Crane group boom anti-collision early warning method and system based on digital twinning

Abstract

The application provides a crane group boom anti-collision early warning method and system based on digital twinning, which comprises the following steps: based on the data collected by a rotary encoder, an angle sensor, an inertial measurement unit, a laser radar and a ship body attitude measuring device, a dynamic three-dimensional digital twinning model is constructed and updated in real time; in the model, real-time path planning is performed on each crane boom, the spatial interference risk with the bulkhead, hatch coaming and other booms is evaluated, and a collision-free optimal action path is generated; the path nodes are dynamically risk evaluated, the minimum distance, relative speed, motion direction angle and ship body swing disturbance are comprehensively considered, and a collision risk score is generated; according to the score, hierarchical intervention control is implemented, the collision is avoided, and the attitude stability of the suspended goods is maintained. The application realizes the safety early warning and intelligent regulation and control of multi-machine cooperative operation under high dynamic sea conditions, and effectively improves the operation efficiency and safety.

Description

Technical Field

[0001] This invention belongs to the field of intelligent port equipment, and in particular relates to a method and system for anti-collision early warning of crane booms based on digital twins. Background Technology

[0002] With the growth of global bulk commodity trade and the continuous expansion of pulp transportation, heavy-lift pulp carriers equipped with multiple cranes have gradually become the industry's mainstay. However, the internal structure of such vessels is narrow and complex, with bulkheads, cargo yards, and hatch coamings creating irregular obstacle spaces. When multiple cranes operate simultaneously inside the hold, interference between the booms, hooks, cargo, and the ship's structure is highly likely. Traditional operating methods rely on operators' experience to judge distances and maintain collaboration through intercom communication. However, under conditions of rolling waves, insufficient nighttime lighting, or continuous high-intensity operation, operators find it difficult to accurately assess the dynamic distances between booms and between booms and obstacles, easily leading to misoperation, delayed judgment, and blind spots. Adding to the complexity, in the marine environment, the ship's roll and pitch, cargo sway, and nonlinear swaying of the wire ropes all contribute to affecting the spatial position of the booms, causing deviations between the actual movement trajectory and the operator's expectations. Existing anti-collision systems relying on single sensors or fixed limiters cannot effectively capture these dynamic changes, resulting in frequent false alarms and missed alarms. In addition, many existing collision avoidance systems are usually based on static modeling or two-dimensional limiters, which can only handle some working conditions. They do not have the overall perception capability of the coupled motion of the boom, cargo and ship, nor can they predict the spatial collision trend that may occur in the next few hundred milliseconds, let alone provide linkage decision-making between multiple booms. Summary of the Invention

[0003] The purpose of this invention is to design a collision avoidance and early warning method and system for crane booms based on digital twins. This system can maintain high-precision spatial perception under complex disturbance conditions such as wind and waves at sea, cargo swaying, and ship swaying, and can identify the current danger distance. It can also identify potential collision trends that may occur in future boom movements.

[0004] To achieve the above objectives, a method for anti-collision early warning of crane booms in a crane group based on digital twins is provided in a first aspect of the present invention, the method comprising: A dynamic three-dimensional digital twin model synchronized with the actual working environment is constructed. The digital twin model is based on the synchronous data collected by the rotary encoder installed on the base of each crane boom, the angle sensor at the root of the boom, the inertial measurement unit at the top of the hook, and the lidar array around the cargo hatch, and is updated in real time in combination with the ship's attitude information. In the digital twin model, real-time path planning is performed for each crane boom. During the path planning process, the spatial interference risk between the boom, the suspended cargo and the bulkhead, hatch coaming and other booms is dynamically evaluated, and the optimal action path without collision is generated. Dynamic risk assessment is performed on each path node in the optimal motion path, taking into account the minimum distance between the boom and the obstacle, the relative motion speed, the angle of motion direction, and the hull roll and pitch disturbance factors to generate a collision risk score. The crane control commands are subject to graded intervention based on the collision risk score. The intervention strategy is determined based on the risk level, the motion redundancy of the current path node, and the degree of coordination dependence with other booms, and the attitude stability of the suspended cargo is maintained during the intervention process.

[0005] Furthermore, the dynamic three-dimensional digital twin model adopts a hybrid structure modeling of skeleton-flexible cable-rigid body group, in which the boom is represented by skeleton nodes, the cable is simulated by flexible line segments, and the cargo is represented by rigid body nodes bound to the bottom of the cable. All objects are bound to the ship coordinate system that is dynamically updated with the ship's attitude.

