A low-altitude unmanned aerial vehicle cooperative control method under an asynchronous joint connectivity topology

Abstract

The application provides a low-altitude unmanned aerial vehicle cooperative control method under an asynchronous joint communication topology, comprising: constructing a speed layer model and an attitude layer model of a fixed-wing unmanned aerial vehicle, converting the model, setting a reference instruction vector of the unmanned aerial vehicle cluster, constructing an asynchronous switching observer, designing a topology switching rule, designing a thrust event trigger control unit, designing a rudder event trigger control unit, and finally forming a complete unmanned aerial vehicle cluster control; the application proposes a new second-order distributed asynchronous switching observer, improves the accuracy and stability of state estimation; the application establishes a topology switching rule, ensures the stability of the system under the joint communication topology environment; the application also discloses a dynamic event trigger mechanism, so that when the system performance is lower than an acceptable threshold, the controller increases the update frequency to enhance the system stability; when the system is in a good operating state, the update frequency is reduced to reduce energy consumption, so that more stable and efficient cluster control is realized.

Description

Technical Field

[0001] This invention belongs to the field of unmanned aerial vehicle (UAV) formation control, specifically relating to a cooperative control method for low-altitude UAVs under an asynchronous joint connected topology. Background Technology

[0002] With the rapid development of the low-altitude economy, drone swarms are increasingly widely used in low-altitude scenarios such as logistics delivery, urban patrol, and emergency rescue. However, low-rise buildings, terrain obstruction, and multipath effects lead to unstable links between drones, making it difficult to maintain traditional synchronous communication. Therefore, a collaborative control method under asynchronous joint connectivity topology is urgently needed to ensure the swarm maintains consistency even under intermittent communication. Furthermore, drone swarms need to coordinate efficiently within limited spaces, and traditional centralized control is ill-suited to the distributed needs of low-altitude environments. Although patent CN108196583A, "Drone Swarm Control Method," provides a control method, it only considers ensuring swarm stability under a fixed communication topology. In practical applications, the communication network of multi-drone systems is susceptible to electromagnetic interference, environmental changes, and other adverse factors, making it difficult to guarantee communication reliability. Specifically, node failures, link disconnections, and signal attenuation can trigger frequent switching of communication topologies, affecting the stable coordination of the swarm. In complex environments, network signal interaction is also affected by uncertainties such as transmission delays, sensor failures, and bandwidth fluctuations, leading to asynchronous topology switching problems for drone swarms. When the switching time of the communication topology is inconsistent with the switching time of the controller or observer, the system may experience problems such as control signal transmission failure, increased state estimation deviation, or even control instability.

[0003] Current UAV topology switching control methods (Y. Yu, C. Chen, J. Guo, M. Chadli and Z. Xiang, “Adaptive Formation Control for Unmanned Aerial Vehicles With Collision Avoidance and Switching Communication Network,” IEEE Transactions on Fuzzy Systems, vol. 32, no. 3, pp. 1435-1445, 2024.) all assume that topology and controller / observer switching are synchronous. Patent CN118394098A, “A Consistency Control Method for Rotary-Wing UAV Swarms under Asynchronous Topology Switching,” presents a UAV control method for asynchronous topology switching, but all communication topology graphs in it are connected graphs, which imposes stringent requirements on UAV swarms.

[0004] Furthermore, traditional UAV control strategies typically employ a small, fixed sampling period to ensure control accuracy and system stability. However, this continuous high-frequency sampling mode introduces additional computational and communication burdens, increasing energy consumption and causing unnecessary wear on actuators, thus shortening their lifespan. To address this issue, researchers have proposed a series of event-triggered control methods (Y. Li, S. Dong and K. Li, “Fuzzy Adaptive Finite-Time Event-Triggered Control of Time-Varying Formation for Nonholonomic Multirobot Systems,” IEEE Transactions on Intelligent Vehicles, vol. 9, no.1, pp. 725-737, 2024.) and (Y. Tan, Y. Yuan, X. Xie, E. Tian and J. Liu, “Observer-Based Event-Triggered Control for Interval Type-2 Fuzzy Networked System With Network Attacks,” IEEE Transactions on Fuzzy Systems, vol. 31, no. 8, pp. 2788-2798, 2023.). These methods trigger control when specific conditions are met, reducing unnecessary control execution, thereby lowering energy consumption and extending system uptime. Patent CN119644759A, "Fixed-Time Fuzzy Adaptive Output Feedback Formation Collision Avoidance Control Method for Multi-UAV Systems," proposes an event-triggered approach to address the collision avoidance problem in multi-UAV systems. However, this triggering method still suffers from fixed triggering strategies and insufficient environmental adaptability, limiting its application in complex dynamic environments. Patent CN119440041A, "Improved Consistency Control Algorithm," ensures that the communication topology of the UAV swarm maintains consistent speed in the case of a minimum spanning tree, but it does not consider the attitude angles of the UAVs, which does not reflect actual UAV flight conditions. Furthermore, since low-altitude flight inevitably alters the communication structure of UAVs when passing through buildings or terrain obstructions, fixed-topology control methods can lead to UAV swarm instability in such situations. Summary of the Invention

[0005] To address the problems existing in current technologies, this invention proposes a cooperative control method for low-altitude unmanned aerial vehicles (UAVs) under an asynchronous jointly connected topology. The basic design idea is as follows: for a six-DOF fixed-wing UAV swarm system, dynamic models for the velocity and attitude layers are constructed separately; then, an asynchronous switching observer is designed to ensure that each UAV can estimate reference commands; subsequently, topology switching rules are designed based on the asynchronous switching observers involved to ensure swarm stability; finally, an event-triggered mechanism is designed, and the control inputs and adaptive update laws for the velocity and attitude layers are designed by reverse calculation to form an effective adaptive control scheme to achieve the desired system performance.

