Dichotomy consistency control method for unmanned aerial vehicle cluster under lagging full-state constraint

Abstract

The invention discloses a bipartite consistency control method of an unmanned aerial vehicle cluster under lagging full-state constraint, and relates to the technical field of unmanned aerial vehicle cluster cooperative control, comprising the following steps: establishing an unmanned aerial vehicle attitude dynamic model subjected to lagging full-state constraint and input saturation; constructing a coordinate conversion mechanism, and converting an original constrained system into an equivalent unconstrained system; constructing a cooperation-competition coexistence topology communication network, and defining an error dynamic state in combination with a self-adaptive backstepping control technology; an interval type-2 fuzzy logic system and a self-triggering communication protocol are integrated, and a self-adaptive fuzzy self-triggering fixed-time bipartite consistency controller is designed. According to the method, feasibility conditions do not need to be preset, through innovative coordinate conversion and control strategies, the problem of fixed-time dichotomy consistency control of the unmanned aerial vehicle cluster under lagging full-state constraint and input saturation is effectively solved, meanwhile, the network communication load is remarkably reduced by means of a self-triggering mechanism, and the system reliability is improved. And the control precision and the resource utilization efficiency of the system are improved.

Description

Technical Field

[0001] This invention relates to the field of collaborative control technology for unmanned aerial vehicle (UAV) swarms, and more specifically, to a binary consistency control method for UAV swarms under lag-full-state constraints. Background Technology

[0002] Unmanned aerial vehicle (UAV) swarm consensus control technology has become a hot research topic due to its significant advantages in efficiently executing large-scale dynamic tasks and wide-area operations. However, existing research mostly focuses on cooperative communication scenarios within UAV swarms, with little in-depth exploration of the more general cooperative-adversarial coexistence working mode. It is important to note that, based on flight safety regulations and mechanical structure constraints, the dynamic behavior of the UAV's position and attitude subsystems must be constrained. Currently, most finite-time control schemes based on obstacle Lyapunov functions are limited by feasibility constraints, and their stability time upper bound estimation functions usually depend on the system's initial state information, resulting in cumbersome calculations and hindering practical engineering applications. Although adaptive full-state constraint control problems without feasibility conditions have been extensively studied, such methods still struggle to effectively solve the binary consensus control problem of UAV swarms under lag full-state constraints.

[0003] Furthermore, real-time information exchange is essential in collaborative operations of drone swarms. However, traditional time-triggered communication protocols inevitably transmit a large amount of redundant data, continuously occupying network channels and creating a communication burden. This can potentially exacerbate the competition among network nodes for limited airborne bandwidth resources, affecting the collaborative efficiency and overall performance of the drone swarm.

[0004] Therefore, there is an urgent need to invent an adaptive fuzzy self-triggered fixed-time control method for UAV swarms to achieve attitude binary consistency control of UAV swarms under lower hysteresis full-state constraints. Summary of the Invention

[0005] This invention addresses the shortcomings of existing UAV swarm cooperative control technologies by providing an adaptive fuzzy self-triggered fixed-time binary consensus control method for UAV swarm systems subject to lag-based full-state constraints. This method effectively solves the key problems of feasibility condition limitations and low communication resource efficiency in traditional obstacle-based Lyapunov function control schemes. Specifically, it implements lag-based full-state constraints without feasibility conditions by constructing a transformation mechanism based on a time-varying nonlinear transfer function; it designs a self-triggered communication protocol and constructs an adaptive fuzzy self-triggered fixed-time binary consensus controller based on it. This controller enables on-demand, non-periodic updates of control signals, ensuring that all states of the UAV closed-loop attitude subsystem converge within a fixed time and strictly satisfy the given lag-based time-varying constraints.

[0006] The technical solution adopted in this invention is as follows:

[0007] A binary consensus control method for UAV swarms under hysteresis full-state constraints includes the following steps:

[0008] S1. Establish the attitude dynamics model of each UAV in the UAV cluster. The model is subject to hysteresis full-state constraints and input saturation.

[0009] S2. Construct a transfer function and a time-varying nonlinear transformation function to convert the original attitude state under the hysteresis full-state constraint into an unconstrained equivalent state, and establish an equivalent state space equation based on the equivalent state.

