Multi-UUV security cooperative control method in complex environment based on memory event triggering

Abstract

The invention provides a multi-UUV security cooperative control method in a complex environment based on memory event triggering, and designs an intelligent mechanism for triggering communication only when a specific event occurs, namely, a controller for uncertain multi-UUV memory events is designed for UUVs working in a communication scene with data tampering and time delay. Therefore, cooperative control is realized. Compared with a traditional time triggering mode, the method has the advantages that the communication burden can be effectively reduced, and the system can still realize consistency and has better universality and flexibility under the conditions of communication delay and communication errors; meanwhile, the problems of parameter uncertainty, data tampering and the like are better solved through uncertainty constraints and energy constraints of attack signals causing communication errors.

Description

Technical Field

[0001] The present invention belongs to the technical field of event-triggered safety collaborative control, and in particular relates to a multi-UUV safety collaborative control method in a complex environment based on memory event triggering.

[0002] The present invention was carried out with the support of the National Natural Science Foundation of China (Grant No. 62425602). Background Art

[0003] With the rapid development of modern technology, collaborative control of unmanned underwater vehicles (UUVs) is widely used in fields such as ocean exploration, environmental monitoring, and maritime rescue. However, in practical applications, numerous uncertainties and suboptimal communication have become significant factors hindering the coordinated performance of multiple UUVs. These factors include unclear enemy status, unclear friendly status, and unclear environmental status. These intertwined uncertainties make the environment complex, diverse, and challenging.

[0004] There are three main types of uncertainty in the maritime environment. When the enemy's position is unclear, they may employ tactics such as concealment, camouflage, and disinformation to interfere with intelligence acquisition, increasing operational uncertainty. Lack of or uncertainty in intelligence prevents commanders from fully understanding enemy dynamics and making accurate tactical decisions. When the enemy's position is unclear, UUVs may face communication delays, interruptions, and insufficient intelligence, making it difficult for commanders to accurately understand their own forces' actual situation. Furthermore, exposing friendly status information to the enemy increases the risk of attack. Uncertainty in the environment can affect the speed, path, and even stability of UUVs, compromising coordination between them and posing a threat to overall performance and safety.

[0005] Therefore, communication errors and delays are critical in such uncertain multi-UUV systems. To address these challenges, an event-triggered safety consistency controller has been introduced into the design of uncertain multi-UUVs. This controller employs an intelligent communication strategy that triggers communication only when specific events occur, such as a significant change in the UUV's state or the fulfillment of a preset trigger condition. For example, in maritime search and rescue operations, communication delays or errors can lead to delays or failures in the rescue operation, potentially endangering personnel safety. When a UUV detects a target or exceeds a preset range from a neighboring UUV, it triggers an event to transmit its position and sensor data to other UUVs. The receiving UUVs then update their understanding of the group's state and adjust their search path to avoid duplicate searches and maintain formation consistency. If a UUV is attacked by a cyberattack, it initiates safety control protocols, such as reducing communication frequency, switching to a more secure communication channel, or requesting support from other UUVs. Compared to continuous communication, this controller effectively reduces unnecessary communication overhead, mitigates the risk of communication congestion, and improves the real-time and accuracy of information transmission. In order to solve this problem, an event-triggered safe collaborative control technology for uncertain multi-UUVs is generated to achieve the goal of multi-UUV collaborative and safe operation in communication-limited and uncertain environments.

[0006] In the event-triggered safety control of multiple UUVs, the more realistic and common situations involve uncertainty in the system model, as well as communication delays and errors, making controller design more difficult. To achieve event-triggered safety control of uncertain multi-UUV systems that accounts for data tampering and communication delays, this paper proposes an intelligent communication strategy, time-stamping and elastic control algorithms, robust design, and fault-tolerant mechanisms to ensure the coordinated and safe operation of multiple UUVs in complex and changing marine environments. With the continuous advancement of technology and the expansion of application scenarios, this controller will play an even more important role in fields such as ocean exploration, environmental monitoring, and maritime rescue. Summary of the Invention

[0007] In order to solve the problems of communication delay and communication error, the present invention provides a multi-UUV safe collaborative control method in a complex environment based on memory event triggering. While processing communication data tampering and communication delay, it reduces the network communication burden and has stronger robustness and better flexibility.

