Suspension cooperative steering and fault-tolerant method of wave adaptive unmanned surface vehicle

Abstract

This application relates to a suspension-coordinated steering and fault-tolerant method for a wave-adaptive unmanned surface vessel (USV), comprising: real-time acquisition of environmental perception information and its own attitude data of the USV, and fusion processing to generate a multi-source data matrix; prediction of the sea state ahead based on the multi-source data matrix, and when the sea state characteristic parameters meet the preset dynamic adjustment conditions, adjusting the extension and retraction of each set of electromagnetic suspensions within the prediction time window to form a draft gradient that adapts to changes in sea state; upon receiving a steering command, synchronously adjusting the draft difference between the floating bodies on both sides of the USV and the thrust difference between the propulsion systems on both sides to control the steering trajectory and suppress the hull roll; real-time monitoring of the working status of the electromagnetic suspension and propulsion system during navigation, classifying faults, and when an abnormality of the execution unit is detected, triggering a fault-tolerant reconstruction strategy according to the fault type, locking the abnormal component and reconstructing the force distribution or propulsion thrust using the remaining execution units to maintain the navigation capability of the USV.

Description

Technical Field

[0001] This application relates to the field of unmanned surface vessel control and marine intelligent equipment technology, and in particular to a suspension-coordinated steering and fault-tolerant method for wave-adaptive unmanned surface vessels. Background Technology

[0002] With the increasing demands for tasks such as ocean observation, maritime patrol, resource exploration, and maritime safety assurance, unmanned surface vessels (USVs) are gradually becoming an important component of the marine intelligent equipment system due to their characteristics of requiring no personnel and being able to navigate autonomously for extended periods. When performing patrol monitoring, formation operations, or operations in complex sea areas, USVs typically need to maintain stable navigation in varying sea conditions while performing maneuvers such as track tracking and turning. However, due to the continuous effects of waves, currents, and wind in the marine environment, USVs are prone to rolling, pitching, and heaving motions during navigation, which can affect navigation stability and control accuracy.

[0003] Current unmanned surface vessels (USVs) typically rely on propulsion system thrust adjustment or rudder control to adjust their course by altering the thrust differential or rudder angle. However, this traditional approach suffers from a strong nonlinear coupling defect in dynamics: during high-speed maneuvers or large-angle turns, the yaw moment generated by the differential thrust and the centrifugal force generated by the hull rotation inevitably translate into a rolling moment that exacerbates the transient heeling of the main hull.

[0004] Furthermore, while some unmanned surface vessels (USVs) are structurally equipped with floats or roll damping devices, their adjustment mechanisms are mostly passive responses. Passive devices can only generate a restoring torque based on buoyancy differences after the hull has already tilted, resulting in a significant time lag in the physical process. This "lagging response" means that existing equipment cannot actively and synchronously provide anti-roll torque the instant a steering command is received. This makes USVs highly susceptible to severe rolling oscillations or even capsizing risks when performing emergency obstacle avoidance or formation maneuvers in complex sea conditions, severely limiting the high-maneuverability navigation capabilities of marine intelligent equipment under adverse conditions.

[0005] On the other hand, during long-term autonomous navigation of unmanned surface vessels (USVs), critical components such as suspension actuators and propulsion systems inevitably experience performance degradation or malfunctions. Once a critical actuator malfunctions, traditional control systems often lack effective fault reconstruction capabilities, easily leading to decreased navigation stability or even mission interruption. Therefore, how to achieve attitude stability control, maneuvering control, and safe navigation in the event of system failures in complex sea conditions has become a crucial problem that needs to be solved in the current field of USV control technology. Summary of the Invention

[0006] One objective of this application is to provide a suspension-coordinated steering and fault-tolerant method for wave-adaptive unmanned surface vessels, at least to solve the aforementioned problems.

[0007] To achieve the above objectives, some embodiments of this application provide a suspension-coordinated steering and fault-tolerant method for wave-adaptive unmanned surface vessels (USVs), applied to USVs including a main hull, multiple sets of floating body suspension assemblies, and a propulsion system. Each set of floating body suspension assemblies includes a floating body and an electromagnetic suspension connected to the main hull, comprising:

[0008] The system collects environmental perception information and its own attitude data from the unmanned surface vessel in real time, and then performs fusion processing to generate a multi-source data matrix.

[0009] Based on the multi-source data matrix, the sea conditions ahead are predicted. When the sea condition characteristic parameters meet the preset dynamic adjustment conditions, the extension and retraction of each group of electromagnetic suspensions are adjusted within the prediction time window to form a draft gradient that adapts to changes in sea conditions.

[0010] Upon receiving a turning command, the unmanned surface vessel simultaneously adjusts the difference in draft between the two floating bodies on both sides and the difference in thrust between the two propulsion systems to control the turning trajectory and suppress the hull roll.

[0011] During navigation, the working status of the electromagnetic suspension and propulsion system is monitored in real time, and faults are classified. When an abnormality of the actuator is detected, a fault-tolerant reconfiguration strategy is triggered according to the fault type. The abnormal component is locked and the remaining actuators are used to reconfigure the force distribution or propulsion thrust to maintain the navigation capability of the unmanned surface vessel.

[0012] Compared with related technologies, the solution provided in this application solves the problem of balancing navigation stability and maneuverability of unmanned surface vessels (USVs) in complex wave environments by constructing a closed-loop control link of perception-prediction-coordination-fault tolerance. It changes the traditional passive response mode of equipment to waves by using predictive draft adjustment and deep coupling of thrust / suspension to suppress the impact of wave disturbances on the main hull. At the same time, the multi-level fault tolerance mechanism ensures that the USV can still maintain basic navigation capabilities through force reconstruction when key execution units are damaged, thereby improving the survivability and mission reliability of the equipment in harsh conditions in the open sea. Attached Figure Description

[0013] One or more embodiments are illustrated by way of example with reference numerals in the accompanying drawings. These illustrations do not constitute a limitation on the embodiments. Elements with the same reference numerals in the drawings are denoted as similar elements. Unless otherwise stated, the figures in the drawings are not to be limited by scale.

[0014] Figure 1 This is a schematic diagram of the draft adjustment of the unmanned surface vessel buoy provided in an embodiment of this disclosure.

[0015] Figure 2 This is a flowchart of the suspension cooperative steering and fault-tolerant method for wave-adaptive unmanned surface vessels provided in the embodiments of this disclosure.

[0016] Figure 3 This is the main flowchart of the method provided in the embodiments of this disclosure.

[0017] Figure 4 This is a flowchart of the process for generating a multi-source data matrix provided in an embodiment of this disclosure.