[0006] The specific steps of this invention are as follows:

[0007] Step 1: Construct the velocity layer model and attitude layer model of the fixed-wing UAV, that is, construct the velocity layer and attitude layer model of the fixed-wing UAV based on Newton's second law and Euler's momentum equation;

[0008] Step 2: Model conversion, that is, according to the definition of scalar velocity, the vector velocity dynamics model is converted into a scalar velocity dynamics model;

[0009] Step 3: Set the reference command vector for the UAV swarm, that is, construct the reference command vector based on the speed and desired Euler angles of the reference command for the fixed-wing UAV swarm;

[0010] Step 4: Construct an asynchronous switching observer, that is, design an asynchronous switching observer based on the stability theorem, differential equations, and inequality scaling method;

[0011] Step 5: Design topology switching rules, that is, design the topology switching rules of the drone cluster using the maximum dwell time, minimum dwell time, and asynchronous switching method;

[0012] Step 6: Design the thrust event triggering control unit, that is, design the control unit of the UAV velocity layer model based on Lyapunov stability theorem, back-thrust technology and adaptive control method;

[0013] Step 7: Design the control surface event triggering control unit, that is, design the control unit of the UAV attitude layer model based on Lyapunov stability theorem, back-propagation technology and adaptive control method;

[0014] Step 8: Form a complete UAV swarm control system, that is, pass the control signal of the thrust event triggering control unit in Step 6 to the velocity layer model in Step 1, and pass the control signal of the control surface event triggering control unit in Step 7 to the attitude layer model in Step 1, thus completing the control of the entire UAV swarm.

[0015] Compared with the prior art, the advantages of the present invention are as follows:

[0016] (1) This invention proposes a novel second-order distributed asynchronous switching observer to ensure that each UAV can accurately estimate the reference command. In addition, the observer solves the problem of non-differentiability of the virtual controller through second-order derivative processing, thereby improving the accuracy and stability of state estimation.

[0017] (2) In UAV swarm communication, topology switching is asynchronous, which leads to a mismatch between the switching time of the controller / observer and the topology, affecting system stability. This invention conducts an in-depth study on the intrinsic mechanism of asynchronous joint connectivity switching topology, focuses on analyzing the maximum dwell time in the mismatch phase and the minimum dwell time in the matching phase, and establishes topology switching rules to ensure the stability of the system in a joint connectivity topology environment.

[0018] (3) The present invention proposes a dynamic event triggering mechanism. When the system performance is lower than the acceptable threshold, the controller increases the update frequency to enhance system stability; when the system is in a better operating state, the update frequency is reduced to reduce energy consumption, thereby achieving more stable and efficient cluster control. Attached Figure Description

[0019] Figure 1 This is a schematic diagram of the overall control process in the method of the present invention;

[0020] Figure 2 This is a joint connected topology graph in an embodiment of the present invention.

[0021] Figure 3 This is a graph showing the trajectory of the asynchronous observer error in an embodiment of the present invention.

[0022] Figure 4 This is a graph showing the trajectory curves of velocity and attitude angle in an embodiment of the present invention;

[0023] Figure 5 This is a diagram showing the triggering times of various event triggering mechanisms in the embodiments of the present invention;

[0024] Figure 6 This is a three-dimensional diagram of a drone swarm in an embodiment of the present invention. Detailed Implementation

[0025] This invention provides an adaptive event-triggered control method for a fixed-wing UAV swarm, enabling the velocity and attitude of a six-DOF fixed-wing UAV swarm to track reference commands even under asynchronous joint connectivity topology switching. The method's technical concept is as follows: First, dynamic models for the velocity and attitude layers of the six-DOF fixed-wing UAV swarm system are constructed separately. Second, an asynchronous switching observer is designed to ensure that each UAV can estimate the reference command. Subsequently, topology switching rules are designed based on the asynchronous switching observer to ensure swarm stability. Finally, an event-triggered mechanism is designed, and the control inputs and adaptive update laws for the velocity and attitude layers are designed by reverse calculation to form an effective adaptive control scheme to achieve the desired system performance.

