CN119045503B - Adaptive wave disturbance control method for unmanned ships based on digital twin - Google Patents
Adaptive wave disturbance control method for unmanned ships based on digital twin Download PDFInfo
- Publication number
- CN119045503B CN119045503B CN202411545019.4A CN202411545019A CN119045503B CN 119045503 B CN119045503 B CN 119045503B CN 202411545019 A CN202411545019 A CN 202411545019A CN 119045503 B CN119045503 B CN 119045503B
- Authority
- CN
- China
- Prior art keywords
- wave
- unmanned ship
- speed
- new
- frequency
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- G05D1/40—Control within particular dimensions
- G05D1/43—Control of position or course in two dimensions
Landscapes
- Engineering & Computer Science (AREA)
- Aviation & Aerospace Engineering (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Feedback Control In General (AREA)
- Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)
Abstract
The invention provides a digital twin unmanned ship-based adaptive wave interference control method, which is characterized in that wave information is acquired in real time through a sensor, main frequencies are extracted by utilizing frequency domain analysis, and the future wave state is predicted by adopting an LSTM model in combination with filtered signals and physical characteristics. The navigation speed and the heading of the unmanned ship are adjusted in real time through the ROS nodes, the changes of the height, the speed and the direction of the wave are dynamically dealt with, and the navigation stability is ensured. And realizing virtual-real synchronous feedback by using a high-dimensional Kalman filter, and correcting a control strategy. The system comprises a wave information acquisition module, a data processing and prediction module, a control decision module and a feedback and optimization module, and ensures stable navigation of the unmanned ship in a wave mutation environment.
Description
Technical Field
The application belongs to the technical field of unmanned ship control, and particularly relates to a digital twin-based unmanned ship self-adaptive wave interference control method.
Background
The unmanned ship is a water surface robot which can independently navigate on the water surface according to a preset task by means of accurate satellite positioning and self sensors without remote control. The water surface robot integrates multiple technologies such as ships, communication, automatic control, remote monitoring, networking systems and the like, and achieves multiple functions such as autonomous navigation, intelligent obstacle avoidance, remote communication, video real-time transmission, networking monitoring and the like. As a new technical means, unmanned ships have been widely used in the fields of marine surveys, offshore defence, and the like.
During the navigation of unmanned ships, the sudden changes of waves caused by wind, waves and other factors are frequently encountered. In this case, the rapid changes in wave height, speed and direction present a great challenge to the stability of the unmanned ship, which is prone to capsizing, stalling or violent swaying. Therefore, the unmanned ship is effectively controlled in the wave abrupt change and high dynamic environment, and the running safety of the unmanned ship and the smooth completion of the tasks are directly related. The current unmanned ship control method can guarantee stable sailing under low sea conditions, but in the complex environment of wave mutation, the current control method cannot carry out rapid self-adaptive adjustment on the unmanned ship according to real-time sea conditions, and is difficult to realize rapid and accurate adjustment of the heading and the navigational speed of the unmanned ship, so that stable running of the unmanned ship under the wave mutation cannot be guaranteed.
Therefore, in view of the instability of unmanned ship sailing in the wave abrupt change environment and the limitations of the existing control method, there is a need for a control method capable of sensing the change of sea conditions in real time and adaptively adjusting the heading and speed of the unmanned ship. The method can ensure that the unmanned ship realizes stable sailing in the complex environment with abrupt wave change, and ensure the running safety and the smooth completion of the tasks.
Disclosure of Invention
Based on the problems, the application provides a digital twin unmanned ship self-adaptive wave interference control method, which has the following technical scheme:
An unmanned ship self-adaptive wave interference control method based on digital twinning comprises the following steps:
s1, acquiring wave signals through a wave radar sensor, and acquiring time domain information of wave height, wave length, wave speed and wave direction;
S2, preprocessing the acquired wave signals, removing noise, extracting main frequency and dominant period characteristics, and forming a wave characteristic matrix X to reflect the dynamic characteristics of waves;
S3, inputting the wave characteristic matrix X into an LSTM model, training by utilizing historical wave data, learning characteristics and variation trend in wave time sequence data, predicting the height, speed and direction of future waves, minimizing errors between predicted values and actual values by optimizing a loss function of the model, and continuously adjusting model parameters;
S4, according to the predicted future wave state, the controller acquires a wave prediction result and adjusts the heading and the navigational speed of the unmanned ship in real time, and when the wave height, speed or direction exceeds a set threshold value, an adjusted control command is dynamically generated, so that the unmanned ship is adaptively adjusted and stably navigated;
s5, feeding the actual state of the unmanned ship back to the virtual model by using a high-dimensional Kalman filter, optimizing a control strategy according to feedback information, guaranteeing synchronization of virtual and actual states, and guaranteeing stability and sailing efficiency of the unmanned ship under wave interference.
