a minimum bounding box is constructed to initially determine the spatial range of the pipe material. Based on the extracted pipe point cloud clusters, principal component analysis (PCA) is used to calculate the central axis of the pipe material. PCA decomposes the covariance matrix of the pipe point cloud into eigenvalues, and the eigenvector with the largest eigenvalue is used as the direction of the central axis. Combined with the centroid coordinates, the spatial position of the central axis is determined. Finally, based on the central axis and bounding box dimensions, a 3D wireframe model that reflects the actual shape of the pipe material is fitted and generated.

[0092] Dynamic safety distance calculation: The dynamic safety distance is dynamically adjusted according to the real-time oscillation angular velocity of the pipe and the ambient wind speed. The calculation formula is shown in equation (1):

[0093]

[0094] In formula (1):

[0095] The dynamic safety distance is the minimum distance between the surface of the dynamic safety envelope and the outer surface of the pipe. Its value is updated in real time as the pipe swings and wind speed changes.

[0096] The basic safety distance is the minimum safety margin set for pipes in a static state to avoid static interference with surrounding objects. Its value is preset according to the diameter and length of the pipes and industry hoisting safety standards to ensure basic safety under static working conditions.

[0097] This is the oscillation angular velocity weighting coefficient, reflecting the degree of influence of the pipe's oscillation angular velocity on the safety distance. Its value is related to the pipe's material and length—the higher the pipe density and the longer the length, the greater the inertia during oscillation. The larger the value, the more sufficient the safety distance is reserved;

[0098] The real-time angular velocity of the pipe, i.e., the rotational angular velocity of the pipe about its central axis perpendicular to the axis, is calculated from point cloud data collected by the pipe attitude sensor. This can be obtained by comparing the change in the pipe's attitude angle with the time interval at adjacent moments. ;

[0099] This is the wind speed weighting coefficient, reflecting the degree of influence of ambient wind speed on the safe distance. Its value is related to the windward area of ​​the pipe—the larger the windward area, the stronger the driving effect of wind speed on the pipe's sway. The larger the value;

[0100] The real-time wind speed is directly collected by a digital anemometer.

[0101] Envelope generation: The fitted 3D wireframe model of the pipe is expanded outward by a dynamic safety distance along its surface normal direction (i.e., the direction perpendicular to the pipe surface). The surface of the expanded model is discretized using a triangular meshing algorithm to generate a dynamic safety envelope based on a triangular mesh; the volume and shape of this envelope vary with... and The changes are updated in real time, and the space that the pipe may touch during dynamic swing is always included in the protection.

[0102] Collision risk prediction: By fusing dynamic safety envelope data with environmental obstacle data, a continuous collision detection algorithm is used to predict collision risk. The specific process is as follows:

[0103] Model unification and trajectory prediction: The triangular mesh model of the dynamic safety envelope and the 3D point cloud / boundary box model of environmental obstacles are unified into a local coordinate system with the geometric center of the lifting device as the origin through a spatiotemporal synchronization algorithm. The spatiotemporal synchronization algorithm eliminates spatiotemporal deviations between different sensors through timestamp alignment and coordinate transformation. Based on the unified model, a physics-based dynamic model is used to predict the motion trajectory of both in the future time interval: For the lifting device and the pipe, rigid body kinematics equations are used, combined with motion commands output by the control module and lifting device status data to predict their future position and attitude; For obstacles, the motion trajectory of static obstacles is fixed coordinates, and the motion trajectory of dynamic obstacles is uniform motion model, combined with the speed and direction detected by millimeter-wave radar to predict their future position.

[0104] Continuous collision detection and shortest collision time calculation: The split axis theorem algorithm is used for continuous collision detection. The split axis theorem algorithm traverses all possible contact surfaces between the dynamic safety envelope and the obstacle to determine whether there is an axis on which the projections of the two objects do not overlap. If the projections on all axes overlap, a collision risk is determined. At the same time, the shortest collision time (TTC) is calculated, and the calculation formula is shown in Equation (2):

[0105]

[0106] In formula (2):

[0107] The shortest collision time is the shortest time from the current moment for the dynamic safety envelope to collide with the obstacle. The smaller the value, the higher the collision risk.