[0018] Figure 5 This is a flowchart of sea state prediction and draft gradient adjustment provided in the embodiments of this disclosure.

[0019] Figure 6 This is a flowchart of the suspension-coordinated steering control provided in the embodiments of this disclosure.

[0020] Figure 7 This is a flowchart of fault detection classification and fault-tolerant reconstruction provided in the embodiments of this disclosure.

[0021] Figure 8 This is a flowchart illustrating the auxiliary functions and closed-loop optimization provided in the embodiments of this disclosure. Detailed Implementation

[0022] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0023] The terms "first," "second," etc., used in the specification, claims, and accompanying drawings of this disclosure are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate for the embodiments of this disclosure described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion.

[0024] In this disclosure, the terms "upper," "lower," "inner," "middle," "outer," "front," and "rear," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. These terms are primarily for better description of the embodiments of this disclosure and their implementations, and are not intended to limit the indicated devices, elements, or components to having a specific orientation, or to require them to be constructed and operated in a specific orientation. Furthermore, some of the aforementioned terms may be used to indicate other meanings besides orientation or positional relationship; for example, the term "upper" may in some cases indicate a dependency or connection relationship. Those skilled in the art can understand the specific meaning of these terms in the embodiments of this disclosure according to the specific circumstances.

[0025] Furthermore, the terms "set up," "connect," and "fix" should be interpreted broadly. For example, "connection" can be a fixed connection, a detachable connection, or an integral structure; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, or it can be an internal connection between two devices, components, or parts. Those skilled in the art can understand the specific meaning of the above terms in the embodiments of this disclosure according to the specific circumstances.

[0026] Unless otherwise stated, the term "multiple" means two or more.

[0027] In this embodiment of the disclosure, the character " / " indicates that the objects before and after it are in an "or" relationship. For example, A / B means: A or B.

[0028] The term "and / or" describes an association between objects, indicating that three relationships can exist. For example, A and / or B means: A or B, or A and B.

[0029] It should be noted that, unless otherwise specified, the embodiments and features described in the present disclosure can be combined with each other.

[0030] Combination Figures 1 to 8 As shown in the embodiments of this disclosure, a suspension cooperative steering and fault-tolerant method for a wave-adaptive unmanned surface vessel (USV) is provided. This method is applied to an USV comprising a main hull, multiple sets of floating body suspension assemblies, and a propulsion system. Each floating body suspension assembly includes a floating body and an electromagnetic suspension connected to the main hull, comprising:

[0031] The system collects environmental perception information and its own attitude data from the unmanned surface vessel in real time, and then performs fusion processing to generate a multi-source data matrix.

[0032] Based on the multi-source data matrix, the sea state ahead is predicted. When the sea state characteristic parameters meet the preset dynamic adjustment conditions, the extension and retraction of each group of electromagnetic suspensions are adjusted within the prediction time window to form a draft gradient that adapts to changes in sea state.

[0033] Upon receiving a turning command, the unmanned surface vessel simultaneously adjusts the difference in draft between the two floating bodies on both sides and the difference in thrust between the two propulsion systems to control the turning trajectory and suppress the hull roll.

[0034] During navigation, the working status of the electromagnetic suspension and propulsion system is monitored in real time, and faults are classified. When an abnormality of the actuator is detected, a fault-tolerant reconfiguration strategy is triggered according to the fault type. The abnormal component is locked and the remaining actuators are used to reconfigure the force distribution or propulsion thrust to maintain the navigation capability of the unmanned surface vessel.

[0035] The suspension-coordinated steering and fault-tolerant method for wave-adaptive unmanned surface vessels (USVs) provided in this disclosure solves the problem of balancing navigation stability and maneuverability in complex wave environments by constructing a closed-loop control link of perception-prediction-coordination-fault tolerance. It changes the traditional passive wave response mode of equipment by using predictive draft adjustment and deep coupling of thrust / suspension to suppress the impact of wave disturbances on the main hull. At the same time, the multi-level fault-tolerant mechanism ensures that the USV can still maintain basic navigation capabilities through force reconstruction when critical execution units are damaged, greatly improving the survivability and mission reliability of the equipment in harsh conditions in the open sea.

[0036] like Figure 1 As shown, the electromagnetic suspension consists of the electromagnetic suspension body and springs ( k =5000N / m) and consists of upper and lower control arms. The upper control arm is connected to the main hull and the floating body through hinge points A and D respectively, and the lower control arm is connected to the main hull and the floating body through hinge points B and C respectively. The distance between AB and CD is equal, and the distance between AD and BC is equal. The electromagnetic suspension body is connected to the main hull and the floating body through hinge points F and E respectively. When the floating body moves up and down due to external excitation, the electromagnetic suspension body and spring are stretched and compressed, outputting control force to suppress the movement of the main hull. Among them, the component at the EDC connection is a rigid body and is rigidly connected to the floating body at its bottom, so BEF can be regarded as a triangle, such as Figure 1 In the triangle shown, BF = 301mm, BE = 356mm, FE = b, and the angle between FE and FB is... a In the equation, by the Law of Cosines, we can obtain... a and b The specific relationship, the length of FE b The adjustment can be made by a ball screw driven by a motor.

[0037] ;

[0038] When the unmanned surface vessel (USV) is launched, the initial length b of USV is... ,angle a = At this point, the initial draft of the buoy is When the length of FE is from Change to The angle is from Change to At this time, the change in the draft of the buoy. , and floating body draft They are respectively:

[0039] ;

[0040] .

[0041] Optionally, generating a multi-source data matrix includes: acquiring obstacle distance, wave characteristic parameters, hull three-dimensional attitude, suspension displacement, and navigation positioning data through a sensor array; performing Kalman filtering on the attitude data; and establishing a rasterized probability model of the operating water area based on Bayesian estimation to form an environment, state, and task data matrix.

[0042] By employing multi-source fusion and a gridded model of sensor arrays, the problems of single sensors being susceptible to clutter interference, experiencing large data fluctuations, and having detection blind zones in dynamic marine environments are addressed. A combination of Kalman filtering and Bayesian estimation logic is used to filter out spurious signals caused by sea spray, rain, and fog, providing a high-confidence environment and state matrix for subsequent control logic and ensuring the robustness of the unmanned surface vessel decision-making system under extreme weather conditions.