[0006] Furthermore, the synchronization data is transmitted via industrial Ethernet or controller area network bus and fused using a multi-source Kalman filter method with a time sliding window to eliminate fluctuations caused by sensor sampling delay and environmental interference.

[0007] Furthermore, the real-time path planning employs an improved A* search algorithm, which uses a risk-weighted cost function that includes obstacle distance and hull sway angle as the evaluation criterion in three-dimensional spatial nodes, and excludes path nodes that overlap with the predicted envelope of other booms within the next two seconds.

[0008] Furthermore, the dynamic risk assessment employs a soft-border mechanism, dynamically expanding the three-dimensional envelope boundary of the boom and cargo based on the cargo swing amplitude and ship sway, in order to anticipate the additional space occupied by disturbances.

[0009] Furthermore, when the collision risk scores of three consecutive path nodes exceed a preset threshold, an early warning event is triggered and an early warning structure containing the risk node index, state variables, and triggering reasons is generated.

[0010] Furthermore, the graded intervention includes four levels of control strategies: maintaining the original speed, gently limiting the speed, slowing down the movement, and suspending the main degree of freedom. Each level of strategy corresponds to a different collision risk score range.

[0011] Furthermore, before pausing the main degree of freedom, the estimated inertial path offset of the cargo, which is jointly determined by the hook velocity vector, the sling length, and the cargo swing angle, is calculated. If the estimated value exceeds the safety threshold, the execution cycle of the current node is extended and a dynamic buffer window is inserted to stabilize the cargo attitude.

[0012] Furthermore, the ship's attitude information is acquired by two inertial measurement units located at the center of the bridge and main deck, and output in the form of three-dimensional attitude quaternions to drive the affine transformation of all objects in the digital twin model.

[0013] A second aspect of the invention provides a crane group boom anti-collision early warning system based on digital twins, the system comprising: The digital twin modeling module is used to construct and update a dynamic three-dimensional digital twin model that is synchronized with the actual working environment based on the synchronous data collected by the rotary encoder of the crane boom base, the angle sensor at the root of the boom, the inertial measurement unit at the top of the hook, the lidar array around the cargo hatch, and the ship's attitude information. The path planning module is used to perform real-time path planning for each crane boom in the digital twin model, dynamically assess the spatial interference risk between the boom, the suspended cargo and the bulkhead, hatch coaming and other booms, and generate the optimal action path without collision. The risk assessment module is used to perform dynamic risk assessment on each path node in the optimal action path, taking into account the minimum distance between the boom and the obstacle, the relative motion speed, the angle of motion direction, and the hull roll and pitch disturbance factors, to generate a collision risk score. The control intervention module is used to implement graded intervention on the crane control commands based on the collision risk score. The intervention strategy is determined based on the risk level, the motion redundancy of the current path node and the degree of coordination dependence with other booms, and maintains the posture stability of the suspended cargo during the intervention process.

[0014] The beneficial technical effects of the present invention are at least as follows: To address the aforementioned issues, this invention provides a collision avoidance and early warning method and system for crane booms based on digital twins. It reconstructs the real spatial relationship between the boom, cargo, and ship by fusing data from multiple sensors, and on this basis, achieves a closed-loop system encompassing path planning, dynamic risk assessment, and real-time control intervention. This invention maps the actual motion state of the boom through a dynamic twin model, enabling the system to maintain high-precision spatial awareness under complex disturbances such as sea waves, cargo swaying, and ship rolling. By incorporating sway extension boundaries and future motion prediction mechanisms tailored to sea conditions into the twin model, the system can not only identify the current danger distance but also anticipate potential collision trends in future boom movements. By introducing a comprehensive risk scoring model that integrates distance, speed, angle, and ship disturbance factors, the risk assessment aligns with the actual dynamic characteristics of offshore lifting operations. By introducing factors such as motion redundancy, cooperative coupling, and cargo attitude maintenance during the control command generation stage, the system possesses adjustable continuous control capabilities, preventing imbalances in cooperative operations caused by excessive motion restrictions. This invention is the first to achieve the overall integration of a four-stage real-time closed loop of "dynamic space modeling - path decision-making - risk prediction - control intervention", providing a forward-looking, interpretable and engineering-feasible intelligent anti-collision method for multi-arm cabin operations, which significantly improves the safety and efficiency of hoisting operations in complex cabin environments. Attached Figure Description

[0015] The present invention will be further described with reference to the accompanying drawings, but the embodiments in the drawings do not constitute any limitation on the present invention. For those skilled in the art, other drawings can be obtained based on the following drawings without creative effort.