[0108] For the surface of the dynamic safety envelope, the first The triangular facet and the surface of the obstacle. The current distance between the triangular faces is obtained by calculating the projection length of the line connecting the centroids of the two triangular faces onto their common normal.

[0109] It represents the relative velocity component of the two triangular faces along the line connecting them, which is the vector difference between the velocity of the dynamic safety envelope driven by the lifting device and the velocity of the obstacle along the line connecting them. A positive value indicates that the two are moving closer to each other, and a negative value indicates that they are moving further apart.

[0110] To find the minimum function, calculate the minimum value of all triangular facet combinations. Take the smallest value among them as This is to reflect the scenarios where collisions are most likely to occur.

[0111] Hierarchical obstacle avoidance command generation: Employs a hierarchical autonomous obstacle avoidance decision-making mechanism, based on... Different levels of obstacle avoidance commands are generated based on the type of obstacle, as follows:

[0112] Threshold setting: Preset first threshold With the second threshold ,and Both are time constants, whose values ​​are preset based on the speed of the lifting equipment, the complexity of the working area, and safety redundancy requirements. This is the starting threshold for collision warning. This is the threshold for triggering emergency braking.

[0113] Hierarchical decision-making:

[0114] when At this time, the system is in normal operating condition: the risk of collision is extremely low, the control module only sends visual warning signals to the remote control terminal, and does not interfere with the normal operation of the spreader, so as to ensure work efficiency;

[0115] when When the system initiates intervention-level obstacle avoidance: the control module uses the artificial potential field method to generate an obstacle avoidance path—the artificial potential field method treats the dynamic safety envelope as a "particle", the obstacle as a "repulsive potential field source", and the target position as an "attractive potential field source", and determines a smooth path without collision by calculating the potential field force; at the same time, the control execution module reduces the lifting, traveling and luffing speed of the spreader, drives the spreader to move along the planned path, and realizes active obstacle avoidance;

[0116] when When the system triggers emergency braking and obstacle avoidance: the control module immediately generates a stop command, cuts off the power source of the lifting, traveling and luffing mechanisms of the spreader, and simultaneously drives the brakes to stop the spreader in an emergency, thus avoiding a collision.

[0117] Dynamic obstacle threshold adjustment: For dynamic obstacles, since their positions change over time, dynamic adjustment is required. and ,in The adjustment formula is shown in equation (3):

[0118]

[0119] In formula (3):

[0120] The adjusted first threshold;

[0121] The initial first threshold;

[0122] The real-time speed of the dynamic obstacle is collected by millimeter-wave radar.

[0123] The maximum permissible speed of the spreading device is determined by its mechanical properties and safety standards.

[0124] when When it increases, This increases the time available for early warning, allowing the system more time to avoid obstacles.

[0125] Execution module's action response:

[0126] The execution module includes the hoisting mechanism, traveling mechanism, and luffing mechanism of the spreader. Each mechanism is equipped with a drive motor and a braking device, and its action response process is as follows:

[0127] When receiving an intervention-level obstacle avoidance command, the hoisting mechanism adjusts the hoisting speed according to the path plan, the traveling mechanism adjusts the horizontal movement direction and speed, and the luffing mechanism adjusts the boom's elevation angle and amplitude. The three work together to ensure that the spreader moves smoothly along the planned path.

[0128] When an emergency braking command is received, the drive motors of each mechanism are immediately de-energized, and the braking device achieves full braking within a preset time to ensure that the spreader stops moving within the shortest distance. At the same time, the execution module feeds back the action status signal to the control module to ensure closed-loop control of command execution.

