Method for hierarchical collaborative motion control of vessel dynamic positioning and wave-compensated embarkation system

Abstract

The application discloses a layered cooperative motion control method for a ship power positioning and wave compensation boarding system, and comprises the following steps: obtaining historical motion data of a ship; fusing the historical motion data of the ship to obtain a current ship body posture and speed; generating a prediction value of a disturbance suffered by the ship in a future preset time through a prediction model; extracting a low-frequency slow-drift motion component; calculating a total propeller output force; and controlling the propeller to work based on the total propeller output force of the ship power positioning system to compensate for the low-frequency slow-drift motion component. A Stewart platform in the wave compensation boarding system controls the work of the supporting legs based on the target length of the supporting legs to compensate for the low-frequency motion remaining after compensation by the ship power positioning system and the high-frequency motion that is not compensated. The remaining disturbance after compensation of the Stewart platform is compensated through a gangway. The application can realize an energy distribution mode in which low-frequency disturbance is compensated by the ship power positioning system and high-frequency disturbance is compensated by the wave compensation boarding system.

Description

Technical Field

[0001] This invention relates to the field of marine engineering equipment technology, and in particular to a hierarchical coordinated motion control method for a ship's dynamic positioning system and wave compensation boarding system under wave disturbance conditions. Background Technology

[0002] Wind power is a high-quality, renewable, and clean energy source. Actively promoting the wind power industry and utilizing wind energy can effectively alleviate problems such as energy shortages, environmental pollution, and climate change. Although offshore wind power is becoming increasingly mature in terms of design and construction, with the continuous expansion of deep-sea wind power development, the ability of maintenance vessels to perform high-precision boarding and maintenance tasks in wave-disturbed environments has become a key technical bottleneck. The six-degree-of-freedom motion caused by wave disturbances affects the end-position stability of the wave-compensated boarding system. Therefore, how to achieve high-precision dynamic compensation and coordinated control between the vessel and the wave-compensated boarding system in complex external disturbance environments is a significant technical challenge in the field of offshore operation equipment.

[0003] Existing research can be divided into two categories: (1) Single-system compensation control strategy: This type of method is based on the ship's dynamic positioning (DP) system and achieves low-frequency attitude and position stability of the ship through propeller control. The dynamic positioning system performs well in stabilizing the ship's attitude when subjected to low-frequency disturbances, but its ability to control high-frequency disturbances such as heave, pitch and roll is limited, and these residual disturbances will be transmitted to the wave compensation boarding system, affecting the stability of the actuator at the end of the gangway; (2) Independent compensation strategy for wave compensation boarding system: This type of method focuses on the motion compensation of the parallel mechanism (such as the Stewart platform) and the serial mechanism (such as the three-degree-of-freedom gangway) itself, and achieves local high-frequency attitude adjustment through sensor feedback. However, this type of method generally lacks coordination with the overall motion of the ship and is prone to problems such as large errors, compensation redundancy and energy waste.

[0004] In addition, traditional wave compensation boarding systems mostly adopt single-layer control or cascade control structures, which cannot make full use of the dynamic coupling relationship between the ship and the wave compensation boarding system. This leads to the system being prone to over-response and compensation lag under complex sea conditions, and thus cannot meet the requirements of large disturbance suppression and high-precision control.

[0005] In summary, existing technologies lack a motion control method that enables coordinated motion control between a ship's dynamic positioning system and a wave compensation boarding system in a wave-turbulent environment. Summary of the Invention

[0006] To address at least one of the problems existing in the prior art, this invention provides a hierarchical cooperative motion control method for a ship's dynamic positioning system and a wave compensation boarding system under wave disturbances. This method enables information exchange, temporal complementarity, and dynamic coordination between the ship and the wave compensation boarding system, thereby effectively separating ship disturbance components of different frequency bands. It achieves an energy distribution method where low frequencies are compensated by the dynamic positioning system and high frequencies are compensated by the wave compensation boarding system. By using short-term wave prediction, it enhances the feedforward response capability and dynamic robustness of the ship's dynamic positioning system and the wave compensation boarding system to future disturbances.

[0007] This method simultaneously possesses: short-term prediction capability for the future six-degree-of-freedom motion of a ship; a multi-level coordination mechanism between the ship's dynamic positioning system and the wave compensation boarding system; and time-domain compensation for high-frequency and low-frequency disturbances, enabling the dynamic positioning system to compensate for the ship's horizontal three-degree-of-freedom low-frequency motion under the influence of wind, waves, and currents, and the wave compensation boarding system to compensate for the wave compensation boarding system's six-degree-of-freedom high-frequency (wave frequency) motion under the influence of wind, waves, and currents.

[0008] To achieve the objective of this invention, the present invention provides a layered cooperative motion control method for a ship dynamic positioning and wave compensation boarding system, comprising the following steps: Acquire historical motion data of the ship, including the ship's acceleration signal, angular velocity signal, position, speed and heading information; By fusing historical motion data of the ship, the current hull attitude and speed can be obtained; The prediction model generates predicted values ​​of the disturbances that the ship will experience at a preset time in the future, based on the ship type response amplitude operator, the current ship attitude and speed. The low-frequency slow drift motion component is extracted based on the predicted value of the disturbances experienced by the ship; the total output force of the thruster is calculated, and the ship's dynamic positioning system controls the operation of the thruster based on the total output force of the thruster to compensate for the low-frequency slow drift motion component; Based on the disturbances and low-frequency slow drift motion components experienced by the ship, the remaining low-frequency motion after compensation by the ship's dynamic positioning system and the uncompensated high-frequency motion are obtained. The target lengths of each outrigger of the Stewart platform are obtained by solving the inverse kinematics of the Stewart platform. The Stewart platform in the wave-compensated boarding system controls the operation of the outriggers based on the target lengths of the outriggers to compensate for the low-frequency motion remaining after compensation by the ship's dynamic positioning system and the uncompensated high-frequency motion. The speed at the end of the gangway is optimized so that the actual speed at the end of the gangway in the wave-compensated boarding system is close to the desired speed of the gangway so that the gangway can compensate for the remaining disturbances after compensation by the Stewart platform.

