Wave self-adaptive unmanned ship, self-adaptive control system and control method

Abstract

The invention relates to the technical field of intelligent unmanned ships, in particular to a wave self-adaptive unmanned ship, a self-adaptive control system and a control method.The wave self-adaptive unmanned ship comprises a main ship body, a front arch, a rear arch, a suspension system, a propelling nacelle, a rigid sliding plate and a buoy, the main ship body is connected with the rear arch and the front arch, and the rigid sliding plate is fixed to the buoy; the propulsion nacelle is installed at the tail of the rigid sliding plate, the rear arch and the rigid sliding plate are installed through two bearings perpendicular to each other, the suspension system comprises a rocker arm, a plate spring and a TPMLM actuator, the rocker arm is connected to the rigid sliding plate, the plate spring and the TPMLM actuator are connected between the rocker arm and the rigid sliding plate and support the rocker arm to be in an upward inclined posture, and the rear arch is connected to the rigid sliding plate. The front arch is connected with the rocker arm, the plate spring and the TPMLM actuator are installed in parallel in the vertical direction, the TPMLM actuator is used for outputting active control force for restraining movement of the ship body, and the unmanned ship, the self-adaptive control system and the self-adaptive control method have the advantages that wave loads can be actively counteracted, and therefore movement restraining is achieved.

Description

Technical Field

[0001] This invention relates to the field of intelligent unmanned surface vessel technology, and in particular to a wave-adaptive unmanned surface vessel, an adaptive control system, and a control method. Background Technology

[0002] Traditional wave-adaptive vessels (WACs) often employ passive spring-damped suspension systems or hydraulically interconnected passive suspension systems (such as the American WAM-V and the Australian Nauti-Craft) to improve seakeeping. While these systems are relatively simple in structure, they have significant drawbacks: 1. Poor adaptability and low control precision: The damping and stiffness parameters of the passive suspension are fixed and cannot be adjusted according to real-time changes in sea conditions. When the sea conditions exceed the design conditions, the vibration reduction effect drops sharply, leading to increased heave, pitching, and rolling motions of the hull, which seriously affects operational safety and crew comfort.

[0003] For example, in known literature, the inventors disclosed a wave-adaptive emergency rescue boat in Chinese Patent Publication No. CN108116635A, published on June 5, 2018. The rescue boat specifically includes: a main hull; a front arch and a rear arch supporting the main hull; a float assembly; a rigid sliding plate; a suspension system; and a propulsion nacelle. The rigid sliding plate is mounted on the float assembly, and the rigid sliding plate is connected to the front arch through the suspension system. The suspension system includes a rocker arm, leaf springs, and gas springs. The rocker arm is supported by leaf springs and gas springs arranged in parallel and movably connected between the sliding plate and the rocker arm. The front arch is movably supported on the rocker arm. The suspension system passively stores and dissipates the energy of the hull during wave motion through the leaf springs and gas springs, thereby enabling the hull to adapt to wave motion. In this suspension system, the damping and stiffness parameters of the leaf springs and gas springs are fixed and cannot be adjusted according to real-time changes in sea conditions. Therefore, the vibration reduction flexibility is poor and it cannot be applied to complex sea conditions.

[0004] 2. Limited adjustment capability of semi-active system: Some improvement schemes use semi-active dampers such as magnetorheological and electrorheological dampers, which have limited damping force adjustment range. When faced with large amplitude and low frequency wave excitation, the control force is insufficient and the response is lag.

[0005] Third, active control systems suffer from high energy consumption and low integration: While existing fully active control systems (such as electro-hydraulic servo actuators) offer good control performance, they consume enormous amounts of energy and are complex and bulky, making them difficult to widely apply on ships where space and energy are limited. Furthermore, existing control strategies are mostly based on single algorithms, which cannot maintain optimal performance across all sea states.

[0006] Therefore, at present, how to establish a simplified, low-energy adaptive system that enables unmanned vessels to actively and in real time adapt to sea conditions in waves has become an urgent problem to be solved. In addition, how to improve the existing adaptive control algorithm so that the wave adaptation capability of unmanned vessels can maintain optimal performance in the entire sea state range has become another urgent problem to be solved. Summary of the Invention

[0007] The technical problem to be solved by the present invention is to overcome the shortcomings of the prior art and provide a wave-adaptive unmanned surface vessel, an adaptive control system and a control method. The unmanned surface vessel, adaptive control system and method have the advantage of being able to actively counteract wave loads, thereby achieving motion suppression.

[0008] To achieve the above objectives, the technical solution adopted by the present invention is as follows: This application first provides a wave-adaptive unmanned surface vessel (USV), including a main hull, a forward arch, a rear arch, a suspension system, a propulsion nacelle, a rigid sliding plate, and floats. The main hull is fixedly connected to the rear arch, and the main hull and the forward arch are connected in the longitudinal direction by a ball joint mechanism. The floats extend along the longitudinal direction of the USV, the rigid sliding plate is fixed to the floats, the propulsion nacelle is installed at the tail of the rigid sliding plate, and the rear arch and the rigid sliding plate are mounted by two mutually perpendicular bearings. The suspension system includes a rocker arm, a leaf spring, and a TPMLM actuator. The rocker arm is connected to the rigid sliding plate by a transverse pin. The leaf spring and the TPMLM actuator are respectively connected between the rocker arm and the rigid sliding plate by transverse pins and support the rocker arm in an upward tilting posture. The forward arch and the rocker arm are connected in the vertical direction by a ball joint mechanism. The leaf spring and the TPMLM actuator are installed in parallel in the vertical direction. The TPMLM actuator is used to output an active control force to suppress the motion of the hull.