[0008] A safe collaborative control method for multiple UUVs in complex environments based on memory event triggering. Each UUV obtains its own control variable only at a specific triggering moment, allowing each UUV to complete collaborative control driven by its own control variable.

[0009] Among them, the next triggering moment of any UUV is obtained according to the memory-type dynamic event triggering mechanism, and the method for obtaining the next triggering moment of each UUV according to the memory-type dynamic event triggering mechanism is as follows:

[0010]

[0011] in, is the latest triggering moment of the i-th UUV, k is the number of the triggering moment, is the next triggering moment of the i-th UUV, is the latest sampling time of the i-th UUV, l is the time interval between the latest sampling time and the latest trigger time, and l∈N, N is a natural number set, is the i-th UUV at the latest triggering moment The state of is the trigger error of the i-th UUV with respect to the s-th historical trigger event, and s=1,2,…,m, m is the total number of historical trigger events that have occurred so far, is the i-th UUV at the latest sampling time The state of is the i-th UUV at the historical triggering time The state of is the i-th UUV at the latest sampling time The trigger threshold when is the event trigger function of the i-th UUV; min is the minimum value;

[0012] The control amount of any UUV is output by the controller in attack mode, and the method for outputting the control amount by the controller of any UUV is as follows:

[0013]

[0014] in, is the controller of the i-th UUV, θ i (t) is a Bernoulli distribution variable describing whether the i-th UUV is attacked, and θ i (t) = 0 means that the i-th UUV is not attacked, θ i (t) = 1 means the i-th UUV is attacked, K s is the memory feedback control gain of the sth historical trigger event, is the i-th UUV at the historical triggering time The data packets that need to be transmitted, is the attack signal to the i-th UUV at the historical triggering time Data packets at the time The data packet is obtained by random tampering.

[0015] Furthermore, the i-th UUV at the latest sampling time The trigger threshold The calculation method is as follows:

[0016]

[0017] Among them, α i is the basic threshold scalar of the i-th UUV, β i is the basic threshold scalar α that characterizes the i-th UUV i The weighting factor of the degree of change, is the trigger error of the i-th UUV with respect to the s-th historical trigger event The weight factor, ω i is the setting constant of the i-th UUV, T is the transpose sign, and tanh is the inverse tangent function.

[0018] Furthermore, the event trigger function of the i-th UUV The calculation method is as follows:

[0019]

[0020] in, is the trigger error of the memory-type dynamic event trigger mechanism of the i-th UUV with respect to the s-th historical trigger event The weight factor, T is the transpose sign, Φ i is the symmetric positive definite matrix of the i-th UUV, is the state mean of the i-th UUV in the m historical triggering events that have occurred so far.

[0021] Furthermore, the state mean The calculation method is as follows:

[0022]

[0023] in, is the neighbor set consisting of neighbor UUVs that fall within the communication range of the i-th UUV, is the jth UUV in the neighbor set of the i-th UUV at the historical triggering time The state at the time, k′-s+1 is selected to calculate the state mean The number of the historical triggering moment of the j-th UUV, and the constant to be determined is the historical triggering moment of the j-th UUV, r is the number of the historical triggering moment of the j-th UUV, a ijis the weight coefficient between the i-th UUV and the j-th UUV, x0(t) is the state of the preset leader at time t, c i is the state of the i-th UUV The weight coefficient of the error between the state x0(t) and the leader.

[0024] Furthermore, the data packet Specifically:

[0025]

[0026] in, is the neighbor set consisting of neighbor UUVs that fall within the communication range of the i-th UUV, a ij is the weight coefficient between the i-th UUV and the j-th UUV, τ i (t) is the time delay of the i-th UUV at time t, τ j (t) The time delay of the jth UUV at time t, is the trigger error with time delay of the i-th UUV with respect to the s-th historical trigger event, and when used to calculate the data packet, The calculation method is is the error tracking vector generated by the i-th UUV being driven by its own control quantity, z j (t-τ j (t)) is the error tracking vector generated by the j-th UUV after being driven by its own control quantity, is the trigger error with time delay of the j-th UUV with respect to the s-th historical trigger event, c i is the state of the i-th UUV The weight coefficient of the error between the state x0(t) and the leader.