[0043] In some optional embodiments, the multi-source data matrix further includes an estimate of the wave propagation direction. The main propagation direction is determined by calculating the wave phase gradient, and the damping coefficient of the corresponding electromagnetic suspension is adjusted in advance based on this direction. By introducing the wave propagation direction estimate into the multi-source data matrix and using the wave phase gradient to identify the main propagation direction, the unmanned surface vessel (USV) can specifically adjust the damping characteristics of the corresponding electromagnetic suspension before the wave arrives, thus ensuring that the suspension system's response direction matches the wave's action direction. This approach helps reduce hull roll and additional impacts caused by lateral wave action, making the suspension's absorption of wave energy more effective and improving the USV's attitude stability and sailing comfort in complex sea conditions.

[0044] Optionally, the sea state is classified according to the multi-source data matrix, and corresponding reference draft and adjustment threshold are set for different sea states; when the predicted wave disturbance exceeds the preset range, the windward electromagnetic suspension is controlled to increase the draft, while the leeward draft is decreased to form a wave-resistant draft gradient.

[0045] By classifying and predicting sea states, a "preemptive intervention" wave-resistant effect was achieved. By constructing an asymmetric draft gradient, the unmanned surface vessel (USV) was positioned in the optimal stress posture before the waves arrived. The increased draft depth on the windward side provided downward pressure stability to counteract the wave lift force, effectively reducing the impact frequency and vertical acceleration of the main hull during high-speed navigation, and protecting the precision mission payloads on board from impact damage.

[0046] The sea state is dynamically classified based on a multi-source data matrix, and corresponding reference draft and adjustment thresholds are set for different sea state levels. When it is determined that a wave interference with a characteristic frequency exceeding the preset range is about to be encountered, the windward electromagnetic suspension is controlled in advance to increase the draft, while the leeward draft is reduced simultaneously to actively build an anti-wave draft gradient. When the channel width or the distance to an obstacle is detected to be less than the preset safety limit, the draft of the entire floating body is reduced and the output speed of the propulsion system is adjusted downward in coordination to reduce the impact of downwash and improve navigation safety.

[0047] For example, sea state classification and prediction divides sea states into four categories based on fused data:

[0048] Calm waters: wave height ≤ 0.5m, wave frequency ≤ 0.3Hz;

[0049] Medium wave: wave height 0.5m-1.5m, wave frequency 0.3Hz-1Hz;

[0050] Giant waves: wave height > 1.5m, wave frequency > 1Hz;

[0051] High-flow conditions: Flow velocity > 1.2 m / s.

[0052] The prediction window is dynamically adjusted according to the complexity of the sea conditions, set to 0.5-1s for calm waters and 0.8-1.2s for large waves / strong currents.

[0053] When the wave sensor detects a wave height increase of ≥50% within 3 seconds and a wave frequency exceeding 1Hz, the system determines that a giant wave is imminent. At this time, 0.8 seconds in advance, the electromagnetic suspension of the floating body on the windward side (such as the front left or rear left) is controlled to adjust the articulated section length from the initial value of 460mm to 495mm (draft increases from 150mm to 170mm); the leeward side is adjusted to 450mm (draft decreases to 140mm), forming a wave-resistant draft gradient and improving anti-roll capability.

[0054] When the millimeter-wave radar detects an obstacle at a distance of less than 5m or the channel width is less than 8m, immediately adjust the length of all float articulated sections to 450mm, reduce the draft to ≤180mm (e.g., 160mm), and simultaneously reduce the propeller speed by 20%-30% (from 1500r / min to 1000r / min) to reduce the downwash and avoid bottoming out or silt blockage.

[0055] When the flow velocity is greater than 1.2 m / s, the draft of the floating body on the downstream side is increased by 8%-12% and decreased by 3%-5% on the upstream side. The asymmetric water resistance is used to counteract the flow-induced drift and maintain the stability of the navigation trajectory.

[0056] When performing marine monitoring missions, the draft is finely adjusted based on the depth feedback from water quality sensors to ensure that the horizontal deviation of the main hull is ≤5°; when operating in wind farms, the draft distribution is adjusted according to the array spacing to improve the trajectory tracking accuracy by 10%.

[0057] In some optional embodiments, the wave energy spectrum is estimated in real time during sea state prediction, and the target natural frequency of the electromagnetic suspension is adjusted according to the peak position of the energy spectrum, so that the suspension system avoids the main wave frequency range. By estimating the wave energy spectrum in real time and identifying the peak position of the energy spectrum, the electromagnetic suspension system can actively adjust its target natural frequency according to changes in sea state, thereby avoiding the main wave frequency range. This method can effectively avoid resonance between the suspension system and wave excitation, improve the wave resistance of the suspension, and reduce the periodic load on the hull structure, which is beneficial to improving the structural safety and long-term operational stability of the unmanned surface vessel in medium and high sea states.

[0058] Optionally, the steering control adopts a multi-objective collaborative optimization strategy, including: establishing a three-degree-of-freedom dynamic prediction model for the unmanned surface vessel; constructing an optimization function that includes trajectory tracking error, roll angle constraint, and energy consumption weight; and solving the control sequence in a rolling manner when executing steering commands, and outputting the extension and retraction of the electromagnetic suspensions on both sides and the thrust difference of the propulsion system.

[0059] By employing a multi-objective collaborative optimization strategy, the nonlinear coupling conflict between trajectory accuracy, roll suppression, and energy consumption allocation during the unmanned surface vessel's (USV) turning process was resolved. Through rolling solution of the optimal control sequence, phase synchronization of the thruster differential thrust and suspension active torque on the time axis was achieved, avoiding attitude oscillations caused by independent responses of actuators in traditional control systems and ensuring smooth trajectory under complex steering maneuvers.

[0060] The three-degree-of-freedom dynamic model of the unmanned surface vessel is discretized and used as a prediction model, and a prediction step size is set. A multi-objective optimization function is constructed, which includes trajectory tracking deviation, roll angle suppression and system energy consumption weights, and international maritime collision avoidance rules are incorporated as constraints. During the execution of steering commands, the control sequence is solved in a rolling manner using an optimization algorithm, and the extension and retraction of the electromagnetic suspension on both sides of the floating body and the thrust difference of the propulsion system on both sides are output simultaneously to achieve decoupled coordination between steering maneuverability and stability.

[0061] For example, the three-degree-of-freedom nonlinear dynamic model of drift, pitch, and yaw is discretized and used as a prediction model, with a prediction step size of 5-8 steps. Specifically, to accurately characterize the auxiliary effect of electromagnetic suspension draft adjustment on steering, time-varying hydrodynamic parameters based on draft are introduced into the three-degree-of-freedom nonlinear dynamic model, and its yaw dynamic equation is expressed as:

[0062] ;

[0063] in, Let be the moment of inertia of the unmanned surface vessel about its vertical axis. Yaw acceleration, Based on the basic hydrodynamic yaw moment This is the conventional control torque generated by the thrust difference between the two propulsion systems.