[0010] S3. Construct a cooperative-competitive coexistence topology communication network for the UAV cluster, and define a coordinate transformation equation to characterize the system error based on the adaptive backstepping control method;

[0011] S4. Based on the coordinate transformation equation, design an adaptive fuzzy self-triggering fixed-time binary consensus controller, including: using an interval type II fuzzy logic system to approximate the unknown nonlinear function online, designing a parameter update law to adjust the weights of the fuzzy logic system in real time, and designing a self-triggering communication protocol to realize the non-periodic update of the controller.

[0012] Further, in step S1, establishing the attitude dynamics model of each UAV specifically involves: representing the attitude dynamics model of each UAV as a second-order system, the state vector of which includes roll angle, pitch angle, yaw angle, and their respective angular velocities and angular accelerations; the hysteresis full-state constraint means that each dimension of the system's state variables is confined between a pair of lower and upper bound functions that are time functions, forming a time-varying constraint boundary; the input saturation means that the control torque input applied to the UAV is restricted within a known saturation boundary value.

[0013] Further, in step S2, the design of the time-varying nonlinear transformation function is specifically as follows: First, a transfer function is constructed and multiplied with the constrained original attitude state to obtain an intermediate transformation state; finally, based on the intermediate transformation state and the upper and lower bound functions of the hysteresis full state constraint, the time-varying nonlinear transformation function is constructed, thereby transforming the original constrained attitude dynamics system into an unconstrained equivalent state-space system.

[0014] Further, in step S3, constructing a cooperative-competitive coexistence topology communication network and defining a coordinate transformation equation specifically involves: using a directed graph that includes cooperative and competitive relationships to represent the communication topology of the UAV swarm; the coordinate transformation equation defines a first error variable and a second error variable, wherein the first error variable is a weighted combination of the equivalent state of the current UAV, the equivalent state of its neighboring UAVs, and the leader's trajectory, and the weight sign depends on the cooperative or competitive relationship between individuals; the second error variable is defined as the difference between the time derivative of the equivalent state of the current UAV and the virtual controller and auxiliary signals.

[0015] Furthermore, the first step of the adaptive backstepping control method includes designing the virtual controller, specifically: constructing a first Lyapunov function for the first error variable; using a first interval type II fuzzy logic system to approximate the unknown nonlinear term in the derivative of the first Lyapunov function online; the virtual controller is designed to include four parts added together, which are: a nonlinear feedback term related to the first error variable, a power term to ensure fixed-time convergence, an adaptive parameter estimation term for the unknown weights of the first interval type II fuzzy logic system, and a coupling term to compensate for the dynamic influence of adjacent UAVs.

[0016] Furthermore, the second step of the adaptive backstepping control method includes designing the adaptive fuzzy self-triggered fixed-time binary consensus controller, specifically: constructing a second Lyapunov function for the second error variable; using a second interval type-II fuzzy logic system to approximate the unknown nonlinear term in the derivative of the second Lyapunov function online; designing the auxiliary signal to actively suppress the nonlinear effects caused by input saturation; and finally, combining the self-triggered communication protocol, constructing the specific form of the controller based on the intermediate control signal, wherein the controller is composed of a nonlinear function based on the second error variable multiplied by a gain function.

[0017] Furthermore, the steps of the design parameter update law are as follows: the weight update laws of the first interval type II fuzzy logic system and the second interval type II fuzzy logic system, as well as the update laws of other adaptive parameters, are all derived based on Lyapunov stability theory; the parameter update law of the fuzzy logic system weights is designed to be proportional to the product of the corresponding error variable and the fuzzy basis function, and includes a σ correction term to enhance robustness, thereby ensuring that all adaptive estimation parameters are bounded.

[0018] Furthermore, the steps for designing the self-triggering communication protocol are as follows: at each triggering moment, a control signal is calculated and applied to the UAV, and the next triggering moment is determined based on the current dynamic information of the system; the next triggering moment is obtained by adding the current triggering moment to a dynamically changing execution interval; the calculation of the execution interval includes: dividing the hyperbolic tangent function with the current control signal as input by the maximum value of the current control signal's time change rate, and being adjusted by a preset positive design parameter to ensure that the interval between two consecutive triggering events is always positive.