[0016] Figure 1 This is a flowchart of the anti-collision early warning method for crane booms based on digital twins according to the present invention.

[0017] Figure 2 This is a framework diagram of the crane group boom anti-collision early warning system based on digital twins according to the present invention. Detailed Implementation

[0018] Embodiments of the present invention are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and should not be construed as limiting the present invention.

[0019] In one or more embodiments, such as Figure 1 As shown, a collision avoidance early warning method for crane booms based on digital twins is disclosed, the method comprising the following: S1: Construct a dynamic three-dimensional digital twin model that is synchronized with the actual working environment. The digital twin model is based on the synchronous data collected by the rotary encoder installed on the base of each crane boom, the angle sensor at the root of the boom, the inertial measurement unit at the top of the hook, and the lidar array around the cargo hatch, and is updated in real time in combination with the ship's attitude information. Specifically, the input comes from sensor systems deployed on each marine crane, including: a rotary encoder mounted on the boom base to acquire the slewing angle; an angle sensor located at the boom root to measure the amplitude angle in real time; an IMU module fixed to the top of the hook to provide triaxial acceleration and angular velocity at the lifting point; and a lidar array distributed around the cargo hatch to detect the relative spatial relationships of cargo stacks, hatch coamings, obstacles, etc. All sensors are connected to the ship's central control unit via industrial Ethernet or CAN bus, and are synchronously sampled according to a unified timestamp, with a time resolution configurable from 50ms to 100ms. The collected data is fed into the data fusion module in real time as input for updating the twin environment.

[0020] To ensure the spatiotemporal consistency between the states of objects in the virtual space and the actual operational scenario, a multi-source Kalman filter method with a time sliding window is used to fuse the above inputs, eliminating data fluctuations caused by delay asynchrony and environmental interference, and obtaining continuous state estimation results. At the current moment... , No. The spatial state vector of a crane boom Updated by the following function: ; in, It indicates the dynamic state of the boom in three-dimensional space, including the base rotation angle, boom amplitude angle, hook three-axis position and speed, etc. This is the state estimation result from the previous moment; These are the observations jointly provided by various sensors during this period, including raw quantities such as angle, voltage, current, and linear velocity; To control the cycle interval, it is generally set to 100ms. (Function) The system integrates two parts: state prediction and observation update. The state prediction is based on a simplified rigid body model of the boom, while the observation update is based on a weighted fusion of sensor accuracy, which has good stability under navigation environmental disturbances.

[0021] Each boom is bound to a set of spatial skeleton nodes in the twin space, with each skeleton corresponding one-to-one with the physical boom structure. Cargo is represented by a flexible line segment of equal length extending downwards from the hook node, simulating a sling, with the bottom node bound to the cargo's center point. The cargo's state is determined by the sling length, the lifting point position, and the swing angle calculated by the IMU. The ship's attitude is measured by dual IMU modules positioned at the center of the bridge and main deck, outputting a three-dimensional attitude quaternion. The twin environment binds all models to the ship's coordinate system, ensuring that the models accurately reflect relative spatial relationships even under ship roll and pitch conditions. For example, when the ship pitches more than 3°, the positions of all booms, cargo, and obstacles in the model will undergo an affine transformation with the entire ship's coordinate system, while maintaining the relative rigid connections between objects.

[0022] The updated 3D twin model is visualized using the Unity engine, maintaining a real-time frame rate of over 10 frames per second, for use by subsequent systems such as the operation monitoring interface, path planning algorithm module, and collision detection module. Current moment The twin model outputs a set of state vectors containing each boom and its suspended load. The collection of data serves as the spatial basis for path optimization and safety assessment. Taking a typical operation as an example, when lifting a bundle of pulp, if the hook IMU measures a Z-axis acceleration that continuously deviates from the direction of gravity by more than 0.3g, and the system calculates that the cargo swing angle exceeds 15°, then the cargo node in the twin model will deviate from its stationary position and shift in the opposite swing direction by a certain length, while simultaneously tilting as a whole. The corresponding collision calculation module will dynamically expand the boundary of the cargo's enclosing volume. This highly efficient and consistent twin space reconstruction mechanism provides crucial dynamic support for subsequent action planning.