[0009] To further optimize the above plan, the following measures were also taken: As one of the preferred embodiments, the TPMLM actuator includes a primary sleeve, a primary coil, a damper shaft, a secondary shaft, and a permanent magnet. The primary sleeve is connected to a rigid slide plate via a pin. The primary coil is installed inside the primary sleeve. The permanent magnet and the secondary shaft are sleeved inside the primary coil and are fixedly connected. The damper shaft is fixedly connected to the secondary shaft, and one end of the damper shaft extends out of the primary sleeve and is connected to the rocker arm via a pin.

[0010] As a preferred embodiment, the suspension system further includes a base plate detachably connected to a rigid slide plate, a rocker arm connected to the base plate via a transverse pin, and a leaf spring and a TPMLM actuator connected between the rocker arm and the base plate via transverse pins, respectively.

[0011] As one preferred embodiment, the rocker arm includes a rectangular bottom frame and triangular connecting frames located on both sides of the bottom frame. The bottom frame is tilted upwards, and a connecting sleeve is fixed between the apex corners of the two connecting frames. The connecting sleeve is connected to the TPMLM actuator via a pin.

[0012] This application also provides an adaptive control system for the wave-adaptive unmanned surface vessel based on the above, including a sensor network, a sea state recognition module, a multi-mode switching controller, a time delay compensator, and a power driver. The sensor network includes inertial measurement units installed on the main hull and the two side pontoons, and linear grating rulers installed on each TPMLM actuator. The inertial measurement units are used to measure the heave acceleration, pitch angle and roll angle of the hull, and the linear grating rulers are used to measure the relative displacement and velocity between the main hull and the rigid sliding plate. The sea state identification module receives the raw motion data from the inertial measurement unit, performs multi-scale decomposition using wavelet packet transform, extracts the energy characteristics of the motion signal in different frequency bands, and estimates the effective wave height Hs and peak period Tp of the current sea state in real time. The multi-mode switching controller has a pre-stored sea state classification and control strategy mapping table. It classifies the current sea state according to the sea state parameters estimated by the received sea state identification module, and calls the corresponding control algorithm according to the corresponding control strategy to calculate the target control force command. The time delay compensator receives the target control force command output by the multi-mode switching controller, performs forward time delay correction on it, and then sends it to the power driver. The power driver drives the TPMLM actuator to generate an active control force to suppress the ship's movement.

[0013] This application also provides a control method for the adaptive control system of the wave-adaptive unmanned surface vessel, comprising the following steps: S1. Collect motion signals of the unmanned vessel in real time through a sensor network; S2. Analyze the motion signal based on wavelet packet transform to identify the current sea state parameters in real time; S3. Classify the sea state according to the identified sea state parameters, and calculate the target control force command by adopting the corresponding control algorithm according to the pre-stored sea state classification and control strategy mapping table. S4. Use a state prediction algorithm to compensate for time delays in the target control commands; S5 outputs the control signal of the target control command after time delay correction to the TPMLM actuator, driving it to output the target active control force to suppress the hull motion.

[0014] Furthermore, the sea state classification and control strategy mapping table classifies sea states and executes corresponding control strategies in the following manner: (1) When Hs < 1.5m and Tp < 6s, it is judged to be a calm sea state. At this time, the LQG control algorithm is adopted, and the state weight matrix in the LQR controller is designed. Q and control weight matrix R This causes the controller to tend to calculate smaller control forces and allows for larger fluctuations in the system state. (2) When 1.5m≤Hs≤4m and 6s≤Tp≤8s, it is judged to be a medium sea state. At this time, the LQG-PI composite control algorithm is adopted, in which the LQG control algorithm is used as the outer loop control algorithm. The state weight matrix in the LQR controller is designed. Q and control weight matrix R The PI algorithm, as the inner-loop control algorithm, is used to make the controller tend to maintain the stability of the system state and allow the calculation of a large control force. It is used to make the actual active control force output by the TPMLM actuator follow the ideal control force calculated by the controller in real time. (3) When Hs>4m and Tp>8s, it is judged to be a severe sea state. At this time, the robust PI control algorithm is adopted. By increasing the proportional gain and integral gain parameters, the controller has sufficient sensitivity and calculates sufficient control force to maintain the stability of the system state.

[0015] Furthermore, under calm sea conditions, the LQG control algorithm specifically includes the following steps: Step a: Establish the system state-space equations: Linearize the unmanned vessel suspension system near the equilibrium point to obtain the following state-space model: in, x It is a state vector, determined by the heave displacement / velocity of the main hull, the heave displacement / velocity of the pontoons, and the pitch angle / angular velocity of the pontoons. u It is the control input vector, determined by the control force generated by the TPMLM actuator; w and v These are process noise and measurement noise, respectively, both of which are assumed to be Gaussian white noise with a mean of 0. y It is the measurement output vector, determined by the acceleration measured by the inertial measurement unit and the relative displacement measured by the grating ruler; A, B, C, and G are system matrices, which can be calculated based on the state-space model equations; Step b, Design the LQR controller: (1) Define the quadratic cost function J As shown below: in: Q It is the state weight matrix. R It is the control weight matrix; (2) Design a smaller state weight matrix Q and a large control weight matrix R This causes the controller to tend to calculate smaller control forces and allows for larger fluctuations in the system state. (3) The optimal state feedback gain matrix is ​​obtained by solving the Riccati equation of the controller. K , making J The minimum optimal control law is: u = - Kx Step c, Design the state observer: State estimation equation As shown below: in: This is the state estimation vector; L It is the Kalman gain matrix, obtained by solving the observer Riccati equation, which minimizes the covariance of the state estimation error; Step d: Combining the LQR controller and the state observer, the LQG control law is determined as follows: Based on this, the state estimation vector is used. To calculate the output control force.