[0064] In particular, The additional yaw moment caused by the difference in draft between the two buoys in this application is calculated using the following formula:

[0065] ;

[0066] In the formula, and The real-time draft of the left and right floats are respectively. h and current speed v The nonlinear fluid resistance experienced by the buoy. Since fluid resistance is positively correlated with the wetted surface area of ​​the submerged buoy, the resistance changes nonlinearly when the draft of the electromagnetic suspension changes. B The lateral distance between the centers of the left and right floating bodies (i.e., the lever arm).

[0067] This dynamic model serves as the underlying logic of the prediction model, enabling the controller to no longer solely rely on thrust difference during the solution process. Instead, it can be achieved by directly changing the draft of the left and right floating bodies. and Actively adjust the fluid resistance distribution at the bottom to generate additional torque. This provides a cooperative steering assist torque without increasing the additional load on the propulsion system.

[0068] A multi-objective optimization function is constructed, incorporating weights for trajectory tracking bias, tilt angle suppression, and system energy consumption. The prediction step size... N p Inside, define a quadratic cost function. J :

[0069] ;

[0070] In the formula, and They are respectively k The penalty weight for trajectory tracking deviation between the actual trajectory position and the reference trajectory position at any given time is set as follows: ; for k The main hull heel angle at any given time, and its suppression weight is set to... ; The control increment, which includes the thrust change rate and draft regulation rate, represents the dynamic control energy consumption of the system, and its weight is set as follows: .

[0071] Steering execution logic (taking a left turn as an example): When a left turn command is triggered at the target radius, the system acquires the initial roll angle (e.g., ...). φ 0 = 3.2°).

[0072] The two left-side buoy articulated sections were adjusted to 490mm (draft approximately 165mm), and the two right-side sections were adjusted to 455mm (draft approximately 142mm), actively generating anti-roll torque.

[0073] The thrust of the left propeller is increased from 100N to 300N, while that of the right propeller is reduced to 50N, thus creating a differential steering torque.

[0074] By optimizing the software to call the IPOPT solver and continuously optimizing the control sequence, the response time is less than 50ms.

[0075] In this way, the yaw angle can be reduced to 2.3° within 5 seconds, the turning radius can be stabilized at 50±2.5m, the trajectory tracking deviation can be reduced from 3.32m to 1.82m, and the overall energy consumption can be reduced by 12%.

[0076] In some alternative embodiments, while adjusting the extension and retraction of each set of electromagnetic suspensions, the pitch angle and heave acceleration of the main hull are simultaneously calculated, and longitudinal coupling motion is suppressed by adjusting the draft difference between the aft and forward floating body suspensions. By calculating the pitch angle and heave acceleration of the main hull while adjusting the extension and retraction of each set of electromagnetic suspensions, and compensating for this with the draft difference between the aft and forward floating body suspensions, longitudinal coupling motion of the hull can be suppressed while controlling lateral stability. This strategy enables the unmanned surface vessel (USV) to reduce the mutual amplification effect between pitch and heave during upwind or downwind navigation, thereby reducing the amplitude of hull attitude fluctuations and improving navigation stability and equipment operational reliability.

[0077] In some optional embodiments, the multi-objective cooperative optimization strategy further includes a hull roll rate constraint to limit the magnitude of roll rate variation during turning. By incorporating a hull roll rate constraint into the multi-objective cooperative optimization strategy, the rate of hull roll change can be limited while ensuring steering responsiveness. This constraint helps avoid sudden and drastic attitude changes during rapid turns, resulting in smoother control of the suspension and propulsion systems, thereby improving the stability of the unmanned surface vessel (USV) during maneuvering and the safety of its onboard equipment.

[0078] In some alternative embodiments, when adjusting the draft of the floats, the lateral stabilizing moment is calculated based on the float spacing and hull width, and the extension / retraction amount of the electromagnetic suspension is determined accordingly. By combining the float spacing and hull width in the calculation of the lateral stabilizing moment when adjusting the float draft, and determining the extension / retraction amount of the electromagnetic suspension accordingly, the suspension adjustment can be matched with the overall stability characteristics of the hull. This method can achieve reasonable draft distribution under different structural layout conditions, improve the utilization efficiency of the float support force, and enhance the unmanned surface vessel's resistance to lateral disturbances.

[0079] Optionally, fault classification is determined by monitoring the drive current, extension rate, and response delay of the actuator; when a single-side suspension power failure is detected, the remaining suspensions are used to compensate for the draft and adjust the thrust of the corresponding side propulsion system to counteract the yaw moment; when mechanical jamming is detected, the locking mechanism is triggered to lock the suspension position and the remaining suspensions are used to redistribute the draft.

[0080] Through multi-dimensional monitoring of physical characteristics, accurate fault classification and dynamic compensation are achieved. This embodiment can take differentiated remedial measures for two very different types of faults: power failure and mechanical jamming. It utilizes the performance redundancy of the remaining normal actuators to counteract the yaw or tilting torque generated at the fault point, preventing local faults from evolving into systemic loss of control and achieving the engineering goal of operating with faults.

[0081] For example, the drive current of the actuator, the extension and retraction rate of the actuator, and the feedback response delay are monitored in real time to determine whether there is a power failure or mechanical jamming. If it is determined that there is a power failure of one side of the suspension, the remaining normally functioning suspensions are controlled to perform draft compensation according to the preset redundancy gradient, and the thrust of the corresponding side propulsion system is increased simultaneously to counteract the yaw moment. If it is determined that there is mechanical jamming, while triggering the emergency locking mechanism to lock the current position, the wave energy recovery unit in the system is called to supply power, and the remaining suspensions are used to quickly reconstruct the draft distribution to maintain the balance of the hull.

[0082] For example, fault monitoring and classification:

[0083] Power failure: The motor current is 0 and the push rod does not move for more than 0.3 seconds;

[0084] Mechanical jamming: The push rod extension / retraction rate is <0.5mm / s and the draft adjustment delay is >1s;

[0085] Communication failure: Communication delay > 1 second or packet loss rate > 5%;

[0086] Propulsion system malfunction: propeller thrust output deviation > 15%.

[0087] If the right front suspension fails, the system controls the other three floats to adjust in a gradient manner: "increase the draft of the side opposite the failure (left rear) by 15%-20% and the same side (right rear) by 8%-12%", and simultaneously compensate the corresponding propeller thrust by 10%-15%. Actual measurements show that this strategy can reduce the roll angle from 6.1° to 3.0°.