[0129] Visualization of 5G Private Network Communication and Remote Control Terminal:

[0130] 5G Private Network Communication: The system uses a 5G private network to achieve communication between the remote control terminal and the lifting device. Its network architecture includes a terminal access layer, an edge computing layer, and a cloud control layer. The terminal access layer is deployed at the lifting device site, using 5G industrial modules to realize data transmission between the sensing and control modules, supporting high bandwidth and low latency. The edge computing layer is deployed on the site edge server, responsible for running dynamic safety envelope modeling and collision risk prediction algorithms, reducing the amount of data transmission and computing load in the cloud. The cloud control layer is connected to the edge computing layer through a dedicated fiber optic line, providing an access interface for the remote control terminal to realize collaborative management of multiple lifting devices. To ensure the reliability of control commands, a redundant transmission protocol is adopted: temporal redundancy—important control commands (such as emergency stop commands and path update commands) are repeatedly sent three times to avoid loss in a single transmission; spatial redundancy—multi-antenna diversity reception technology is used to improve the stability of signal reception and reduce the impact of interference.

[0131] Remote control terminal visualization: The visualization interface is built based on digital twin technology, and its specific functions include:

[0132] Establish high-fidelity 3D models of lifting equipment, pipes, and the working environment, and receive real-time status data of physical entities through a 5G private network to ensure that the virtual model is synchronized with the movement status of the physical entities;

[0133] The system renders a dynamic safety envelope in real time and displays it with a semi-transparent red highlight to visually present the dynamic protection range of the pipe. At the same time, obstacle location markers and the shortest collision time value are overlaid on the interface to provide operators with clear risk warnings.

[0134] Provides operator interaction functions, including one-click emergency stop, manual route planning, and system status monitoring;

[0135] The integrated alarm log system automatically records all obstacle avoidance events and sensor anomalies, supports historical data playback, and facilitates post-operation analysis and system optimization.

[0136] Operation of the fault prediction and health management module:

[0137] The fault prediction and health management module is integrated into the control module, and its operation process is as follows:

[0138] Motor status monitoring: Real-time acquisition of three-phase current waveform data of the drive motor in the execution module, frequency domain analysis of the current waveform through fast Fourier transform, extraction of characteristic frequency components—the fundamental amplitude reflects the load status of the motor, and the harmonic amplitude reflects abnormalities such as winding faults and bearing wear of the motor;

[0139] Health score calculation: The extracted feature frequency components are input into a convolutional neural network model. The model structure includes an input layer, a convolutional layer, a pooling layer, and a fully connected layer. The health score ranges from 0 to 100, with a higher score indicating a better motor health status.

[0140] Early warning and communication backup: When the health score is lower than the preset threshold, the module generates a predictive maintenance alarm and prompts the operator to replace the faulty parts or adjust the motor operating parameters through the remote control terminal; at the same time, it monitors the signal strength of the 5G private network in real time, and automatically switches to the backup communication channel when the signal strength is lower than the set threshold to ensure uninterrupted system communication.

[0141] In summary, this embodiment achieves remote wireless control of an automated pipe lifting device by constructing a complete closed loop of sensing, computation, decision-making, execution, and monitoring. Its core innovations lie in: constructing a dynamic safety envelope based on the pipe's dynamic posture, adjusting to the swing angular velocity and wind speed, thus overcoming the limitations of existing static safety boundaries; employing a hierarchical obstacle avoidance decision-making mechanism, combined with adjusting the warning threshold based on the speed of dynamic obstacles, enhancing the ability to handle collision risks in complex environments; relying on the low latency and redundant transmission of the 5G private network to ensure the reliability of remote control; and achieving transparency in the operation process and predictability in equipment maintenance through digital twin visualization and fault prediction. This system effectively solves the problems of low accuracy and high safety risks in existing remote control of pipe lifting, significantly improving operational efficiency and safety.

[0142] Example 2

[0143] like Figure 1 As shown in Example 1, this example elaborates on the specific steps of the remote wireless control system for automatic pipe lifting devices during operation. The specific steps are as follows:

[0144] 1. System startup and initialization:

[0145] The spreader body is powered on, and each module performs a self-test; the remote control terminal starts up and establishes a connection with the spreader through the 5G private network; the visual interface loads the digital twin model and displays the initial status of the spreader, pipes and working environment.