[0009] Furthermore, the ship's historical motion data can be fused using any of the following algorithms: extended Kalman filtering, unscented Kalman filtering, particle filtering, capacitive Kalman filtering, or a state estimator based on optimization theory.

[0010] Furthermore, the prediction model employs an autoregressive model, or any one of the following network models: Long Short-Term Memory Network and Temporal Convolutional Network.

[0011] Furthermore, the calculation method for the total output force of the propellers includes: obtaining the total output force of the propellers based on the propeller configuration matrix and the thrust components generated by all the propellers of the ship.

[0012] Furthermore, the target length of each leg of the Stewart platform The calculation formula is:

[0013] in, and These are the hinge point coordinates of the lower platform and the upper platform, respectively. Let the position vector of the upper platform relative to the lower platform of the Stewart platform be the pose of the upper platform of the Stewart platform. get, It is a rotation matrix.

[0014] Furthermore, the desired velocity at the end of the gangway is obtained based on the residual disturbance after compensation by the Stewart platform, i.e., the residual disturbance generated by the Stewart platform on the gangway, and the gangway end pose reference trajectory.

[0015] Furthermore, the optimization steps for the speed at the end of the gangway include: Construct and solve the quadratic programming problem: ,in It is a decision variable. , is the joint velocity value of the gangway; It is the quadratic term matrix of the objective function; It is a linear term vector of the objective function, and the actual velocity at the end of the gangway is... The optimization objective is to find an optimal joint velocity. This results in the actual end velocity. Desired speed approaching the end of the gangway And meet safety and smoothness requirements, Indicates the position of the gangway joint and the actual speed at the end of the gangway. .

[0016] The present invention also provides a hierarchical coordinated motion control system for a ship dynamic positioning and wave compensation boarding system.

[0017] The present invention also provides a computer device.

[0018] The present invention also provides a computer-readable storage medium.

[0019] Compared with the prior art, the present invention can achieve at least the following beneficial effects: (1) A hierarchical collaborative motion control architecture is proposed: a three-layer compensation control structure of “external disturbance prediction layer - low-frequency motion compensation layer - high-frequency motion compensation layer” is constructed in the field of marine equipment control to realize time-domain decoupling and coordinated motion control of low-frequency and high-frequency disturbances.

[0020] (2) An active feedforward compensation mechanism based on short-time prediction is adopted: the prediction model is used to fuse the data of the inertial measurement unit (IMU) sensor to realize the short-time prediction of the future six degrees of freedom motion of the ship, and to provide feedforward signals for the subsequent low-frequency motion compensation layer and high-frequency motion compensation layer control, thereby reducing control lag.

[0021] (3) Establish a collaborative division of labor mechanism between dynamic positioning and wave compensation boarding system: low-frequency disturbances are compensated by dynamic positioning system, and high-frequency disturbances are compensated by wave compensation boarding system composed of Stewart platform and gangway, forming a control strategy with optimal energy distribution and phase consistency, avoiding the problems of "repetitive control" and "mutual cancellation" in traditional methods.

[0022] (4) Achieve system integration and significantly improve the compensation performance of each system: In the simulation platform verification, the method of the present invention can control the relative motion error of the wave compensation boarding system at the millimeter level under typical medium sea conditions, reduce the energy consumption of each system during the boarding operation, and fully demonstrate the practicality and innovation of the method in the dynamic marine environment. Attached Figure Description

[0023] Figure 1 This is a schematic diagram of the overall architecture of a six-degree-of-freedom wave compensation boarding system control method applicable to complex sea conditions, as described in an embodiment of the present invention. The diagram illustrates the three-layer structure of the system: an external disturbance prediction layer, a low-frequency motion compensation layer, and a high-frequency motion compensation layer, and shows the information interaction and signal transmission paths.

[0024] Figure 2 This is a flowchart illustrating the steps of the control method provided in an embodiment of the present invention.

[0025] Figure 3 This is a schematic diagram of the Stewart platform and its three-degree-of-freedom gangway. The diagram shows the layout of the Stewart platform's six legs and the connection between the upper and lower platforms, as well as the pitch, telescopic, and slewing joints of the three-degree-of-freedom gangway.

[0026] Figure 4This is a probability density diagram of the steady-state position error in an embodiment of the present invention.