Preferably, in step S2:
S21, adopting generalized Fourier transform to carry out wave signal Performing frequency domain analysis to convert into wave frequency spectrum in frequency domain;
S22, extracting main frequency through spectrum analysisI.e. the frequency in the spectrum at which the energy is most concentrated, first, in the spectrumIn finding a frequency range in which the energy distribution is concentratedThe dominant frequency is then determined by means of weighted averaging:
;
Wherein, Is the amount of energy corresponding to the frequency,As frequency variableMinimum of the value of the sum of the values,As frequency variableMaximum value.
Preferably, in step S2, the extracted dominant frequency is calculatedTo obtain the dominant period of the waveAnd calculates the response time window of the unmanned ship according to the dominant periodThe response time window determines the minimum time for the unmanned ship to complete the adjustment before the wave impact.
Preferably, the training process of the LSTM model includes:
s3.1 wave characteristic matrix Inputting LSTM model, whereinFor the filtered sensor time domain signal,Representing characteristic parameters extracted from the physical model, including a dominant frequency and a dominant period, describing the dynamic characteristics of the wave;
S3.2 processing input feature matrix through multilayer LSTM cells Calculating hidden statesRecursively capturing wave time sequence characteristics, and performing forward and backward processing on sequence data by using a bidirectional LSTM (least squares) to obtain a complete hidden state;
S3.3 to hide the state Inputting fully connected layers, generating predictions of future wave statesIncluding the height, speed and direction of the waves;
s3.4 calculating the predicted value And true wave stateThe error between the two modes is optimized by minimizing the loss function;
And S3.5, adjusting model parameters by using a back propagation algorithm, and performing gradient descent optimization on the weight matrix.
Preferably, in step S4:
S41, calculating future wave state by using LSTM model The controller issues wave prediction instructions to the control nodes in the form of ROS messages for controlling and adjusting the unmanned ship;
s42, the control node acquires a wave prediction instruction, and the heading and the navigational speed of the unmanned ship are adaptively adjusted according to the wave prediction result;
S43, unmanned ship receives heading And navigational speedAnd the instruction is used for dynamically adjusting the rudder angle and the propeller speed of the ship body.
Preferably, in step S4, the unmanned ship self-adaptive adjustment is as follows:
Condition one, when the wave height is Greater than a high threshold(E.g., 1.5 m), the unmanned ship should reduce the speed to reduce the impact of the waves on the hull, new speedFor the current speed of navigationSubtracting a certain proportion:
;
Wherein the method comprises the steps of Is a deceleration coefficient (e.g., 0.3);
Second condition is when the wave height is lower than the low threshold (E.g., 0.5 m), the wave has less impact on sailing, and the speed of sailing can be increased appropriately to accelerate the journey:
;
Wherein the method comprises the steps of Is a speed-increasing coefficient (e.g., 0.1);
Condition III, when the wave is in the direction Heading of unmanned shipPhase difference is greater than a set angle thresholdAt (e.g., 30 °), new headingIs adjusted to be consistent with or close to the wave direction so as to reduce the lateral stress:
;
condition IV, when the wave speed is Exceeding a speed threshold(E.g., 2 m/s), it is shown that the wave energy is higher and has a greater impact on hull stability. At this point the speed should be reduced to enhance hull control:
;
Wherein the method comprises the steps of Is a deceleration coefficient (e.g., 0.2);
preferably, the unmanned ship issues its actual state in the form of ROS message to the control node during execution of the control command, to form the actual state of the unmanned ship 。
Preferably, the actual state of the unmanned ship is fed back in real time by utilizing the multidimensional real-time feedback of the high-dimensional Kalman filterFeedback to virtual model, comparing virtual model state estimationAnd actual stateAnd calculating errors and correcting the virtual model to ensure that the virtual model is consistent with the physical unmanned ship.