[0027] Furthermore, the error tracking vector z of the i-th UUV i (t-τ i (t)) is calculated as follows:

[0028] z i (t-τ i (t)) = x i (t-τ i (t))-x0(t-τ0(t))

[0029] Among them, x i (t-τ i (t)) is the time at which the i-th UUV i (t) is the state of the leader UUV at time t, τ0(t) is the time delay of the leader UUV at time t, and x0(t-τ0(t)) is the state of the leader UUV at time t-τ i(t) Status at the moment;

[0030] The error tracking vector z of the jth UUV j (t-τ j (t)) is calculated as follows:

[0031] z j (t-τ j (t)) = x j (t-τ j (t))-x0(t-τ0(t))

[0032] Among them, x j (t-τ j (t)) is the jth UUV at t-τ j (t) The state at time moment.

[0033] Furthermore, a multi-UUV safety collaborative control method in a complex environment based on memory event triggering further includes the following steps:

[0034] Determine the error tracking vector z i (t-τ i (t))、z j (t-τ j (t)) are both less than the set threshold, if not, then adjust the feedback control gain K s The trigger moment and control amount are re-obtained until the error tracking vectors are all smaller than the set threshold, thus achieving coordinated control.

[0035] Beneficial effects:

[0036] The present invention provides a method for safe collaborative control of multiple UUVs in a complex environment based on memory event triggering. It designs an intelligent mechanism that triggers communication only when a specific event occurs. That is, an uncertain multi-UUV memory event controller is designed for UUVs working in communication scenarios with data tampering and time delay, so as to achieve collaborative control. Compared with the traditional time triggering method, the present invention can effectively reduce the communication burden, and the system can still achieve consistency in the case of communication delay and communication error, with better universality and flexibility. At the same time, the present invention also better solves problems such as parameter uncertainty and data tampering through uncertainty constraints and energy constraints of attack signals that cause communication errors. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 Flowchart of the present invention.

[0038] Figure 2 FIG. 4 is a topological diagram of an embodiment.

[0039] Figure 3is the state tracking error trajectory diagram.

[0040] Figure 4 The trigger time and interval diagram. DETAILED DESCRIPTION

[0041] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application.

[0042] A safe collaborative control method for multiple UUVs in complex environments based on memory event triggering. Each UUV obtains its own control variable only at a specific triggering moment, allowing each UUV to complete collaborative control driven by its own control variable.

[0043] It should be noted that before completing the collaborative control, the communication status of each UUV can be determined, including the dynamics and system uncertainty parameters of the leader and each follower, as follows:

[0044] S11. For a directed graph G consisting of a leader and N followers, define its Laplacian matrix as L.

[0045] S12, the uncertainty parameter of the system is ΔA(t)=D A F A (t)E A , ΔB(t)=D B F B (t)E B ,and Where I is the identity matrix, D A , E A , D B , E B is a deterministic matrix, and F A (t) and F B (t) is a time-varying uncertain matrix.

[0046] S13. Construct a conventional system controller as follows:

[0047]

[0048] in, represents the neighbor set of the i-th (i=1,2,…,N) UUV, x i (t) is the state of the i-th (i=0,1,…,N) UUV, K is the feedback control gain, a ij is the adjacency weight between UUVs, c i is the coupling weight between the follower and the leader.

[0049] Based on this, the next trigger moment of any UUV is obtained according to the memory-type dynamic event trigger mechanism, and the method for obtaining the next trigger moment of each UUV according to the memory-type dynamic event trigger mechanism is as follows:

[0050]

[0051] in, is the latest triggering moment of the i-th UUV, k is the number of the triggering moment, is the next triggering moment of the i-th UUV, is the latest sampling time of the i-th UUV, l is the time interval between the latest sampling time and the latest trigger time, and l∈N, N is a natural number set, is the i-th UUV at the latest triggering moment The state of is the trigger error of the i-th UUV with respect to the s-th historical trigger event, and s=1,2,…,m, m is the total number of historical trigger events that have occurred so far, is the i-th UUV at the latest sampling time The state of is the i-th UUV at the historical triggering time The state of is the i-th UUV at the latest sampling time The trigger threshold when is the event trigger function of the i-th UUV; min is the minimum value;