[0027] Figure 5 This is a probability density diagram of the steady-state angle error in an embodiment of the present invention. Detailed Implementation

[0028] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0029] In the prior art, the ship dynamic positioning system and the wave compensation boarding system usually operate independently: the ship dynamic positioning system is mainly used to suppress low-frequency slow drift motion, and the control accuracy is limited by the uncertainty of the dynamic response of the propeller and wave disturbance; the wave compensation boarding system mostly relies on passive or single-layer feedback control, which can only compensate for high-frequency attitude disturbances to a limited extent, and the compensation effect is easily affected by the large-scale motion of the ship. This kind of separate control method architecture leads to the lack of coordination mechanism between the two systems in the time domain and energy level, and there are the following main problems: (1) The ship dynamic positioning system and the wave compensation boarding system may repeatedly or in reverse compensate for the same disturbance component, which will lead to phase interference and unstable response delay; (2) The ship's high-energy-consuming propeller is used for small attitude correction for a long time during operation, while the wave compensation boarding system has a limited compensation range and low overall system energy efficiency; (3) The wave compensation boarding system is usually not designed to be coupled with the overall dynamics of the ship, resulting in poor multibody motion coordination and unstable compensation effect. In order to at least solve one of the above-mentioned technical problems in the prior art, the embodiments of the present invention adopt a hierarchical architecture design and a predictive-driven cooperative motion control mechanism to realize multi-timescale coordination and optimal energy allocation of the ship's dynamic positioning system and boarding compensation system under wave disturbance environment. This breaks through the limitations of the prior art that "the dynamic positioning system and the wave compensation boarding system are independent, mutually disturbing and lagging", and forms a motion control method that can achieve cooperative compensation under strong wave disturbance conditions.

[0030] Please see Figure 1 and Figure 2 The present invention provides a hierarchical cooperative motion control method for a ship's dynamic positioning system and wave compensation boarding system under wave disturbance conditions, comprising the following steps: Step S1: Initialization (t=0~0.2s, single execution) After the system is powered on, the main control computer 201 completes coordinate system initialization, loads the dynamic positioning vessel type response amplitude operator (RAO) data, and mechanism geometric parameters / constraints (the upper / lower platform radius, outrigger travel and angular limits of the Stewart platform; the travel and velocity / acceleration upper limit of the gangway's three joints, and the minimum collision avoidance distance). Then, it sequentially wakes up the inertial measurement unit (IMU) 101 and the global positioning system (GNSS) module 102, performs power-on self-test and zero-bias calibration, and generates the initial hull attitude. With speed estimation .

[0031] In this step, the reference for the operation and docking plan is loaded as part of the configuration and used to align the QP target weighting in subsequent steps S5-B with the error assessment in step S7. This includes: obtaining the end-effector reference trajectory from the local configuration file or online interface to form the gangway end-effector pose reference trajectory. , This represents the desired position vector of the gangway end in the inertial coordinate system; Let be the desired attitude vector of the gangway end in the inertial coordinate system. It is a time variable, representing time. End-effector pose reference.

[0032] Step S2: External disturbance information collection and fusion (20ms / time) The ship's six-degree-of-freedom historical motion data were acquired using an inertial measurement unit (IMU) (100 Hz) and a global positioning system (GNSS) module (10 Hz), and the current hull attitude was obtained by fusion using an extended Kalman filter (EKF). With speed The results are used in steps S3, S4, and S5, and written to shared memory for short-term prediction, low-frequency separation of the ship's dynamic positioning system (DP), and kinematic / velocity mapping. At the same time, timestamps and equipment health status are recorded in the log for subsequent quality traceability.

[0033] Step S3: Short-term wave and ship motion prediction The current hull attitude is obtained by fusing the hull response amplitude operator RAO and the extended Kalman filter EKF. and speed Call the autoregressive model ( AR(p) Generate predicted values ​​of the disturbances experienced by the ship at a future preset time (in one embodiment, the preset time is 3-5 seconds). p represents the order of the autoregressive model, which is used as the signal input for step S4 and archived.

[0034] Step S4: Ship motion compensation ( (20-50ms beat) In each main cycle, the main control computer 201 of the low-frequency motion compensation layer predicts the disturbances experienced by the ship. A low-pass filter with a cutoff frequency of approximately 0.03 Hz is applied to extract the low-frequency slow-drift motion components. (Mainly corresponding to the low-frequency components of sway, roll, and pitch); Subsequently, the main control computer 201 performs thrust allocation solution at a preset time (in one embodiment, the time is 50 ms), wherein the thrust allocation solution is based on the thruster configuration matrix. With thrust, α represents the azimuth angle of the thruster, and the total output force of the thruster is calculated. And it is sent to the thruster controller via the bus.

[0035] In one embodiment, the low-frequency slow drift motion component The extraction process involves using a digital filter to extract the predicted full-band motion signal, i.e., the predicted value of the disturbance experienced by the ship. Low-frequency components are extracted. In one embodiment, a Butterworth low-pass filter with a flat passband and steep roll-off characteristics is selected:

[0036] in, It is the predicted low-frequency slow drift motion component; It is the filter order; cutoff frequency. Used to define the range of wave frequency motion and slow drift motion, with a value between 0.1 and 0.2 Hz.

[0037] In one embodiment, the thruster outputs total force The calculation process is as follows: The thruster configuration matrix is ​​as follows:

[0038] in Let be the azimuth angle of the i-th thruster; Let be the position of the i-th thruster in the ship's coordinate system; ; Therefore, the formula for calculating the total output force of the thruster is: (3) in The thrust component generated by all the ship's propulsion systems.

[0039] At the end of this step, two key outputs can be generated: one is the low-frequency motion remaining after compensation by the ship's dynamic positioning system, and the other is the uncompensated high-frequency motion. It represents the disturbance component that, even after low-frequency suppression, will still act on the wave compensation boarding system installed on the ship's deck; the second is the updated hull condition. With speed Updated hull status With speed The downstream kinematics chain and velocity mapping will be used for time-domain quantization. Full-step operation ensures the thruster does not experience abrupt changes, maintaining heading within ±1° and horizontal displacement within ±0.5 m of the operating range.