Preferably, based on real-time errorsAndThe control node (/ wave_adaptive_control) dynamically adjusts the control strategy;
When the speed of the ship is error Exceeding the speed of the journeyAt 15% of (C), the reduction coefficientThe speed of the unmanned ship is reduced by 0.1-0.2, so that the stable sailing is ensured. Decreasing the deceleration coefficient or increasing the acceleration coefficient when the speed error is less than 5% of the desired speedTo improve the speed of unmanned ship and enhance the sailing efficiency;
when heading error When the heading adjustment sensitivity is increased by more than 15 degrees, the adjustment step length is increased by 10-30%, and the unmanned ship is accelerated to turn to meet waves. When the heading error is smaller than 5 degrees, the sensitivity is reduced, excessive adjustment is prevented, and the unmanned ship stably navigates.
Unmanned ship self-adaptive wave interference control system based on digital twin includes:
The wave information acquisition module is used for acquiring the height, speed, direction and other data of the waves in real time and providing basic information of sea conditions around the unmanned ship;
the data preprocessing and feature extraction module is used for filtering and denoising the acquired wave data, extracting main frequency and feature parameters, and constructing a wave feature matrix for the subsequent model prediction;
the wave state prediction module is used for training and predicting the wave feature matrix by utilizing the LSTM model, generating a future wave state prediction result and supporting real-time decision of the unmanned ship;
the unmanned ship control decision-making module is used for making a course and speed adjustment strategy of the unmanned ship according to the wave prediction result, so that the unmanned ship can adapt to the wave environment and keep stable sailing;
And the virtual-real feedback and optimization module is used for feeding the actual sailing state of the unmanned ship back to the digital twin model, realizing the synchronization of the virtual-real state through the high-dimensional Kalman filter, and realizing the self-adaptive optimization control of the unmanned ship according to the optimized control strategy.
Compared with the prior art, the application has the following beneficial effects:
1. by introducing an LSTM model to train the wave feature matrix, accurate prediction of the wave height, speed and direction is realized. Compared with the traditional control method based on experience and preset rules, the method can predict the future wave state in real time, dynamically adjust the heading and the navigational speed of the unmanned ship according to the prediction result, and realize more efficient self-adaptive control.
2. And extracting key characteristic parameters such as main frequency, dominant period and the like in the wave signal by utilizing a filtering technology, combining the key characteristic parameters with the physical model characteristics, constructing an input matrix, and further optimizing a wave prediction model. Compared with simple wave data analysis, the wave characteristic multi-level analysis method fully considers the wave characteristic multi-level, and improves the sensitivity and prediction accuracy of the model to wave interference.
3. And a high-dimensional Kalman filter is introduced to feed back the actual state of the unmanned ship to the digital twin model, the error between the virtual model and the actual state is compared, and a control strategy is dynamically adjusted, so that virtual-real synchronization and self-optimization control are realized. The closed loop feedback mechanism improves the control precision and the real-time response capability of the unmanned ship, so that the unmanned ship can keep stable sailing in complex sea conditions.
4. According to the real-time wave state, the method sets the threshold value for different wave heights, directions and speeds, and dynamically generates the adjustment strategy. For example, when the wave height is too high, the unmanned ship decelerates to ensure stability, and when the wave direction deviation is large, the heading is adjusted in time to run against the waves. The self-adaptive adjustment strategy is more flexible than the existing method based on fixed control parameters, and ensures that the unmanned ship can still keep high-efficiency sailing under variable sea conditions.
5. The unmanned ship is ensured to run at the optimal speed and heading by monitoring the speed error and heading error of the unmanned ship in real time and adjusting the speed reduction coefficient, the speed increase coefficient and the heading sensitivity according to the error. The navigation efficiency of the unmanned ship in the wave environment is improved, and the navigation safety is ensured.
Drawings
FIG. 1 is a flow chart of the present application.
FIG. 2 is a wave treatment and prediction flow chart of the present application.
Fig. 3 is a flow chart of the adaptive control strategy of the present application.