[0016] Furthermore, under moderate sea states, the LQG algorithm specifically includes increasing the state weight matrix. Q The value of the control weight matrix is ​​reduced. R The value of makes the system tend to maintain a stable state and allows the system to calculate a larger control force for the output.

[0017] Furthermore, under severe sea conditions, the robust PI control algorithm is based on the following control law: in: e(t) It is an error signal, that is, the difference between the actual output value of the system and the set value; K p It is proportional gain; K i It is integral gain; u(t) It controls the input quantity; In rough sea conditions, K p and K i Perform parameter tuning to improve K p andK i Gain parameters are used to suppress violent hull movements by increasing the system's sensitivity to errors and to eliminate hull attitude deviations by integrating and accumulating errors to output control force.

[0018] Because of the above-described solutions, one or more technical solutions provided in this application embodiment have at least the following technical effects or advantages: 1. High control precision and significant vibration reduction effect: Using TPMLM as the fully active actuator, it has a large output force and fast response. It can actively apply the driving force that is perfectly matched with the control target, which can reduce the vertical acceleration of the hull by more than 45% and significantly reduce the roll and pitch angles.

[0019] 2. Full sea state adaptive: Through real-time sea state identification and multi-mode control strategy switching, the system can automatically adopt the optimal control algorithm regardless of whether the sea state is calm, moderate or severe, thus solving the problem of poor adaptability of a single control strategy.

[0020] 3. Compact structure and optimized energy consumption: TPMLM directly outputs linear force, eliminating the need for complex transmission conversion mechanisms required by traditional active systems such as hydraulic and rotary motors. Combined with leaf springs bearing static loads and LQG's energy-saving mode in low sea states, the overall energy consumption of the system is effectively controlled.

[0021] 4. Fast response speed and good dynamic performance: By introducing a time delay compensation mechanism based on Kalman prediction, the system control lag is compressed to less than 20ms, which greatly improves the ability to suppress high-frequency wave excitation. Attached Figure Description

[0022] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings of the embodiments will be briefly described below. Obviously, the drawings described below only involve some embodiments of this application and should not be construed as limiting this application.

[0023] Figure 1 This is a schematic diagram of the overall structure of the unmanned vessel in this embodiment; Figure 2 This is a schematic diagram of the suspension system in this embodiment; Figure 3 This is a schematic diagram of the TPMLM actuator in this embodiment; Figure 4 This is a schematic diagram of the control principle of the adaptive control system in this embodiment; Figure 5 This is a logic flowchart of the adaptive control method in this embodiment.

[0024] Figure label: 1-1. Main hull; 1-2. Forward arch; 1-3. Aft arch; 1-4. Suspension system; 1-5. Propulsion nacelle; 1-6. Rigid sliding plate; 1-7. Float; 2-1. Rocker arm; 2-2. Leaf spring; 2-3. TPMLM actuator; 2-4. Bottom plate; 3-1. Primary sleeve; 3-2. Primary coil; 3-3. Vibration damper shaft; 3-4. Secondary shaft; 3-5. Permanent magnet; 4-1. First inertial measurement unit; 4-2. Second inertial measurement unit; 4-3. Linear grating ruler; 4-4. Central processing unit. Detailed Implementation

[0025] To enable those skilled in the art to better understand the technical solutions of this invention, the technical solutions of the embodiments of this invention will be clearly and completely described below with reference to the accompanying drawings, so as to more clearly understand the purpose, features and advantages of this invention. It should be understood that the embodiments shown in the drawings are not intended to limit the scope of this invention, but are only for illustrating the essential spirit of the technical solutions of this invention. Obviously, the described embodiments are only a part of the embodiments of this invention, and not all of them. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this invention.

[0026] Unless the context requires otherwise, throughout the specification and claims, the word “comprising” and its variations, such as “including” and “having”, shall be understood to have an open, inclusive meaning, that is, to be interpreted as “including, but not limited to”.

[0027] Throughout this specification, references to "an embodiment" or "an embodiment" indicate that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one embodiment. Therefore, the appearance of "in an embodiment" or "an embodiment" in various places throughout the specification does not necessarily refer to the same embodiment. Furthermore, a particular feature, structure, or characteristic may be combined in any manner in one or more embodiments.

[0028] The singular forms “a” and “the” used in this specification and the appended claims include plural references unless otherwise expressly stated herein. It should be noted that the term “or” is generally used to mean “and / or” unless otherwise expressly stated herein.

[0029] In the following description, in order to clearly demonstrate the structure and working method of the present invention, a number of directional terms will be used. However, terms such as "front", "back", "left", "right", "outside", "inside", "outward", "inward", "up", and "down" should be understood as convenient terms and not as limiting terms.

[0030] The implementation details of the embodiments of the present invention will be described in detail below with reference to the accompanying drawings. The following content is only for the convenience of understanding the implementation details and is not necessary for implementing this solution.

[0031] The embodiments in this specification provide a wave-adaptive unmanned surface vessel, an adaptive control system, and an adaptive control method. These aim to solve the problems of existing adaptive vessels where the damping and stiffness parameters of the suspension system cannot be adjusted in real time according to sea conditions, and the adaptive capability cannot maintain optimal performance across all sea states. This not only enables the suspension system to actively and adaptively adjust according to sea conditions in real time, enhancing seakeeping, but also achieves high-precision, adaptive, and active suppression of heave, pitch, and roll motions across all sea states, maintaining optimal performance. Furthermore, the system is compact, consumes little energy, and its applicability is further expanded.

[0032] refer to Figures 1 to 2 As shown, in this embodiment, the wave-adaptive unmanned surface vessel includes a main hull 1-1, a front arch 1-2, a rear arch 1-3, a suspension system 1-4, a propulsion nacelle 1-5, a rigid sliding plate 1-6, and a float 1-7. The main hull 1-1 is fixedly connected to the rear arch 1-3, and the main hull 1-1 and the front arch 1-2 are connected in the longitudinal direction by a ball joint mechanism. In one case, the ball joint mechanism includes a ball joint rod mounted on the main hull and a ball socket mounted on the front arch 1-2. (Refer to...) Figure 2 As shown, it illustrates the case where the ball joint is set on the rocker arm 2-1. The ball joint and the ball joint are connected as one unit along the longitudinal axis of the hull, thereby allowing a certain angle of rotational movement between the main hull 1-1 and the front arch 1-2. In this way, the main hull 1-1 and the front arch 1-2 can absorb and buffer at least some wind and wave loads and / or external stresses through the movement of the connection point, avoiding the generation of excessive structural internal forces. The pontoon 1-7 extends along the longitudinal axis of the unmanned vessel. The rigid sliding plate 1-6 is fixed to the pontoon 1-7. The propulsion nacelle 1-5 is installed at the stern of the rigid sliding plate 1-6. The aft arch 1-3 and the rigid sliding plate 1-6 are connected by two mutually perpendicular bearings. One bearing allows the aft arch 1-3 to rotate at a certain angle relative to the pontoon 1-7 in the horizontal plane, and the other bearing allows the aft arch 1-3 to rotate at a certain angle relative to the pontoon 1-7 in the longitudinal plane. The specific connection structure between the aft arch 1-3 and the rigid sliding plate can be referred to in the prior application CN108116635A, and will not be elaborated here. In this way, the aft arch 1-3 and the rigid sliding plate 1-6 are allowed to have two degrees of freedom: rotation in the longitudinal plane along the length of the vessel and rotation in the horizontal plane.

[0033] Here, for reference Figure 2As shown, the suspension system 1-4 is an active suspension system, specifically including a rocker arm 2-1, a leaf spring 2-2, and a TPMLM actuator 2-3. The rocker arm 2-1 is connected to the rigid slide plate 1-6 via a transverse pin and can rotate along the longitudinal plane of the ship's length. The leaf spring 2-2 and the TPMLM actuator 2-3 are respectively connected between the rocker arm 2-1 and the rigid slide plate 1-6 via transverse pins, supporting the rocker arm 2-1 in an upward tilting posture. The front arch 1-2 is connected to the rocker arm 2-1 in the vertical direction using a ball joint mechanism. The leaf spring 2-2 and the TPMLM actuator 2-3 are installed in parallel in the vertical direction. The TPMLM actuator 2-3 is used to output an active control force to suppress the ship's motion. The main function of the leaf spring 2-2 is to bear the static load of the ship and provide basic positioning stiffness for the system, thereby significantly reducing the static power consumption of the TPMLM. On the other hand, when waves excite the buoy, the leaf spring 2-2 buffers the wave impact and stores some of the wave's kinetic energy. When the buoy 1-7 is excited by waves, the front arch 1-2 and the rigid slide plate 1-6 compress the rocker arm 2-1, the rocker arm 2-1 rotates around the pin, and compresses the leaf spring 2-2 and the TPMLM actuator 2-3.

[0034] refer to Figure 3 As shown, in this embodiment, the TPMLM actuator 2-3 is a cylindrical permanent magnet linear motor, specifically including a primary sleeve 3-1, a primary coil 3-2, a damper shaft 3-3, a secondary shaft 3-4, and a permanent magnet 3-5. The primary sleeve 3-1 is connected to the rigid slide plate 1-6 by a pin. The primary coil 3-2 is installed inside the primary sleeve 3-1. The permanent magnet 3-5 and the secondary shaft 3-4 are fitted inside the primary coil 3-2 and are fixedly connected. The damper shaft 3-3 is fixedly connected to the secondary shaft 3-4, and one end of the damper shaft 3-3 extends out of the primary sleeve 3-1 and is connected to the rocker arm 2-1 by a pin.

[0035] When waves act on the pontoon 1-7, causing it to move vertically relative to the main hull 1-1, the central processing unit 4-4 of the control system calculates the required optimal control force, i.e., the target active control force, based on sensor information. The primary coil 3-2 of the TPMLM actuator 2-3 is driven by the control system, generating a magnetic field. This controls the permanent magnet 3-5 to generate an electromagnetic control force of a specified magnitude and direction, which is equivalent to significantly changing the magnetic damping value in the traditional passive suspension. This suppresses the movement of the main hull 1-1, thereby improving the vertical movement of the main hull 1-1 and enhancing the ride comfort of the main hull 1-1.

[0036] In this embodiment, the suspension system 1-4 further includes a base plate 2-4, which is detachably connected to the rigid slide plate 1-6. The rocker arm 2-1 is connected to the base plate 2-4 via a transverse pin. The leaf spring 2-2 and the TPMLM actuator 2-3 are respectively connected between the rocker arm 2-1 and the base plate 2-4 via transverse pins. This allows the suspension system 1-4 to be installed and replaced independently, facilitating subsequent maintenance.

[0037] In this embodiment, the rocker arm 2-1 includes a rectangular bottom frame and triangular connecting frames located on both wings of the bottom frame. The bottom frame is tilted upwards. A connecting sleeve is fixed between the apex corners of the two connecting frames. The connecting sleeve is connected to the TPMLM actuator 2-3 by a pin. The bottom frame has a first end and a second end along the longitudinal axis. The first end is connected to the base plate 2-4 by a pin, and the other end is connected to the front arch 1-2 support leg by a ball joint mechanism.

[0038] refer to Figure 4 As shown, this embodiment also provides an adaptive control system based on the aforementioned wave-adaptive unmanned surface vessel, specifically including a sensor network, a sea state recognition module, a multi-mode switching controller, a time delay compensator, and a power driver. The sensor network includes inertial measurement units (IMUs) installed on the main hull 1-1 and the two side floats 1-7, and linear grating rulers 4-3 installed on each TPMLM actuator 2-3. The IMUs specifically include a first IMU 4-1 installed on the main hull 1-1 and a second IMU 4-2 installed on the floats 1-7. The IMUs are used to measure the heave acceleration, pitch angle, and roll angle of the hull. The linear grating ruler 4-3 is used to measure the relative displacement and velocity between the main hull 1-1 and the rigid sliding plate 1-6. The sea state identification module receives the raw motion data from the inertial measurement unit, performs multi-scale decomposition using db4 wavelet packet transform, extracts the energy characteristics of the motion signal in different frequency bands, and estimates the effective wave height Hs and peak period Tp of the current sea state in real time. The multi-mode switching controller has a pre-stored sea state classification and control strategy mapping table, and the logic table is shown in Table 1 below. The current sea state is classified according to the sea state parameters estimated by the sea state identification module, and the corresponding control algorithm is called according to the corresponding control strategy to calculate the target control force command.

[0039] Table 1 Here, the multi-mode switching controller automatically switches between three control modes based on the identified sea state parameters. The control logic is as follows: Figure 5 As shown: 1. Calm sea state mode Hs<1.5m, Tp<6s: Linear quadratic Gaussian LQG control is adopted, with weighting coefficients biased towards energy optimization, minimizing actuator energy consumption while ensuring basic comfort.

[0040] 2. Medium sea state mode 1.5m ≤ Hs ≤ 4m, 6s≤Tp≤8s: LQG-PI composite control is adopted, in which the outer loop LQG controller is responsible for global optimization, and the inner loop PI controller is used to quickly track the ideal control force output by LQG, so as to achieve a balance between accuracy and energy consumption.

[0041] 3. Severe sea state mode Hs>4m, Tp>8s: Switch to robust PI control, using larger proportional and integral gains to prioritize system stability and maximum motion suppression capability.

[0042] The time-delay compensator receives the target control force command output by the multi-mode switching controller, performs forward-looking time-delay correction on it, and then sends it to the power driver. Specifically, it uses a Kalman filter-based state observer to predict the system's motion state for the next 20ms, using the predicted value as the controller input to compensate for the time delay throughout the entire process from sensor sampling and controller calculation to TPMLM force output, ensuring that the control force is synchronized with the wave excitation phase. The power driver drives the TPMLM actuators 2-3 to generate an active control force to suppress the ship's motion.

[0043] Here, the sea state identification module, multi-mode switching controller, time delay compensator, and power driver can be integrated into the central processing unit 4-4, such as... Figure 4 As shown, this makes the control system tend to be more compact and smaller.

[0044] Combination Figure 4 The control system operates as follows: Motion signals collected by the first inertial measurement unit 4-1, the second inertial measurement unit 4-2, and the linear encoder 4-3 are sent to the central processing unit 4-4. The sea state identification module within the processor analyzes the signals and identifies the current sea state level. The multi-mode switching controller calls the corresponding control algorithm LQG, LQG-PI, or PI control algorithm based on the sea state level to calculate the target control force command. This command is then forward-looking corrected by the time-delay compensator and sent to the power driver, ultimately driving the TPMLM actuator 2-3 to generate precise active control force.

[0045] Combined with appendix Figure 5The control mode switching logic is as follows: After the system starts, it continuously identifies the sea state. If Hs < 1.5m and Tp < 6s, it enters the calm sea state mode. At this time, the LQG control algorithm is used to calculate and output the target control force command. If 1.5m ≤ Hs ≤ 4m and 6s ≤ Tp ≤ 8s, it enters the moderate sea state mode. At this time, the LQG-PI control algorithm is used to calculate and output the target control force command. If Hs > 4m or Tp > 8s, it immediately switches to the severe sea state mode. The robust PI control algorithm is used to calculate and output the target control force command to ensure absolute stability under severe sea conditions.

[0046] The design of this control system follows the principle of "hierarchical decision-making and zoned optimization". Its core is to dynamically switch between three control algorithms based on the sea state identification results to adapt to different performance requirements.