[0040] Step S5-A: Coarse compensation on the Stewart platform (20 ms / time) The Stewart platform compensation uses the ship's dynamic positioning system output from step S4 to account for the remaining low-frequency motion and the uncompensated high-frequency motion. Using the input as input, solve the inverse kinematics of the Stewart platform to obtain the target lengths of each leg of the Stewart platform. Then, within the safety boundaries such as the ball joint angle limit, the hydraulic cylinder is driven to complete attitude compensation; the residual disturbance generated by the Stewart platform on the gangway As a direct input to step S5-B, the current pose state of the Stewart platform is simultaneously recorded. With speed The data is sent to step S6 for end-effector pose and velocity quantization. This step does not calculate errors or control gains, but only ensures that the physical execution meets geometric and mechanical constraints and is updated stably at 20 ms ticks.

[0041] Specifically, by controlling the changes in the outriggers of the Stewart platform, the upper platform maintains an ideal and stable posture in the inertial coordinate system. (E represents the inertial coordinate system, t represents the upper platform frame, and des represents the pose); however, due to the physical constraints of the Stewart platform itself (such as the leg length limit and the ball joint rotation angle limit), the ideal stable pose calculated through inverse kinematics is difficult to achieve. This may exceed the reachability space of the Stewart platform. Therefore, when the ideal stable pose is unreachable, a "degraded" pose that is "closest" to the relative pose but is guaranteed to be reachable is calculated. The residual disturbance generated by the Stewart platform on the gangway = - .

[0042] Step S5-B: Fine-tuning of the gangway (20 ms / time) Residual disturbance from the gangway caused by the Stewart platform Converted into the desired velocity at the end of the gangway The specific transformation process is as follows: at time t, the actual pose of the gangway base in the inertial coordinate system is obtained from the pose coordinate system transformation relationship between the ship, Stewart, and the gangway end. It is the residual disturbance generated by the Stewart platform on the gangway. Decision; based on actual pose and gangway end position reference trajectory The difference in phase yields the positional error. and attitude error The error vector consisting of position error and attitude error divided by the time step Obtain the desired velocity value at the end of the gangway .

[0043] An optimization method based on task space velocity control optimizes velocity commands in the task space to achieve the desired actual end velocity of the gangway. Desired speed approaching the end of the gangway It also meets safety and smoothness requirements to achieve gangway compensation. E represents the inertial coordinate system, tip represents the gangway, and act represents the actual speed to distinguish it from the desired speed. Construct and solve the quadratic programming problem: ,in It is a decision variable. , is the joint velocity value of the gangway; It is a quadratic term matrix of the objective function, reflecting the "cost weights"; This is the linear term vector of the objective function. The optimal joint velocity is obtained within a 20 ms period. The execution result is synchronously sent to step S5 for quantization recording; the gangway level does not output error or control gain. Indicates the location of the gangway joint. This indicates the speed of the gangway joint.

[0044] Actual speed at the end of the gangway Speed ​​of the gangway joint The two are linked by the gangway geometric Jacobian matrix: , The geometric Jacobian matrix representing the gangway is a known quantity. The vector representing the joint space variables of the gangway.

[0045] Step S6: Kinematic chain and velocity mapping ( (20ms / time, used only for quantization and recording) After step S4, the wave-compensated boarding system invokes the disturbance transmission chain (ship disturbed → dynamic positioning system → Stewart → gangway) and Jacobian mapping to perform a recorded-only quantitative calculation of the end-effector pose and velocity: based on the hull state from step S4. ,speed With the platform / gangway attitude and velocity from step S6 , , , As input, the pose of the gangway end is first synthesized according to the geometric relationship of the mechanism's kinematic chain. The joint velocities of the gangway are then obtained using a structured Jacobian expression. The above calculations are completed within a 20 ms period, with an end-to-end delay of no more than 10 ms.

[0046] The generated end pose Joint speed with the gangway It only writes to the log for subsequent evaluation and does not affect any execution instructions; at the same time, it records key intermediate quantities (such as the contribution rate of each Jacobian quantity) to facilitate the differentiation of the dominant frequency band and energy share of different subsystems to the terminal speed.

[0047] First, kinematic and dynamic calculations are performed on the gangway end: First, an end-effector kinematic chain is established. The pose of the gangway end in the inertial coordinate system can be obtained by the following matrix multiplication, used to describe the pose coordinate system transformation relationship between the ship, Stewart, and gangway end:

[0048] in This represents the transformation from the inertial coordinate system to the ship's coordinate system; This indicates the transformation from the ship's hull coordinate system to the Stewart platform coordinate system; This represents the transformation from the lower platform coordinate system to the upper platform coordinate system of the Stewart platform; This indicates the transformation from the Stewart platform system to the gangway base system; This indicates the transition from the gangway base system to the terminal system. This indicates the position and orientation of the gangway end in the inertial coordinate system.

[0049] The pose of the gangway end is obtained based on the pose coordinate system transformation relationship. ; Secondly, the velocity propagation relationship is analyzed. For solving the linear velocity at the end of the gangway, the gangway includes multiple moving joints such as pitch, extension, and rotation. Its instantaneous linear velocity at the end is affected not only by the angular velocities of each joint but also by the coupling effect of the motion between the upper Stewart platform and the ship's base. Therefore, based on multibody dynamics modeling, this embodiment of the invention expresses the velocity at the end of the gangway as a linear combination of the generalized coordinate velocities of each subsystem. The specific form of the structured Jacobian expression is as follows:

[0050] in The velocity at the end of the gangway in the inertial coordinate system; Let be the geometric Jacobian matrix of the ship; For the geometric Jacobian matrix of the Stewart platform; Let be the geometric Jacobian matrix of the gangway; , , These are the generalized velocity vectors for the ship, Stewart platform, and gangway joint, respectively.