[0019] Furthermore, the aforementioned binary consistency control method also includes: approximating the saturated nonlinearity using a smooth function with a known upper bound approximation error; and, in the design process of adaptive backstepping control, actively compensating for the impact of the approximation error by designing an auxiliary signal system. The dynamics of the auxiliary signal system are driven by itself, the nonlinear term related to the control error, and the difference between the actual control input and the approximate control input, thereby ensuring the stability of the closed-loop system.

[0020] Furthermore, the aforementioned binary consistency control method also includes: constructing an overall Lyapunov function composed of the first and second Lyapunov functions of all UAVs; determining the gain parameters of the controller to be designed by differentiating the overall Lyapunov function and combining it with the inequality scaling method for stability analysis; and proving that under the action of the designed controller, all signals in the closed-loop system are consistent and eventually bounded, and that the synchronization error of the UAV swarm can converge to the residual set near the origin within a fixed time.

[0021] Compared with the prior art, the present invention has the following beneficial effects:

[0022] (1) Unlike traditional control methods based on obstacle Lyapunov functions, this method innovatively introduces a coordinate transformation mechanism consisting of a transfer function and a time-varying nonlinear transformation function, which cleverly transforms a complex system with time-varying full-state constraints with lag into an equivalent unconstrained system for control design. This method fundamentally avoids the stringent requirement that the initial state of the system must meet the constraint conditions, that is, it does not require any pre-set feasibility conditions, thus solving a major bottleneck in the engineering application of existing technologies and greatly expanding the applicable scenarios of full-state constraint control methods.

[0023] (2) The adaptive fuzzy self-triggering fixed-time binary consensus controller designed in this invention can not only cope with unknown nonlinear dynamics and external disturbances in the system, but also handle the control input saturation problem. More importantly, it can ensure that the attitude synchronization error of the UAV swarm converges to the residual set in the neighborhood of the origin within a fixed time independent of the initial state, and provides an explicit analytical solution for the stable time upper bound. Compared with finite-time control, fixed-time control provides stronger convergence performance guarantee. At the same time, through the design of adaptive laws and auxiliary signals, it effectively suppresses approximation errors and saturation effects, and significantly improves the anti-interference capability and control accuracy of the closed-loop system.

[0024] (3) This invention proposes a self-triggering communication protocol, which eliminates the need for fixed time periods or continuous monitoring of measurement errors in updating control signals. Instead, it dynamically and non-periodically triggers the updates on demand based on the system state. This protocol not only fundamentally avoids the continuous monitoring problem inherent in event-triggered mechanisms but also effectively reduces redundant data transmission and significantly lowers network bandwidth occupancy. This is crucial for UAV swarm systems with limited bandwidth resources, effectively alleviating network congestion, improving swarm collaboration efficiency and overall task performance, and making theoretical results more aligned with the needs of high-real-time engineering scenarios.

[0025] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, embodiments of the present invention are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0026] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the embodiments. It should be understood that the following drawings only illustrate certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other relevant drawings can be obtained based on these drawings without paying any creative work.

[0027] Figure 1 This is a flowchart illustrating the binary consistency control method for UAV swarms under lag full-state constraints in an embodiment of the present invention.

[0028] Figure 2 This is a communication topology diagram of the drone cluster in an embodiment of the present invention;

[0029] Figure 3 This is an attitude angle trajectory diagram of the follower in the drone swarm in an embodiment of the present invention;

[0030] Figure 4 This is a diagram showing the angular velocity trajectory of a follower in a drone swarm in an embodiment of the present invention.

[0031] Figure 5This is a synchronization error trajectory diagram of the drone swarm follower in an embodiment of the present invention;

[0032] Figures 6-9 The control input curves for the UAV swarm followers QUAV1-QUAV4 in this embodiment of the invention are shown respectively.

[0033] Figures 10-13 The diagrams show the control input triggering intervals for the UAV cluster followers QUAV1-QUAV4 in this embodiment of the invention. Detailed Implementation

[0034] 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 some embodiments of the present invention, but not all embodiments.

[0035] Please refer to Figure 1 This invention provides a binary consensus control method for UAV swarms under hysteresis full-state constraints, comprising the following steps:

[0036] Step S1: Establish an attitude dynamics model of the UAV subject to hysteresis full-state constraints and input saturation.