[0052] The control amount of any UUV is output by the controller in attack mode, and the method for outputting the control amount by the controller of any UUV is as follows:

[0053]

[0054] in, is the controller of the i-th UUV, θ i (t) is a Bernoulli distribution variable describing whether the i-th UUV is attacked, and θ i (t) = 0 means that the i-th UUV is not attacked, θ i (t) = 1 means the i-th UUV is attacked, K s is the memory feedback control gain of the sth historical trigger event, is the i-th UUV at the historical triggering time The data packets that need to be transmitted, is the attack signal to the i-th UUV at the historical triggering time Data packets at the time The data packet is obtained by random tampering.

[0055] In addition, the present invention can further determine the error tracking vector z i (t-τ i (t))、z j (t-τ j (t)) are both less than the set threshold, if not, then adjust the feedback control gain K s The trigger moment and control amount are re-obtained until the error tracking vectors are all smaller than the set threshold, thus achieving coordinated control.

[0056] Furthermore, the i-th UUV at the latest sampling time The trigger threshold The calculation method is as follows:

[0057]

[0058] Among them, α i is the basic threshold scalar of the i-th UUV, β i is the basic threshold scalar α that characterizes the i-th UUV i The weighting factor of the degree of change, is the trigger error of the i-th UUV with respect to the s-th historical trigger event The weight factor, ω i is the setting constant of the i-th UUV, T is the transpose sign, and tanh is the inverse tangent function. It should be noted that the trigger threshold is a function that changes dynamically in the interval (0, 1). In addition, the constant ω i >0, and β i >0 and Describes the basic threshold scalar α i the degree and direction of change.

[0059] Furthermore, the event trigger function of the i-th UUV is designed using the m historical trigger signals The calculation method is as follows:

[0060]

[0061] in, is the trigger error of the memory-type dynamic event trigger mechanism of the i-th UUV with respect to the s-th historical trigger event The weight factor, T is the transpose sign, Φ i is the symmetric positive definite matrix of the i-th UUV, is the state mean of the i-th UUV in the m historical triggering events that have occurred so far.

[0062] Furthermore, the state mean The calculation method is as follows:

[0063]

[0064] in, is the neighbor set consisting of neighbor UUVs that fall within the communication range of the i-th UUV, is the jth UUV in the neighbor set of the i-th UUV at the historical triggering time The state at time k ′ -s+1 is selected to calculate the state mean The number of the historical triggering moment of the j-th UUV, and the constant to be determined is the historical triggering moment of the j-th UUV, r is the number of the historical triggering moment of the j-th UUV, a ij is the weight coefficient between the i-th UUV and the j-th UUV, x0(t) is the state of the preset leader at time t, c i is the state of the i-th UUV The weight coefficient of the error between the state x0(t) and the leader.

[0065] Furthermore, the data packet Specifically:

[0066]

[0067] in, is the neighbor set consisting of neighbor UUVs that fall within the communication range of the i-th UUV, a ij is the weight coefficient between the i-th UUV and the j-th UUV, τ i (t) is the time delay of the i-th UUV at time t, τ j (t) The time delay of the jth UUV at time t, is the trigger error with time delay of the i-th UUV with respect to the s-th historical trigger event, and when used to calculate the data packet, The calculation method is is the error tracking vector generated by the i-th UUV being driven by its own control quantity, z j (t-τ j (t)) is the error tracking vector generated by the j-th UUV after being driven by its own control quantity, is the trigger error with time delay of the j-th UUV with respect to the s-th historical trigger event, c i is the state of the i-th UUV The weight coefficient of the error between the state x0(t) and the leader.

[0068] Furthermore, the error tracking vector z of the i-th UUV i (t-τ i (t)) is calculated as follows:

[0069] z i (t-τ i (t)) = x i (t-τ i (t))-x0(t-τ0(t))

[0070] Among them, x i (t-τ i (t)) is the time at which the i-th UUV i (t) is the state of the leader UUV at time t, τ0(t) is the time delay of the leader UUV at time t, and x0(t-τ0(t)) is the state of the leader UUV at time t-τ i (t) Status at the moment;

[0071] The error tracking vector z of the jth UUV j (t-τ j (t)) is calculated as follows:

[0072] z j (t-τ j (t)) = x j (t-τ j (t))-x0(t-τ0(t))

[0073] Among them, x j (t-τ j (t)) is the jth UUV at t-τ j (t) The state at time moment.

[0074] The following takes a leader and four follower UUVs as an example to further illustrate the multi-UUV safe collaborative control method in a complex environment based on memory event triggering provided by the present invention.

[0075] like Figure 1 As shown, this embodiment proposes a technology for generating a multi-UUVs safety collaborative control strategy in a complex environment, including the following steps:

[0076] Step S1: Determine the communication status of each UUV, including the dynamics and system uncertainty parameters of the leader and each follower, specifically:

[0077] Consider a dynamic model of a linear multi-agent system

[0078]

[0079] Each agent is a UUV. The system consists of a leader and four follower UUVs. The network topology is as follows: Figure 2 As shown, the system matrix is

[0080]

[0081] A is the adjacency matrix, B is the output gain matrix of the system controller;

[0082] The uncertainty parameter is

[0083]

[0084] Step S2: Design a distributed memory-based dynamic event triggering mechanism for each follower UUV, specifically including:

[0085] Set the trigger function buffer capacity to m = 3, and the trigger weight of the i-th UUV is and Select the threshold parameter δ = diag {0.0098, 0.0083, 0.0081, 0.0067}, then δ ≤ 0.01. Assume the sampling period T = 0.01s, the upper bound of the communication delay τ M =2T.

[0086] Step S3: Provide the trigger interval division based on the communication delay and the delay trigger controller in the interval at the current moment. The specific implementation process is as follows:

[0087] S31. Dividing the trigger interval based on delay It is the extended time of the latest event trigger state in the network, and the event trigger error under delay is obtained τ i (t) is the time delay at time t.

[0088] S32. Define the error tracking vector z i (t-τ i (t)) = x i (t-τ i (t))-x0(t-τ0(t)), for any The controller designed for each follower under the time delay error system is:

[0089]

[0090] Among them, K s (s=1,2,…,m) is the memory-based feedback control gain matrix,

[0091] For example, assuming the simulation time is 0-3s, the upper bound of the communication delay τ M =2T=0.02s. The controller under the time delay error system of each following UUV is:

[0092]

[0093] Step S4: Designing a distributed controller with tampered data using Bernoulli distribution variables, specifically including the following steps:

[0094] S41. Establish an attack model and transmit dynamic data packets to the control end The attack signal is randomly tampered with

[0095] With Bernoulli distribution variable θ i (t) describes the attack, where θ i (t) = 0 means no attack occurs, and θ i (t) = 1 means that the actual signal of the i-th UUV transmitted to the control end is The expected assumptions of a spoofing attack are as follows:

[0096]

[0097] where θ i , is a known constant. Since the attacker has limited energy, his attack signal is constrained:

[0098]

[0099] and is a known constant matrix,

[0100] S42. The controller of the i-th UUV can be described as follows when it is attacked by this type of attack:

[0101]

[0102] S43. Define augmented vector

[0103]

[0104] θ(t)=diag{θ1(t), θ2(t),…,θ N (t)},

[0105] θ=diag{θ1, θ2,…,θ N},

[0106]

[0107] From the Kronecker matrix, we can get the control protocol The closed-loop system is:

[0108]

[0109] in And the matrix H=L+C, L is the Laplace matrix, C is a i0 (i=1,2,…,N) is a diagonal matrix. z(t+1) is the state of the controlled object output by the closed-loop system;

[0110] In addition, the extended form of the trigger function of the memory-type dynamic event trigger is

[0111]

[0112] where δ=diag{max{δ1(t)},…,max{δ N (t)}}, Φ=diag{Φ1,…,Φ N}, and the matrix

[0113] Finally, with the help of Lyapunov stability theory, safe collaborative control of uncertain multiple UUVs in complex environments is realized, and the control gain and event trigger matrix are solved through linear matrix inequalities.