[0051] This method transforms complex multibody coupled motion relationships into a unified linear velocity superposition expression by modularizing and structuring the Jacobian matrix. This expression can be directly embedded into motion planning and control allocation processes to achieve velocity coordination and constraint consistency control among different subsystems, thereby improving the real-time response accuracy and compensation coordination at the gangway end.

[0052] In order to accurately describe the dynamic impact of wave disturbances on the ship and the wave compensation boarding system, an analysis model of the transmission characteristics of external disturbances is established for the analysis of the transmission characteristics of external disturbances on the ship. This model can reveal the transmission law of external wave excitation in the ship-Stewart platform-gangway multibody system through frequency domain response analysis.

[0053] The analysis model for the transmission characteristics of external disturbances on the ship is as follows: The response amplitude operator (RAO) is introduced into the study to analyze the incident wave hyperspectral. The ship's response spectrum in each degree of freedom Establish a quantitative mapping relationship: (9) in The ship response spectrum describes the intensity of the motion response caused by wave excitation in each degree of freedom. The high-power spectrum of the incident wave reflects the energy distribution characteristics of the external sea state; The frequency domain transfer function (or RAO) of a ship in a specific degree of freedom includes the ship's hydrodynamic damping, added mass, and restoring force characteristics. These are the wave frequencies and the ship's response frequencies. Based on this, the multibody transfer characteristic function of the disturbance from the ship to the end of the wave-compensated boarding system is further derived, and the local dynamic characteristics of the platform and gangway are considered to make the relative motion spectrum at the end... It can be represented as: (10) in This is the integrated compensation transfer function for the wave-compensated boarding system, used to describe the collaborative compensation characteristics of the Stewart platform and the gangway.

[0054] Through the above analysis of transmission characteristics, this embodiment of the invention achieves hierarchical identification and channel allocation of wave frequency domain energy: low-frequency energy is mainly suppressed by the ship's dynamic positioning system; mid-to-high-frequency energy is actively compensated by the wave compensation boarding system composed of the Stewart platform and gangway. This method can not only quantitatively assess the impact of disturbances on boarding accuracy under different sea states, but also provide a frequency domain weighting basis for subsequent control law design, realizing the modeling and optimization of the complete energy transfer chain from wave excitation → ship response → boarding end disturbance.

[0055] By establishing a multibody dynamics model of the ship-Stewart platform-gangway and a kinematic chain at the gangway end, continuous motion mapping and Jacobian matrix coordination from hull motion to gangway end are achieved, ensuring the continuity and accuracy of end attitude compensation.

[0056] Step S7: Assessment of gangway end attitude and error (S5 substep, 20ms / time) Within the same sampling period, immediately following step S6, the wave-compensated boarding system reads the gangway end pose reference trajectory from the operation and docking planning module in step S1. (The two values ​​represent position and attitude, respectively), and are compared with the gangway end pose from step S6. Perform item-by-item alignment and calculate the positional error of the gangway end under a unified reference frame. Attitude error (logarithmic mapping) , Indicates a reference position. The reference attitude is indicated. All results are written to the log every 20 ms as the sole data source for statistical evaluation and are not used in the control link or parameter adjustments. This method uses the attitude error at the gangway end as an indicator to evaluate the method's effectiveness.

[0057] Step S8: Performance Evaluation In one embodiment, the evaluation layer takes the real-time error (including position error and attitude error) obtained from step S5-A of the most recent 120 seconds as input every 10 seconds, and constructs probability density maps of position error magnitude and attitude error magnitude respectively using the normalized histogram method. The probability density map is used as the final presentation of the method performance of the embodiment of the present invention. This evaluation process is completely decoupled from the execution link and does not write back any control information to step S4 / S6 or S5. It is only used for the objective quantification and documentation of the method.

[0058] In one embodiment, Figure 4 This is the probability density map of the steady-state position error. Figure 5 The steady-state angle error probability density plot, obtained from step S7, shows that the error histogram in the steady state exhibits a Gaussian-like distribution. Figure 4The positional error is mainly concentrated in the range of 150-400 mm, and the probability of an error greater than 400 mm is low. Figure 5 The angular error is mainly concentrated in the range of 0.5°-2°, with a low probability of errors exceeding 2°. This indicates that the system's error characteristics are predictable and stable, rather than randomly diverging. This demonstrates that allocating disturbances according to their spectral characteristics to different actuators (DP handles low frequencies, Stewart handles mid-to-high frequencies, and the gangway handles residual errors) can effectively utilize the advantages of each subsystem, achieving the highest compensation accuracy with minimal execution cost.

[0059] The coordinate systems that may be involved are described below: Inertial coordinate system Take any point on the sea surface as the origin. , The axis points due north. The axis points due east. The axis points to the Earth's center; ship coordinate system The origin of the coordinate system is located at the center of the ship's moment of inertia in order to establish the equations of motion. For the ship's center of gravity, Vertically downwards, Pointing to the right, Pointing in the direction of travel, and , , Mutually perpendicular; Stewart platform coordinate system. Using the lower platform coordinate system, For the coordinate system of the upper platform, The coordinate system is fixed to the hull. The origin of the coordinate system on the lower platform is... And the origin of the coordinate system of the platform Located at the centers of the upper and lower platforms respectively; the gangway coordinate system is set with the center of the base bottom as the reference coordinate system. The origin position is the center of the connection point between the base and the gangway pitch axis. , The origin of the coordinate system is established at the center of the contact point between the gangway pitch axis and the top of the telescopic shaft. A coordinate system is established at the end of the gangway telescopic shaft. .