[0023] The twin modeling in this step does not employ a general modeling platform or CAD preloading method. Instead, it uses a hybrid modeling structure of "skeleton-flexible cable-rigid body group" designed specifically for the operational characteristics of marine cranes. This achieves the synchronous representation of the physical connections and real-time motion characteristics between cargo, cables, and boom. Furthermore, to address the coordinate drift issue caused by changes in ship attitude, a multi-point IMU is introduced, and a dynamic coordinate system update mechanism is established. This ensures that the twin model can stably track the actual operational status even in highly dynamic environments such as when the ship is anchored or berthed at a dock.

[0024] S2: Real-time path planning is performed for each crane boom in the digital twin model. During the path planning process, the spatial interference risk between the boom, the suspended cargo and the bulkhead, hatch coaming and other booms is dynamically evaluated, and the optimal action path without collision is generated. Specifically, the input comes from the digital twin model generated in step 1, the core of which is the time of each boom. dynamic state set Each of them This includes the boom's three-dimensional position, attitude angle, angular velocity, and real-time relative distances to obstacles such as cargo, bulkheads, and hatch coamings in twin space. This dataset is obtained by synchronously collecting data from on-site lidar, IMU, and rotary encoders, and then fusing it using Kalman filtering. It has already undergone dynamic coordinate calibration in step 1 and can be directly used as input for the path planning and collision avoidance warning modules.

[0025] The core objective of path planning is to generate a motion path for each crane that does not spatially interfere with other cranes, cargo, or bulkheads, while ensuring physical accessibility. Considering the unique three-degree-of-freedom coupled motion (slewing, luffing, and hoisting) of cranes and the overall coordinate system disturbance caused by ship rolling, path planning cannot use the conventional Euclidean shortest path. Instead, it must construct a workspace grid with dynamic risk weights using a twin model. In this space, each node represents an executable crane attitude point, and each node is accompanied by a risk label read in real-time from the twin model to determine whether it is close to a danger zone. For example, when the ship rolls, the bulkheads will periodically shift in position in the twin space, significantly reducing the number of safe nodes in the path; therefore, the risk labels need to be updated in real time.

[0026] In this scenario, this step employs a dynamically risk-weighted path cost model, constructing the total cost of each candidate path using the following formula: ; in, Indicates the first Platform crane boom at all times The total cost of the path; For path number The spatial distance from the node to the target location is calculated by subtracting the target coordinates from the three-dimensional coordinates of the node in the twin model; The collision risk coefficient of this node is calculated by back-calculating the closest distance to surrounding entities in the twin model. If the distance between the path node and the nearby obstacle is less than a set threshold (such as 1.5m), the risk coefficient will increase in a linear decay manner. and These represent the ship's roll and pitch angles, respectively, and are provided directly by the IMU installed on the ship's hull to increase the operational cost during intense ship movements. To fix the adjustment coefficient, it is set according to the values ​​tested on site. For example, in the case of three cranes working together, It is usually set to a higher value to suppress the amplitude of path movement when the hull rolls greatly.

[0027] The cost function described above differs from traditional path planning in that it incorporates a ship sway term. This feature is specifically designed for offshore lifting environments and can significantly improve the safety factor of path planning in rough seas. For example, when the ship rolls at 5°, this feature will automatically eliminate high-amplitude paths, effectively avoiding the risk of cargo hitting the bulkhead due to the combined effect of boom sway and ship roll.

[0028] To ensure the operational feasibility of path nodes, the system restricts the range of three degrees of freedom changes at each step when constructing the state-action diagram, making it consistent with the response capability of an actual boom electro-hydraulic control system. For example, the boom luffing angle cannot exceed 2° within one sampling period, the slewing angular velocity cannot exceed 80% of the rated angular velocity, and the hoisting speed must be maintained within the range of 50% to 100% of the rated speed. The generation of each feasible node is based on the operational constraint table of the twin model, which is constructed from field survey data and equipment manuals.

[0029] The path search employs an improved A* strategy, using the 3D spatial nodes in the twin model as the search space, and a cost function... As a basis for evaluation. During the search process, the system dynamically reads the state vectors of other booms. ( The system calculates the corresponding collision envelope of each candidate node. If a candidate node overlaps with the envelope of any other arm, it is immediately eliminated to avoid potential interference when the arms are crossed or turned inward. Furthermore, the system provides a "two-second prediction window," meaning that each time a path node is expanded, the current action is extended to several future sampling periods using linear prediction. If the predicted path intersects with the future envelope of another arm, it is determined to be an infeasible node. This mechanism is particularly crucial for the coordinated movement of the two arms when lifting pulp bundles, preventing mutual restraint caused by asynchronous movements.

