Method for switching safe dynamic positioning of unmanned ship based on attack detection-compensation mechanism

Abstract

The invention discloses a switching unmanned ship safe dynamic positioning method based on an attack detection-compensation mechanism, and the method comprises the steps: building an initial unmanned ship model, constructing a switching LPV networked unmanned ship system through a switching LPV modeling method based on the number of shipborne equipment, eliminating the frequent switching problem of an existing switching modeling method, and reducing the network burden; constructing an event triggering mechanism of the ship body-shore-based channel; establishing a spoofing attack detection observer of a hull-shore-based channel, and designing a first detection standard; acquiring an estimated controller mode, establishing a safety output and a safety controller mode according to the first detection standard and the estimated controller mode, and designing a safety controller based on the safety output; an event triggering mechanism of a shore-ship channel is established, and network resources are saved; and a second detection standard is designed, a control input compensator is constructed, the asynchronous switching behavior duration caused by tampering quantity parameter vectors (namely switching signals) due to spoofing attacks is reduced, the influence of the spoofing attacks on the output of the safety controller is resisted, and finally unmanned ship dynamic positioning is realized.

Description

Technical Field

[0001] This invention relates to the field of unmanned vessel cooperative dynamic positioning technology, and in particular to a method for switching unmanned vessel safe dynamic positioning based on an attack detection-compensation mechanism. Background Technology

[0002] In recent years, the demand for unmanned surface vessels (USVs) has surged due to the need to avoid casualties and improve mission efficiency. Compared to manned ocean-going vessels, USVs are more likely to enhance flexibility in practical applications. They are small, stealthy, highly maneuverable, low-cost, and have low energy consumption, making them widely applicable in fields such as hydrological surveying, waterway exploration, and modern warfare.

[0003] Due to the openness of networks, data transmitted through the network in network switching systems is vulnerable to malicious tampering by deception attacks. Deception attacks involve attackers injecting false data into the communication channel, modifying or replacing transmitted data, preventing the controller from receiving accurate data, and ultimately impacting system stability and performance. Compared to other types of network attacks, deception attacks are more covert, difficult to detect, prevent, and handle. Therefore, reliable attack detection schemes are crucial for establishing secure contactless communication systems. Representative detection methods in existing technologies include residual-based detection methods, Bayesian-based detection methods, and artificial intelligence-based detection methods. Besides attack detection, defense against deception attacks has also been extensively studied. However, most existing technologies do not consider the impact of deception attacks on switching signals during network transmission. As a critical characteristic of network switching systems, switching signals are inevitably susceptible to tampering during network transmission. After being attacked, the switching signal may exhibit asynchronous switching behavior due to modal mismatch between the system and the controller, affecting the controller's output and impacting system stability. Summary of the Invention

[0004] This invention provides a method for switching the dynamic positioning of unmanned vessels based on an attack detection-compensation mechanism to overcome the above-mentioned technical problems.

[0005] To achieve the above objectives, the technical solution of the present invention is as follows: A method for safe dynamic positioning of unmanned surface vessels based on an attack detection-compensation mechanism, comprising the following steps: S1. Establish an initial unmanned vessel model, and construct a switching LPV networked unmanned vessel system using a switching LPV modeling method based on the number of onboard equipment and the initial unmanned vessel model; S2. Construct an event triggering mechanism for the hull-shore channel, wherein the event triggering mechanism for the hull-shore channel is used to control the adjustment output and the transmission protocol of the quantity parameter vector from the hull to the shore system in the switching LPV networked unmanned vessel system. S3. Establish a deception attack detection observer for the ship-shore channel, and design a first detection standard based on the deception attack detection observer. The first detection standard is used to determine whether the regulation output and quantity parameter vector received by the shore-based system have been tampered with by a deception attack. S4. Obtain the estimated controller mode, establish a safety output and a safety controller mode based on the first detection standard and the estimated controller mode, and design a safety controller based on the safety output; S5. Establish an event triggering mechanism for the shore-based-ship channel, wherein the event triggering mechanism for the shore-based-ship channel is used to control the shore-based system to transmit the output of the safety controller and the transmission protocol of the safety controller mode back to the unmanned ship system. S6. Design a second detection standard, which is used to determine whether the output of the security controller and the security controller mode have been tampered with by a deception attack, and construct a control input compensator based on the second detection standard; S7. Based on the aforementioned switching LPV networked unmanned vessel system, the event triggering mechanism of the hull-shore channel, the deception attack detection observer, the first detection standard, the event triggering mechanism of the shore-hull channel, the second detection standard, and the control input compensator, the unmanned vessel dynamic positioning is achieved.