[0060] like Figure 1As shown in the figure, this invention discloses a hierarchical cooperative motion control system for a ship's dynamic positioning system and a wave-compensated boarding system under wave disturbances. The system includes three main functional layers and several sensors (including an inertial measurement unit 101 and a global positioning system GNSS module 102), specifically including: a ship external disturbance prediction layer 1, which is responsible for short-term prediction of ocean waves and the ship's future six-degree-of-freedom motion; a low-frequency motion compensation layer 2, which is responsible for receiving the ship's future motion information from the ship external disturbance prediction layer and separating the low-frequency slow drift motion component (mainly referring to the sway, roll, and bow roll on the horizontal plane), and the ship's dynamic positioning system actively compensates for this large-scale, low-frequency motion by controlling the ship's own propellers. The main goal of this layer is to reduce the disturbance amplitude and energy that the wave-compensated boarding system needs to handle, creating a more "stable" working basis for it; and a high-frequency motion compensation layer 3, including a Stewart platform 101 and a three-degree-of-freedom gangway 102, which is responsible for fine compensation of high-frequency disturbances and residual errors after compensation by the ship's dynamic positioning system.

[0061] The overall operating principle of the three functional layers is as follows: External disturbance prediction layer → Low-frequency motion compensation layer → High-frequency motion compensation layer (Stewart + gangway) → Stability control output at the end of gangway.

[0062] The ship external disturbance prediction layer 1 includes an inertial measurement unit (IMU) 101, a global positioning system (GNSS) module 102, a data fusion calculation module 103, and a prediction module. The IMU 101 is used to collect the ship's acceleration and angular velocity signals; the GNSS module 102 provides position, velocity, and heading information; and the data fusion calculation module 103 fuses the acceleration, angular velocity, position, velocity, and heading information to obtain the current ship attitude. With speed In the prediction module, the prediction model generates predicted values ​​of the disturbances that the ship will experience at a preset time in the future, based on the ship type response amplitude operator, the current ship attitude and speed.

[0063] In one embodiment, the IMU 101+Global Positioning System GNSS module 102 can be replaced with a high-precision attitude measurement device based on a laser inertial navigation system (LiDAR+IMU) or a differential RTK-GNSS system; in areas without satellite signals, a wave radar or a Doppler acoustic current profiler (ADCP) can be installed.

[0064] In one embodiment, the data fusion calculation module 103 performs data fusion using the Extended Kalman Filter (EKF) algorithm, which effectively balances computational complexity and estimation accuracy. In other embodiments, it can be replaced with any of the following algorithms: Unscented Kalman Filter (UKF), Particle Filter (PF), Capacitive Kalman Filter (CKF), or a state estimator based on optimization theory.

[0065] The data fusion process using the Extended Kalman Filter (EKF) algorithm is as follows: This invention employs the Extended Kalman Filter (EKF) algorithm to fuse data. In complex sea conditions, the ship's state variables exhibit nonlinear characteristics. To achieve accurate online state estimation of the ship's position, attitude, and velocity, a state-space model based on the Extended Kalman Filter is constructed.

[0066] The process of constructing the state-space model is as follows: The ship's state vector is defined as ,in These are slowly varying external perturbation parameters of sensors (including IMU and GNSS); The state-space equation of the ship in continuous time can be derived from the ship's state vector as follows:

[0067] Its display expands to , , It is process noise, which can be assumed to be Gaussian white noise. This represents the ship's velocity in the inertial coordinate system E. It refers to the speed under the ship system. Velocity mapped to inertial coordinate system E The velocity transformation matrix, It refers to the ship's condition. For control input (for thruster thrust). Represents acceleration in a ship system. The total inertia matrix, It is the thruster that controls the resultant force. It is external disturbance force (including wind, waves and currents). It is the vector of the Coriolis force and the centripetal force. It is the fluid damping force vector. It is a vector of static restoring force and torque. It is a ship dynamics and observation model.

[0068] Let the sampling period be ,exist and Sampling interval pair Discretization using the first-order Euler method yields the nonlinear discrete state equations:

[0069] Expanding the above equation into , , .

[0070] If the measurement value at time k (obtained by GNSS) can be directly used as... The observations correspond to the position and velocity. The raw signal from the IMU is usually not directly used as... The data is preprocessed through the data fusion module (103) (e.g., calibration, timestamp alignment, coordinate transformation, and construction of observation vectors corresponding to the state vectors. The process of constructing the observation vectors corresponding to the state vectors is a known method). Then the state-space model is:

[0071] in This refers to the type of sensor.

[0072] Based on the aforementioned state-space model, an extended Kalman filter is used to iteratively calculate at each time step to obtain a smooth and reliable estimate of the ship's motion state, i.e., the current hull attitude. With speed . The real-time ship motion state vector, i.e., the current ship attitude, is obtained from the above data fusion method: , , , It is the position of the ship in the inertial coordinate system. Indicates the ship's heading angle, with starboard being positive. The roll angle, The pitch angle; and the velocity. .

[0073] In one embodiment, the prediction model employs an autoregressive model. The prediction module, based on the ship type response amplitude operator, the current ship attitude, and speed, uses an autoregressive model to output the predicted values ​​of future disturbances experienced by the ship within a prediction window of 3-5 seconds. The autoregressive model is an existing model and will not be elaborated upon here. In other embodiments, the autoregressive model can be replaced by a short-term wave prediction algorithm based on a Long Short-Term Memory (LSTM) network or a Temporal Convolutional Network (TCN); such algorithms automatically extract temporal features through deep learning structures and can achieve higher prediction accuracy in more complex sea conditions (the average error can be controlled within ±5%).

[0074] The low-frequency motion compensation layer is performed by the ship's dynamic positioning system to perform disturbance compensation tasks. Its main purpose is to keep the ship's attitude in a predetermined steady-state area by suppressing low-frequency motions (sway, roll, bow roll) on the horizontal plane, thereby reducing the disturbance intensity that the wave compensation boarding system needs to handle.

[0075] The low-frequency motion compensation layer includes a main control computer 201, an attitude sensor interface 202, and a thruster actuator unit 203. The attitude sensor interface 202 is responsible for integrating attitude / position / velocity information measured by the inertial measurement unit (IMU) and the global positioning system (GNSS) module 102 into the system, providing the available current hull state. ,speed The commands are provided to the main control computer 201 for real-time feasibility and safety constraint checks during the generation and issuance of thrust distribution commands for the ship's dynamic positioning system. The main control computer 201 receives predicted values ​​of disturbances experienced by the ship from the external disturbance prediction layer. A digital filter is used for low-frequency filtering to extract the low-frequency slow-drift motion component. The thruster execution unit 203 (thruster controller) is responsible for receiving thruster commands from the main control computer 201 and converting them into specific thruster thrust. The thruster controller operates according to the thruster configuration matrix. Calculate the total output force of the thruster under thrust / rate of change constraints. The thruster is based on the total output force of the thruster. It works to counteract this low-frequency, slow-drift motion component.

[0076] The high-frequency motion compensation layer includes a parallel Stewart platform 301 and a series three-degree-of-freedom gangway 302, which are used to perform graded and fine compensation for residual disturbances.

[0077] (1) Stewart platform 301 proactive compensation The compensation objective of the Stewart platform is to compensate for the remaining low-frequency motion and uncompensated high-frequency motion after compensation of the ship's dynamic positioning system, under the condition of ship motion. Compensation is performed by adjusting the pose of the upper platform of the Stewart platform to maintain a fixed and ideal stable pose in the inertial coordinate system. .

[0078] Based on the kinematic chain relationship, the upper platform pose of the Stewart platform It can be obtained by the following formula:

[0079] in It represents the amount of motion of the ship's hull over time, and the calculation uses uncompensated high-frequency motion. value; It is the rigid connection matrix from the hull to the lower platform of the Stewart platform. This represents the ideal, stable pose. The calculated relative pose... The position vector of the upper platform relative to the lower platform can be extracted from the data. and rotation matrix Furthermore, by employing the inverse kinematics solution method of the Stewart platform, the relative pose of the driving outrigger can be calculated. Required outrigger length :

[0080] in and These are the hinge coordinates of the lower platform and the upper platform, respectively.

[0081] (2) Active compensation for the three-degree-of-freedom gangway 302 As the final link in the compensation chain, the task of the gangway is to eliminate residual errors transmitted from all upstream links, ensuring that the end of the gangway is precisely maintained at the preset target point in the inertial coordinate system. The control method used in this section is speed mapping control based on the Jacobian matrix, which achieves the desired speed at the end of the gangway to maintain it at the preset target point. The specific implementation process is as follows: an optimization method based on task space speed control is adopted to optimize the speed command in the task space, so that the controller can more directly realize the "tracking of the target point at the end of the gangway" and reduce the indirect error transmission from the joint space to the task space.

[0082] The optimization method based on task space velocity control includes: The optimal decision variable is the joint velocity of the gangway. The actual speed at the end of the gangway is The optimization objective of this method for motion control is to find an optimal joint velocity. This results in the actual end velocity. Approaching the desired speed And satisfying the requirements of safety and smoothness: Construct and solve the quadratic programming problem: ,in It is a decision variable. , is the joint velocity value of the gangway; It is a quadratic term matrix of the objective function, reflecting the "cost weights"; It is the vector of first-order terms of the objective function.

[0083] In one embodiment, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the methods described in the foregoing embodiments.

[0084] In one embodiment, a computer-readable storage medium stores a computer program that, when executed by a processor, implements the methods described in the foregoing embodiments.

[0085] This invention, through the construction of a three-layer collaborative control architecture of "prediction-compensation-fine-tuning," achieves full-time-domain dynamic coordination between the ship's dynamic positioning system and the boarding compensation system under wave disturbance environments. This method significantly improves upon existing technologies in terms of system stability, response speed, energy efficiency, and end-point attitude accuracy. Specific technical effects are as follows: (i) Significantly improve boarding safety and dynamic stability (1) Significant effect of hierarchical disturbance compensation: Through the coordinated operation of the external disturbance prediction layer, the low-frequency motion compensation layer, and the high-frequency motion compensation layer, the system can decompose and compensate for disturbances at different time scales. The low-frequency slow drift motion is controlled by the ship's dynamic positioning system, reducing attitude error; the high-frequency wave disturbance is jointly compensated by the Stewart platform and the gangway, reducing the amplitude of end attitude jitter. (2) Prevention of structural overload and phase conflict of the wave compensation boarding system: In the traditional independent control method, the ship's dynamic positioning system (DP) and the wave compensation boarding system may cancel each other out or superimpose in phase, resulting in unsatisfactory control effect. This invention adopts a hierarchical planning method to divide external disturbances into low-frequency disturbances and high-frequency disturbances, and transmits these disturbance information to the low-frequency motion compensation layer and the high-frequency motion compensation layer through the external disturbance prediction layer. By staggering the control actions of low-frequency disturbances and high-frequency disturbances and rationally distributing energy, reverse compensation and mechanical shock problems are avoided. This method significantly extends the service life of the gangway and platform connection mechanism.

[0086] (ii) Significantly improve energy utilization efficiency and control coordination: (1) Optimized energy distribution: Layered control achieves energy decoupling between different systems. The ship's dynamic positioning system undertakes low-frequency displacement energy (about 70% of the total compensation power), while the Stewart platform and gangway undertake medium- and high-frequency energy (about 30%). Overall energy consumption is reduced by about 25%–30% compared to traditional parallel control. (2) Significantly improved control execution stability. The speed fluctuation amplitude of the propeller, hydraulic cylinder and gangway drive is reduced by an average of about 40%, effectively reducing system fatigue wear and maintenance frequency. (3) The external disturbance prediction layer provides a unified future state estimate to the two lower layers in real time, making the control commands of the entire system continuous in time and matched in amplitude, realizing consistent and coordinated control of "multi-device, single-objective".

[0087] The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined in this invention may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A hierarchical cooperative motion control method for a ship's dynamic positioning and wave compensation boarding system, characterized in that, Includes the following steps: Acquire historical motion data of the ship, including the ship's acceleration signal, angular velocity signal, position, speed and heading information; By fusing historical motion data of the ship, the current hull attitude and speed can be obtained; The prediction model generates predicted values ​​of the disturbances that the ship will experience at a preset time in the future, based on the ship type response amplitude operator, the current ship attitude and speed. Low-frequency slow drift motion components are extracted based on the predicted values ​​of disturbances experienced by the ship. The total output force of the propeller is calculated, and the ship's dynamic positioning system controls the operation of the propeller based on the total output force of the propeller to compensate for the low-frequency slow drift motion component; Based on the disturbances and low-frequency slow drift motion components experienced by the ship, the remaining low-frequency motion after compensation by the ship's dynamic positioning system and the uncompensated high-frequency motion are obtained. The target lengths of each outrigger of the Stewart platform are obtained by solving the inverse kinematics of the Stewart platform. The Stewart platform in the wave-compensated boarding system controls the operation of the outriggers based on the target lengths of the outriggers to compensate for the low-frequency motion remaining after compensation by the ship's dynamic positioning system and the uncompensated high-frequency motion. The speed at the end of the gangway is optimized so that the actual speed at the end of the gangway in the wave-compensated boarding system is close to the desired speed of the gangway so as to compensate for the remaining disturbances after compensation by the Stewart platform through the gangway. Among them, the target length of each leg of the Stewart platform The calculation formula is: in, and These are the hinge point coordinates of the lower and upper platforms of the Stewart platform, respectively. Let the position vector of the upper platform relative to the lower platform of the Stewart platform be the pose of the upper platform of the Stewart platform. get, It is a rotation matrix; The optimization steps for the speed at the end of the gangway include: Construct and solve the quadratic programming problem: ,in It is a decision variable. , is the joint velocity value of the gangway; It is the quadratic term matrix of the objective function; It is a linear term vector of the objective function, and the actual velocity at the end of the gangway is... The optimization objective is to find an optimal joint velocity. This results in the actual end velocity. Desired speed approaching the end of the gangway And meet safety and smoothness requirements, Indicates the position of the gangway joint and the actual speed at the end of the gangway. .

2. The hierarchical cooperative motion control method for a ship dynamic positioning and wave compensation boarding system according to claim 1, characterized in that, Ship historical motion data can be fused using any of the following algorithms: extended Kalman filter, unscented Kalman filter, particle filter, capacitive Kalman filter, or state estimator based on optimization theory.

3. The hierarchical cooperative motion control method for a ship dynamic positioning and wave compensation boarding system according to claim 1, characterized in that, The prediction model adopts an autoregressive model, or any one of the network models of long short-term memory network and temporal convolutional network.

4. The hierarchical cooperative motion control method for a ship dynamic positioning and wave compensation boarding system according to claim 1, characterized in that, The calculation method for the total output force of the propellers includes: obtaining the total output force of the propellers based on the propeller configuration matrix and the thrust components generated by all the propellers of the ship.

5. The hierarchical cooperative motion control method for a ship dynamic positioning and wave compensation boarding system according to claim 1, characterized in that, The desired velocity at the end of the gangway is obtained based on the residual disturbance after compensation by the Stewart platform, i.e., the residual disturbance generated by the Stewart platform on the gangway, and the gangway end pose reference trajectory.

6. A layered coordinated motion control system for a ship's dynamic positioning and wave compensation boarding system, characterized in that, The system for implementing the method of any one of claims 1-5 comprises: The ship external disturbance prediction layer is used to acquire historical ship motion data, which includes the ship's acceleration signal, angular velocity signal, position, speed and heading information; the historical ship motion data is fused to obtain the current ship attitude and speed; and the prediction model generates the predicted value of the disturbance that the ship will be subjected to at a preset time in the future based on the ship type response amplitude operator, the current ship attitude and speed. The low-frequency motion compensation layer is used to extract the low-frequency slow drift motion component based on the predicted value of the disturbances experienced by the ship, calculate the total output force of the propeller, and control the operation of the propeller based on the total output force of the propeller through the ship's dynamic positioning system to compensate for the low-frequency slow drift motion component. The high-frequency motion compensation layer is used to solve for the target length of each outrigger of the Stewart platform. The Stewart platform controls the outrigger operation based on the target length of the outrigger to compensate for the remaining low-frequency motion and uncompensated high-frequency motion after the ship's dynamic positioning system is compensated. The gangway end speed is optimized so that the actual end speed of the gangway in the wave-compensated boarding system is close to the desired speed of the gangway so as to compensate for the remaining disturbance through the gangway.

7. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the method according to any one of claims 1-5.

8. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the method described in any one of claims 1-5.

Images