[0026] The specific steps of the method of the present invention are as follows:

[0027] Step 1: Construct the velocity and attitude layer models of the fixed-wing UAV. This involves building the velocity and attitude layer models of the fixed-wing UAV based on Newton's second law and Euler's momentum equation. This step generates the state parameters for subsequent steps. For a UAV swarm system consisting of N fixed-wing UAVs, for the... The velocity and attitude layer models of a fixed-wing UAV are as follows:

[0028] (1);

[0029] In the formula: the first formula describes the relationship between the unmanned vector velocity and spatial position; Indicates drone Spatial position vector, Indicates to Find the first derivative, since Spatial position is described by three-dimensional coordinates, therefore three-dimensional coordinates are defined as follows: ; For drones The attitude angle vector, derived from the roll angle Pitch angle and yaw angle Composition, that is ; This represents the transformation matrix between the body coordinate system and the ground coordinate system; For drones The vector velocity in the body coordinate system can be decomposed into velocities in three directions in the body coordinate system, namely... The second formula is the translation equation derived from Newton's second law. Indicates to Find the first derivative. Let be the attitude angular velocity vector, which is decomposed into angular velocities in three directions, i.e. ; It is the outer product matrix; For thrust vector, For the thrust control of the engine; The gravity vector It is the acceleration due to gravity; Indicates the mass of the drone; Indicates drone The aerodynamic force acting on the object in the airflow coordinate system. ; For dynamic pressure, For drones wing surface area; This is the transformation matrix from the body coordinate system to the airflow coordinate system; These represent three aerodynamic parameters; For bounded external disturbances of the velocity layer; The first term is an unknown function containing uncertainties in aerodynamic parameters; the third formula describes the relationship between the unmanned attitude angle and the attitude angular velocity. Indicates to Find the first derivative; The transformation matrix is ​​given; the fourth formula is derived from Euler's momentum equation. Indicates to Find the first derivative; This represents the aerodynamic torque experienced by the drone in the airflow coordinate system. ; , , There are three aerodynamic parameters; For wingspan; It is the average wing chord length; These are three control vectors for the control surfaces, where the three control surfaces are respectively... , , ; It is the rotational inertia matrix, specifically in the form of:

[0030] ;

[0031] , , , and Constant inertia coefficient. The control gain matrix has the following specific form:

[0032] ;

[0033] , , , and It is a dimensionless parameter. The resistance matrix has the following specific form:

[0034] ;

[0035] , , , and It is a dimensionless parameter. For bounded external disturbances. It is an unknown function term containing uncertainties in rotational inertia and aerodynamic parameters.

[0036] Step 2: Model conversion, i.e., according to the definition of scalar velocity, transforming the vector velocity dynamics model into a scalar velocity dynamics model. The input of this model is the actual thrust, and the output is scalar velocity. (Based on the definition of scalar velocity...) It can be seen that its first derivative is:

[0037] (2);

[0038] In the formula: Vector velocity The corresponding scalar velocity satisfies , It is an unknown function term containing uncertainties in aerodynamic parameters.

[0039] Step 3: Set the reference command vector for the UAV swarm. Its purpose is to enable the fixed-wing UAV swarm to fly collaboratively according to the reference command. The speed of the reference command is... The expected Euler angles are ,definition The reference command vector is a vector consisting of roll, pitch, yaw, and scalar velocity. Symbols are used. This represents the initial signal vector composed of the drone's physical state. Represents the first A drone.

[0040] Step 4: Construct an asynchronous switching observer. This involves designing an asynchronous switching observer based on stability theorems, differential equations, and inequality scaling methods. This observer allows each UAV to estimate the reference command, with its input being the observation state of neighboring UAVs and its output being the observer's state. The asynchronous switching observer model is represented as follows:

[0041] (3);

[0042] In the formula: , Both are in observer state. This represents the first-order observer state. This represents the second-order observer state; and Parameters that are greater than zero; To switch the delay signal; The consistency error between adjacent drones is defined as:

[0043] (4);

[0044] In the formula: To determine the connection weights between drones under a switching latency topology, if the drones Can receive drones The information, ; The connection weights between the UAV and the reference command under the switching latency topology; Indicates by arrive Integers.

[0045] Step 5: Design topology switching rules, specifically using maximum dwell time, minimum dwell time, and asynchronous switching methods to design topology switching rules for the UAV swarm. When the swarm's communication topology satisfies these rules, the UAV swarm can fly stably. The UAV communication topology is a jointly connected topology, meaning the union of all topologies in the topology graph set forms a valid spanning tree. To enable the UAV's asynchronous switching observer to asymptotically track reference commands, the topology switching rules are designed as follows:

[0046] (5);

[0047] In the formula: , and For design parameters, satisfy , and ; Indicates topology switching Minimum stay time; Indicates topology switching The maximum stay time, and ; The minimum dwell time for the matching phase, which is the phase in which the drone has successfully identified the topology; This represents the maximum dwell time during the non-matching phase, which is the phase in which the drone is identifying the topology. , The transition matrix for the reconstructed non-matching stage is designed as follows: That is, by The matrix formed, where: when satisfy:

[0048] (6);

[0049] when Timely satisfaction ; and To match the phase parameters, the following conditions must be met:

[0050] (7);

[0051] In the formula: .

[0052] Step 6: Design the thrust event triggering control unit, that is, design the control unit of the UAV velocity layer model based on Lyapunov stability theorem, back-thrust technology and adaptive control method. Its input is the output unit of Step 1, Step 2 and Step 3. The function of this step is to design the thrust control unit of the UAV.

[0053] Step 6.1: Design the transition control signal, i.e., design a continuous control signal based on Lyapunov's stability theorem, back-calculation techniques, and adaptive control methods. Its input is the output unit of steps 1, 2, and 3. Design the transition control signal. :

[0054] (8);

[0055] In the formula:

[0056] ; , , ; The control gain is greater than zero; , and There are three positive parameters, and their values ​​range from 1 to 2. Inside; For the radial basis functions of the fuzzy logic system. The following update rules must be met:

[0057] (9);

[0058] In the formula: For positive design parameters, express The initial state value, that is, in hour The value, This is the initial moment of the entire system.