[0146] 2. Real-time acquisition of sensor data:

[0147] The spreader status sensors (navigation satellite system receiver, inertial measurement unit) continuously collect the spreader's position, velocity, acceleration, and attitude angle data;

[0148] Pipe attitude sensors (multi-line LiDAR, vision sensors) scan the suspended pipes to acquire point cloud and image data, and identify the real-time motion attitude of the pipes.

[0149] Environmental sensors (3D LiDAR, millimeter-wave radar) scan the work area to detect static and dynamic obstacles; digital anemometers measure environmental wind speed and direction in real time.

[0150] 3. Data transmission and preprocessing:

[0151] Sensing data is transmitted to the edge computing layer through the terminal access layer of the 5G private network;

[0152] The control module verifies and filters the received data to remove outliers and ensure data validity.

[0153] 4. Construction of dynamic safety envelope:

[0154] The control module preprocesses the pipe point cloud data, including noise removal and downsampling;

[0155] The central axis of the pipe was extracted by point cloud segmentation and principal component analysis, and a three-dimensional wireframe model was fitted.

[0156] Based on the real-time angular velocity of the pipe and the ambient wind speed, the safe distance is dynamically calculated, and a dynamic safety envelope represented by a triangular mesh is generated.

[0157] 5. Collision risk prediction:

[0158] By unifying the dynamic safety envelope and obstacle model into the same coordinate system, future motion trajectories can be predicted.

[0159] A continuous collision detection algorithm is used to calculate the shortest collision time (TTC) between the dynamic safety envelope and obstacles.

[0160] 6. Obstacle avoidance decision generation:

[0161] Compare TTC with preset thresholds (T1, T2):

[0162] If TTC > T1, the system will only send a visual warning to the remote control terminal and will not intervene in the operation.

[0163] If T1≥TTC>T2, the system initiates intervention-level obstacle avoidance, generates a collision-free path, and reduces the speed of the spreader.

[0164] If TTC≤T2, the system triggers emergency braking, cuts off the power source, and performs an emergency stop.

[0165] For dynamic obstacles, the threshold is dynamically adjusted according to their speed.

[0166] 7. Command execution and spreader control:

[0167] The execution module receives obstacle avoidance commands and drives the hoisting, traveling, and luffing mechanisms to perform corresponding actions:

[0168] During intervention-level obstacle avoidance, the spreader moves smoothly along the planned path;

[0169] When emergency braking is applied, the lifting equipment stops moving immediately.

[0170] 8. Status feedback and visual updates:

[0171] The execution module feeds back the action status to the control module;

[0172] The remote control terminal's visual interface updates the dynamic safety envelope, obstacle location, TTC value, and spreader status in real time, and records alarm logs.

[0173] 9. Cyclic monitoring and adaptive adjustment:

[0174] The system continuously executes steps 2 to 8 in a loop to achieve real-time monitoring, decision-making, and adjustment.

[0175] The fault prediction module monitors the motor health status and communication signal strength, and triggers an early warning or switches the backup channel when an abnormality occurs.