[0037] Specifically, the attitude dynamics model of each UAV is characterized as a second-order system. The state vector of this system includes roll angle, pitch angle, yaw angle, and their respective angular velocities and angular accelerations. The hysteresis full-state constraint means that each dimension of the system's state variables is confined between a pair of lower and upper bound functions that are time functions, forming a time-varying constraint boundary. Input saturation means that the control torque input applied to the UAV is limited within a known saturation boundary value.

[0038] The attitude dynamics model of the i-th UAV is:

[0039]

[0040] Where i = 1, 2, ..., Q, Q is the number of drones in the drone swarm; φ i , as well as Let θ be the roll angle, angular velocity, and angular acceleration of the i-th UAV; i , as well as Let ψ be the pitch angle, angular velocity, and angular acceleration of the i-th UAV; i , as well as Let Φ be the yaw angle, angular velocity, and angular acceleration of the i-th UAV; iφΦ iθ and Φ iψ Let δ be the control torque of the i-th UAV; ix δ iy and δ iz Let be the inertia coefficient of the i-th UAV. as well as Let be the aerodynamic damping coefficient of the i-th UAV. as well as This is due to an unknown external disturbance.

[0041] To facilitate controller design and stability analysis, the attitude dynamics model of the i-th UAV is rewritten as follows:

[0042]

[0043] in,

[0044]

[0045] Assume the state of the attitude subsystem of the i-th UAV Need to be Converging to the following compact set:

[0046]

[0047] Where α = 1, 2, 3, and It is a time-varying constraint boundary function with lag, and satisfies and

[0048] Suppose that the i-th drone is subject to the following input saturation constraint:

[0049]

[0050] in, Given the saturation boundary, ∈ iα This is the actual control input.

[0051] To address the input saturation constraint, the following approximation strategy is designed:

[0052]

[0053] in, And satisfy

[0054] Step S2: Construct the transfer function and the time-varying nonlinear transformation function, and establish the equivalent state-space equation.

[0055] Specifically, first, a transfer function is constructed and multiplied with the constrained original attitude state to obtain an intermediate transition state; finally, based on the intermediate transition state and the upper and lower bound functions of the hysteresis full state constraint, a time-varying nonlinear transfer function is constructed, thereby transforming the original constrained attitude dynamics system into an unconstrained equivalent state-space system.

[0056] Transfer function Designed as follows:

[0057]

[0058] in, The preset time, and meets the following conditions The design parameters are positive and satisfy the following conditions:

[0059] n is the system order. and The initial value is at time t=0.

[0060] The time-varying nonlinear transformation function is designed as follows:

[0061]

[0062] right Taking the time derivative, we get:

[0063]

[0064] Where α = 1, 2, 3,

[0065]

[0066] Based on the designed time-varying nonlinear transformation function, the equivalent state-space equation is defined as:

[0067]

[0068] in,

[0069]

[0070] Step S3: Construct a cooperative-competitive topology communication network and define coordinate transformation equations based on adaptive backstepping control technology.

[0071] Using directed graphs This represents the topology communication network of Q drones in a drone swarm, where and Let represent the set of points and the set of edges, respectively.

[0072] Let represent the neighborhood set of the i-th drone, and (s,i)∈ε represent that the i-th drone can receive information from the s-th drone; a is >0 indicates that the i-th drone and the s-th drone are in a cooperative relationship, a is <0 indicates that the i-th drone and the s-th drone are in competition.

[0073] Adjacency matrix is ​​defined as

[0074] The in-degree matrix is ​​defined as in

[0075] Directed graph Laplace matrix Defined as

[0076] If point set Binary satisfaction relation and binary set and Make a is ≥0, a is ≤0, Then it is called a directed graph. Structural equilibrium.

[0077] Diagonal matrix This represents the dichotomous property of the structural equilibrium diagram, where

[0078] definition in This indicates that the i-th drone is capable of receiving information from the leader. and Let i represent the i-th drone and the leader as having a cooperative relationship and a competitive relationship, respectively.

[0079] Based on adaptive backstepping control technology, the coordinate transformation equation is defined as:

[0080]

[0081] in, For the leader's output trajectory; For virtual controllers; As an auxiliary signal, β iα1 and β iα2 These are the first error variable and the second error variable, respectively.