[0006] Beneficial effects: The present invention has the following improvements and effects: (1) A switching LPV networked unmanned vessel system was constructed by using a switching LPV modeling method based on the number of shipboard equipment and an initial unmanned vessel model, which eliminated the frequent switching problem of the existing switching modeling method and reduced the network burden; (2) An event triggering mechanism for the ship-shore channel and an event triggering mechanism for the shore-ship channel were constructed, saving network resources; (3) Design a deception attack detection observer and design a first detection standard to determine whether the regulation output and quantity parameter vector received by the shore-based system have been tampered with by a deception attack. Obtain the controller mode based on the deception attack detection observer, and establish a safety output and safety controller mode according to the first detection standard and the controller mode. Control the transmission process of the designed safety controller output and safety controller mode from the shore-based system to the unmanned ship system through the event triggering mechanism of the shore-based-ship channel. Design a second detection standard to determine whether the designed safety controller output has been tampered with by a deception attack, and construct a control input compensator based on the second detection standard to reduce the duration of asynchronous switching behavior caused by the tampering of the quantity parameter vector (i.e., switching signal) by the deception attack, and resist the impact of the deception attack on the output of the safety controller. Attached Figure Description

[0007] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0008] Figure 1 This is a flowchart of a method for switching the safe dynamic positioning of an unmanned vessel based on an attack detection-compensation mechanism, as described in this invention. Figure 2 This is a structural block diagram of the switching LPV networked unmanned surface vessel system in an embodiment of the present invention; Figure 3 This is a schematic diagram illustrating the change of surge mass over time in an embodiment of the present invention; Figure 4 This is a schematic diagram illustrating the changes of the mass, modal, and number parameter vectors of the unmanned vessel over time in an embodiment of the present invention. Figure 5 The above are the modal variation diagrams of the unmanned vessel based on the mass partition switching modeling technology listed in the embodiments of the present invention. Figure 6 This is a schematic diagram showing the triggering time and triggering interval of the event triggering mechanism in the hull-shore channel in an embodiment of the present invention; Figure 7 This is a graph showing the estimated controller modal values ​​and their mean square error variations in an embodiment of the present invention. Figure 8 This is a diagram showing the detection function and detection results of the deception attack in this embodiment of the invention; Figure 9 This is a diagram showing the modes of the safety controller and their asynchronous duration variations in an embodiment of the present invention; Figure 10 This is a diagram showing the controller modes and their asynchronous duration changes in an embodiment of the invention without using a mode estimator. Figure 11 This is a diagram showing the detection results based on the second detection standard in an embodiment of the present invention; Figure 12 This is a comparison diagram of control inputs with and without a control input compensator in an embodiment of the present invention; Figure 13 This is a state curve diagram of the switching LPV networked unmanned vessel system in an embodiment of the present invention. Detailed Implementation

[0009] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0010] This embodiment provides a method for switching the safe dynamic positioning of unmanned surface vessels based on an attack detection-compensation mechanism, such as... Figure 1 and Figure 2 As shown, the specific steps include: S1. Establish an initial unmanned vessel model, and construct a switching LPV networked unmanned vessel system using a switching LPV modeling method based on the number of onboard equipment and the initial unmanned vessel model; In a specific embodiment, the initial unmanned vessel model is represented as follows: (1) Where, vector , For surge speed, For swing speed, Yaw angular velocity; vector , , For position coordinates, Yaw angle; control input vector ; The disturbance signal is represented by an external system, that is: (2) in, S It is the system matrix of the external system, used to describe the dynamic characteristics of ocean wind, wave, and current disturbances, and S The real part of all eigenvalues ​​is zero; Specifically, in ship control, the external system is used to model and compensate for the influence of the external environment on ship motion, such as wind, waves, and currents. In this embodiment, to describe the continuous impact of wind, waves, and currents on the unmanned vessel, it is assumed that... S The real part of all eigenvalues ​​is zero, meaning that the external system (2) is neutral and stable; The inertia matrix is ​​represented as: , in, These are, respectively, the increase in mass due to hydrodynamics, Represents the longitudinal components of different vectors from the origin to the center of mass in a fixed reference frame. It is the moment of inertia. The total mass of the unmanned vessel, And satisfy , and These are the total mass of the unmanned vessel when it is empty and fully loaded, respectively. This represents the payload of the unmanned vessel as it changes over time. Indicates the surge mass as it changes over time; The damping matrix is ​​expressed as: , in, These are the linear damping coefficients; In the dynamic positioning process of an unmanned surface vessel, the transformation matrix used for coordinate system transformation is represented as follows: , In the dynamic positioning process of unmanned surface vessels (USVs), such as stopping and anchoring when performing tasks like scientific characterization and exploration, the yaw angle is typically used. It is small enough. Therefore, without loss of generality, we can assume... ,thereby , ,at this time , I It is an identity matrix.

[0011] In a specific embodiment, the specific steps for constructing a switching LPV networked unmanned vessel system using a switching LPV modeling method based on the number of onboard devices and an initial unmanned vessel model include: Setting up unmanned ships to carry Type of shipborne equipment, the first The unit mass and quantity of the shipborne equipment are respectively and , When the first When shipborne equipment is released or recovered, The surge may decrease or increase accordingly; simultaneously, considering the impact of the surge on the total mass of the unmanned vessel, the unit mass of the surge is set. The number of surges ; To eliminate the frequent switching problem caused by using quality partition-based switching modeling techniques, the changes in the total number of onboard devices on the unmanned vessel are used to reflect the changes in the unmanned vessel's modes. The unmanned vessel modes are set as follows: (3) in, This is the maximum number of devices that can be loaded onto a fully unmanned vessel. , ; Based on the aforementioned unmanned vessel modes, the total mass of the unmanned vessel is... Represented as: (4) in, ; Based on equation (4), the matrix in equation (1) and Reconstructed as: , in, These are the unmanned surface vessel's empty weight. The corresponding hydrodynamic increase in mass, The unmanned surface vessel's empty mass The corresponding inertia matrix, These are the unmanned surface vessel's empty weight. The corresponding linear damping coefficient; These are the shipboard equipment mass The corresponding hydrodynamic increase in mass, Is the quality of shipboard equipment The corresponding inertia matrix, These are the shipboard equipment mass The corresponding linear damping coefficient; Based on the reconstructed matrix above, and the initial unmanned vessel model (1) and external system (2), a switching LPV networked unmanned vessel system is constructed, which is represented as: (5) , in, and All are predefined matrices; It is the measurement output. .

[0012] Specifically, when the sum of the number of onboard devices of the unmanned vessel meets the following conditions... At that time, unmanned ship mode At this point, the output adjustment of the LPV networked unmanned vessel system may be unsolvable, meaning that the state of the unmanned vessel model may diverge due to factors such as thruster failure. When the sum of the number of unmanned shipboard devices from Change to At that time, unmanned ship mode Depend on Become At this point, the change in the sum of the number of onboard devices of the unmanned vessel is taken as the mode switching point of the unmanned vessel; at the change in the sum of the number of onboard devices of the unmanned vessel, the state of the unmanned vessel model may show a divergent / convergent trend, that is, the unstable / stable mode switching point. Unmanned ship mode exist Number of switches on Equal to stable subsequence Number of switching and unstable subsequences Number of switching The sum of Unmanned ship mode Corresponding switching sequence .

[0013] Specifically, this embodiment proposes a switching LPV model establishment method based on the number of onboard equipment of the unmanned vessel. This method can adapt to the characteristic of a large number of onboard devices on the unmanned vessel. It reflects the change in the total mass of the unmanned vessel through the change in the number of onboard devices, and constructs a switching LPV networked unmanned vessel system. This eliminates the frequent switching problem of the mass partition switching modeling method and reduces the conservatism of the fixed mass value switching modeling method, which requires a large number of preset masses. Compared with the mass partition switching modeling method, due to the setting of equation (3), the change in total mass caused by the two factors of surge and the unmanned vessel simultaneously releasing and recovering the same number of onboard devices will not change the unmanned vessel mode. Specifically, assuming time... unmanned ship mode , No. Type of shipborne equipment and the first The quality of the shipborne equipment is , Their respective quantities are , ,in When the unmanned boat releases one While recovering a ship-borne equipment When it comes to ship-borne equipment, reduce At the same time add At this point, due to equation (3), Still However, under the quality partition switching modeling method, if at time... The total mass of the unmanned vessel and So, releasing one at the same time on the unmanned ship Ship-borne equipment and recycling one Ship-borne equipment or surge quality Even after the increase, Unmanned ship mode It will still change to Compared to the fixed-mass-value switching modeling method, the switching LPV modeling method proposed in this embodiment only requires knowledge of the quantity of each type of shipboard equipment. If a fixed mass value switching modeling method is used to describe the total mass of the unmanned vessel during the release / recovery of shipboard equipment... If there are changes, then it is necessary to pre-plan them. Quality value.

[0014] It should also be noted that the modeling method based on the number of unmanned ship onboard equipment can also reduce the network load, since equation (3) establishes the unmanned ship mode. and quantity parameter vector The relationship between the unmanned ships requires transmission. To shore-based systems. However, in traditional discussions of switched LPV systems, modal... and quantity parameter vector All of them need to be transmitted through network channels.

[0015] S2. Construct an event triggering mechanism for the hull-shore channel, wherein the event triggering mechanism for the hull-shore channel is used to control the adjustment output and the transmission protocol of the quantity parameter vector from the hull to the shore system in the switching LPV networked unmanned vessel system. To reduce network load and shorten asynchronous duration, an event-triggered mechanism for the ship-shore channel is constructed, represented as follows: (6) in (7) (8) In the formula, , For output error, and Triggering time The next Each sampling time is , The weight matrix is ​​unknown, and , Given a threshold, and ; Specifically, error condition (7) is used to ensure that when the triggering time is reached... Subsequent sampling time Output With trigger time Output When the error between them exceeds the given conditions, the hull-shore event triggering mechanism (6) can update the data in a timely manner and transmit the data from the hull to the shore-based system; condition (8) is used to ensure that the onboard equipment of the unmanned vessel is released, recovered, or the unmanned vessel mode is activated. After the change occurs, the event triggering mechanism (6) of the hull-shore channel can update the data in a timely manner and transmit the data from the hull to the shore system. S3. Establish a deception attack detection observer for the ship-shore channel, and design a first detection standard based on the deception attack detection observer. The first detection standard is used to determine whether the regulation output and quantity parameter vector received by the shore-based system have been tampered with by a deception attack. S4. Obtain the estimated controller mode, establish a safety output and a safety controller mode based on the first detection standard and the estimated controller mode, and design a safety controller based on the safety output; In a specific embodiment, the specific steps for establishing a deception attack detection observer for the ship-shore channel include: The network latency during data transmission via the ship-to-shore channel is set as follows: Therefore, adjust the output. and quantity parameter vector exist The system can switch from a networked LPV unmanned surface vessel system to a shore-based system at any time. in, satisfy , , ; Specifically, during transmission, the adjustment output and quantity parameter vectors are susceptible to spoofing attacks. When a spoofing attack occurs, the adjustment output is set... Deceived and attacked Altered to: (9) Among them, due to energy limitations, , It is a given matrix; Set quantity parameter vector Deceived and attacked Altered to: (10) The deception attack detection observer is set based on the tampered adjustment output and quantity parameter vector received by the shore-based system, and is represented as follows: (11a) (11b) in, It is to switch the state observation values ​​of the LPV networked unmanned surface vessel system. It outputs the observed values. The gain matrix to be determined is... For the estimated controller modes, To adjust the output residual signal, and .

[0016] In a specific embodiment, the specific steps for designing the first detection standard based on the deception attack detection observer include: The deception attack detection observer acquires the residual signal of the adjustment output in the switched LPV networked unmanned surface vessel system. , To Obtained by variable substitution; Selecting a time window using time window chi-square detection technique And based on the residual signal The detection function is set as follows: , The first detection criterion is set based on the detection function, expressed as: (12) in, The detection threshold, and . Specifically, when the detection function At that time, the test results This indicates that the output and quantity parameter vectors have been tampered with by a deception attack; if the detection result... This indicates that the output and quantity parameter vectors have not been tampered with by a deception attack; based on the detection criteria, the first... The activation interval of a deception attack The dormancy interval is ,in It is the first one detected The duration of a deception attack, and .

[0017] In a specific embodiment, the specific steps for obtaining the estimated controller mode and establishing the safety output and safety controller mode based on the first detection criterion and the estimated controller mode include: Introducing virtual input vectors , , ; In a specific embodiment, the virtual input vector elements in The estimation process includes: use Reconstruct equation (11a) in the deception attack detection observer as follows: (13) in, , , , , in, , It is a positive constant; Based on equation (13) for virtual input vector elements in The estimation process includes the following steps: An auxiliary observer is established, represented as: (14) in, It is the state of the auxiliary observer. It is the input vector of the auxiliary observer to be designed. , It is the dimension of the state observations; To make the auxiliary observer state Approximate the state observations as closely as possible Introducing an error, it is expressed as: , , And order: (15) At the same time, In The construction is as follows: (16) in, , , , It is to satisfy At that moment, , It is a given parameter; , , Substitute (13) and (14) into (15) And combined with (16), we get: (17) By rearranging terms in (17), we obtain the index function: (18) The virtual input vector is set based on the aforementioned index function. elements in The value can be: (19) Based on virtual input vector elements in The estimated controller mode is expressed as: (20) Based on the first detection criterion and the estimated controller mode, the safety output and safety controller mode are constructed, and are respectively expressed as: (twenty one) (twenty two) in, The detection result is a deception attack. It outputs the observed values.

[0018] Specifically, when the regulation output (9) and the quantity parameter vector (10) are detected to have been tampered with by a deception attack, the security output is selected according to equation (11b), and the security controller mode is selected according to equation (20); otherwise, the security output is selected... The selection of the safety controller mode is determined based on the mode obtained by summing all elements in the quantity parameter vector (10). Specifically, the safety controller mode... exist Number of switches on Equal to stable subsequence Number of switching and unstable subsequences Number of switching The sum of , Safety controller mode Corresponding switching sequence .

[0019] Based on the aforementioned safety output, a safety controller is designed, represented as follows: (twenty three) in, This is the gain matrix of the safety controller to be designed; Indicates safe output; S5. Establish an event triggering mechanism for the shore-based-ship channel, wherein the event triggering mechanism for the shore-based-ship channel is used to control the shore-based system to transmit the output of the safety controller and the transmission protocol of the safety controller mode back to the unmanned ship system. In a specific embodiment, to save network resources, the event triggering mechanism of the established shore-to-ship channel is represented as follows: (twenty four) in, Indicates the first A trigger moment, It is a set positive scalar that triggers the counter. The value is incremented by 1 after the event triggering mechanism in the hull-shore passage is activated, and ;when hour, .

[0020] S6. Design a second detection standard, which is used to determine whether the output of the security controller and the security controller mode have been tampered with by a deception attack, and construct a control input compensator based on the second detection standard; S7. Based on the aforementioned switching LPV networked unmanned vessel system, the event triggering mechanism of the hull-shore channel, the deception attack detection observer, the first detection standard, the event triggering mechanism of the shore-hull channel, the second detection standard, and the control input compensator, the unmanned vessel dynamic positioning is achieved.

[0021] Specifically, this embodiment addresses bidirectional spoofing attacks by constructing a dual-channel spoofing attack defense mechanism based on controller mode estimation. In the ship-shore channel, this mechanism estimates reliable controller modes for use when tampering of transmitted data is detected, thereby mitigating the asynchronous behavior duration caused by tampering of quantity parameters in spoofing attacks. In the shore-ship channel, a control input compensator based on quantity parameter vectors is designed to resist the impact of spoofing attacks on the output of the security controller.

[0022] In a specific embodiment, the specific steps for designing the second detection standard include: The network latency during data transmission via the shore-to-ship channel is set to... Therefore, safety controller mode and the output of the safety controller exist The system can switch from a shore-based system to a networked unmanned surface vessel (LPV) system in real time. ; in, , , , ; Specifically, when a deception attack occurs, the output of the security controller is configured. Deceived and attacked Altered to: (25) Setting the safety controller mode Deceived and attacked Altered to: (26) in, ; Therefore, a second testing standard is designed, expressed as: (27) in, It is a vector of quantity parameters.

[0023] Specifically, if the safety controller modal and quantity parameter vector In If the sums are not equal, then the test results are... This indicates the output of the safety controller. If the control input is tampered with by a deception attack, compensation is required; otherwise, it indicates that the output of the security controller has not been tampered with by a deception attack.

[0024] The control input compensator constructed based on the second detection standard is represented as follows: (28) Combining (6), (21), and (23), the control input compensator of the switched LPV networked unmanned vessel system is converted to: (29) in, The maximum event trigger interval to be determined Previous control input compensator; Specifically, control input compensator By switching the actuators of the LPV networked unmanned surface vessel system to the unmanned surface vessel, when a spoofing attack is detected, the control input compensator uses... To compensate for the output of the safety controller .

[0025] In a specific embodiment, in order to reduce network resource consumption and improve the compensation performance of the control input compensator (28), this embodiment selects a suitable sensor sampling period. and maximum event trigger interval On the one hand, the sensor sampling period The number of events triggered by the event triggering mechanisms (6) and (24) is determined, thus affecting the consumption of network resources. On the other hand, the control signals of the security controller (23) Deceived and attacked After tampering, it is necessary to ensure that the data used for compensation is secure, thereby guaranteeing the good compensation performance of the control input compensator (28). Therefore, the present invention needs to select an appropriate sampling period. and maximum event trigger interval Execute two nested loops using the given parameters. In the first loop, candidate parameters are... Increment and record the values ​​that lead to faster stability. Value; in the second loop, add candidate parameters. And record the methods that can achieve faster stability. The value of . See Algorithm 1 for details:

[0026] in, Indicates the adjustment of output under the current parameter conditions. At the first moment, variables res1, res2, and cop are temporary parameters that store temporary values. and They are and Each of their respective step values.

[0027] This embodiment is for describing the triggering time. With sampling time Output error between , will the interval Divide into sub-intervals, represented as:

[0028] in, ; Then, using the input delay method, we can derive... , Therefore, the control input (29) can be rewritten as: , (30) To prove the effectiveness of observer (11), we combine (5) and (11) and substitute (30) to obtain the error system as follows: , (31) To demonstrate that the dynamic positioning control objective of the constructed unmanned vessel switching LPV networked unmanned vessel system is transformed into the output regulation problem of a closed-loop system (33), this embodiment introduces a coordinate transformation matrix within the output regulation framework. and To eliminate the switching between LPV networked unmanned surface vessel systems and error systems The following assumptions are made: Assumption 1: There exists a matrix , , Make: (32) By simultaneously switching the LPV networked unmanned vessel system (5) and the error system (31), the closed-loop system is obtained as follows: , (33) Among them, initial conditions , , , , , , , , , , , , . in, This represents the state matrix of the closed-loop system. This represents the state delay matrix of the closed-loop system. , , , , This represents the intermediate transformation matrix of the closed-loop system. This represents the error matrix of the closed-loop system. This represents the input matrix of the closed-loop system. This represents the output matrix of the closed-loop system.

[0029] Specifically, the construction of the safety controller mode (22) and network latency cause asynchronous behavior between the shore-based system's safety controller (23) and the switching LPV networked unmanned vessel (5). This is due to the detection of a deception attack activation interval. Duration and unmanned vessel mode arbitrary activation interval The uncertainty of the relationship between durations, intervals The duration may be greater or less than the interval. Duration. During the interval Duration less than interval Time, interval In the interval Different internal positions will lead to different safety controller modes Different values ​​of result in complex asynchronous behavior. Therefore, it is necessary to discuss the different scenarios of asynchronous behavior in detail. Without loss of generality, assume that Above, unmanned ship mode ;exist Above, unmanned ship mode , The estimator (20) is active in the detected deception attack region. Estimated controller modes , .according to and The changes can be categorized into three cases: Scenario A: In It was detected on the top This is a deception attack, and none of the deception attacks cover the controller mode switching time. That is Due to the construction of the security controller mode (22), in the activation region where a spoofing attack is detected, Equal to the estimated controller mode In the dormant interval where no deception attack was detected, Equal to unmanned ship mode Therefore, modality as follows: exist superior, , ; exist superior, ; exist superior, .

[0030] Scenario B: In It was detected on the top The second deception attack, and the third The secondary deception attack covers the controller's mode switching time. That is Similar to case A, the modality as follows: exist superior, , ; exist superior, ; exist superior, .

[0031] Case C: Detected spoofing attack activation range cover ,Right now At this moment, the entire On the interval Both are equal to the estimated controller modes Therefore, modality as follows: exist superior, .

[0032] To prove that the output regulation problem of the closed-loop system (33) is solvable, it is necessary to prove: a) when At that time, the closed-loop system (33) is exponentially stable; b) When When, for any initial conditions The regulation output of the closed-loop system (33) satisfies .

[0033] Based on the deception attack detection and control input compensator, the following sufficient conditions are given to ensure the safe dynamic positioning of the constructed switching LPV networked unmanned vessel system (5): Theorem 1: Under Assumption 1, for a given positive parameter , , , , , , , , Dynamic positioning of networked unmanned surface vessels (USVs) based on the switching of the number of onboard devices is achievable if: 1) There exists a matrix , , , , and ,and , , so that: (34) (35) 2) For a given parameter , , , and There exists a matrix , , satisfy: (36) (37) in,

[0034] 3) Modes of unmanned ships The required number of stable / unstable unmanned surface vessel switching points needs to be met. Number of stable / unstable controller switching points The quantity restrictions between them are expressed as: (38) And the average length of stay criteria: (39) in, dimensional block matrix It consists of the following block matrix:

[0035] , , , , , , , , , , , , , , , , , , , , , ; The remaining positions are zero matrices of appropriate dimension, and the parameters in the block matrix are as follows: , , , , , , , , , , , , , , Proof: First, the candidate Lyapunov function is selected as follows: (40) in, ,

[0036]

[0037] Combining the lemma and conditions (34)-(35), we can obtain The derivative along the closed-loop system (33) and The relationship between them is: (41) in , .

[0038] First, we analyze the Lyapunov function in arbitrary intervals under three cases A, B, and C. Changes on: Scenario A: For , At this point, the switching between the LPV networked unmanned vessel system (5) and the safety controller (23) exhibits asynchronous behavior. , Integrating both sides of (41) yields: (42) for , Select , Using the same integration method as (42), we get: (43) for , , Select , ,have: (44) for , Select , ,have: (45) for , Select , ,have: (46) for , Select , ,have: (47) for , Select , ,have: (48) exist The memory experiences controller mode switching caused by network latency and spoofing attacks, so the timing of mode switching needs to be discussed. , , The changes in the Lyapunov function before and after. For Due to different coordinate transformations, Combining (40) and ,have:

[0039] Depend on get According to (36), Using the same method, there exists , , .therefore, (49) Using the same derivation method, for , ,have: (50) for , ,have: (51) for , ,have: (52) for , ,have: (53) By combining (42) and (53), we can obtain any interval. Above: (54) Case B: Similar to Case A, for , Select , ,have: (55) for , Select , ,have: (56) for , , Select , ,have: (57) for , , Select , ,have: (58) for , Select , ,have: (59) Due to the A deception attack activation zone Controller mode switching time caused by overlay network latency Therefore, in this case, we only need to discuss the mode switching time. , The changes in the Lyapunov function before and after, for , ,have: (60) for , ,have: (61) for , ,have: (62) for , ,have: (63) By combining (53) and (61), we can obtain any interval. Above: (64) Case C: Due to the first A deception attack activation zone Covering the entire interval Therefore, for , Select , ,have: (65) Then, the unmanned surface vessel modes were analyzed. Switching time The relationship between the Lyapunov functions before and after is obtained from condition (36): (66) Finally, based on (54) and (64)-(66), it can be deduced that in Above: (67) in, , , Representing an interval The asynchronous duration caused by internal delays and deception attacks. , , , , This represents the number of times a deception attack was detected. According to conditions (38) and (39), , Then, from (40) and (67): (68) in, , , Thus, a) above is proven; Second, due to the lemma: and Make This means Due to the closed-loop system (33) , Thus, b) is proven; In summary, it has been proven that the output regulation of the closed-loop system (33) under the external system (2) is solvable, that is, the dynamic positioning of the constructed switching LPV networked unmanned vessel system (5) is solvable.

[0040] In this embodiment, the effectiveness of the proposed model conversion method, detection method, and safety control scheme is verified through an example of an unmanned surface vessel with varying mass. The unmanned surface vessel's empty mass is considered. Generally, the payload capacity of an unmanned surface vessel (USV) is designed based on its own weight, typically ranging from 30% to 100% of its weight. Therefore, the full-load mass of the USV must be considered. The first type of shipborne equipment quality The second type of shipborne equipment quality their respective quantities .in addition, Figure 2 Surge quality in Its corresponding quantity parameter The specific changes in the quantity, mass, and modal characteristics of the unmanned surface vessel's onboard equipment during navigation are shown in Table 1 and... Figure 3 The unmanned vessel mode is shown in the diagram. Based on the modal setting strategy (3), the unmanned vessel mode... The total mass of the unmanned vessel can be obtained from equation (4). for: , , .