[0176] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any simple modifications, equivalent changes and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. A remote wireless control system of a pipe automatic sling, comprising a sling body, a remote control terminal, and a sensing module, a control module and an execution module arranged on the sling body, the remote control terminal being connected with the sling body through a wireless communication network, characterized in that: the sensing module comprises a sling state sensor, a pipe attitude sensor and an environment sensor, the sling state sensor is used to collect state data of the sling, including position, speed, acceleration and attitude angle of the sling, the pipe attitude sensor is used to collect motion attitude data of the pipe, including point cloud data and image data of the pipe, and the environment sensor is used to collect environmental obstacle data and environmental wind speed data; the control module comprises a processor and a memory, the memory stores a computer program, and the processor realizes the following steps when executing the computer program: real-time receiving sensing data from the sensing module; based on the pipe motion attitude data, a dynamic safety envelope is constructed through a point cloud processing algorithm and an attitude recognition algorithm, the dynamic safety envelope is a continuum in three-dimensional space, a surface of the dynamic safety envelope maintains a dynamic safety distance with an outer surface of the pipe, and the dynamic safety distance is dynamically adjusted according to a real-time swing angular velocity of the pipe and an environmental wind speed; the dynamic safety envelope is fused with the environmental obstacle data, a collision risk between the dynamic safety envelope and the obstacle is predicted through a continuous collision detection algorithm, and a shortest collision time is calculated; when the shortest collision time is less than or equal to a preset threshold, an obstacle avoidance control instruction is generated; the execution module receives the obstacle avoidance control instruction, and drives a hoisting mechanism, a walking mechanism and a luffing mechanism of the sling to execute an obstacle avoidance action; the wireless communication network adopts a 5G private network, and the remote control terminal provides a visual interface; the step of constructing the dynamic safety envelope specifically comprises: preprocessing point cloud data collected by the pipe attitude sensor, including outlier removal, smoothing filtering and voxel grid downsampling; extracting pipe point cloud through a point cloud segmentation algorithm, the point cloud segmentation algorithm is based on region growing and clustering method, identifies pipe point cloud clusters and calculates their centroids and bounding boxes; based on the extracted pipe point cloud, a principal component analysis algorithm is used to calculate a center axis of the pipe, and a three-dimensional wireframe model of the pipe is fitted; According to the real-time swing angular velocity of the pipe material and the environmental wind speed , the dynamic safety distance is calculated , and the calculation formula is: wherein is a base safety distance, is a swing angular velocity weight coefficient, is a wind speed weight coefficient, , and is a preset positive real number; Expanding a fitted three-dimensional wireframe model of a pipe along its surface normal direction by a dynamic safety distance generating a dynamic safety envelope represented by a triangular mesh, the volume and shape of the dynamic safety envelope being updated in real-time and response to changes in the three-dimensional wireframe model the step of predicting the collision risk specifically comprises: representing the dynamic safety envelope as a triangular mesh model, and representing the environmental obstacle data as a three-dimensional point cloud or a bounding box model; unifying the dynamic safety envelope model and the obstacle model to the same coordinate system through a space-time synchronization algorithm, and predicting motion trajectories of the dynamic safety envelope and the obstacle in a future time interval; the motion trajectory prediction adopts a physics-based dynamics model, for the sling and the pipe, a rigid body kinematics equation is used, and for the obstacle, a uniform motion model is used; continuous collision detection is performed to calculate the shortest collision time between the dynamic safety envelope and the obstacle, the continuous collision detection algorithm adopts a separating axis theorem algorithm to traverse all possible contact surfaces of the dynamic safety envelope and the obstacle. The shortest collision time The formula for calculating is: wherein is the current distance between the surface of the dynamic safety envelope and the surface of the obstacle, is the current distance between the surface of the dynamic safety envelope and the surface of the obstacle, is the current distance between the surface of the dynamic safety envelope and the surface of the obstacle, is the component of the relative velocity between the two in the direction of the line, denotes taking the minimum over all pairs of facets.

2. The remote wireless control system for an automatic pipe sling as claimed in claim 1, wherein The spreader state sensor comprises a navigation satellite system receiver and an inertial measurement unit, the navigation satellite system receiver is used to acquire position coordinates of the spreader, the inertial measurement unit is used to measure angular velocity and linear acceleration of the spreader, and data fusion is performed through a Kalman filtering algorithm, and real-time three-dimensional position, velocity, acceleration, pitch angle and roll angle of the spreader are output; The pipe posture sensor comprises a multi-line laser radar and a visual sensor, the multi-line laser radar is installed below the spreader, scans the suspended pipe in a high-speed scanning mode, and generates high-density point cloud data, and the visual sensor comprises a high-resolution binocular camera, depth information of the pipe is extracted through a stereo vision algorithm, and the depth information is fused with the point cloud data of the multi-line laser radar, and a point cloud registration and feature matching method is adopted in the fusion algorithm; The environment sensor comprises a three-dimensional laser radar and a millimeter wave radar, the three-dimensional laser radar is arranged around the spreader and the station, is used to construct a real-time three-dimensional point cloud map of the working environment, and is used to identify static obstacles and dynamic obstacles, and the millimeter wave radar is used to detect the motion speed and direction of the obstacles; The sensing module further comprises a digital wind speed and direction meter, which is used to measure the environmental wind speed and direction in real time.