[0082] Step S4: Construct the first Lyapunov function V based on the coordinate transformation equation. iα1 Constructing a Type II fuzzy logic system for the first interval Online approximation of the unknown function f iα Design parameter update law and virtual controller

[0083] Specifically, in the first step of the adaptive backstepping control technology, the design of the virtual controller is as follows: First, for the first error variable, a first Lyapunov function is constructed; then, a first interval type II fuzzy logic system is used to approximate the unknown nonlinear term in the derivative of the first Lyapunov function online; the virtual controller is designed to contain four parts added together, which are: a nonlinear feedback term related to the first error variable, a power term to ensure fixed-time convergence, an adaptive parameter estimation term for the unknown weights of the first interval type II fuzzy logic system, and a coupling term to compensate for the dynamic influence of adjacent UAVs.

[0084] Construct the first Lyapunov function V based on the coordinate transformation equation. iα1 for:

[0085]

[0086] in, Λ is a positive definite matrix iα1 The inverse matrix.

[0087] For the first Lyapunov function V iα1 By taking the derivative and calculating, we can obtain:

[0088]

[0089] in,

[0090] Constructing a Type II fuzzy logic system for the first interval Online approximation of the unknown function f iα We can obtain:

[0091]

[0092] Where, μ iα1 (c iα1 ) represents the fuzzy basis function vector. Approximation error o iα1 (c iα1 )satisfy The parameter is an unknown positive parameter.

[0093] Furthermore, the steps for designing the parameter update law are as follows: the weight update law of the first interval type II fuzzy logic system and the second interval type II fuzzy logic system, as well as the update law of other adaptive parameters, are all derived based on Lyapunov stability theory; specifically, the parameter update law of the fuzzy logic system weights is designed to be proportional to the product of the corresponding error variable and the fuzzy basis function, and includes a σ correction term to enhance robustness, thereby ensuring that all adaptive estimation parameters are bounded.

[0094] Parameter update law Designed as follows:

[0095]

[0096] Among them, Λ iα 1 is a positive definite matrix, d iα1 For positive design parameters, For unknown weights λ iα1 The estimated value.

[0097] Virtual Controller Designed as follows:

[0098]

[0099] in, and h iα1 For positive design parameters, 0 < c < 1.

[0100] Equation (13) and the parameter update law and virtual controller Substituting into equation (12), we can simplify to obtain:

[0101]

[0102] Step S5: Construct the second Lyapunov function V based on the coordinate transformation equation. iα2 Constructing an interval-type II fuzzy logic system Online approximation of unknown functions Design auxiliary signal Parameter update law Parameter update law intermediate control signal And a self-triggering communication protocol, and based on this, an adaptive fuzzy self-triggering fixed-time binary consensus controller ε is constructed. iα .

[0103] Specifically, firstly, a second Lyapunov function is constructed for the second error variable; then, an unknown nonlinear term in the derivative of the second Lyapunov function is approximated online using a second interval type-II fuzzy logic system; and an auxiliary signal is designed to suppress the influence of input saturation nonlinearity; finally, combined with a self-triggering communication protocol, a specific form of an adaptive fuzzy self-triggering fixed-time binary consensus controller is designed, which consists of a nonlinear function based on the second error variable, a gain function, and an intermediate control signal.

[0104] Based on the coordinate transformation equation in step S3, construct the second Lyapunov function V. iα2 for:

[0105]

[0106] in, Λ is a positive definite matrix iα2 The inverse matrix.

[0107] To address the input saturation problem, a smooth function with a known upper bound approximation error is used to approximate the saturation nonlinearity. Furthermore, in the design process of adaptive backstepping control, an auxiliary signal system is designed to actively compensate for the impact of this approximation error. The dynamics of this auxiliary signal system are driven by itself, a nonlinear term related to the control error, and the difference between the actual control input and the approximate control input, thereby ensuring the stability of the closed-loop system.

[0108] In this embodiment, the following auxiliary signals are designed. To suppress the effects of saturation nonlinearity:

[0109]

[0110] For the second Lyapunov function V iα2 Taking the derivative, we get:

[0111]

[0112] in,

[0113] Constructing a second-interval type-2 fuzzy logic system Approximate unknown function We can obtain:

[0114]

[0115] Where, μ iα2 (c iα2 ) represents the fuzzy basis function vector. Approximation error o iα2 (c iα2 )satisfy The parameter is an unknown positive parameter.