[0088] Once mechanical jamming is detected, the emergency locking mechanism is immediately triggered to lock the current position within 50ms; then the wave energy recovery unit is activated to provide power, and the remaining suspensions quickly reconstruct the draft distribution within 1 second.

[0089] In some alternative embodiments, when multiple actuator failures are detected simultaneously, the unmanned surface vessel (USV) maintains stable navigation by reducing speed and redistributing suspension load. When multiple actuator failures are detected simultaneously, reducing speed and redistributing suspension load allows the remaining functional actuators to assume stability control functions, thereby maintaining the USV's basic navigation capabilities. This fault-tolerant approach maintains controllable hull attitude under complex fault conditions, preventing rapid stability deterioration due to actuator failure, and improving the safety and mission continuity of the USV system in extreme situations.

[0090] Optionally, a wave energy recovery unit is provided in the floating suspension assembly to convert the mechanical energy generated by the wave action into electrical energy; the electrical energy is preferentially supplied to the electromagnetic suspension and sensor system to realize closed-loop management of energy recovery and control power supply.

[0091] By implementing closed-loop energy management, the issues of endurance and energy efficiency during long-term unmanned surface vessels (USVs) operations have been resolved. The mechanical energy captured by the electromagnetic suspension during vibration suppression is converted into electrical energy in real time and prioritized for supply to the sensing and control systems. This not only achieves active roll reduction but also creates an energy micro-circulation, effectively extending the USV's continuous operational cycle in unresupply environments.

[0092] For example, the wave energy recovery unit uses a magnetic flux decoupling resonant structure to capture wave energy, converting the mechanical energy dissipated by the unmanned surface vessel due to motion suppression into electrical energy; the converted electrical energy is processed and preferentially supplied to the electromagnetic suspension and sensor array to realize the energy feedback cycle of roll reduction control and energy recovery, thereby extending the continuous operation capability of the unmanned surface vessel.

[0093] Optionally, in the multi-vessel formation operation mode, the status information of each unmanned vessel is exchanged through a communication network; when an unmanned vessel in the formation fails, the formation is reconstructed through a cooperative algorithm, and the task of the failed vessel is assigned to other unmanned vessels for execution.

[0094] By actively reconstructing the formation and redistributing tasks, the problem of the impact of single vessel failure on the overall mission progress was solved, ensuring that even when some nodes lose capability, the entire unmanned surface vessel cluster can still complete the predetermined sea area coverage or target search tasks through coordinated cooperation.

[0095] For example, in formation mode, the status information of each vessel is exchanged in real time through a communication network; when a vessel in the formation is found to have a malfunction and cannot maintain its predetermined position, the lead vessel reconstructs the formation through a distributed cooperative algorithm and assigns the predetermined tasks of the malfunctioning vessel to the other unmanned vessels in the formation that are sailing normally, thereby achieving cooperative obstacle avoidance and task redistribution with limited relative position errors.

[0096] For example, when ≥2 suspension faults are detected, the emergency return mode is activated, limiting the speed to ≤3m / s, and prioritizing the safety thresholds of roll angle ≤10° and pitch angle ≤8°.

[0097] When multiple boats are operating in formation, a malfunctioning boat reports its status via underwater acoustic communication or 5G. The lead boat uses a distributed MPC algorithm to reconstruct the formation (e.g., from a "diamond" to a "line") and assigns tasks to other boats, maintaining a relative position error of ≤1 boat length to ensure the integrity of the formation.

[0098] Optionally, when generating the multi-source data matrix, the observation quality of each sensor in the sensor array is evaluated, and its confidence weight is dynamically adjusted.

[0099] By dynamically adjusting the confidence weights of sensors, the technical challenge of sensor performance degradation caused by drastic environmental changes has been solved. In special scenarios such as rain, fog, strong light, or electromagnetic interference, the system can automatically select the most reliable data source for decision-making, preventing misleading information from entering the control loop and improving the stability of the multi-source data matrix under complex backgrounds.

[0100] The system evaluates the observation quality of each sensor in the sensor array under the current sea conditions in real time and dynamically adjusts the confidence weight of each sensor. When the environmental noise exceeds the preset electromagnetic compatibility threshold or the visibility is lower than the preset range, the system automatically increases the weight ratio of the inertial attitude sensor and the wave sensor, and simultaneously decreases the weight ratio of the environmental perception sensor to ensure the robustness of the multi-source data matrix.

[0101] Optionally, the encounter frequency and amplitude parameters of the waves can be extracted through frequency domain analysis; when the encounter frequency is close to the hull's natural frequency or the predicted wave amplitude exceeds a preset threshold, the draft adjustment control is triggered, and the damping coefficient or stiffness coefficient of the electromagnetic suspension is adjusted.

[0102] By using frequency domain analysis and active adjustment of mechanical parameters, the resonance safety hazard of unmanned surface vessels (USVs) at specific wave frequencies has been resolved. By dynamically changing the stiffness and damping of the electromagnetic suspension, the hull's physical parameters can avoid the excitation frequency of the current wave in real time, thus mitigating the risk of structural fatigue or capsizing caused by resonance.

[0103] For example, the encounter frequency and amplitude characteristics of waves are extracted through frequency domain analysis; when it is determined that the encounter frequency enters the resonant frequency range of the unmanned surface vessel hull, or when it is predicted that the wave amplitude will cause the main hull to crash, the predictive adaptive draft adjustment logic is triggered, which changes the damping coefficient and stiffness coefficient of the electromagnetic suspension to enable the unmanned surface vessel to avoid the natural frequency and improve vertical stability.

[0104] In some alternative embodiments, vibration signals of the main hull structure are monitored during navigation. When increased structural vibration is detected, vibration transmission is reduced by adjusting the damping coefficient of the electromagnetic suspension. By monitoring the vibration signals of the main hull structure during navigation and adjusting the damping coefficient of the electromagnetic suspension when vibration intensifies, vibration transmission within the hull structure can be reduced. This method can reduce the accumulation of structural vibration caused by wave impact or maneuvering, which is beneficial for protecting the hull structure and onboard equipment, and improving the structural safety and reliability of the unmanned surface vessel during long-term operation.

[0105] Optionally, actuator constraints can be introduced when solving the control sequence, including electromagnetic suspension travel limits, actuation rate limits, and propulsion system power limits. By filtering out illegal commands such as exceeding the range or power limits in advance when calculating control quantities, the electromagnetic suspension or propulsion system can be prevented from being in an overloaded or fatigued state for extended periods, thus reducing the maintenance costs of the equipment throughout its lifecycle.