Detailed Description
The following description of embodiments of the present invention will be made clearly and fully with reference to the accompanying drawings, in which it is evident that the embodiments described are only some, but not all embodiments of the invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
An unmanned ship self-adaptive wave interference control method based on digital twinning comprises the following steps:
s1, acquiring wave signals through a wave radar sensor, and acquiring time domain information of wave height, wave length, wave speed and wave direction;
s11, acquiring wave signals in real time by utilizing wave radar sensor Including wave heightWavelength ofWave velocityAnd wave directionTime domain information of (1), at this time domain signalReflecting the change in the wave in the time dimension:
。
s2, preprocessing the acquired wave signals, removing noise, extracting main frequency, dominant period and other characteristics, and forming a wave characteristic matrix X so as to reflect the dynamic characteristics of waves;
S21, adopting generalized Fourier transform to carry out wave signal Performing frequency domain analysis to convert into wave frequency spectrum in frequency domain:
;
Is a function of the time window of the time,Is a frequency variable, different frequencies correspond to different oscillation periods in the wave;
s22, extracting main frequency through spectrum analysis I.e. the frequency in the spectrum where the energy is most concentrated. First, in the frequency spectrumIn finding a frequency range in which the energy distribution is concentratedThe dominant frequency is then determined by means of weighted averaging:
;
Wherein, Is the energy level of the corresponding frequency. The formula calculates the energy weighted average frequency in the frequency range, comprehensively considers all frequency energy in the frequency range, and obtains the main frequency reflecting the main oscillation characteristic of the wave.
Calculating the principal frequency of extractionTo obtain the dominant period of the waveAnd calculates the response time window of the unmanned ship according to the dominant periodThe response time window determines the minimum time for the unmanned ship to complete the adjustment before the wave impact.
S23, using a filter to further filter noise, and adjusting the coefficient of the filter according to the real-time wave dataOptimizing the detection effect of the wave signal:
。
S3, inputting the wave characteristic matrix X into an LSTM model, and learning characteristics and variation trend in wave time sequence data through training of historical wave data to predict the height, speed and direction of future waves. And (3) minimizing the error between the predicted value and the actual value by optimizing the loss function of the model, and continuously adjusting the model parameters.
S31, wave characteristic matrixInputting LSTM model thereinFor the filtered time domain signal of the sensor,Representing characteristic parameters extracted from a physical model, including principal frequenciesAnd dominant periodEtc. Wave characteristic matrixReflecting the dynamic characteristics of the waves and forming the input layer.
S32, hiding the layer, namely in an LSTM model, recursively processing input features through a plurality of layers of LSTM units, capturing wave time sequence features, and calculating a hiding state matrixObtained by recursive calculation:
;
Wherein, In order to conceal the layer weight matrix,For the input layer to hidden layer weight matrix,As an input feature for the current time step,In the hidden state of the previous time step,As a result of the bias term,To activate the function.
And the bidirectional LSTM unit carries out forward and backward processing on the sequence data, and the finally output hidden state is the splicing of forward and backward LSTM.
S33, hiding layer output stateConversion to predicted wave state through fully connected layers:
;
wherein:
To conceal the layer-to-output layer weight matrix, Bias terms for the output layer;
for the prediction of future time steps wave conditions, including the height, direction and speed of the wave.
S34, calculating a predicted valueAnd true wave stateMean square error between:
wherein:
the specific calculation formula is as follows, wherein the 2-norm is the difference between the true value and the predicted value:
;
The first to be the true wave state The characteristics of the device are that,Is the corresponding predicted value. By back propagation algorithm, according to errorFor model parametersAnd gradient descent optimization is carried out, prediction errors are gradually reduced, and model prediction accuracy is improved.
S4, according to the predicted future wave state, the controller adjusts the heading and the navigational speed of the unmanned ship in real time. When the height, speed or direction of the wave exceeds a set threshold, dynamically generating an adjusted control command to enable the unmanned ship to adaptively adjust and keep stable sailing;
S41, creating/wave_sensor_data nodes to subscribe real-time sensor data topics related to wave states, and preprocessing the topics to obtain . Input by LSTM modelPredicting characteristics to calculate future wave stateAnd issued in the form of ROS messages to a new topic (/ predicted_wave_state) for control adjustment of the unmanned ship.