[0114] For example, suppose the attack signal is described as:

[0115]

[0116] Bernoulli distribution variable θ i (t) = 0 means no attack occurs, and θ i (t) = 1 means that the actual signal of the i-th UUV transmitted to the control end is The attack sequence occurs on the interval [0.09s, 0.12s]∪[0.19s, 0.24s]∪[0.5s, 0.64s]∪[1s, 1.04s]∪[1.85s, 1.95s]∪[2.2s, 2.25s]. The expected assumptions of this attack are as follows:

[0117]

[0118] The probability of an attack occurring is θ i = 0.003. Since the attacker has limited energy, his attack signal is constrained:

[0119]

[0120] And F i =diag{0.05,0.1}.

[0121] Through the above steps, with the help of Lyapunov stability theory, the safe collaborative control of uncertain multi-UUVs in complex environments can be achieved, where the state tracking error trajectories of the four UUVs and the leader are as follows: Figure 3 As shown in Figure 2, the system achieves stability under network attacks and delays. In addition, the triggering time and interval of the trigger mechanism are as follows: Figure 4 As shown in Figure 2, it effectively reduces the network burden by reducing unnecessary signal transmission. The control gain and event trigger matrix are also solved.

[0122]

[0123] In summary, the present invention provides a method for safe collaborative control of multiple UUVs in a complex environment based on memory event triggering, comprising the following steps: S1. Determine the communication status of each UUV, including the dynamics and system uncertainty parameters of the leader and each follower; S2. Design a distributed memory-type dynamic event triggering mechanism for each follower UUV, eliminating the impact of communication errors and delays caused by network attacks in uncertain environments on the collaborative control of multiple UUVs; S3. Provide a trigger interval division based on communication delay and a delay trigger controller on the interval, reducing the burden of network communication; S4. Design a distributed safety controller with tampered data using Bernoulli distribution variables. The present invention proposes a controller design for uncertain multi-UUV memory events in the presence of data tampering and delay in communication. Through event-triggered elastic control, it eliminates the impact of communication errors and delays caused by network attacks in uncertain environments on the collaborative control of multiple UUVs, while reducing the burden of network communication, and has better universality and flexibility.

[0124] Of course, the present invention may have many other embodiments. Without departing from the spirit and essence of the present invention, those skilled in the art may of course make various corresponding changes and modifications based on the present invention, but these corresponding changes and modifications should all fall within the scope of protection of the claims attached to the present invention.

Claims

1. A multi-UUV safety collaborative control method in a complex environment based on memory event triggering, characterized by: Each UUV obtains its own control quantity only at a specific triggering moment, so that each UUV can complete the coordinated control under the drive of its own control quantity; Among them, the next triggering moment of any UUV is obtained according to the memory-type dynamic event triggering mechanism, and the method for obtaining the next triggering moment of each UUV according to the memory-type dynamic event triggering mechanism is as follows: in, is the latest triggering moment of the i-th UUV, k is the number of the triggering moment, is the next triggering moment of the i-th UUV, is the latest sampling time of the i-th UUV, l is the time interval between the latest sampling time and the latest trigger time, and l∈N, N is a natural number set, is the i-th UUV at the latest triggering moment The state of is the trigger error of the i-th UUV with respect to the s-th historical trigger event, and s=1,2,…,m, m is the total number of historical trigger events that have occurred so far, is the i-th UUV at the latest sampling time The state of is the i-th UUV at the historical triggering time The state of is the i-th UUV at the latest sampling time The trigger threshold when is the event trigger function of the i-th UUV; min is the minimum value; The control amount of any UUV is output by the controller in attack mode, and the method for outputting the control amount by the controller of any UUV is as follows: in, is the controller of the i-th UUV, θ i (t) is a Bernoulli distribution variable describing whether the i-th UUV is attacked, and θ i (t) = 0 means that the i-th UUV is not attacked, θ i (t) = 1 means the i-th UUV is attacked, K s is the memory feedback control gain of the sth historical trigger event, is the i-th UUV at the historical triggering time The data packets that need to be transmitted, is the attack signal to the i-th UUV at the historical triggering time Data packets at the time The data packet is obtained by random tampering.