[0047] Based on this, this embodiment also proposes a control method for the aforementioned adaptive control system of wave-adaptive unmanned surface vessel, which specifically includes the following steps: S1. Collect motion signals of the unmanned vessel in real time through a sensor network; S2. Analyze the motion signal based on wavelet packet transform to identify the current sea state parameters in real time; S3. Classify the sea state according to the identified sea state parameters, and calculate the target control force command by adopting the corresponding control algorithm according to the pre-stored sea state classification and control strategy mapping table. S4. Use a state prediction algorithm to compensate for time delays in the target control commands; S5 outputs the control signal of the target control command after time delay correction to the TPMLM actuator 2-3, driving it to output the target active control force to suppress the hull movement.

[0048] In this embodiment, the sea state classification and control strategy mapping table classifies sea states and executes corresponding control strategies in the following manner: 1. When Hs < 1.5m and Tp < 6s, the sea state is considered calm. In this case, the LQG control algorithm is adopted, and the state weight matrix in the LQR controller is designed. Q and control weight matrix R This causes the controller to tend to calculate smaller control forces and allows for larger fluctuations in the system state. 2. When 1.5m ≤ Hs ≤ 4m and 6s ≤ Tp ≤ 8s, it is judged to be a moderate sea state. In this case, the LQG-PI composite control algorithm is adopted, in which the LQG control algorithm is used as the outer loop control algorithm. The state weight matrix in the LQR controller is designed. Q and control weight matrix RThe controller tends to maintain a stable system state and allows for the calculation of a larger control force. The PI algorithm, as the inner-loop control algorithm, is used to make the actual active control force output by the TPMLM actuator 2-3 follow the ideal control force calculated by the controller in real time. 3. When Hs>4m and Tp>8s, it is judged to be a severe sea state. At this time, a robust PI control algorithm is adopted. By increasing the proportional gain and integral gain parameters, the controller has sufficient sensitivity and calculates sufficient control force to maintain the stability of the system.

[0049] Specifically, the LQG control algorithm assumes that the system is linear, the performance index is a quadratic form of the state, and that Gaussian white noise exists in the system. The LQG controller is a combination of the linear quadratic regulator of the LQR controller and the state observer of the Kalman filter.

[0050] The LQR controller is responsible for calculating the optimal control force to minimize performance metrics in a noise-free environment. The Kalman filter is responsible for optimally estimating all system states, such as displacement and velocity, from noisy sensor measurements. In calm sea conditions, the LQG control algorithm, which prioritizes energy optimization, includes the following steps: Step a: Establish the system state-space equations: Linearize the unmanned vessel suspension system 1-4 near the equilibrium point to obtain the following state-space model: in, x It is a state vector, determined by the heave displacement / velocity of the main hull, the heave displacement / velocity of the pontoons, and the pitch angle / angular velocity of the pontoons. u It is the control input vector, determined by the active control force generated by TPMLM actuator 2-3; w and v These are process noise and measurement noise, respectively, both of which are assumed to be Gaussian white noise with a mean of 0. y It is the measurement output vector, determined by the acceleration measured by the inertial measurement unit and the relative displacement measured by the grating ruler; A, B, C, and G are system matrices, which can be calculated based on the state-space model equations; Step b, Design the LQR controller: 1. Define the quadratic cost function J As shown below, the goal is to find the optimal control law. u This results in the cumulative cost from the initial state to the final state. J Minimum.

[0051] in: Q It is the state weight matrix. R It is the control weight matrix; 2. Design a smaller state weight matrix Q and a large control weight matrix R The controller is made to tend to calculate smaller control forces and allow larger fluctuations in the system state, that is, energy saving is the primary goal, but a certain degree of fluctuation in the system state is allowed. 3. By solving the Riccati equations of the controller, the optimal state feedback gain matrix is ​​obtained. K , making J The minimum optimal control law is: u = - Kx Step c, Design the state observer: State estimation equation As shown below: in: The state estimation vector output by the state observer; L It is the Kalman gain matrix, obtained by solving the observer Riccati equation, which minimizes the covariance of the state estimation error; Step d: Combining the LQR controller and the state observer, the LQG control law is determined as follows: Based on this, the state estimation vector is used. To calculate and output the optimal control force.

[0052] Specifically, the medium sea state mode aims to balance the optimization capabilities of LQG with the robust tracking capabilities of PI. It employs a cascade control structure: The outer control loop and the main control loop both employ an LQG controller, which uses the overall system state as input to calculate an "ideal control force". u desired This force is calculated based on the global optimization objective.

[0053] The inner control loop uses a PI controller, which is responsible for quickly and accurately tracking the ideal control force command given by the outer loop LQG. Its error signal is... e force (t) = u desired (t) - u actual (t)， in u actual (t) The actual control force output by the system at time t.

[0054] The derivation process is as follows: The LQG control algorithm includes designing the LQG controller and adjusting the weight matrix. Q andR The steps for obtaining the value are specifically to improve the value under moderate sea state conditions. Q Value, reducing the value under moderate sea state R The value, which makes the system more inclined to maintain a stable state compared to calm sea states, allows the system to increase its output control force to achieve a balance between stability and energy consumption. It outputs the ideal control force of the system at time t. u desired (t) .

[0055] The control law of the PI controller is: K P_force This is the proportional gain of the inner-loop PI controller; K i_force This is the integral gain of the inner-loop PI controller; The function of this PI controller is to control the actual output force of the TPMLM. u actual Follow closely u desired It can compensate for the dynamic delay and nonlinearity of the TPMLM actuator itself, ensuring that the ideal control strategy can be faithfully executed.