[0059] Step 6.2: Design the event triggering mechanism. Based on event triggering theory, design the events proposed in this step so that the output of the transition control signal will be transmitted to the thrust unit of the dynamics in Step 1. The selection of the thrust controller and the design of the events are as follows:

[0060] (10);

[0061] In the formula: the first row of the formula represents the continuous transition control signal. The actual control input will only be updated if the second formula is met; the second formula describes the designed event. For a piecewise function to satisfy:

[0062] (11);

[0063] In the formula: , ; ; , and There are two positive parameters, and their values ​​range from 1 to 2. ; and There are two positive parameters, and their values ​​range from 1 to 2. ; ; The threshold parameters are for the design. Furthermore, according to... As can be seen from the dynamics, it dynamically adjusts according to the magnitude of the thrust. The value of this parameter, as shown in the first formula of this section, is directly related to the thrust; that is, the magnitude of the thrust affects its value. However, for traditional event-triggered mechanisms... It is a fixed value that does not change according to the magnitude of the thrust. Therefore, the designed event triggering mechanism is a dynamic event triggering mechanism, which solves the problems of fixed triggering strategies and insufficient environmental adaptability, and can be applied to complex dynamic environments.

[0064] Step 7: Design the control surface event triggering control unit, that is, design the control unit of the UAV attitude layer model based on Lyapunov stability theorem, backpropagation technology and adaptive control method. Its input is the output unit of Step 1 and Step 4. The purpose of this step is to design the control surface control unit for the UAV.

[0065] Step 7.1: Design the transient control signal, i.e., design a continuous control signal based on Lyapunov's stability theorem, back-calculation techniques, and adaptive control methods. Its input is the output unit from steps 1 and 4. Design the transient control signal. :

[0066] (12);

[0067] In the formula: , , ; To meet The positive definite matrix of the condition; For positive design parameters, select a parameter range of... ; , , For the radial basis functions of the fuzzy logic system, , For positive design parameters, select a parameter range of... ; The error state is defined as follows: , As a virtual controller, it is designed as follows:

[0068] (13);

[0069] In the formula: for A positive definite matrix, This is an error state.

[0070] Design an adaptive update law with the following parameters:

[0071] (14);

[0072] In the formula: , For positive design parameters, express The initial state value.

[0073] Step 7.2: Design the event triggering mechanism. Based on event triggering theory, design the events proposed in this step so that the output of the transition control signal will be transmitted to the control surface unit of the dynamics in Step 1. The following formula introduces the selection of the control surface controller and the design of the events. Since the selection of the controller and the design of the events for the three control surfaces are in the same form, a symbol is defined. It replaces the superscript or subscript of the symbols in the following formula. ,For example It can be expressed as Three controllers. The control surface controller and the event controller are designed as follows:

[0074] (15);

[0075] In the formula: the first formula describes the selection of the controller, that is, when a continuous transient control signal... The actual control input will only be updated if the second formula condition is met; the second formula is the designed event. For a piecewise function to satisfy:

[0076] (16);

[0077] In the formula: satisfy , , ; and For design parameters; and For design parameters; , Each corresponds to a different attitude angle, i.e., when When, it corresponds to the roll angle, when When, it corresponds to the pitch angle, when At that time, the corresponding angle is the yaw angle; The threshold parameters are for the design. Furthermore, according to... As can be seen from the definition, it dynamically adjusts according to the size of the control surface. The numerical value indicates that the designed event triggering mechanism is a dynamic event triggering mechanism, which solves the problems of fixed triggering strategies and insufficient environmental adaptability, and can be applied to complex dynamic environments.

[0078] Step 8: Form a complete UAV swarm control system. This involves transmitting the control signal from the thrust event triggering control unit in Step 6 to the velocity layer model in Step 1, and transmitting the control signal from the control surface event triggering control unit in Step 7 to the attitude layer model in Step 1, thus completing the control of the entire UAV swarm. First, through the topology switching rules in Step 5 and the switching observer in Step 4, each UAV can estimate the reference command from Step 3. This estimated reference command is then input to the control units in Steps 6 and 7. Furthermore, Steps 6 and 7 simultaneously obtain the UAV states from Steps 1 and 2, i.e. Steps 6 and 7 design transition control signals. Based on the event triggering mechanism designed in steps 6 and 7, when an event is met, the control signal is transmitted to the velocity layer model and attitude layer model from step 1 to complete the control of the entire UAV swarm.

[0079] Figure 1This is a schematic diagram of the overall control process of the present invention. Under the switching rules designed in step 5, the UAV attitude layer model composed of step 1 and the velocity layer model converted by step 2 can estimate the reference command in step 3 through the switching observer in step 4. Then, the controllers in steps 6 and 7 feed back to the UAV attitude layer model and velocity layer model in step 1 to complete the entire closed-loop control.

[0080] According to the specific implementation steps of the technical solution of the present invention, the following embodiment is given. The software testing environment of this embodiment is MATLAB 2022b, and the hardware testing environment is a computer with an Intel(R) Core(TM) i7-14700KF@3.40GHZ CPU, NVIDIA Geforce RTX4060 GPU, 8GB RAM, and a 1TB SATA hard drive. Consider a cluster system consisting of five drones, where... Figure 2 The communication topology of five UAVs is described, where the reference command refers to the UAV needing to reach the reference angle and velocity. The black arrows represent the second-order observer state calculated by equation (4). The drone represented by the beginning of the black arrow can transmit data to the drone represented by the end of the black arrow.

[0081] Step 1: Construct a six-degree-of-freedom kinematic and dynamic model of the fixed-wing UAV. This involves constructing the six-degree-of-freedom kinematic and dynamic model of the fixed-wing UAV based on Newton's second law and Euler's momentum equation. This step generates the state parameters for subsequent steps. Establish a six-degree-of-freedom fixed-wing UAV swarm system model as shown in equation (1). The model parameters are as follows: , , , , , , air density wing surface area ,span average wing chord length , , , , , , , and For angle of attack and sideslip angle, , , , , , , , , , .

[0082] Step 2: Model conversion, i.e., according to the definition of scalar velocity, transforming the vector velocity dynamics model into a scalar velocity dynamics model. The input of this model is the actual thrust, and the output is scalar velocity. (Based on the definition of scalar velocity...) It can be seen that its first derivative is:

[0083] (17);

[0084] In the formula, all parameters and symbols are described in step 1.

[0085] Step 3: Set the reference commands for the drone swarm. This enables the drone swarm to fly collaboratively according to the reference commands. Assuming 5 drones are performing a turn and climb mission, the selected reference commands are:

[0086] (18);

[0087] For a more concise description, the initial state vector of the drone swarm is set as follows: , , , , .

[0088] Step 4: Construct an asynchronous switching observer, i.e., design an observer that, to ensure each UAV can estimate the reference command, takes the observation state of neighboring UAVs as its input and outputs the observer state as its output. This observer is independent of the model and is designed independently. Establish a distributed asynchronous switching observer according to equation (3), where the parameters are selected as follows: , ; Figure 2 This diagram shows the joint connectivity switching topology, with the communication topology based on... Figure 2 The four communication topology scenarios are switched in the following order: It can be seen that the union of the four communication topologies forms a directed spanning tree.

[0089] Step 5: Design topology switching rules, that is, design the topology switching rules for the UAV swarm. When the communication topology of the swarm meets these rules, the UAV swarm can fly stably. Establish the topology switching rules of equation (5), with the parameters selected as follows: , , , , , , , .

[0090] Step 6: Design the thrust event trigger control unit, that is, design the input unit for the UAV speed. Its input is the output unit of Step 1, Step 2 and Step 3. The purpose of this step is to design the thrust control unit for the UAV.

[0091] Establish an adaptive cooperative thrust event triggering controller based on equation (8), where the parameters are selected as follows: , , , , , .

[0092] Establish the parameter adaptive update law of equation (9), where the parameter is chosen as follows: , Furthermore, the initial value for the adaptive selection parameters is... .

[0093] Establish the thrust event triggering mechanism of equation (10), where the parameter is selected as follows: , , , , , .

[0094] Step 7: Design the control surface event triggering control unit, that is, design the input unit for the UAV attitude angle. Its input is the output unit of Step 1 and Step 3. The purpose of this step is to design the control surface control unit for the UAV.

[0095] Establish an adaptive cooperative control surface event-triggered controller based on equation (12), where the parameters are selected as follows: , , , , Furthermore, the parameters of the virtual controller (13) are selected as follows: , , .

[0096] Establish the parameter adaptive update law of formula (14), where the parameter is selected as follows. , Furthermore, the initial value of the selected parameter is adaptive. .

[0097] Establish the rudder surface event triggering mechanism of equation (15), where the parameter is selected as follows: , , , , , .

[0098] Step 8: Establish complete UAV control. First, using the topology switching rules in Step 5 and the switching observer in Step 4, each UAV can estimate the reference command from Step 3. This estimated reference command is then input to the control units in Steps 6 and 7. Furthermore, Steps 6 and 7 simultaneously obtain the UAV states from Steps 1 and 2, i.e. Steps 6 and 7 design transition control signals. Based on the event triggering mechanism designed in steps 6 and 7, when the event is met, the control signal is transmitted to the dynamics system in step 1, completing the control of the entire UAV.

[0099] Figure 3 The estimated error trajectory formed by the asynchronous observer in this invention is shown, wherein Figure 3 The purpose of all subplots is to more clearly observe the changes in the trajectory. Furthermore, the x and y coordinates, curve styles, and trajectories of all subplots are consistent with their corresponding images. Figure 3 (a) shows the observer velocity error trajectory, representing the estimate of the reference velocity command for each UAV. The horizontal axis represents time, and the vertical axis represents the error value of the UAV for velocity observation. The smaller the error value, the better the estimation effect of the observer. Figure 3 (b) shows the observer roll angle error trajectory, which represents the estimate of each UAV to the reference roll angle command. The horizontal axis represents time, and the vertical axis represents the error value of the UAV for the roll angle observation. The smaller the error value, the better the estimation effect of the observer. Figure 3 (c) shows the observer pitch angle error trajectory, which represents the estimated value of each UAV for the reference pitch angle command. The horizontal axis represents time, and the vertical axis represents the error value of the UAV for the pitch angle observation. The smaller the error value, the better the estimation effect of the observer. Figure 3 (d) shows the yaw angle error trajectory of the observer, which represents the estimated value of each UAV for the reference yaw angle command. The horizontal axis represents time, and the vertical axis represents the error value of the UAV for the yaw angle observation. The smaller the error value, the better the estimation effect of the observer. Figure 3(e) shows the velocity error trajectory of a conventional observer, representing the estimated value of each UAV for the reference velocity command when using a conventional observer. The horizontal axis represents time, and the vertical axis represents the error value of the UAV's velocity observation. The smaller the error value, the better the estimation effect of the observer. Based on... Figure 3 (a) shows that the traditional observer cannot track the reference velocity command when the command changes, indicating that the traditional observer cannot adapt to the case of asynchronous joint connected topology. Figure 3 (f) shows the roll angle error trajectory of the conventional observer, which represents the estimated value of each UAV for the reference roll angle command when using the conventional observer. The horizontal axis represents time, and the vertical axis represents the error value of the UAV for the roll angle observation. The smaller the error value, the better the estimation effect of the observer. Figure 3 (g) shows the pitch angle error trajectory of the conventional observer, which represents the estimated value of each UAV for the reference pitch angle command when using the conventional observer. The horizontal axis represents time, and the vertical axis represents the error value of the UAV for the pitch angle observation. The smaller the error value, the better the estimation effect of the observer. Figure 3 (h) shows the yaw angle error trajectory of the traditional observer, which represents the estimated value of each UAV for the reference yaw angle command when using the traditional observer. The horizontal axis represents time, and the vertical axis represents the error value of the UAV for the yaw angle observation. The smaller the error value, the better the estimation effect of the observer. Compared with the existing traditional switching control method, the method of the present invention can be better applied to the asynchronous joint connected topology switching scenario. Figure 4 The speed / attitude trajectory of the drone is shown, where Figure 4 The purpose of all subplots is to more clearly observe the changes in the trajectory. Furthermore, the x and y coordinates, curve styles, and trajectories of all subplots are consistent with their corresponding images. Figure 4 (a) shows the speed trajectory of the UAV. The horizontal axis represents time, and the vertical axis represents the actual speed value of the UAV. The closer the speed is to the reference speed, the better the control effect. Figure 4 (b) shows the roll angle trajectory of the UAV. The horizontal axis represents time, and the vertical axis represents the actual roll angle value of the UAV. The closer it is to the reference roll angle, the better the control effect. Figure 4 (c) shows the pitch angle trajectory of the UAV. The horizontal axis represents time, and the vertical axis represents the actual pitch angle value of the UAV. The closer it is to the reference pitch angle, the better the control effect. Figure 4 (d) shows the yaw angle trajectory, with the horizontal axis representing time and the vertical axis representing the actual yaw angle value of the UAV. The closer the yaw angle is to the reference yaw angle, the better the control effect. Figure 4 It can be seen that the speed and attitude of the drone swarm can converge to the reference command. Figure 5The diagram shows a comparison of the tracking errors of UAV 1 using the proposed dynamic event triggering mechanism and the traditional event triggering mechanism, respectively. Figure 5 The purpose of all subplots is to more clearly show the observation trigger moment. Furthermore, the x and y coordinates, curve styles, and trajectories of all subplots are consistent with the corresponding images. Figure 5 (a) shows a comparison of the speed tracking performance of UAV 1 using the proposed dynamic event triggering mechanism and the traditional event triggering mechanism. The horizontal axis represents time and the vertical axis represents speed error. The smaller the error value, the better the tracking performance of the event triggering mechanism. Figure 5 (b) shows a comparison of the roll angle tracking performance of UAV 1 using the proposed dynamic event triggering mechanism and the traditional event triggering mechanism. The horizontal axis represents time and the vertical axis represents the roll angle error. The smaller the error value, the better the tracking performance of the event triggering mechanism. Figure 5 (c) shows a comparison of the pitch angle tracking performance of UAV 1 using the proposed dynamic event triggering mechanism and the traditional event triggering mechanism. The horizontal axis represents time and the vertical axis represents the roll angle error. The smaller the error value, the better the tracking performance of the event triggering mechanism. Figure 5 (d) shows a comparison of the yaw angle tracking performance of UAV 1 using the proposed dynamic event triggering mechanism and the traditional event triggering mechanism. The horizontal axis represents time, and the vertical axis represents the yaw angle error. The smaller the error value, the better the tracking performance of the event triggering mechanism. Figure 5 It can be seen that the method of the present invention has a better tracking effect compared with the existing traditional event triggering mechanism method. Figure 6 This illustrates the coordinated flight trajectory of a fixed-wing unmanned aerial vehicle (UAV) swarm. Figure 6 It can be seen that even under asynchronous joint connectivity topology, drone swarms can still fly collaboratively under specified commands.

[0100] from Figures 2 to 6 It can be seen that the speed and attitude errors of the UAV eventually converge to a small neighborhood of the origin, that is, the method of the present invention can enable the speed and attitude of a six-degree-of-freedom fixed-wing UAV swarm system to meet the consistency requirements.

[0101] The advantages of this invention are:

[0102] (1) This invention proposes a novel second-order distributed asynchronous switching observer to ensure that each UAV can accurately estimate the reference command, thereby improving the accuracy and stability of state estimation.

[0103] (2) This invention has studied the underlying mechanism of asynchronous joint connectivity switching topology in depth, focused on analyzing the maximum dwell time of the unmatched phase and the minimum dwell time of the matched phase, and established topology switching rules to ensure the stability of the system in the joint connectivity topology environment.

[0104] (3) The present invention proposes a dynamic event triggering mechanism. When the system performance is lower than the acceptable threshold, the controller increases the update frequency to enhance system stability; when the system is in a better operating state, the update frequency is reduced to reduce energy consumption, thereby achieving more stable and efficient cluster control.

Claims

1. A cooperative control method for low-altitude unmanned aerial vehicles (UAVs) under an asynchronous jointly connected topology, comprising: Step 1: Construct the velocity layer model and attitude layer model of the fixed-wing UAV, that is, construct the velocity layer and attitude layer model of the fixed-wing UAV based on Newton's second law and Euler's momentum equation; Step 2: Model conversion, that is, according to the definition of scalar velocity, the vector velocity dynamics model is converted into a scalar velocity dynamics model; Step 3: Set the reference command vector for the UAV swarm, that is, construct the reference command vector based on the speed and desired Euler angles of the reference command for the fixed-wing UAV swarm; Step 4: Construct an asynchronous switching observer, that is, design an asynchronous switching observer based on the stability theorem, differential equations, and inequality scaling method; Step 5: Design topology switching rules, that is, design the topology switching rules of the drone cluster using the maximum dwell time, minimum dwell time, and asynchronous switching method; Step 6: Design the thrust event triggering control unit, that is, design the control unit of the UAV velocity layer model based on Lyapunov stability theorem, back-thrust technology and adaptive control method; Step 7: Design the control surface event triggering control unit, that is, design the control unit of the UAV attitude layer model based on Lyapunov stability theorem, back-propagation technology and adaptive control method; Step 8: Form a complete UAV swarm control system, that is, pass the control signal of the thrust event triggering control unit in Step 6 to the velocity layer model in Step 1, and pass the control signal of the control surface event triggering control unit in Step 7 to the attitude layer model in Step 1, thus completing the control of the entire UAV swarm.

2. The method for cooperative control of low-altitude unmanned aerial vehicles under an asynchronous jointly connected topology as described in claim 1, characterized in that: The drone swarm system consisting of N fixed-wing drones in step 1, for the... The velocity and attitude layer models of a fixed-wing UAV are as follows: ； In the formula: the first formula describes the relationship between the unmanned vector velocity and spatial position; Indicates drone Spatial position vector, Indicates to Find the first derivative, since Spatial position is described by three-dimensional coordinates, therefore three-dimensional coordinates are defined as follows: ; For drones The attitude angle vector, derived from the roll angle Pitch angle and yaw angle Composition, that is ; This represents the transformation matrix between the body coordinate system and the ground coordinate system; For drones The vector velocity in the body coordinate system can be decomposed into velocities in three directions in the body coordinate system, namely... ; The second formula is the translation equation obtained through Newton's second law; Indicates to Find the first derivative. Let be the attitude angular velocity vector, which is decomposed into angular velocities in three directions, i.e. ; It is the outer product matrix; For thrust vector, This refers to the thrust control parameters of the engine. The gravity vector It is the acceleration due to gravity; Indicates the mass of the drone; Indicates drone The aerodynamic forces acting on the object in the airflow coordinate system. ; For dynamic pressure, For drones wing surface area; This is the transformation matrix from the body coordinate system to the airflow coordinate system; These represent three aerodynamic parameters; For bounded external disturbances of the velocity layer; The first term is an unknown function containing uncertainties in aerodynamic parameters; the third formula describes the relationship between the unmanned attitude angle and the attitude angular velocity. Indicates to Find the first derivative; The transformation matrix is ​​given; the fourth formula is derived from Euler's momentum equation. Indicates to Find the first derivative; This represents the aerodynamic torque experienced by the drone in the airflow coordinate system. ; , , There are three aerodynamic parameters; For wingspan; It is the average chord length; These are three control vectors for the control surfaces, where the three control surfaces are respectively... , , ; It is the rotational inertia matrix, specifically in the form of: ； , , , and Constant inertia coefficient The control gain matrix has the following specific form: ； , , , and For dimensionless parameters, The resistance matrix has the following specific form: ； , , , and For dimensionless parameters, For bounded external disturbances. It is an unknown function term containing uncertainties in rotational inertia and aerodynamic parameters.

3. The method for cooperative control of low-altitude unmanned aerial vehicles under an asynchronous jointly connected topology as described in claim 1, characterized in that: In step 2, the scalar velocity dynamics model takes the actual thrust as input and outputs a scalar velocity, based on the definition of scalar velocity. It can be seen that its first derivative is: ； In the formula: Vector velocity The corresponding scalar velocity satisfies , It is an unknown function term containing uncertainties in aerodynamic parameters.

4. The method for cooperative control of low-altitude unmanned aerial vehicles under an asynchronous jointly connected topology as described in claim 1, characterized in that: The speed of the reference instruction in step 3 is The expected Euler angles are ,definition The reference command vector is a vector consisting of roll angle, pitch angle, yaw angle, and scalar velocity.