[0116] Design parameter update law and parameter update law for:

[0117]

[0118] Where, d iα2 , as well as For positive design parameters, It is a time-varying bounded function. For unknown weights λ iα2 The estimated value, Unknown parameters The estimated value.

[0119] The following design incorporates a self-triggering communication protocol: At each trigger moment, a control signal is calculated and applied to the UAV, and the next trigger moment is calculated based on the current dynamic information of the system. The next trigger moment is determined by adding a dynamic execution interval to the current trigger moment. The execution interval is calculated by dividing a hyperbolic tangent function with the current control signal as input by the maximum value of the current control signal's time change rate, and is adjusted by a preset positive design parameter. This design ensures that the interval between two consecutive trigger events is always positive, thereby avoiding the occurrence of Zeno's phenomenon.

[0120] In this embodiment, to reduce the burden on airborne bandwidth and improve the utilization rate of network bandwidth resources, a self-triggered communication protocol is designed as follows:

[0121]

[0122] in, 0 < g iα <1, and F iα For positive design parameters, The execution interval between two consecutive trigger signals. To control the rate of change of the interval, max{·} represents taking the maximum value, ∈ iα This is denoted as an adaptive fuzzy self-triggered fixed-time binary consensus controller.

[0123] Design intermediate control signals for:

[0124]

[0125] in, h iα2 , as well as These are positive design parameters.

[0126] for definition:

[0127]

[0128] in, And |Δ iα (t)|<1.

[0129] Equation (20) and parameter update law Parameter update law and intermediate control signals Substituting into equation (19), we can simplify to obtain:

[0130]

[0131] Step S6: Construct the overall Lyapunov function V, analyze the stability of the closed-loop attitude control system based on Lyapunov stability theory, determine the controller gain to be designed, and realize the fixed-time attitude binary consistency control of the UAV swarm under the lag full-state constraint.

[0132] In this embodiment, according to steps S5 and S6, a global Lyapunov function V, composed of the first and second Lyapunov functions of all UAVs, is constructed as follows:

[0133]

[0134] Combining equations (16) and (27), and differentiating the global Lyapunov function V, we obtain:

[0135]

[0136] in,

[0137] Based on Young's inequality, the calculation yields:

[0138]

[0139] Using the inequality lemma, we can calculate:

[0140]

[0141] Substituting equations (30) to (33) into equation (29), we can calculate:

[0142]

[0143] in,

[0144]

[0145] Based on Lyapunov stability theory, it can be concluded that the UAV closed-loop attitude subsystem is stable in a fixed time, and the residual set of the system solution is...

[0146] The upper bound of the steady-state time can be estimated as follows:

[0147] Error variable β iαj It will converge to the following interval:

[0148] It should be noted that the transfer function For interval The function is monotonically increasing and satisfies Due to virtual control signals and auxiliary signals It is bounded, and Therefore (0. Further, we can obtain that for the interval [0, ), for

[0149] Therefore, it can be concluded that Therefore, the invented control method achieves control over the unmanned aerial vehicle system state. Lag time-varying constraints.

[0150] definition

[0151] as well as Based on the topology communication network, we can obtain in, For the Kronecker product. Because Since it is neither singular nor singular, we can conclude that:

[0152]

[0153] in, For matrix The minimum singular value is obtained by selecting an appropriate controller gain, which ensures that the synchronization error converges to a small neighborhood near the origin within a fixed time.

[0154] According to equation (23), It is bounded. Therefore, the trigger interval between two consecutive executions can be derived. satisfy Therefore, the designed self-triggering communication protocol will not experience the Zeno phenomenon.

[0155] The effectiveness and feasibility of the binary consensus control method for UAV swarms under lag full-state constraints designed in this invention are explained in detail through simulation experiments conducted in the simulation software MATLAB R2020a / SIMULINK.

[0156] Assume a drone swarm system contains one leader and four followers, labeled "L0" and "QUAV1-QUAV4" respectively, with the following communication topology: Figure 2 As shown.

[0157] The parameters of the UAV attitude dynamics system and the external disturbances are selected as follows:

[0158] δ ix =δ iy =0.082 kg·m 2 δ iz =0.149 kg·m 2 ,

[0159] The initial conditions for the drone are selected as follows:

[0160] [φ1(0),θ1(0),ψ1(0)]=[-1.8,2.2,-2.1], [φ2(0),θ2(0),ψ2(0)]=[3,-2.2,1.6],

[0161] [φ3(0),θ3(0),ψ3(0)]=[1.1,-0.5,2.4], [φ4(0),θ4(0),ψ4(0)]=[-1.5,-2,2.2].

[0162] The leader's trajectory is selected as follows:

[0163] The time-varying constraint boundary is selected as follows:

[0164]

[0165] The controller parameters are selected as follows:

[0166] c = 997 / 1001 h iα1 =0.02, h iα2 =3, G iα =3, g iα =0.1, F iα =10.

[0167] The fuzzy membership function is selected as follows:

[0168]

[0169] The simulation results were plotted on Figures 3 to 13 . Figure 3 and Figure 4 The attitude angles (φ) of the follower drones QUAV1-QUAV4 in the drone swarm were plotted respectively. i ,θ i ,ψ i ) and angular velocity The trajectory diagram, based on simulation results, shows that the UAV swarm has achieved binary consensus control, and the UAV attitude subsystem state is within a specified time. Converging to the interval And it always remains within that area. Figure 5 Let β be the synchronization error of each follower in the drone swarm. iα1 Trajectory diagram. Simulation results show that the synchronization error achieved fixed-time convergence under different initial conditions. Figures 6 to 9 The control input curves for followers QUAV1-QUAV4 are shown. The simulation results show that the control signal is updated non-periodically based on a given self-triggering communication protocol under input saturation constraints. Figures 10 to 13 A trigger interval diagram of drone swarm followers was drawn, and the invented self-triggering communication protocol strictly avoids Zeno's phenomenon.

[0170] Based on simulation results, it can be concluded that the invented control method can achieve time-varying full-state constraint control of UAV swarm attitude under no feasibility constraints; it can achieve intermittent updates of control signals by deploying a self-triggering communication protocol, thereby reducing communication load; and it can strictly guarantee fixed-time binary consistency control of UAV swarm attitude under time-varying full-state constraints, input saturation, and bandwidth-limited constraints.

[0171] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A binary consensus control method for UAV swarms under lag-full-state constraints, characterized in that, Includes the following steps: S1. Establish the attitude dynamics model of each UAV in the UAV cluster. The model is subject to hysteresis full-state constraints and input saturation. S2. Construct a transfer function and a time-varying nonlinear transformation function to convert the original attitude state under the hysteresis full-state constraint into an unconstrained equivalent state, and establish an equivalent state space equation based on the equivalent state. S3. Construct a cooperative-competitive coexistence topology communication network for the UAV cluster, and define a coordinate transformation equation to characterize the system error based on the adaptive backstepping control method; S4. Based on the coordinate transformation equation, design an adaptive fuzzy self-triggering fixed-time binary consensus controller, including: using an interval type II fuzzy logic system to approximate the unknown nonlinear function online, designing a parameter update law to adjust the weights of the fuzzy logic system in real time, and designing a self-triggering communication protocol to realize the non-periodic update of the controller.

2. The binary consensus control method for UAV swarms under hysteresis full-state constraints according to claim 1, characterized in that, In step S1, establishing the attitude dynamics model of each UAV specifically involves: representing the attitude dynamics model of each UAV as a second-order system, the state vector of which includes roll angle, pitch angle, yaw angle, and their respective angular velocities and angular accelerations; the hysteresis full-state constraint means that each dimension of the system's state variables is confined between a pair of lower and upper bound functions that are time functions, forming a time-varying constraint boundary; the input saturation means that the control torque input applied to the UAV is restricted within a known saturation boundary value.

3. The binary consensus control method for UAV swarms under hysteresis full-state constraints according to claim 1, characterized in that, In step S2, the design of the time-varying nonlinear transformation function is as follows: First, a transfer function is constructed and multiplied with the constrained original attitude state to obtain an intermediate transformation state; finally, based on the intermediate transformation state and the upper and lower bound functions of the hysteresis full state constraint, the time-varying nonlinear transformation function is constructed, thereby transforming the original constrained attitude dynamics system into an unconstrained equivalent state-space system.