[0106] For example, when solving the control sequence, the maximum travel limit of the electromagnetic suspension, the actuation rate limit, and the power saturation constraint of the propulsion system are introduced; by setting a soft constraint penalty function, the electromagnetic suspension is prevented from frequently being in the extreme displacement state, and the mechanical fatigue life of the actuator is extended while ensuring the trajectory tracking accuracy.

[0107] In some embodiments, during the turning process, an anti-rolling moment is generated by increasing the draft of the inner float and decreasing the draft of the outer float; at the same time, the thrust difference between the two propulsion systems is adjusted.

[0108] By actively constructing an inward tilting attitude, the problem of centrifugal force cancellation during high-speed turns of the unmanned surface vessel (USV) was solved. By coordinating the anti-rolling moment generated by the difference in draft on both sides, the USV is allowed to maneuver at a greater turning angular velocity while maintaining the main hull nearly horizontal, significantly improving its dynamic response speed in tactical evasion or emergency obstacle avoidance scenarios.

[0109] Before entering the steady-state steering phase, the inner float suspension generates active downforce, increasing the depth of the inner float, while the outer float generates an upward compensating force, creating an anti-roll moment opposite to the steering centrifugal force. Specifically, the anti-roll moment... Provided by the asymmetric buoyancy difference generated by the change in draft between the two floating bodies, its calculation model is expressed as:

[0110] ;

[0111] In the formula, The density of seawater, g It is the acceleration due to gravity. The area of ​​the waterline on one side of the floating body. The increased draft required to turn the inner float. The reduced draft required to turn to the outer float (usually a negative value to match the inner increment). B The distance between the centers of the left and right groups of floats is the lateral distance between them.

[0112] Furthermore, the control system keeps the thrust difference adjustment curve and the draft difference adjustment curve strictly phase synchronized on the time axis to eliminate the response delay between the thruster yaw moment and the suspension recovery moment.

[0113] In some optional embodiments, during steering control, the thrust distribution ratio of the propulsion system is dynamically adjusted based on the hydrodynamic differences experienced by the two floating bodies. This dynamic adjustment ensures that the propulsion system output aligns with the hydrodynamic changes generated by suspension adjustment. This method reduces hydrodynamic asymmetry caused by draft changes, lowers the additional yaw moment generated during steering, makes the unmanned surface vessel's steering trajectory smoother and more stable, and improves heading control accuracy.

[0114] In some alternative embodiments, during steering control, the actuation speed of the electromagnetic suspension is adjusted according to the rate of change of heading to match the suspension response with the steering dynamics. Adjusting the actuation speed of the electromagnetic suspension according to the rate of change of heading during steering control ensures that the dynamic response of the suspension system is consistent with the steering action of the propulsion system. This approach reduces attitude fluctuations caused by suspension response lag, coordinating the suspension adjustment process with the steering maneuver, thereby improving the stability and control precision of the unmanned surface vessel during steering.

[0115] In some embodiments, when a power system failure is detected that causes the system power to be limited, the system enters a low-power operation mode and adjusts the power supply priority of each unit.

[0116] In emergency situations where power is limited, a priority management strategy ensured the continuity of core functions. When the total battery or total power was insufficient to support the operation of the entire system, priority was given to ensuring the power supply to the navigation control and communication units, achieving optimal resource allocation and creating an opportunity for rescue or a low-speed return to base.

[0117] When a partial failure of the power system is detected, resulting in a limitation of total power, the system enters a low-power survival mode. At this time, the power supply priority of non-critical sensing payloads is reduced, the backup power stored in the wave energy recovery unit is concentrated to supply the critical navigation units, and the maximum speed is limited and the turning slope is reduced to ensure that the unmanned surface vessel can maintain basic track keeping capability until returning to port.

[0118] In some embodiments, after the fault is cleared, the control output of the suspension and propulsion system is gradually restored by setting a transition time window until the system returns to normal control mode.

[0119] By setting a smooth transition time window, the problem of sudden jumps in system state after fault elimination is solved, avoiding mechanical shocks or power grid fluctuations caused by instantaneous high-power actions during the reset process of the execution unit, and ensuring a smooth transition when the system switches from fault-tolerant mode back to normal mode.

[0120] Specifically, after the fault condition is cleared, the system does not immediately switch back to the standard control mode. Instead, it gradually reduces the fault reconstruction supply by setting a smooth transition time window and simultaneously observes the response consistency of each suspension and thruster. Only when the feedback data of each actuator recovers to the allowable range of coordination deviation does the system resume the fully autonomous cooperative operation mode.

[0121] In some embodiments, the electromagnetic suspension has variable stiffness and variable damping control functions, and dynamically adjusts its mechanical parameters according to the dominant wave frequency.

[0122] Seamlessly switching between hard support and soft energy absorption modes based on the main sea state frequency ensures both load-bearing capacity in large waves and shock absorption comfort in small waves, thus enhancing the adaptability of the unmanned surface vessel to different sea environments.

[0123] The wave energy spectrum characteristics in the multi-source data matrix are analyzed in real time, and the damping coefficient and stiffness coefficient of the electromagnetic suspension are dynamically adjusted according to the main wave frequency. When the wave frequency is close to the natural frequency of the hull, frequency avoidance control is achieved by actively changing the suspension stiffness, and the wave impact energy is absorbed by increasing the damping ratio, realizing the transformation from single position adjustment to adaptive mechanical characteristic adjustment.

[0124] In some alternative embodiments, a virtual center of mass offset of the main hull is constructed by adjusting the support reactions of each float suspension to compensate for additional torques caused by load changes or flow field disturbances. This resolves attitude deviations caused by load offsets or continuous unilateral flow field disturbances. Hull balance can be achieved without physical counterweight adjustments, giving the unmanned surface vessel (USV) more flexible mission configuration capabilities and effectively addressing special operational requirements such as asymmetric loading.

[0125] For example, monitor the load distribution changes on the main hull and external dynamic load disturbances, and calculate the support reaction moment of each floating body suspension combination on the main hull; when performing steering or anti-wave actions, construct a virtual center of mass offset by asymmetrically adjusting the output of each electromagnetic suspension, under the premise that the physical center of gravity remains unchanged, in order to counteract the additional steering torque caused by eccentric loads or non-uniform flow fields.