[0041] Since the surge mass is negligible compared to the mass of the networked unmanned surface vessel (USV), its impact on the coefficient matrix can be ignored. Therefore, using the USV parameters in Table 2, the matrix... and for: , , , , , , Table 1:

[0042] Table 2:

[0043] Thus, the unmanned vessel (1) is transformed into a switching LPV networked unmanned vessel system comprising two subsystems:

[0044] in, coefficient matrix , , for: , , , , , , matrix It is a 6-dimensional identity matrix. and Selected as: , , Without loss of generality, choose the initial state Other parameters , , , , , , , , , , , , Based on the given parameters and according to condition (26), the maximum and minimum average dwell times can be calculated. , Select the range that meets this average dwell time. The switching rules, such as Figure 3 As shown, the number of onboard devices on networked unmanned ships changes. Within this range, the gain matrix of the deception attack detection observer (12) can be calculated based on assumption 3 as follows: , , Applying Theorem 1, the gain matrix of the safety controller (24) is calculated as follows: , , , Specifically, the simulation part considers the dynamic positioning of a single unmanned vessel. Figure 3 It shows the changes in the quality of the external surge. Figure 4 This demonstrates the changing number of onboard devices on networked unmanned ships. Within the specified interval is acceptable. The switching model establishment method based on quality partitioning uses a specified dividing point. mass range Divided into two equal parts, corresponding networked unmanned surface vessel modes like Figure 5 As shown, it can be seen that due to the quality of networked unmanned surface vessels... At 27.5 Frequent fluctuations in the vicinity, networked unmanned vessel mode Frequent switching occurred. However, modeling methods based on the number of shipboard devices can eliminate this frequent switching problem. Figure 6 The demonstration shows the triggering times and intervals of the event triggering mechanism in the ship-shore passageway. Figure 7 In the middle, the estimated The values ​​are given, along with estimates and networked unmanned surface vessel modes. The mean square error between the two values ​​indicates that the accuracy of the estimation is reliable. Figure 8 A deception attack detection function is given. and test results This means that the detection standard can detect the activation range of deception attacks. Figure 9 Demonstrates safety controller modality The asynchronous duration between the networked unmanned surface vessel mode and the networked unmanned surface vessel mode. For comparison, Figure 10 This demonstrates the controller modes without using controller mode estimation. The asynchronous duration, as observed, shows a reduction in the asynchronous duration between the safety controller based on controller mode estimation and the networked unmanned surface vessel. Figure 11 In the second detection standard, the security controller modality passing through the network is compared. With the number of networked unmanned ships and The test results were obtained. By comparison Figure 12 Whether there is a control input compensator based on a quantitative parameter vector is clearly demonstrated by the fact that the proposed control input compensator effectively eliminates control input fluctuations caused by deception attacks during unmanned surface vessel operation. Figure 13 The position error shown and Yaw angle Surge speed sway speed yaw rate The curve approaching zero indicates that the dynamic positioning of the LPV-connected unmanned vessel can be achieved. Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A switching unmanned ship safety dynamic positioning method based on attack detection-compensation mechanism, characterized in that, The specific steps include: S1. Establishing an initial unmanned ship model, constructing a switching LPV networked unmanned ship system by using a switching LPV modeling method based on the number of shipborne devices and the initial unmanned ship model; S2. Constructing an event-triggered mechanism of a ship-shore channel, which is used to control the adjustment output in the switching LPV networked unmanned ship system and the transmission protocol of the number parameter vector from the ship to the shore system; S3. Establishing a spoofing attack detection observer of the ship-shore channel, designing a first detection criterion based on the spoofing attack detection observer, and determining whether the adjustment output and the number parameter vector received by the shore system are tampered by a spoofing attack; S4. Obtaining an estimated controller mode, establishing a safe output and a safe controller mode according to the first detection criterion and the estimated controller mode, and designing a safe controller based on the safe output; S5. Establishing an event-triggered mechanism of a shore-ship channel, which is used to control the output of the safe controller and the transmission protocol of the safe controller mode from the shore system to the unmanned ship system; S6. Designing a second detection criterion for judging whether the output of the safe controller and the safe controller mode are tampered by a spoofing attack, and constructing a control input compensator based on the second detection criterion; S7. Realizing unmanned ship dynamic positioning based on the switching LPV networked unmanned ship system, the event-triggered mechanism of the ship-shore channel, the spoofing attack detection observer, the first detection criterion, the event-triggered mechanism of the shore-ship channel, the second detection criterion and the control input compensator.

2. The switching uncrewed ship safety dynamic positioning method based on attack detection-compensation mechanism according to claim 1, characterized in that, The established initial unmanned ship model is represented as: (1) where the vector , is the surge velocity, is the sway velocity, is the yaw angular velocity; the vector , , is the position coordinate, is the yaw angle; the control input vector ; is the disturbance signal, expressed through the outer system, i.e.: (2) wherein S is a system matrix of the outer system, and S the real part of all eigenvalues of is zero; is the inertia matrix, denoted as: ， wherein, are the water force added masses, denote the longitudinal components of different vectors from the origin to the center of gravity in the body-fixed reference frame, is the moment of inertia, is the total mass of the unmanned ship, and satisfies , and are the total masses of the unmanned ship when it is empty and fully loaded, respectively, denotes the load of the unmanned ship varying with time, denotes the surge mass varying with time; is the damping matrix, represented as: ， wherein, are linear damping coefficients, respectively.

3. The switching uncrewed ship safety dynamic positioning method based on attack detection-compensation mechanism according to claim 2, characterized in that, The specific steps for constructing the switching LPV networked unmanned ship system by using the switching LPV modeling method based on the number of shipborne devices and the initial unmanned ship model include: Setting unmanned ship carrying A shipborne device, first The unit mass and quantity of the shipborne device are respectively And , When the first The shipborne device is released or recovered, Correspondingly reduced or increased; Considering the influence of the surge on the total mass of the unmanned ship, the unit mass of the surge is set , and the number of surges ; The unmanned ship mode is set as: (3) wherein is the maximum number of devices on a fully loaded unmanned ship, , ; based on the unmanned ship modal, the total mass of the unmanned ship is determined is re-expressed as: (4) wherein ; Based on equation (4), the matrix in equation (1) is restructured as and ​ ， wherein, Munspareis the unmanned ship's bare hull mass Munspareis the unmanned ship's bare hull mass Munspareis the unmanned ship's bare hull mass Munspareis the unmanned ship's bare hull mass Munspareis the unmanned ship's bare hull mass Munspareis the unmanned ship's bare hull mass Munspareis the unmanned ship's bare hull mass Munspareis the unmanned ship's bare hull mass Munspareis the unmanned ship's bare hull mass Munspareis the unmanned ship's bare hull mass Munspareis the unmanned ship's bare hull mass Munspareis the unmanned ship's bare hull mass Based on the above-mentioned restructured matrix, the initial unmanned ship model (1) and the external system (2), a switching LPV networked unmanned ship system is constructed, which is represented as: (5) ， wherein and are both set matrices; is a measured output, .

4. The switching uncrewed ship safety dynamic positioning method based on attack detection-compensation mechanism according to claim 3, characterized in that, The constructed event-triggered mechanism of the ship-shore channel is represented as: (6) Wherein, (7) (8) wherein , is the output error, and is the triggering time instant , the sampled time instant after , is an unknown weight matrix, and , is a given threshold, and .

5. The switching uncrewed ship safety dynamic positioning method based on attack detection-compensation mechanism according to claim 4, characterized in that, The specific steps for establishing the spoofing attack detection observer of the ship-shore channel include: Setting the network delay existing in the data transmission process based on the hull-shore base channel as Therefore, the output is adjusted And the quantity parameter vector At The moment from the switching LPV networked unmanned ship system to the shore-based system; wherein satisfies , , ; When a spoofing attack occurs, set the adjustment output Spoofed attack Tampering is: (9) wherein , is a given matrix; Set quantity parameter vector Spoofing attacks Tampering is: (10) The spoofing attack detection observer is set based on the tampered adjustment output and the number parameter vector received by the shore system, which is represented as: (11a) (11b) wherein is the state observation value of the switching LPV networked unmanned ship system, is the output observation value, is the gain matrix to be solved, is the estimated controller mode, is the residual signal of the adjusted output, and .

6. The switching uncrewed ship safety dynamic positioning method based on attack detection-compensation mechanism according to claim 5, characterized in that, The specific steps for establishing the first detection criterion based on the spoofing attack detection observer include: A residual signal of a regulated output in the switching LPV networked unmanned ship system is acquired by using the deception attack detection observer , Substitute variables for to obtain selecting a time window using a time window chi-square detection technique and based on the residual signal setting the detection function as: ， The first detection criterion is set based on the detection function, which is represented as: (12) wherein is a detection threshold, and .

7. The safety dynamic positioning method of switching unmanned ship based on attack detection-compensation mechanism according to claim 6, characterized in that, The specific steps for obtaining the estimated controller mode and establishing the safe output and the safe controller mode according to the first detection criterion and the estimated controller mode include: Introducing a virtual input vector , , ; Based on virtual input vectors in the element The estimated controller modes are obtained, denoted as: (13) The safe output and the safe controller mode are constructed based on the first detection criterion and the estimated controller mode, which are respectively represented as: (14) (15) wherein is a detection result of a spoofing attack, is an output observation; The safe controller is designed based on the safe output, which is represented as: (16) wherein, is the safety controller gain matrix to be designed.

8. The switching uncrewed ship safety dynamic positioning method based on attack detection-compensation mechanism according to claim 7, characterized in that, The established event-triggered mechanism of the shore-ship channel is represented as: (17) wherein, denotes the triggering time, is a positive scalar set, the trigger counter is incremented after the event trigger mechanism of the ship-shore passage has been triggered, and ; when , .

9. The switching uncrewed ship safety dynamic positioning method based on attack detection-compensation mechanism according to claim 8, characterized in that, The specific steps for designing the second detection criterion include: Setting the network delay existing in the data transmission process based on the shore-hull passage as Therefore, the safety controller mode And the output of the safety controller At The switching LPV networked unmanned ship system arrives from the shore-based system, that is ; wherein , , , ; Setting the output of the security controller when a spoofing attack occurs Spoofing attack Tampering is: (18) Setting safety controller mode spoofing attacks tampering (19) wherein ; Therefore, a second detection criterion is designed, denoted as: (20) wherein is a vector of quantity parameters.

10. The safety dynamic positioning method of switching unmanned ship based on attack detection-compensation mechanism according to claim 9, characterized in that, A control input compensator constructed based on the second detection criterion is denoted as: (21) By combining (6), (14) and (16), the control input compensator is converted into: (22) wherein, is the maximum event-triggered interval to be sought previous control input compensator.

Images