4. The binary consensus control method for UAV swarms under hysteresis full-state constraints according to claim 1, characterized in that, In step S3, constructing a cooperative-competitive coexistence topology communication network and defining the coordinate transformation equation specifically involves: using a directed graph that includes cooperative and competitive relationships to represent the communication topology of the UAV swarm; the coordinate transformation equation defines a first error variable and a second error variable, wherein the first error variable is a weighted combination based on the equivalent state of the current UAV, the equivalent states of its neighboring UAVs, and the leader's trajectory, and the weight sign depends on the cooperative or competitive relationship between individuals; the second error variable is defined as the difference between the time derivative of the current UAV's equivalent state and the virtual controller and auxiliary signals.

5. The binary consensus control method for UAV swarms under hysteresis full-state constraints according to claim 4, characterized in that, The first step of the adaptive backstepping control method includes designing the virtual controller, specifically: constructing a first Lyapunov function for the first error variable; using a first interval type II fuzzy logic system to approximate the unknown nonlinear term in the derivative of the first Lyapunov function online; the virtual controller is designed to contain four parts added together, which are: a nonlinear feedback term related to the first error variable, a power term to ensure fixed-time convergence, an adaptive parameter estimation term for the unknown weights of the first interval type II fuzzy logic system, and a coupling term to compensate for the dynamic influence of adjacent UAVs.

6. The binary consensus control method for UAV swarms under hysteresis full-state constraints according to claim 5, characterized in that, The second step of the adaptive backstepping control method includes designing the adaptive fuzzy self-triggering fixed-time binary consensus controller. Specifically, it involves: constructing a second Lyapunov function for the second error variable; using a second interval type-II fuzzy logic system to approximate the unknown nonlinear term in the derivative of the second Lyapunov function online; designing the auxiliary signal to actively suppress the nonlinear effects caused by input saturation; and finally, combining the self-triggering communication protocol, constructing the specific form of the controller based on the intermediate control signal. This controller is composed of a nonlinear function based on the second error variable multiplied by a gain function.

7. The binary consensus control method for UAV swarms under hysteresis full-state constraints according to claim 6, characterized in that, The steps of the design parameter update law are as follows: the weight update law of the first interval type II fuzzy logic system and the second interval type II fuzzy logic system, as well as the update law of other adaptive parameters, are all derived based on Lyapunov stability theory; the parameter update law of the fuzzy logic system weights is designed to be proportional to the product of the corresponding error variable and the fuzzy basis function, and includes a σ correction term to enhance robustness, thereby ensuring that all adaptive estimation parameters are bounded.

8. The binary consensus control method for UAV swarms under hysteresis full-state constraints according to claim 1, characterized in that, The steps for designing the self-triggering communication protocol include: at each triggering moment, calculating a control signal and applying it to the UAV, and determining the next triggering moment based on the current dynamic information of the system; the next triggering moment is obtained by adding the current triggering moment to a dynamically changing execution interval; the calculation of the execution interval includes: dividing the hyperbolic tangent function with the current control signal as input by the maximum value of the current control signal's time change rate, and being adjusted by a preset positive design parameter to ensure that the interval between two consecutive triggering events is always positive.

9. The binary consensus control method for UAV swarms under hysteresis full-state constraints according to claim 1, characterized in that, Also includes: The saturated nonlinearity is approximated by a smooth function with a known upper bound approximation error. Furthermore, in the design process of adaptive backstepping control, an auxiliary signal system is designed to actively compensate for the influence of this approximation error. The dynamics of this auxiliary signal system are driven by itself, the nonlinear term related to the control error, and the difference between the actual control input and the approximate control input, thereby ensuring the stability of the closed-loop system.

10. The binary consensus control method for UAV swarms under hysteresis full-state constraints according to claim 1, characterized in that, Also includes: Construct an overall Lyapunov function composed of the first and second Lyapunov functions of all UAVs; By differentiating the overall Lyapunov function and combining it with the inequality scaling method for stability analysis, the gain parameters of the controller to be designed are determined. It is also proven that under the action of the designed controller, all signals in the closed-loop system are uniformly bounded, and the synchronization error of the UAV swarm can converge to the residual set near the origin within a fixed time.

Images