S42, creating a control node (/ wave_adaptive_control), subscribing/predicting_wave_state theme, and adaptively adjusting the heading and the navigational speed of the unmanned ship according to the wave prediction result.
Condition one, when the wave height isGreater than a high threshold(E.g., 1.5 m), the unmanned ship should reduce the speed to reduce the impact of the waves on the hull, new speedFor the current speed of navigationSubtracting a certain proportion:
;
Wherein the method comprises the steps of Is a deceleration coefficient (e.g., 0.3).
Second condition is when the wave height is lower than the low threshold(E.g., 0.5 m), the wave has less impact on sailing, and the speed of sailing can be increased appropriately to accelerate the journey:
;
Wherein the method comprises the steps of Is a speed-increasing coefficient (e.g., 0.1).
Condition III, when the wave is in the directionHeading of unmanned shipPhase difference is greater than a set angle thresholdAt (e.g., 30 °), new headingIs adjusted to be consistent with or close to the wave direction so as to reduce the lateral stress:
。
condition IV, when the wave speed is Exceeding a speed threshold(E.g., 2 m/s), it is shown that the wave energy is higher and has a greater impact on hull stability. At this point the speed should be reduced to enhance hull control:
;
Wherein the method comprises the steps of Is a deceleration coefficient (e.g., 0.2).
S43, generating adjusted heading and navigational speed control commands by the aid of the/wave_adaptive_control control node, and issuing the commands to a controller (/ cmd_control) in the form of ROS messages. And further to the motion controller of the unmanned ship. Unmanned ship according to received expected headingAnd desired navigational speedAnd dynamically adjusting the rudder angle and the propeller speed of the ship body to realize the expected heading and the expected speed.
S5, feeding the actual state of the unmanned ship back to the virtual model by using a high-dimensional Kalman filter, optimizing a control strategy according to feedback information, guaranteeing synchronization of virtual and actual states, and guaranteeing stability and sailing efficiency of the unmanned ship under wave interference.
S51, in the process of executing a control instruction, the unmanned ship issues/shifts_state subjects in real time states (including current heading, navigational speed, attitude and the like) in the form of ROS messages to form the real state of the unmanned ship。
S52, utilizing multidimensional real-time feedback of a high-dimensional Kalman filter to enable the actual state of the unmanned ship to be achievedFeedback to virtual model, comparing virtual model state estimationAnd actual stateCalculating an error and correcting the virtual model according to the real-time errorAndThe control node (/ wave_adaptive_control) dynamically adjusts the control strategy.
When the speed of the ship is errorExceeding 15% of the desired speed, the coefficient of deceleration is reducedThe speed of the unmanned ship is reduced by 0.1-0.2, so that the stable sailing is ensured. Decreasing the deceleration coefficient or increasing the acceleration coefficient when the speed error is less than 5% of the desired speedTo promote unmanned ship speed, strengthen navigation efficiency.
When heading errorWhen the heading adjustment sensitivity is increased by more than 15 degrees, the adjustment step length is increased by 10-30%, and the unmanned ship is accelerated to turn to meet waves. When the heading error is smaller than 5 degrees, the sensitivity is reduced, excessive adjustment is prevented, and the unmanned ship stably navigates.
Unmanned ship self-adaptive wave interference control system based on digital twin includes:
The wave information acquisition module is used for acquiring the height, speed, direction and other data of the waves in real time and providing basic information of sea conditions around the unmanned ship;
the data preprocessing and feature extraction module is used for filtering and denoising the acquired wave data, extracting main frequency and feature parameters, and constructing a wave feature matrix for the subsequent model prediction;
the wave state prediction module is used for training and predicting the wave feature matrix by utilizing the LSTM model, generating a future wave state prediction result and supporting real-time decision of the unmanned ship;
The unmanned ship control decision-making module outputs an unmanned ship control instruction, and makes a heading and navigational speed adjustment strategy of the unmanned ship according to a wave prediction result, so that the unmanned ship can adapt to a wave environment and keep stable navigational;
And the virtual-real feedback and optimization module is used for feeding the actual sailing state of the unmanned ship back to the digital twin model, realizing the synchronization of the virtual-real state through the high-dimensional Kalman filter, and realizing the self-adaptive optimization control of the unmanned ship according to the optimized control strategy.