[0030] The optimal path obtained based on the above mechanism is formed after the calculation. It contains several executable attitude nodes, each with its target position, target angle, and expected arrival timestamp. The path is fed back to the trajectory following module at fixed intervals to achieve the actual boom motion control. The output consists of two parts: one is the... Optimal path of the crane boom Secondly, there is the set of dynamic risk trajectories corresponding to this path, used for dynamic collision assessment and real-time control intervention in the next step. Each node in the path can be directly used for target position or attitude commands issued by the control system.

[0031] S3: Perform dynamic risk assessment on each path node in the optimal motion path, taking into account the minimum distance between the boom and the obstacle, the relative motion speed, the angle of motion direction, and the hull roll and pitch disturbance factors, and generate a collision risk score. Specifically, the input comes from two outputs of step 2: one is the optimal path for each boom within the current work cycle. The first part consists of multiple attitude nodes ordered by timestamps, each containing control targets such as three-dimensional coordinates, rotation angle, amplitude angle, and lift height; the second part is the set of dynamic risk trajectories along the path, which contains the risk trends initially calculated for each node during the planning phase. Simultaneously, the continuously updated twin state vector from step 1... These parameters are directly incorporated into this step, allowing the system to utilize the latest boom attitude, cargo swing angle, hull roll angle, and the current actual position of obstacles during risk assessment. This step relies on these three inputs to construct a dynamic collision risk model and to identify and issue warnings in real time for various hazardous states during path execution.

[0032] In crane group operations, the cabin space is narrow, the bulkheads are irregularly tilted, and cargo stacking, such as pulp bales, can cause localized bulges. Furthermore, when the crane booms operate across multiple cabins, there is a risk of interference due to the booms swerving inwards or outwards. Therefore, risk assessment cannot rely on a single indicator; it needs to utilize multiple factors simultaneously, including distance, speed, structural morphology, and ship disturbance. The system performs node-by-node analysis within the path execution cycle (100ms). The node state is predicted, and combined with The provided real-time geometric model for the entire scene constructs a 3D envelope for the current node. This envelope employs a "soft border" mechanism: an additional virtual boundary is added outside the static envelope, dynamically stretched based on cargo swing and ship sway. This allows risk assessment to anticipate additional space occupancy caused by complex disturbances. For example, when wind speed causes the cargo swing angle to be between 2° and 5°, the soft boundary automatically expands around the cargo and crane boom, enabling the system to issue a warning before the cargo reaches its maximum swing angle.

[0033] Based on these dynamic spatial relationships, this step proposes a composite collision risk scoring formula applicable to multi-arm collaborative lifting operations at sea, the structure of which is as follows: ; in, The minimum boundary distance between the current node's 3D soft envelope and the nearest obstacle is obtained from the physical bounding box in the twin scene; The relative velocity between the boom's direction of motion and the obstacle's approach direction is obtained by dividing the displacement difference between nodes by the predicted time interval. The angle between the tangential vector of the boom path and the normal to the obstacle surface is used to identify "head-on collision" actions; This is the hull roll enhancement term, composed of the roll and pitch angles magnified by a factor. This term simulates spatial instability caused by large hull rolls. For example, when the roll angle exceeds 4°, the soft boundary may increase significantly, necessitating consideration of this magnification factor in the assessment of safe distances. Four coefficients are included. The weights of different risk factors are controlled separately, and the specific values ​​are adjusted based on field experience. For example, in cross-containment operations, the proportion of cargo swaying will increase. It will increase during this stage.

[0034] The scoring formula incorporates a "term related to external force disturbance". This design is specifically tailored to the random disturbances inherent in shipboard operating environments. It not only reflects the geometric offset effects of hull rolling but also significantly improves risk scores when the boom's movement direction resonates with the rolling direction. For example, if the boom is rotating at high speed, and the hull's rolling direction is consistent with the boom's movement direction, then... The contribution will significantly increase, resulting in a higher score. The warning line was breached in advance to prevent the boom from hitting the bulkhead when the sway peak was reached.