[0126] In some optional embodiments, the prediction model parameters are corrected online by comparing predicted sea state data with actual attitude response data. Closed-loop verification between predicted data and measured feedback enables online evolution of the control model. By correcting hydrodynamic parameters, the model inaccuracy caused by water temperature, salinity, or biofouling on the hull is resolved, ensuring that the algorithm maintains extremely high prediction accuracy throughout the entire voyage.

[0127] For example, the predicted sea state data within the prediction time window is compared in real time with the attitude response data actually experienced by the unmanned surface vessel to calculate the prediction deviation value; the fluid drag coefficient and lift coefficient constraints in the prediction model are dynamically corrected using an online parameter identification algorithm, and the fusion weight of environmental perception information in the multi-source data matrix is ​​adjusted simultaneously to realize the self-learning adaptation of the control logic to different sea salinity, density and flow velocity environments.

[0128] In some alternative embodiments, a mapping relationship between the buoy's draft and navigation resistance is established, and the propulsion system thrust is simultaneously compensated when adjusting the draft difference. By establishing a coupled compensation model for draft and resistance, the technical challenge of the side effects of attitude adjustment on speed is solved. By simultaneously fine-tuning the thrust, the additional wave-making drag caused by the increased draft is offset, achieving complete decoupling between speed and navigation attitude, and ensuring a constant navigation rhythm during wave-fighting maneuvers.

[0129] For example, a mapping model of the relationship between the draft of the float and the navigation resistance is established; while adjusting the difference in draft between the two floats to suppress the roll, the additional wave-making resistance and frictional resistance caused by the increase in draft are calculated, and the thrust increment of the corresponding side is compensated by a preset proportion to eliminate the instantaneous yaw rate fluctuation caused by the suspension adjustment and maintain the smoothness of the steering trajectory.

[0130] In some optional embodiments, health parameters are calculated based on the operating load and temperature rise of the electromagnetic suspension, and the suspension actuation weights are adjusted according to the health status during the control process. This health evaluation system achieves a leap from "passive fault tolerance" to "active protection." By prioritizing load tasks to high-health execution units, the deterioration rate of fatigued units is slowed, and the flexible adjustment of software algorithms extends the continuous operational life of the entire vessel.

[0131] For example, the mechanical fatigue load and motor temperature rise status of each electromagnetic suspension group are evaluated in real time to generate a health factor; when the health of a certain suspension group is determined to be lower than the preset threshold, the actuation penalty weight of that suspension is increased in the multi-objective optimization function, and the load task is redistributed to the suspension with high health, thereby extending the overall task reliability of the system by actively degrading the performance.

[0132] In some alternative embodiments, when sea state parameters exceed a preset safe operating range, the electromagnetic suspension is controlled to enter a high-damping state, and the propulsion system maintains the unmanned surface vessel's attitude stability. Through the coordination of maximum damping locking and dynamic positioning, the unmanned surface vessel can maintain itself within the safe operating envelope with minimal energy consumption, preventing structural damage under the impact of large waves.

[0133] When the sea state characteristic parameters in the multi-source data matrix exceed the system's safe operating envelope, a ship-wide lock-up command is triggered; all electromagnetic suspensions are switched to maximum damping, and each float is adjusted to its maximum draft to increase the system's rotational inertia. At the same time, the propulsion system is switched to dynamic positioning mode, utilizing the physical energy dissipation characteristics of the suspension and the fine-tuning of thrust to maintain the unmanned surface vessel's survivability in polar sea conditions.

[0134] In some optional embodiments, the actuation history data of the electromagnetic suspension is recorded during navigation, and a suspension actuation frequency model is established based on the historical data to reduce repetitive high-frequency actuations. By recording the actuation history data of the electromagnetic suspension and establishing the suspension actuation frequency model, high-frequency repetitive actuations of the suspension system during navigation can be identified and appropriately suppressed during control. This method can reduce ineffective and frequent movements of the suspension actuators, reduce mechanical fatigue and energy consumption, thereby extending the service life of the electromagnetic suspension and improving the overall operational reliability of the system.

[0135] Example 1: Predictive Active Wave Resistance Scenario under Complex Sea Conditions

[0136] While conducting a long-range meteorological monitoring mission in a certain sea area, the unmanned surface vessel (USV) was traveling at its rated speed. The front-end sensor array scanned the waters ahead in real time using millimeter-wave radar and wave sensors, generating environmental perception information which was then processed and incorporated into a multi-source data matrix. When the system extracted wave characteristics through frequency domain analysis and determined that the frequency of the wave approaching the hull's resonance range and the wave amplitude exceeding a safe threshold, the control core immediately issued a command within the predicted time window before reaching the wave crest. At this point, the electromagnetic suspension system no longer passively absorbed shocks but actively adjusted the displacement of the four buoys according to sea conditions: controlling the draft of the electromagnetic suspension on the windward side to "dive" into the water and increase downward stability, while simultaneously reducing the travel of the suspension on the leeward side to counteract the tilting moment. This actively constructed anti-wave draft gradient, combined with the real-time switching of the high-damping state of the electromagnetic suspension, enabled the USV to maintain the horizontal attitude of the main hull even under the impact of large waves, effectively protecting the high-precision sensors on top from severe pitching.

[0137] Example 2: Suspension-coordinated steering scenario during high-maneuverability obstacle avoidance

[0138] When the unmanned surface vessel (USV) was conducting autonomous inspections in a near-shore waterway dense with obstacles, it received an emergency large-angle left turn command to avoid sudden obstacles. The system immediately invoked a multi-objective collaborative optimization strategy, breaking the old logic of independent thrust and suspension adjustment. Simultaneously, as the left propeller power surged and the right propeller power decreased, generating a steering torque, the system calculated the mapping relationship between draft and drag, instructing the two sets of electromagnetic suspensions on the left to actively retract, increasing the draft of the inner floating body and using the resulting water resistance to assist in steering; at the same time, the right suspension extended to support the main hull. Through this phase-synchronized adjustment of thrust and draft differences, the USV constructed an "inward lean" posture, similar to a motorcycle leaning into a corner, offsetting the centrifugal force generated by the turn and keeping the heel angle within an extremely low range. Furthermore, the propulsion system simultaneously compensated for the additional wave-making drag caused by the increased draft, ensuring stable speed throughout the turn and significantly reducing the turning radius.