5. The method for cooperative control of low-altitude unmanned aerial vehicles under an asynchronous jointly connected topology as described in claim 1, characterized in that: The asynchronously switched observer model in step 4 is represented as follows: ； In the formula: , Both are in observer state. This represents the first-order observer state. This represents the second-order observer state; and Parameters that are greater than zero; To switch the delay signal; The consistency error between adjacent drones is defined as: ； In the formula: To determine the connection weights between drones under a switching latency topology, if the drones Can receive drones The information, ; The connection weights between the UAV and the reference command under the switching latency topology; Indicates by arrive Integers.

6. The method for cooperative control of low-altitude unmanned aerial vehicles under an asynchronous jointly connected topology as described in claim 1, characterized in that: The topology switching rules in step 5 are designed as follows: ； In the formula: , and For design parameters, satisfy , and ; Indicates topology switching Minimum stay time; Indicates topology switching The maximum stay time, and ; The minimum dwell time for the matching phase, which is the phase in which the drone has successfully identified the topology; This represents the maximum dwell time during the non-matching phase, which is the phase in which the drone is identifying the topology. , The transition matrix for the reconstructed non-matching stage is designed as follows: That is, by The matrix formed, where: when satisfy: ； when Timely satisfaction ; and To match the phase parameters, the following conditions must be met: ； In the formula: .

7. The method for cooperative control of low-altitude unmanned aerial vehicles under an asynchronous jointly connected topology as described in claim 1, characterized in that: Step 6 includes: Step 6.1: Design the transient control signal, i.e., design the transient control signal based on Lyapunov's stability theorem, back-calculation techniques, and adaptive control methods. : ； In the formula: ; ; , , ; The control gain is greater than zero; , and There are three positive parameters, and their values ​​range from 1 to 2. Inside; For the radial basis functions of the fuzzy logic system, The following update rules must be met: ； In the formula: For positive design parameters, express The initial state value, that is, in hour The value, This is the initial moment of the entire system; Step 6.2: Design an event triggering mechanism, that is, design events based on event triggering theory so that the output of the transition control signal is transmitted to the thrust unit of the dynamic model in Step 1. The selection of the thrust controller and the design of the events are as follows: ； In the formula: the first row of the formula represents the continuous transition control signal. The actual control input will only be updated if the second formula is met; the second formula describes the designed event. For a piecewise function to satisfy: ； In the formula: , ; ; , and There are two positive parameters, and their values ​​range from 1 to 2. ; and There are two positive parameters, and their values ​​range from 1 to 2. ; ; The threshold parameters are for the design.

8. The method for cooperative control of low-altitude unmanned aerial vehicles under an asynchronous jointly connected topology as described in claim 1, characterized in that: Step 7 includes: Step 7.1: Design the transient control signal, i.e., design the transient control signal based on Lyapunov's stability theorem, back-calculation techniques, and adaptive control methods. : ； In the formula: , , ; To meet The positive definite matrix of the condition; For positive design parameters, select a parameter range of... ; , , For the radial basis functions of the fuzzy logic system, , For positive design parameters, select a parameter range of... ; The error state is defined as follows: , As a virtual controller, it is designed as follows: ； In the formula: for A positive definite matrix, This is an error state; Design an adaptive update law with the following parameters: ； In the formula: , For positive design parameters, express The initial state value; Step 7.2: Design an event triggering mechanism. Based on event triggering theory, design events that enable the output of the transition control signal to be transmitted to the control surface unit of the dynamics in Step 1. The control surface controller and the events are designed as follows: ； In the formula: the first formula describes the selection of the controller, that is, when a continuous transient control signal... The actual control input will only be updated if the second formula condition is met; the second formula is the designed event. For a piecewise function to satisfy: ； In the formula: satisfy , , ; and For design parameters; and For design parameters; , Each corresponds to a different attitude angle, i.e., when When, it corresponds to the roll angle, when When, it corresponds to the pitch angle, when At that time, the corresponding angle is the yaw angle; The threshold parameters are for the design.

9. The method for cooperative control of low-altitude unmanned aerial vehicles under an asynchronous jointly connected topology as described in claim 1, characterized in that: In step 8, the topology switching rules from step 5 and the switching observer from step 4 are used to estimate the reference command for step 3 for each UAV. The estimated reference command is then input to the control unit in steps 6 and 7. Steps 6 and 7 simultaneously obtain the UAV states from steps 1 and 2, i.e. Steps 6 and 7 design transition control signals; then, based on the event triggering mechanism designed in steps 6 and 7, when the event is met, the control signal is transmitted to the velocity layer model and attitude layer model in step 1 to complete the control of the entire UAV cluster.

10. The method for cooperative control of low-altitude unmanned aerial vehicles under an asynchronous jointly connected topology as described in claim 9, characterized in that: The model parameters in step 1 are as follows: , , , , , , air density wing surface area ,span average wing chord length , , , , , , , and For angle of attack and sideslip angle, , , , , , , , , , ; In step 3, the five drones perform a turning and climbing mission, and the reference command is selected as follows: ； The initial state vector of the drone swarm is set as follows: , , , , ; In step 4, the structure , ; The communication topology switching sequence is as follows The union of the four communication topologies forms a directed spanning tree; In step 5 , , , , , , , ; In step 6 , , , , , ; , ; ; , , , , , ; In step 7 , , , , ; , , ; , ; ; , , , , , .

Images