[0056] Specifically, the robust PI control algorithm is based on the following control law: in: e(t) It is an error signal, that is, the difference between the actual output value of the system and the set value; K p It is proportional gain; K i It is integral gain; u(t) It controls the input quantity; In rough sea conditions, K p and K i Parameter tuning improves performance compared to other sea states. K p and K i Gain parameters are used to suppress violent hull movements by increasing the system's sensitivity to errors and to eliminate hull attitude deviations by accumulating errors through integration and outputting control forces. This "high-gain" PI control strategy, although energy-intensive, provides strong restoring power, prioritizing the stability of the ship under extreme conditions.

[0057] Specifically, this embodiment employs the following time-delay compensation mechanism: A pre-installed time-delay compensation module is included in all control modes. The principle is to use a Kalman filter as a state predictor. The Kalman filter not only provides the current optimal estimate of the state but also predicts the state over a future period, such as 20ms, based on the current model. The controller, whether LQG or PI, no longer uses the current estimated state but instead uses this predicted state to calculate the control force. Thus, when the control force, after calculation and execution delays, actually acts on the system, it is perfectly synchronized with the system's future state, thereby offsetting the effect of the time delay.

[0058] In summary, in this embodiment, the active levitation system 1-4 based on a cylindrical permanent magnet linear motor (TPMLM) integrates the TMLM as the sole actuator between the main hull 1-1 and the rigid sliding plate 1-6. By precisely controlling the electromagnetic force output by the TMLM, wave loads are actively counteracted, thereby achieving motion suppression. Through the application of an adaptive control system and its control methods, high-precision, adaptive, and active suppression of heave, pitch, and roll motions can be achieved for the unmanned vessel across all sea states.

[0059] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or basic characteristics. Therefore, the embodiments should be considered illustrative and non-limiting in all respects. The scope of the invention is defined by the appended claims rather than the foregoing description. Therefore, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the present invention, and no reference numerals in the claims should be construed as limiting the scope of the claims.

[0060] Furthermore, it should be understood that although this specification describes embodiments, not every embodiment contains only one independent technical solution. This narrative style is merely for clarity. Those skilled in the art should consider the specification as a whole, and the technical solutions in each embodiment can also be appropriately combined to form other embodiments that can be understood by those skilled in the art.

Claims

1. A wave adaptive unmanned ship, characterized in that, The unmanned ship comprises a main hull (1-1), a front arch (1-2), a rear arch (1-3), a suspension system (1-4), a propulsion nacelle (1-5), a rigid slide plate (1-6) and a float (1-7); The main hull (1-1) is fixedly connected with the rear arch (1-3), the main hull (1-1) and the front arch (1-2) are connected through a spherical hinge mechanism in the longitudinal direction, the float (1-7) extends along the longitudinal direction of the unmanned ship, the rigid slide plate (1-6) is fixed on the float (1-7), the propulsion nacelle (1-5) is installed at the tail of the rigid slide plate (1-6), and the rear arch (1-3) and the rigid slide plate (1-6) are installed through two mutually perpendicular bearings. The suspension system (1-4) comprises a rocker arm (2-1), a plate spring (2-2) and a TPMLM actuator (2-3), the rocker arm (2-1) is connected to the rigid slide plate (1-6) through a transverse pin shaft, the plate spring (2-2) and the TPMLM actuator (2-3) are connected between the rocker arm (2-1) and the rigid slide plate (1-6) through transverse pin shafts and support the rocker arm (2-1) in an upwardly inclined posture, the front arch (1-2) and the rocker arm (2-1) are connected through a spherical hinge mechanism in the vertical direction, the plate spring (2-2) and the TPMLM actuator (2-3) are installed in parallel in the vertical direction, and the TPMLM actuator (2-3) is used for outputting a main control force for suppressing the movement of the ship body.

2. The wave adaptive unmanned ship in accordance with claim 1, characterized in that, The TPMLM actuator (2-3) comprises a primary sleeve (3-1), a primary coil (3-2), a damper shaft (3-3), a secondary shaft (3-4) and a permanent magnet (3-5), the primary sleeve (3-1) is connected to the rigid slide plate (1-6) through a pin shaft, the primary coil (3-2) is installed in the primary sleeve (3-1), the permanent magnet (3-5) and the secondary shaft (3-4) are sleeved in the primary coil (3-2) and are fixedly connected, the damper shaft (3-3) is fixedly connected with the secondary shaft (3-4), and one end of the damper shaft (3-3) extends out of the primary sleeve (3-1) and is connected to the rocker arm (2-1) through a pin shaft.

3. The wave adaptive unmanned ship in accordance with claim 1, characterized in that, The suspension system (1-4) further comprises a bottom plate (2-4), the bottom plate (2-4) is detachably connected to the rigid slide plate (1-6), the rocker arm (2-1) is connected to the bottom plate (2-4) through a transverse pin shaft, and the plate spring (2-2) and the TPMLM actuator (2-3) are connected between the rocker arm (2-1) and the bottom plate (2-4) through transverse pin shafts.

4. The wave adaptive unmanned ship in accordance with claim 1, wherein, The rocker arm (2-1) comprises a rectangular bottom frame and triangular connecting frames located at two wings of the bottom frame, the bottom frame is in an upwardly inclined posture, a connecting sleeve rod is fixedly connected between top corners of the two connecting frames, and the connecting sleeve rod is connected to the TPMLM actuator (2-3) through a pin shaft.