Although embodiments of the present application have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made therein without departing from the principles and spirit of the application, the scope of which is defined in the appended claims and their equivalents.
Claims (7)
1. The digital twinning-based unmanned ship self-adaptive wave interference control method is characterized by comprising the following steps of:
s1, acquiring wave signals through a wave radar sensor, and acquiring time domain information of wave height, wave length, wave speed and wave direction;
s2, preprocessing the acquired wave signals, removing noise, extracting main frequency and dominant period characteristics, and forming a wave characteristic matrix X;
S21, carrying out frequency domain analysis on the wave signal S (t) by adopting generalized Fourier transform, and converting the wave signal S (t) into a wave frequency spectrum S (f) in a frequency domain;
S22, extracting a main frequency f dominat, namely the frequency with the most concentrated energy in the frequency spectrum through frequency spectrum analysis, firstly, finding a frequency range [ f lower,fupper ] with concentrated energy distribution in the frequency spectrum [ S (f) [ I ], and then determining the main frequency through a weighted average mode:
Wherein |s (f) | is the energy magnitude of the corresponding frequency, f lower is the frequency variable fmin, and f upper is the frequency variable fmax;
Calculating the reciprocal of the extracted main frequency f dominant to obtain a dominant period T wave of the wave, and calculating a response time window T response of the unmanned ship according to the dominant period, wherein the response time window determines the shortest time for the unmanned ship to finish adjustment before wave impact;
S3, inputting the wave characteristic matrix X into an LSTM model, training by utilizing historical wave data, learning characteristics and variation trend in wave time sequence data, predicting the height, speed and direction of future waves, minimizing errors between predicted values and actual values by optimizing a loss function of the model, and continuously adjusting model parameters;
S4, according to the predicted future wave state, the controller acquires a wave prediction result and adjusts the heading and the navigational speed of the unmanned ship in real time, and when the wave height, speed or direction exceeds a set threshold value, an adjusted control command is dynamically generated, so that the unmanned ship is adaptively adjusted;
s5, feeding the actual state of the unmanned ship back to the virtual model by using a high-dimensional Kalman filter, optimizing a control strategy according to feedback information, guaranteeing synchronization of virtual and actual states, and guaranteeing stability and sailing efficiency of the unmanned ship under wave interference.
2. The digital twinning-based unmanned ship adaptive wave interference control method according to claim 1, wherein the training process of the LSTM model comprises:
S3.1, inputting a wave characteristic matrix X= [ S filtered (t), P ] into an LSTM model, wherein S filtered (t) is a filtered sensor time domain signal, and P represents characteristic parameters extracted from a physical model, including main frequency and dominant period, and describing the dynamic characteristics of waves;
S3.2, processing an input feature matrix X through a multi-layer LSTM unit, calculating a hidden state h t, recursively capturing wave time sequence features, and performing forward and backward processing on sequence data by utilizing a bidirectional LSTM unit to obtain a complete hidden state;
S3.3 inputting the hidden state h t into the fully connected layer to generate a prediction of future wave states Including the height, speed and direction of the waves;
s3.4 calculating the predicted value The error between the model and the real wave state Y is optimized in a mode of minimizing a loss function;
And S3.5, adjusting model parameters by using a back propagation algorithm, and performing gradient descent optimization on the weight matrix.
3. The digital twinning-based unmanned ship adaptive wave interference control method according to claim 1, wherein in step S4,
S41, calculating future wave state by using LSTM modelThe controller issues wave prediction instructions to the control nodes in the form of ROS messages for controlling and adjusting the unmanned ship;
s42, the control node acquires a wave prediction instruction, and the heading and the navigational speed of the unmanned ship are adaptively adjusted according to the wave prediction result;
S43, dynamically adjusting the rudder angle and the propeller speed of the ship body by the unmanned ship according to the received heading theta new and the received navigational speed V new instruction.