[0139] Example 3: Fault-tolerant reconfiguration scenario under sudden actuator failure

[0140] During prolonged cruising, the electromagnetic suspension on the right rear of the unmanned surface vessel (USV) experienced a jamming failure due to mechanical fatigue or foreign object intrusion. The onboard integrated control unit, through real-time monitoring of drive current and response delay, quickly identified this state as an "execution unit anomaly" and accurately classified it as mechanical jamming. The system immediately triggered a fault-tolerant reconfiguration strategy, activating an emergency locking mechanism to lock the faulty suspension in its current position, preventing secondary damage. Subsequently, the control logic dynamically redistributed the support load based on the health parameters of the remaining three normal suspensions: by adjusting the extension gradient of the remaining suspensions and fine-tuning the output reference of the propulsion systems on both sides, a virtual centroid offset was constructed within the controller, thereby compensating for the asymmetric forces generated by the jamming point. The USV did not lose control and yaw due to the localized "leg breakage," but instead maintained basic balance by utilizing the redundancy of the remaining execution units, ensuring the closed-loop execution of the mission.

[0141] Example 4: Extreme Survival Modes and Closed-Loop Energy Management Scenarios

[0142] In extreme situations involving severe sea conditions on the verge of a sudden typhoon and insufficient battery power, the unmanned surface vessel (USV) automatically switches to survival mode. The system comprehensively reduces the power supply priority of non-critical payloads, entering a low-power operation state. At this time, the wave energy recovery unit integrated within the suspension begins to operate efficiently, converting the mechanical energy generated by the hull's movement with the waves into electrical energy, which is then rectified and stored in the energy management system, prioritizing the operation of the sensing matrix and communication modules. All electromagnetic suspensions switch to maximum damping to dissipate wave impact energy, while the propulsion system maintains necessary dynamic positioning with minimal power. Utilizing the physical energy dissipation characteristics of the suspension and the synergy of minimal thrust, without consuming the core power of the main battery, the USV ensures that it is not capsized by the giant waves through attitude fine-tuning while "going with the current," waiting for the severe weather to end.

[0143] The foregoing description and accompanying drawings fully illustrate embodiments of the present disclosure to enable those skilled in the art to practice them. Other embodiments may include structural and other changes. The embodiments represent only possible variations. Individual components and functions are optional unless explicitly required, and the order of operation may vary. Parts and features of some embodiments may be included or substituted for parts and features of other embodiments. Embodiments of the present disclosure are not limited to the structures described above and shown in the accompanying drawings, and various modifications and changes may be made without departing from its scope. The scope of the present disclosure is limited only by the appended claims, and the foregoing embodiments should be considered exemplary and non-limiting.

Claims

1. A suspension-coordinated steering and fault-tolerant method for a wave-adaptive unmanned surface vessel (USV), applied to a USV comprising a main hull, multiple sets of floating suspension assemblies, and a propulsion system, wherein each set of floating suspension assemblies comprises a floating body and an electromagnetic suspension connected to the main hull, characterized in that, include: The system collects environmental perception information and its own attitude data from the unmanned surface vessel in real time, and then performs fusion processing to generate a multi-source data matrix. Based on the multi-source data matrix, the sea conditions ahead are predicted. When the sea condition characteristic parameters meet the preset dynamic adjustment conditions, the extension and retraction of each group of electromagnetic suspensions are adjusted within the prediction time window to form a draft gradient that adapts to changes in sea conditions. Upon receiving a turning command, the unmanned surface vessel simultaneously adjusts the difference in draft between the two floating bodies on both sides and the difference in thrust between the two propulsion systems to control the turning trajectory and suppress the hull roll. During navigation, the working status of the electromagnetic suspension and propulsion system is monitored in real time, and faults are classified. When an abnormality of the execution unit is detected, a fault-tolerant reconfiguration strategy is triggered according to the fault type. The abnormal execution unit is locked and the remaining execution units are used to reconfigure the force distribution or propulsion thrust in order to maintain the navigation capability of the unmanned surface vessel.

2. The method according to claim 1, characterized in that: The generation of the multi-source data matrix includes: The sensor array is used to acquire obstacle distance, wave characteristic parameters, three-dimensional attitude of the hull, suspension displacement and navigation positioning data. Kalman filtering is applied to the attitude data, and a gridded probability model of the operating water area is established based on Bayesian estimation, forming an environment, state, and task data matrix.

3. The method according to claim 1, characterized in that: The sea state is classified according to the multi-source data matrix, and corresponding reference draft and adjustment threshold are set for different sea states. When the predicted wave disturbance exceeds the preset range, the electromagnetic suspension on the windward side is controlled to increase the draft, while the draft on the leeward side is decreased, in order to form a wave-resistant draft gradient.

4. The method according to claim 1, characterized in that: The steering control employs a multi-objective collaborative optimization strategy, including: Establish a three-degree-of-freedom dynamic prediction model for unmanned surface vessels; Construct an optimization function that includes trajectory tracking error, tilt angle constraint, and energy consumption weight; When executing steering commands, the control sequence is solved by rolling and the extension and retraction of the electromagnetic suspension on both sides and the thrust difference of the propulsion system are output.

5. The method according to claim 1, characterized in that: The fault classification is determined by monitoring the drive current, extension rate, and response delay of the execution unit; When a single-sided electromagnetic suspension power failure is detected, the remaining electromagnetic suspensions are used to compensate for the water level and adjust the thrust of the corresponding side propulsion system to counteract the yaw moment. When mechanical jamming is detected, the locking mechanism is triggered to lock the electromagnetic suspension position and the remaining electromagnetic suspension is used to redistribute the draft.

6. The method according to claim 1, characterized in that: The floating suspension assembly is equipped with a wave energy recovery unit to convert the mechanical energy generated by wave action into electrical energy; Electricity is prioritized for supplying the electromagnetic suspension and sensor system to achieve closed-loop management of energy recovery and control power supply.

7. The method according to claim 1, characterized in that: In multi-vessel formation operation mode, the status information of each unmanned vessel is exchanged through a communication network; When a malfunction occurs in one of the unmanned surface vessels (USVs) in the formation, the formation is reconstructed through a collaborative algorithm, and the tasks of the malfunctioning USV are assigned to other USVs.

8. The method according to claim 2, characterized in that: When generating the multi-source data matrix, the observation quality of each sensor in the sensor array is evaluated, and its confidence weight is dynamically adjusted.

9. The method according to claim 1, characterized in that: The encounter frequency and amplitude parameters of the waves are extracted through frequency domain analysis; When the encounter frequency is close to the hull's natural frequency or the predicted wave amplitude exceeds a preset threshold, the draft adjustment control is triggered, and the damping coefficient or stiffness coefficient of the electromagnetic suspension is adjusted.

10. The method according to claim 4, characterized in that: During the turning process, an anti-rolling moment is generated by increasing the draft of the inner float and decreasing the draft of the outer float; at the same time, the thrust difference between the two propulsion systems is adjusted.

Images