5. An adaptive control system for a wave adaptive unmanned ship based on the wave adaptive unmanned ship of claim 1, characterized by, The unmanned ship further comprises a sensor network, a sea state identification module, a multi-mode switching controller, a time lag compensator and a power driver. The sensor network comprises an inertial measurement unit installed on the main hull (1-1) and the two side pontoons (1-7) for measuring the heave acceleration, pitch angle and roll angle of the hull, and a linear grating ruler (4-3) installed on each TPMLM actuator (2-3) for measuring the relative displacement and velocity between the main hull (1-1) and the rigid slide plate (1-6); The sea state identification module receives the original motion data of the inertial measurement unit, performs multi-scale decomposition by wavelet packet transform, extracts the energy features of the motion signal in different frequency bands, and estimates the significant wave height Hs and peak period Tp of the current sea state in real time; The multi-mode switching controller pre-stores a sea state classification and control strategy mapping table, classifies the current sea state according to the sea state parameters estimated by the sea state identification module, and calculates the target control force command according to the corresponding control strategy; The time delay compensator receives the target control force command output by the multi-mode switching controller and sends it to the power driver after performing forward-looking time delay correction; The power driver drives the TPMLM actuator (2-3) to generate active control force to suppress hull motion.

6. A control method of an adaptive control system of a wave adaptive unmanned ship according to claim 5, characterized by, The method comprises the following steps: S1, real-time acquisition of motion signals of the unmanned ship through a sensor network; S2, analysis of the motion signals based on wavelet packet transform to identify the current sea state parameters in real time; S3, classification of the sea state according to the identified sea state parameters, and calculation of the target control force command according to the pre-stored sea state classification and control strategy mapping table; S4, time delay compensation of the target control command using a state prediction algorithm; S5, output of the control signal of the target control command after time delay correction to the TPMLM actuator (2-3) to drive it to output the target active control force to suppress hull motion.

7. The control method of the adaptive control system according to claim 6, characterized by, The sea state classification and control strategy mapping table classifies the sea state and executes the corresponding control strategy according to the following method: (1) When Hs < 1.5m and Tp < 6s, it is judged to be a calm sea state. At this time, the LQG control algorithm is adopted, and the state weight matrix in the LQR controller is designed. Q and control weight matrix R This causes the controller to tend to calculate smaller control forces and allows for larger fluctuations in the system state. (2) When 1.5m≤Hs≤4m and 6s≤Tp≤8s, it is judged as moderate sea state, at this time the LQG-PI compound control algorithm is adopted, wherein the LQG control algorithm is used as the outer loop control algorithm, and the state weight matrix Q and the control weight matrix R in the LQR controller are designed to make the controller tend to maintain the system state stable and allow the calculation of a larger control force, and the PI algorithm is used as the inner loop control algorithm to make the actual output of the TPMLM actuator (2-3) follow the ideal control force calculated by the controller in real time. (3) When Hs>4m and Tp>8s, it is judged as severe sea state, at this time, the robust PI control algorithm is adopted, by increasing the proportional gain and integral gain parameters, the controller has enough sensitivity and calculates enough control force to maintain the system state stable.

8. The control method of the adaptive control system according to claim 7, characterized by, Under calm sea conditions, the LQG control algorithm comprises the following steps: Step a, establish the system state space equation: linearize the unmanned ship suspension system (1-4) near the equilibrium point to obtain the state space model as follows: wherein, x is the state vector determined by the main hull heave displacement / velocity, buoy heave displacement / velocity, buoy pitch angle / angular velocity; u is the control input vector determined by the control force generated by the TPMLM actuator (2-3); w and v are the process noise and measurement noise, respectively, both assumed to be Gaussian white noise with mean 0; y is the measurement output vector determined by the acceleration measured by the inertial measurement unit, the relative displacement measured by the grating ruler; A, B, C, G are system matrices, which can be calculated according to the state space model equation; Step b, design the LQR controller: (1) Defining a Quadratic Cost Function J As follows: wherein: Q is a state weight matrix, R is a control weight matrix; (2) Designing a small state weight matrix Q and a large control weight matrix R causing the controller to favor computing small control forces and allowing large fluctuations in the system states; (3) The optimal state feedback gain matrix is obtained by solving the controller Riccati equation K such that J The minimum optimal control law is u = - Kx Step c, design the state observer: State estimation equation As follows: where: is the state estimation vector; L is the Kalman gain matrix, obtained by solving the observer Riccati equation, which minimizes the covariance of the state estimation error; Step d, combine the LQR controller and the state observer to determine the LQG control law as: Accordingly, the output control force is calculated using the state estimation vector ​ 9. The control method of the adaptive control system according to claim 7, characterized by, In moderate sea conditions, the LQG algorithm specifically includes increasing the values of the state weight matrix Q and decreasing the values of the control weight matrix R , causing the system to tend to maintain state stability and allowing the system to calculate a larger control effort for the output.

10. The control method of the adaptive control system according to claim 7, characterized by, Under severe sea conditions, the robust PI control algorithm is based on the following control law: Where: e(t) is the error signal, i.e. the difference between the actual output value of the system and the set value; K p is a proportional gain; K i is the integral gain; u(t) is a control input quantity; In rough sea conditions, the control system K p and K i is tuned to improve K p and K i gain parameters, suppressing the ship's violent motion by increasing the system's sensitivity to the error and eliminating the ship's attitude deviation by integrating the error output to control the force.

Images