2. The multi-UUV safety collaborative control method in a complex environment based on memory event triggering according to claim 1 is characterized in that: The i-th UUV at the latest sampling time The trigger threshold The calculation method is as follows: Among them, α i is the basic threshold scalar of the i-th UUV, β i is the basic threshold scalar α that characterizes the i-th UUV i The weighting factor of the degree of change, is the trigger error of the i-th UUV with respect to the s-th historical trigger event The weight factor, ω i is the setting constant of the i-th UUV, T is the transpose sign, and tanh is the inverse tangent function.

3. The multi-UUV safety collaborative control method in a complex environment based on memory event triggering according to claim 1 is characterized in that: Event trigger function of the i-th UUV The calculation method is as follows: in, is the trigger error of the memory-type dynamic event trigger mechanism of the i-th UUV with respect to the s-th historical trigger event The weight factor, T is the transpose sign, Φ i is the symmetric positive definite matrix of the i-th UUV, is the state mean of the i-th UUV in the m historical triggering events that have occurred so far.

4. The multi-UUV safety collaborative control method in a complex environment based on memory event triggering according to claim 3 is characterized in that: State mean The calculation method is as follows: in, is the neighbor set consisting of neighbor UUVs that fall within the communication range of the i-th UUV, is the jth UUV in the neighbor set of the i-th UUV at the historical triggering time The state at the time, k′-s+1 is selected to calculate the state mean The number of the historical triggering moment of the j-th UUV, and the constant to be determined r is a natural number}, is the historical triggering moment of the j-th UUV, r is the number of the historical triggering moment of the j-th UUV, a ij is the weight coefficient between the i-th UUV and the j-th UUV, x0(t) is the state of the preset leader at time t, c i is the state of the i-th UUV The weight coefficient of the error between the state x0(t) and the leader.

5. The multi-UUV safety collaborative control method in a complex environment based on memory event triggering according to claim 1 is characterized in that: Data Packet Specifically: in, is the neighbor set consisting of neighbor UUVs that fall within the communication range of the i-th UUV, a ij is the weight coefficient between the i-th UUV and the j-th UUV, τ i (t) is the time delay of the i-th UUV at time t, τ j (t) The time delay of the jth UUV at time t, is the trigger error with time delay of the i-th UUV with respect to the s-th historical trigger event, and when used to calculate the data packet, The calculation method is z i (t-τ i (t)) is the error tracking vector generated by the i-th UUV after being driven by its own control quantity, z j (t-τ j (t)) is the error tracking vector generated by the j-th UUV after being driven by its own control quantity, is the trigger error with time delay of the j-th UUV with respect to the s-th historical trigger event, c i is the state of the i-th UUV The weight coefficient of the error between the state x0(t) and the leader.

6. The multi-UUV safety collaborative control method in a complex environment based on memory event triggering according to claim 5 is characterized in that: The error tracking vector z of the i-th UUV i (t-τ i (t)) is calculated as follows: z i (t-τ i (t))=x i (t-τ i (t))-x0(t-τ0(t)) Among them, x i (t-τ i (t)) is the time at which the i-th UUV i (t) is the state of the leader UUV at time t, τ0(t) is the time delay of the leader UUV at time t, and x0(t-τ0(t)) is the state of the leader UUV at time t-τ i (t) Status at the moment; The error tracking vector z of the jth UUV j (t-τ j (t)) is calculated as follows: z j (t-τ j (t))=x j (t-τ j (t))-x0(t-τ0(t)) Among them, x j (t-τ j (t)) is the jth UUV at t-τ j (t) The state at time moment.

7. The multi-UUV safety collaborative control method in a complex environment based on memory event triggering according to claim 5 is characterized in that: The following steps are also included: Determine the error tracking vector z i (t-τ i (t))、z j (t-τ j (t)) are both less than the set threshold, if not, then adjust the feedback control gain K s The trigger moment and control amount are re-obtained until the error tracking vectors are all smaller than the set threshold, thus achieving coordinated control.

Images