4. The digital twinning-based unmanned ship adaptive wave interference control method according to claim 1, wherein in step S4, the unmanned ship adaptive adjustment is as follows:
the first condition is that when the wave height H wave is greater than the high threshold H threshold_high, the unmanned ship should reduce the speed to reduce the impact of the wave on the hull, and the new speed V new is calculated as follows:
Vnew＝Vcurrent×(1-k1)
wherein k 1 is the deceleration coefficient;
condition two, when the wave height is below the low threshold H threshold_low, the new voyage speed V new is calculated as follows:
Vnew＝Vcurrent×(1+k2)
Wherein k 2 is the acceleration factor;
And thirdly, when the difference between the wave direction D wave and the unmanned ship heading theta ship is larger than a set angle threshold value theta threshold, the new heading theta new is adjusted to be consistent with or close to the wave direction, and the new heading theta new is calculated as follows:
θnew＝Dwave
And fourthly, when the wave speed C wave exceeds the speed threshold C threshold, the condition shows that the wave energy is higher, the influence on the stability of the ship body is larger, and the new navigational speed V new is calculated as follows:
Vnew＝Vcurrent×(1-k3)
Where k 3 is the deceleration coefficient.
5. The digital twinning-based unmanned ship self-adaptive wave interference control method according to claim 1, wherein in step S4, the unmanned ship issues its actual state in the form of ROS message to the control node in the process of executing the control command, so as to form the actual state D physical of the unmanned ship.
6. The digital twinning-based unmanned ship self-adaptive wave interference control method according to claim 5, wherein in step S5, the actual state D physical of the unmanned ship is fed back to the virtual model by using multidimensional real-time feedback of a high-dimensional kalman filter, and the state estimation of the virtual model is comparedAnd the actual state z t, calculating errors and correcting the virtual model, and ensuring that the virtual model keeps consistent with the physical unmanned ship.
7. The digital twinning-based unmanned ship self-adaptive wave interference control method according to claim 6, wherein the control node dynamically adjusts the control strategy according to real-time errors e V＝Vnew-Vactual and e θ＝θnew-θactual, wherein V actual is the actual navigational speed and theta actual is the actual heading;
Increasing the deceleration coefficient k 1,k3 when the speed error e V exceeds 15% of the speed V new, decreasing the deceleration coefficient k 1,k3 or increasing the acceleration coefficient k 2 when the speed error is less than 5% of the speed V new,
When the heading error e θ exceeds 15 deg., the step size is increased.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202411545019.4A CN119045503B (en) | 2024-11-01 | 2024-11-01 | Adaptive wave disturbance control method for unmanned ships based on digital twin |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202411545019.4A CN119045503B (en) | 2024-11-01 | 2024-11-01 | Adaptive wave disturbance control method for unmanned ships based on digital twin |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN119045503A CN119045503A (en) | 2024-11-29 |
| CN119045503B true CN119045503B (en) | 2025-03-18 |
Family
ID=93587435
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202411545019.4A Active CN119045503B (en) | 2024-11-01 | 2024-11-01 | Adaptive wave disturbance control method for unmanned ships based on digital twin |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN119045503B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN119987212B (en) * | 2025-04-11 | 2025-08-05 | 山东省科学院海洋仪器仪表研究所 | Adaptive wave interference control method and system for unmanned ships based on digital twins |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114492217A (en) * | 2022-04-19 | 2022-05-13 | 中国海洋大学 | A typhoon wave height prediction method based on hybrid time series perceptron model |
| CN116736867A (en) * | 2023-08-10 | 2023-09-12 | 湖南湘船重工有限公司 | Unmanned ship obstacle avoidance control system |
| CN117111600A (en) * | 2023-08-03 | 2023-11-24 | 大连海事大学 | Path tracking control method based on digital intelligent double-drive sea wave inversion observation |
| CN117744514A (en) * | 2023-12-04 | 2024-03-22 | 华南理工大学 | Unmanned ship real-time navigation monitoring and forecasting system based on CFD numerical proxy model |
Family Cites Families (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110806692A (en) * | 2019-10-21 | 2020-02-18 | 上海海事大学 | A Wave Compensation Prediction Method Based on CNN-LATM Combination Model |