[0035] The system receives at each node Then, the full path scoring sequence is constructed as follows: And analyze trends in real time. If three consecutive nodes have scores higher than the threshold... The system immediately generates an early warning event and returns it to the control module. The reason for using "three consecutive cycles" is to filter out instantaneous high values ​​caused by minor adjustments to the boom, ensuring stable and reliable early warning behavior. For example, when lifting a bale of pulp with both booms working together, if one boom suddenly shifts laterally due to wire sway, but recovers within 200ms, this instantaneous high-risk value will not trigger an early warning; however, if the swaying continues for more than 300ms, the system determines there is a persistent danger and automatically triggers movement limitation or stops the operation.

[0036] The output consists of two categories: one is the risk scoring sequence. The first is the risk trend of the current boom path execution, which is used for graphical display on the control interface; the second is the early warning event structure, which contains the node index that generates the risk and the state variables of that node. ,score And the triggering reasons, used for control and intervention decisions in the next step.

[0037] S4: Implement graded intervention for crane control commands based on the collision risk score. The intervention strategy is determined based on the risk level, the motion redundancy of the current path node, and the degree of coordination dependence with other booms, and maintains the attitude stability of the suspended cargo during the intervention process. Specifically, the input consists of two parts from the output of step 3: the first is the risk score sequence corresponding to each crane boom path node. Each The first in the corresponding path The first is a control node, whose calculations have taken into account dynamic obstacle distance, relative speed, motion angle, and the impact of hull disturbance; the second is a warning event structure triggered by consecutive high-risk nodes, containing trigger node indices. The state of the crane boom at that node. And warning reasons such as "the cargo swings inward and the inward octagon angle is too small".

[0038] Considering that the boom control system itself has a multi-degree-of-freedom linkage structure, in collision avoidance scenarios, it is not possible to directly freeze or limit all degrees of freedom, otherwise it will cause instability in the cargo's posture. Especially during the coordinated lifting of two booms, excessive braking of a single boom can lead to overall posture imbalance and even cause the cargo to tip over. Therefore, this step designs a strategy framework of "active suppression of the main control degree of freedom + smooth adjustment of the secondary control degree of freedom," using different intervention intensities along different boom control axes to achieve differentiated control intervention.

[0039] The intensity and manner of the intervention are controlled by a multinomial piecewise control function. The formula is generated uniformly, taking into account three factors: risk score, action redundancy, and collaboration dependency. The formula is as follows: ; in, Represents path nodes The corresponding intervention control intensity output, the value range is in between; For the risk scoring normalization item, a linear normalization method is used. ; This indicates the redundancy of the current control node's actions. It is calculated as the percentage of low-risk candidate actions that can be substituted within the local path window. The higher the value, the stronger the substitutability of the actions at that point, making it suitable for applying control. This indicates the degree of coordination dependency between the current node and other booms, calculated by analyzing the temporal and spatial overlap between this node and other boom path nodes. Three adjustment coefficients are also included. , , The weights for risk mitigation, path flexibility utilization, and coordinated intervention penalties should be controlled separately, and it is recommended to set them as follows: , , .

[0040] This control function is not only based on risk scores Instead of determining the intervention intensity, this function takes into account both the "operational substitution flexibility" and "multi-arm cooperative coupling degree" of the path nodes. Especially in high-coupling areas such as cross-lifting of cargo in the hold, large swaying of steel wires, and small redundancy of path space, this function can automatically reduce the intervention intensity, avoid misjudgment that causes false stops or non-cooperative actions, thereby improving the overall collision avoidance robustness of the system.

[0041] when After being calculated, the system will adjust the boom control strategy in multiple levels based on this value: if Maintain the original path speed during execution; if Apply "soft speed limiting" to the current node control signal, that is, compress the speed command to 80%; if Enter "motion slowdown" mode, interpolate to extend the motion execution cycle, and freeze secondary degrees of freedom (such as amplitude pause); if Immediately execute the main degree of freedom pause operation, and at the same time broadcast a freeze signal to the cooperating boom to trigger the synchronization rhythm buffer window.

[0042] This step also specifically introduces a "cargo safety posture maintenance item". This function is used to assess whether a load can remain stable when the boom decelerates or even stops. It calculates a short-term inertial path offset estimate based on the hook's current velocity vector, sling length, and the load's center of mass position, defined as: ; in, The velocity vector of the hook is obtained by fusion of the encoder and IMU; The current swing angle of the cargo around the vertical line is provided by the cargo IMU; the physical meaning of this term is: if the hook speed is high and the cargo has already swung significantly ( In this situation, it is not advisable to abruptly stop the crane's main movement; otherwise, the cargo will continue to sway uncontrollably due to inertia, causing secondary risks. Therefore, when the system determines to execute a forced stop, if... If the set safety threshold is exceeded, the stop action will not be executed immediately. Instead, an interpolation transition will be selected to extend the execution cycle of the current node action and insert a short dynamic buffer window for smooth convergence of the cargo attitude.

[0043] In one or more embodiments, such as Figure 2 As shown, a crane group boom anti-collision early warning system based on digital twin is disclosed, the system comprising: The digital twin modeling module is used to construct and update a dynamic three-dimensional digital twin model that is synchronized with the actual working environment based on the synchronous data collected by the rotary encoder of the crane boom base, the angle sensor at the root of the boom, the inertial measurement unit at the top of the hook, the lidar array around the cargo hatch, and the ship's attitude information. The path planning module is used to perform real-time path planning for each crane boom in the digital twin model, dynamically assess the spatial interference risk between the boom, the suspended cargo and the bulkhead, hatch coaming and other booms, and generate the optimal action path without collision. The risk assessment module is used to perform dynamic risk assessment on each path node in the optimal action path, taking into account the minimum distance between the boom and the obstacle, the relative motion speed, the angle of motion direction, and the hull roll and pitch disturbance factors, to generate a collision risk score. The control intervention module is used to implement graded intervention on the crane control commands based on the collision risk score. The intervention strategy is determined based on the risk level, the motion redundancy of the current path node and the degree of coordination dependence with other booms, and maintains the posture stability of the suspended cargo during the intervention process.

[0044] It is worth noting that the specific workflow of the crane group boom anti-collision early warning system based on digital twin provided in this embodiment of the invention is the same as that of the crane group boom anti-collision early warning method based on digital twin described in the above embodiment, and will not be repeated here.

[0045] This invention also provides a crane group boom collision avoidance and early warning device based on digital twins, including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, it implements the steps described in the above embodiments of the crane group boom collision avoidance and early warning method based on digital twins, for example... Figure 1 The steps S1 to S4 described above; or, when the processor executes the computer program, it implements the functions of each module in the above system embodiments.

[0046] For example, the computer program may be divided into one or more modules, which are stored in the memory and executed by the processor to complete the present invention. The one or more modules may be a series of computer program instruction segments capable of performing specific functions, which describe the execution process of the computer program in the digital twin-based crane boom anti-collision warning device.

[0047] The crane group boom collision avoidance and early warning device based on digital twins can be a computing device such as a desktop computer, laptop, handheld computer, or cloud server. The device may include, but is not limited to, a processor and memory. Those skilled in the art will understand that the device may also include input / output devices, network access devices, buses, etc.

[0048] The processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASACs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor. This processor is the control center of the digital twin-based crane group boom collision avoidance and early warning device, connecting all parts of the device via various interfaces and lines.

[0049] The memory can be used to store the computer programs and / or modules. The processor implements various functions of the digital twin-based crane boom anti-collision warning device by running or executing the computer programs and / or modules stored in the memory, and by calling the data stored in the memory. The memory may mainly include a program storage area and a data storage area. The program storage area may store the operating system, at least one application program required for a function, etc.; the data storage area may store data created based on the operation of the air conditioning controller, etc. In addition, the memory may include high-speed random access memory, and may also include non-volatile memory, such as hard disk, memory, plug-in hard disk, smart media card (SMC), secure digital card (SD card), flash card, at least one disk storage device, flash memory device, or other volatile solid-state storage devices.

[0050] The module integrated into the crane group boom anti-collision warning device based on digital twins, if implemented as a software functional unit and sold or used as an independent product, can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the above embodiments of the present invention can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a portable hard drive, a magnetic disk, an optical disk, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium, etc.

[0051] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. The storage medium can be a magnetic disk, optical disk, read-only memory (ROM), or random access memory (RAM), etc.

[0052] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A method for anti-collision early warning of crane booms in a crane group based on digital twins, characterized in that, The method includes: A dynamic three-dimensional digital twin model synchronized with the actual working environment is constructed. The digital twin model is based on the synchronous data collected by the rotary encoder installed on the base of each crane boom, the angle sensor at the root of the boom, the inertial measurement unit at the top of the hook, and the lidar array around the cargo hatch, and is updated in real time in combination with the ship's attitude information. In the digital twin model, real-time path planning is performed for each crane boom. During the path planning process, the spatial interference risk between the boom, the suspended cargo and the bulkhead, hatch coaming and other booms is dynamically evaluated, and the optimal action path without collision is generated. Dynamic risk assessment is performed on each path node in the optimal motion path, taking into account the minimum distance between the boom and the obstacle, the relative motion speed, the angle of motion direction, and the hull roll and pitch disturbance factors to generate a collision risk score. The crane control commands are subject to graded intervention based on the collision risk score. The intervention strategy is determined based on the risk level, the motion redundancy of the current path node, and the degree of coordination dependence with other booms, and the attitude stability of the suspended cargo is maintained during the intervention process.

2. The crane group boom anti-collision early warning method based on digital twin according to claim 1, characterized in that, The dynamic three-dimensional digital twin model adopts a hybrid structure modeling of skeleton-flexible cable-rigid body group, in which the boom is represented by skeleton nodes, the cable is simulated by flexible line segments, and the cargo is represented by rigid body nodes bound to the bottom of the cable. All objects are bound to the ship coordinate system that is dynamically updated with the ship's attitude.

3. The crane group boom anti-collision early warning method based on digital twin according to claim 1, characterized in that, The synchronization data is transmitted via industrial Ethernet or controller area network bus and fused using a multi-source Kalman filter with a time sliding window to eliminate fluctuations caused by sensor sampling delay and environmental interference.

4. The crane group boom anti-collision early warning method based on digital twin according to claim 1, characterized in that, The real-time path planning uses an improved A* search algorithm, which evaluates three-dimensional spatial nodes based on a risk-weighted cost function that includes obstacle distance and hull sway angle, and excludes path nodes that overlap with the predicted envelope of other cranes within the next two seconds.

5. The crane group boom anti-collision early warning method based on digital twin according to claim 1, characterized in that, The dynamic risk assessment employs a soft-border mechanism, which dynamically expands the three-dimensional envelope boundary of the boom and cargo based on the cargo swing amplitude and ship sway, in order to anticipate the additional space occupied by disturbances.

6. The method for anti-collision early warning of crane booms based on digital twins according to claim 1, characterized in that, When the collision risk scores of three consecutive path nodes exceed a preset threshold, an early warning event is triggered and an early warning structure containing the risk node index, state variables, and triggering reasons is generated.

7. The method for anti-collision early warning of crane booms based on digital twins according to claim 1, characterized in that, The graded intervention includes four levels of control strategies: maintaining the original speed, gently limiting the speed, slowing down the movement, and suspending the main degree of freedom. Each level of strategy corresponds to a different collision risk score range.

8. The method for anti-collision early warning of crane booms based on digital twins according to claim 1, characterized in that, Before pausing the main degree of freedom, calculate the estimated inertial path offset of the cargo, which is jointly determined by the hook velocity vector, sling length, and cargo swing angle. If the estimated value exceeds the safety threshold, extend the execution cycle of the current node and insert a dynamic buffer window to stabilize the cargo attitude.

9. The crane group boom anti-collision early warning method based on digital twin according to claim 1, characterized in that, The ship's attitude information is acquired by two inertial measurement units located at the center of the bridge and main deck, and output in the form of three-dimensional attitude quaternions to drive the affine transformation of all objects in the digital twin model.

10. A crane group boom anti-collision early warning system based on digital twin, characterized in that, The system includes: The digital twin modeling module is used to construct and update a dynamic three-dimensional digital twin model that is synchronized with the actual working environment based on the synchronous data collected by the rotary encoder of the crane boom base, the angle sensor at the root of the boom, the inertial measurement unit at the top of the hook, the lidar array around the cargo hatch, and the ship's attitude information. The path planning module is used to perform real-time path planning for each crane boom in the dynamic three-dimensional digital twin model, dynamically assess the spatial interference risk between the boom, the suspended cargo and the bulkhead, hatch coaming and other booms, and generate the optimal action path without collision. The risk assessment module is used to perform dynamic risk assessment on each path node in the optimal action path, taking into account the minimum distance between the boom and the obstacle, the relative motion speed, the angle of motion direction, and the hull roll and pitch disturbance factors, to generate a collision risk score. The control intervention module is used to implement graded intervention on the crane control commands based on the collision risk score. The intervention strategy is determined based on the risk level, the motion redundancy of the current path node and the degree of coordination dependence with other booms, and maintains the posture stability of the suspended cargo during the intervention process.

Images