3. The remote wireless control system for an automatic pipe sling as claimed in claim 1, wherein The step of generating the obstacle avoidance control instruction adopts a hierarchical autonomous obstacle avoidance decision mechanism, and specifically comprises the following steps: Setting a first threshold and a second threshold , greater than , and is a time constant; When the shortest collision time is greater than , the system is in normal operation state, only sends visual warning signal to remote control terminal, and does not intervene in the spreader operation. When less than or equal to and greater than , the system starts the intervention level obstacle avoidance, automatically generates an obstacle avoidance path planning, the obstacle avoidance path planning adopts the artificial potential field method, calculates a smooth path without collision, and controls the execution module to move the spreader along the smooth path while reducing the spreader running speed; When less than or equal to When the distance is less than or equal to the threshold value, the system triggers an emergency braking level obstacle avoidance, immediately generates a stop instruction, cuts off the power source of the spreader, and makes the brake act to achieve emergency stop. The hierarchical autonomous obstacle avoidance decision mechanism further comprises obstacle type identification, for dynamic obstacles, and the value of which is dynamically adjusted according to the obstacle speed, and the adjustment formula is: wherein is the adjusted first threshold value, is the obstacle speed, is the maximum allowed speed of the spreader.

4. The remote wireless control system for an automatic pipe sling as claimed in claim 1, wherein, The wireless communication network adopts a 5G private network technology, and a network architecture thereof comprises a terminal access layer, an edge computing layer and a cloud control layer; The terminal access layer is arranged on the spreader site, and a 5G industrial module is adopted to realize data transmission of the sensing module and the control module; The edge computing layer is arranged on an edge server of the station, is responsible for running a dynamic safety envelope modeling and collision risk prediction algorithm, reduces the cloud load, and the edge server is connected with the terminal access layer through a 5G core network; The cloud control layer provides a remote control terminal access, realizes multi-spread coordination management, and the cloud control layer is connected with the edge computing layer through a fiber private line; The wireless communication network further adopts a redundant transmission protocol, important control instructions include emergency stop instructions and path update instructions, time redundancy and space redundancy transmission are performed, the time redundancy means that the instructions are repeatedly sent three times, and the space redundancy means that the instructions are received through multi-antenna diversity.

5. The remote wireless control system for an automatic pipe sling as claimed in claim 1, wherein, The visual interface of the remote control terminal is constructed based on a digital twinning technology, and specifically comprises the following steps: High-fidelity three-dimensional models of the spreader, the pipe and the working environment are established, and are kept synchronous with physical entities; A dynamic safety envelope is rendered in real time, is highlighted in a semi-transparent red color, and position information and shortest collision time information of obstacles are superimposed; An operator interaction function is provided, including one-key emergency stop, manual path planning and system state monitoring; The visual interface further integrates an alarm log system, records all obstacle avoidance events and sensor abnormalities, and supports historical data playback; The remote control terminal adopts an industrial-grade tablet computer, is provided with a touch screen and physical buttons, and meets outdoor operation requirements.

6. The remote wireless control system for an automatic pipe sling as claimed in claim 1, wherein The system further comprises a fault prediction and health management module, the fault prediction and health management module is integrated in the control module, and specifically comprises the following steps: Real-time monitoring of current signals of motors in the execution module, and acquisition of three-phase current waveform data; Fast Fourier transform is performed on the current waveform to extract characteristic frequency components, including fundamental and harmonic amplitudes; The characteristic components are input into a deep learning model, which is a convolutional neural network including an input layer, convolutional layers, pooling layers, and fully connected layers, to output a motor health score; When the health score is lower than a preset threshold, a predictive maintenance alert is generated to prompt replacement of components or adjustment of parameters; The fault prediction and health management module also monitors wireless communication network signal strength and automatically switches to a backup communication channel when the signal strength is lower than a set threshold.

Images