| CN111813107B (en) * | 2020-06-17 | 2021-11-02 | 珠海云洲智能科技股份有限公司 | Unmanned ship control method, unmanned ship control device and unmanned ship |
| CN114445634A (en) * | 2022-02-28 | 2022-05-06 | 南京信息工程大学 | Sea wave height prediction method and system based on deep learning model |
| CN116933152B (en) * | 2023-06-07 | 2024-05-03 | 哈尔滨工业大学（威海） | A method and system for predicting ocean wave information based on multidimensional EMD-PSO-LSTM neural network |
| CN117217097B (en) * | 2023-11-07 | 2024-01-26 | 江苏航运职业技术学院 | A platform-based construction method and system for digital twins in the shipbuilding industry |
| CN118015228A (en) * | 2024-01-25 | 2024-05-10 | 江苏科技大学 | A multi-AUV digital twin system, device and storage medium based on mixed reality |
| CN118011839B (en) * | 2024-04-09 | 2024-06-04 | 哈船通航(三亚)技术服务合伙企业(有限合伙) | Unmanned ship active self-righting system regulation and control method based on artificial intelligence |
| CN118533147A (en) * | 2024-06-13 | 2024-08-23 | 广东电网有限责任公司 | A wave information monitoring method and system based on RTK |
-
2024
- 2024-11-01 CN CN202411545019.4A patent/CN119045503B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114492217A (en) * | 2022-04-19 | 2022-05-13 | 中国海洋大学 | A typhoon wave height prediction method based on hybrid time series perceptron model |
| CN117111600A (en) * | 2023-08-03 | 2023-11-24 | 大连海事大学 | Path tracking control method based on digital intelligent double-drive sea wave inversion observation |
| CN116736867A (en) * | 2023-08-10 | 2023-09-12 | 湖南湘船重工有限公司 | Unmanned ship obstacle avoidance control system |
| CN117744514A (en) * | 2023-12-04 | 2024-03-22 | 华南理工大学 | Unmanned ship real-time navigation monitoring and forecasting system based on CFD numerical proxy model |
Also Published As
| Publication number | Publication date |
|---|---|
| CN119045503A (en) | 2024-11-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN115016496A (en) | Path tracking method of surface unmanned vehicle based on deep reinforcement learning | |
| EP2952994A1 (en) | System and method for dynamic positioning | |
| Zhang et al. | Anti-disturbance control for dynamic positioning system of ships with disturbances | |
| CN119045503B (en) | Adaptive wave disturbance control method for unmanned ships based on digital twin | |
| CN110209054A (en) | Unmanned ship course active disturbance rejection control system based on RBF neural network | |
| CN114967702A (en) | Unmanned ship control system and path tracking method | |
| CN118915593B (en) | Ship actuating mechanism motion control method based on fusion optimization algorithm | |
| CN112947504B (en) | Autonomous recovery control system and control method for unmanned underwater vehicle | |
| CN116736709A (en) | A dynamic compensation type self-disturbance heading control method for marine robots | |
| CN114137826A (en) | An Active Disturbance Rejection Control Method for Ship Dynamic Positioning Based on Grey Wolf Algorithm | |
| CN108717263A (en) | Model-free adaption course heading control method of the more body bows of wave glider to fusion | |
| CN114609905A (en) | A ship formation event trigger control method | |
| CN119758736A (en) | A method and device for surface and underwater heterogeneous cooperative control based on preset time interference observer | |
| CN109656142A (en) | A kind of tandem structure model-free adaption method of guidance of unmanned boat | |
| Yu et al. | Adaptive path following control for wave gliders in ocean currents and waves | |
| Dai et al. | Dual closed loop AUV trajectory tracking control based on finite time and state observer | |
| Shen et al. | A novel speed optimisation scheme for unmanned sailboats by sliding mode extremum seeking control without steady-state oscillation | |
| Huo | Research on unmanned ship control system based on fuzzy PID | |
| CN104142626B (en) | Ship dynamic positioning control method based on inverse system and internal model control | |
| Li et al. | Dynamic positioning test for removable of ocean observation platform | |
| CN118605541B (en) | ASV dynamic event triggering course control system under unknown ocean disturbance | |
| CN120161867A (en) | A formation-keeping control method for unmanned boats under communication delay | |
| Unar et al. | A neural network based heading and position control system of a ship | |
| Yuan et al. | Combined depth and heading control and experiment of ROV under the influence of residual buoyancy, current disturbance, and control dead zone | |
| CN113820956B (en) | High-speed AUV motion control method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |