A distributed UAV formation method and system for resisting complex wind field interference

Abstract

The present invention discloses a distributed UAV formation method and system for resisting composite wind field interference, which relates to the technical field of UAV formation control, and comprises the following steps: obtaining composite wind speed, inputting the composite wind speed into a pre-established single quad-rotor UAV dynamic model, and outputting a UAV model with wind interference based on an enhanced adaptive control method, wherein the composite wind speed includes a constant wind speed and a wind shear speed; solving the UAV model with wind interference based on a Lur'e nonlinear estimation method, outputting an estimated interference result of the UAV formation, and realizing time-varying formation control based on the estimated interference result of the UAV formation, wherein non-smooth functions and environmental interference data are introduced in the solution process for auxiliary solution.

Description

Technical Field

[0001] The present invention relates to the technical field of unmanned aerial vehicle formation control, and in particular to a distributed unmanned aerial vehicle formation method and system that resists complex wind field interference. Background Art

[0002] Since the 1950s, a lot of exploration and research has been conducted in the field of UAV formation control at home and abroad. Driven by application scenarios and program development, the demand for time-varying formations is growing. There is an endless stream of research on linear and nonlinear UAV distributed formations. Over time, the focus of controlling multiple UAV formations has shifted from ideal environments to more practical scenarios. Therefore, ensuring the stability of UAVs under unknown interference or different meteorological conditions has become a necessary condition for widespread application.

[0003] As applications and scenarios evolve, the need for dynamic adaptation of relative positioning between UAVs becomes increasingly important. This adaptability allows real-time adjustment of the overall formation configuration based on changing external environmental conditions, ultimately improving mission execution conditions. This evolving necessity highlights the importance of in-depth research on the time-varying formation control challenges faced by multi-UAV systems. One notable approach uses the Euler-Lagrangian system to solve the time-varying tracking problem, while another study further considers external interference to optimize time-varying tracking performance. In actual flight scenarios, UAVs will inevitably encounter a variety of interference factors, which often have nonlinear or even unknown behavioral characteristics. Detecting the upper limit of such interference in a time-varying state is particularly challenging.

[0004] With the in-depth understanding of multi-UAV formations, research has shifted from linear systems to more complex nonlinear systems. For example, for the time-varying linear formation problem in second-order systems, people have designed robust controllers to solve the external disturbance problem using the extended state observer method. Similarly, the problem of designing robust tracking controllers for time-varying linear formations has also been studied considering model parameter uncertainties and external disturbances. In addition, a group of researchers have explored the second-order consensus of multi-intelligent systems with nonlinear dynamics under complex external conditions. On this basis, nonlinear multi-agent formation control under event-triggered conditions has also been explored. Based on the above research, it becomes more challenging to achieve stable flight in nonlinear multi-UAV formations when external disturbances are taken into account. Therefore, studying the coupling between nonlinearity and external disturbances remains a key area to be further explored.

[0005] Environmental factors in actual scenarios, especially weather conditions, can pose significant challenges to UAV formations and interfere with the smooth conduct of formation flight. Despite this, the impact of environmental factors on multiple UAVs has not been widely studied, and traditionally more attention has been paid to uncertain interference. Although some studies have considered the stability of single UAV flight in a wind field environment, the study did not involve the status of multiple UAVs. Therefore, it is of great significance to study multi-UAV formations under the influence of wind field environments and unknown disturbances. In addition, the intricate interactions between various environmental factors exacerbate the complexity, affecting both the UAV formation model and the positioning accuracy of the UAV. This complex coupling relationship poses a major research challenge that requires further attention. Summary of the invention

[0006] In order to solve the deficiencies mentioned in the above background technology, the purpose of the present invention is to provide a distributed UAV formation method and system that can resist complex wind field interference.

[0007] In the first aspect, the purpose of the present invention can be achieved by the following technical solution: a distributed UAV formation method for resisting composite wind field interference, the method comprising the following steps:

[0008] Obtaining a composite wind speed, inputting the composite wind speed into a pre-established single quad-rotor UAV dynamic model, and outputting a UAV model with wind disturbance based on an enhanced adaptive control method, wherein the composite wind speed includes a constant wind speed and a wind shear speed;

[0009] Based on the Lur'e nonlinear estimation method, the UAV model with wind interference is solved, and the UAV formation estimation interference result is output. Time-varying formation control is realized based on the UAV formation estimation interference result, wherein non-smooth functions and environmental interference data are introduced in the solution process for auxiliary solution.

[0010] In combination with the first aspect, in some implementations of the first aspect, the method further includes: the constant wind speed Calculation: represents the wind direction angle, and the constant wind speed is recorded as and the constant wind direction angle is , the constant wind speed and direction are expressed as ;

[0011] Wind shear speed It is expressed as:

[0012]

[0013] in, is the maximum outflow velocity, and By changing the size of these three parameters for the corresponding radius and height, tiny downdrafts of different intensities and ranges are generated. , , x represents the actual radius distance, Indicates the height of a small downdraft;

[0014] Composite wind speed .

[0015] In combination with the first aspect, in certain implementations of the first aspect, the method further includes: setting conditions of the pre-established single quadrotor drone dynamic model:

[0016] A quadrotor drone has a rigid and symmetrical structure that remains unchanged during flight; the center point of the quadrotor drone is precisely aligned with its center of gravity and remains unchanged throughout the flight; the gravity field and air density in the quadrotor drone's operating environment remain unchanged.

[0017] In combination with the first aspect, in certain implementations of the first aspect, the method further includes: the drone model with wind interference is as follows:

[0018]

[0019]

[0020] in, Represents the flight position information of the drone in the inertial coordinate system. It represents the nonlinear speed information of the UAV in the inertial coordinate system. The three-axis speeds are are the yaw angle, roll angle and pitch angle of the drone respectively. represents the wind field resistance coefficient, represents the air resistance coefficient, represents gravity, represents the wind speed, Represents the quality of the drone, are the speeds of the four rotors, represents the inertia moment of a single motor, Represents the residual speed, Represents the elements of the quadrotor UAV's moment of inertia matrix, represents the thrust control amount, represent the control variables of roll, pitch and yaw respectively, represents the lift coefficient of the rotor, represents the anti-torque coefficient of the rotor, Represents the distance from the center of mass to the center of the rotor.

[0021] In combination with the first aspect, in some implementations of the first aspect, the method further includes: the UAV model with wind interference first considers the motion model of the quad-rotor nonlinear UAV system on a two-dimensional plane, Unmanned aerial vehicles, the leader marked , N followers are marked as ,According to the above conditions, the motion model of the quadrotor nonlinear UAV system on the two-dimensional plane is

[0022]

[0023] make , we get the following second-order integral formula , where the input of the drone control is expressed as , the unknown nonlinear part is expressed as , and obtain the results of the UAV second-order integrator model;

[0024] set up For the indivual status, , For the indivual The control input of Then the results of the UAV second-order integrator model are transformed into a nonlinear system as shown below ;

[0025] The nonlinear system is transformed into the following formula ,in , other interferences are as follows , is the approximate small residual of the unknown disturbance in the wind field, which is bounded as ,in is an unknown positive constant, represents the noise received by the drone in the wind field, It is also an independent standard white noise with zero mean.

[0026] In combination with the first aspect, in some implementations of the first aspect, the method further includes: the noise received by the drone in the wind farm ,in represents the change of wind speed over time, indicates that the noise intensity is a positive definite diagonal matrix, is a Gaussian white noise process with a mean of zero, and the boundary is set to ,in is an unknown positive constant.

[0027] In combination with the first aspect, in some implementations of the first aspect, the method further includes: the nonlinear system based on Lur'e nonlinearity is designed as follows: ,in

[0028]

[0029]

[0030] Communication diagram content: The communication network between the drones consists of an undirected graph Description, where . is an edge set, It is a picture The adjacency matrix of . Indicates indivual and indivual Assume that is simple, that is, for The Laplacian matrix is ​​defined as follows ,in is a diagonal matrix with diagonal elements ,in , if there is a path between any two points in an undirected graph, the graph is called a strongly connected graph.

[0031] The control protocol of the drone is designed as follows

[0032]

[0033] in, Represents a constant, , , , is the feedback gain matrix, P is a positive definite matrix, It is estimated The upper limit of the adaptive gain, is the adaptive gain vector, Representative The formation control compensation vector of the UAVs, and , is the expected state of the ith and jth UAVs, is the actual state of the i-th UAV and the j-th UAV, is a non-smooth function.

[0034] In combination with the first aspect, in some implementations of the first aspect, the method further includes: the non-smooth function: the non-smooth function is derived from the sliding mode function , defined as

[0035] ,

[0036] in

[0037] Consider the desired formation configuration ;

[0038] Suppose there is a given positive scalar , and there exists a positive definite matrix with set dimensions , there exists an identity matrix I, the matrix satisfies the following LMI: , the system is stable.

[0039] In combination with the first aspect, in some implementations of the first aspect, the method further includes: the nonlinear utilization formula based on Lur'e To achieve nonlinear control, is a given symmetric positive definite matrix.

[0040] In a second aspect, in order to achieve the above-mentioned purpose, the present invention discloses a distributed UAV formation system for resisting composite wind field interference, comprising:

[0041] A model processing module is used to obtain a composite wind speed, input the composite wind speed into a pre-established single quad-rotor UAV dynamic model, and output a UAV model with wind interference based on an enhanced adaptive control method, wherein the composite wind speed includes a constant wind speed and a wind shear speed;

[0042] The UAV formation module is used to solve the UAV model with wind interference based on the Lur'e nonlinear estimation method, output the UAV formation estimation interference result, and realize time-varying formation control based on the UAV formation estimation interference result, wherein non-smooth functions and environmental interference data are introduced in the solution process for auxiliary solution.

[0043] Beneficial effects of the present invention:

[0044] This invention tailors a control system model for quadrotor drones operating in complex wind environments. The model combines Lur'e nonlinearity, takes into account internal and external nonlinear interference effects, and surpasses existing drone models, thereby enhancing adaptability and robustness; a new method for drone formation control using non-smooth functions is introduced, which estimates wind field interference more accurately and adaptively than neural network-based methods. Our method aims to better conform to the nature of interference, while emphasizing the impact of wind speed on wind field noise to achieve accurate interference estimation; in response to the complex coupling problem between wind force and wind noise, our research proposes an enhanced adaptive control method for effectively controlling drone formations under wind field interference. By adding control parameters and parameter matrices to the control protocol, we enhance the robustness of the control system and its ability to resist external interference, ultimately achieving precise tracking control of the system state. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative work.

[0046] Figure 1 It is a schematic flow chart of the method of the present invention;

[0047] Figure 2 This is a graph showing the trajectory change experiment results of the UAV formation of the present invention;

[0048] Figure 3 This is a graph showing the experimental results of the speed variation trend of the UAV formation in the x-axis direction of the present invention;

[0049] Figure 4 This is a graph showing the experimental results of the speed variation trend of the UAV formation in the y-axis direction of the present invention;

[0050] Figure 5 It is a graph of experimental results of the control process of the x-axis control protocol of the present invention;

[0051] Figure 6 It is a graph of experimental results of the control process of the y-axis control protocol of the present invention;

[0052] Figure 7 It is a diagram of the experimental results of the process of estimating wind field noise using the adaptive law of the present invention;

[0053] Figure 8 It is a diagram of the experimental results of the process of adaptive gain compensation for unknown interference of the present invention;

[0054] Fig. 9 It is a schematic diagram of the system structure of the present invention. DETAILED DESCRIPTION

[0055] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0056] Embodiment 1:

[0057] The following is an introduction to the relevant terms involved in the embodiments of the present application:

[0058] UAV formation: UAV formation flight is a technology that has been pursued by military forces and scientific researchers in various countries. Its key technical issues mainly include formation design, aerodynamic coupling, dynamic adjustment of formation, trajectory planning, information exchange and formation flight control strategy, which are the research hotspots of various UAV research and development powers.

[0059] There are many benefits to drone formation flying. The mode of multiple drones flying in formation and coordinating reconnaissance and combat can improve the success probability of a single aircraft in a single combat mission to a certain extent. In terms of military reconnaissance, target strike, communication relay, electronic countermeasures, battlefield assessment and harassment temptation, drone formation flying can improve the efficiency of completing a single mission. For example, in a target strike mission, multiple attack / combat drones can simultaneously carry out all-round attacks on the same target from different angles, expand the hit range, and improve the lethality and hit rate; they can also attack multiple enemy targets at the same time, disrupt the enemy's air defense system, and improve the timeliness of the battle; and when performing reconnaissance missions, by adjusting the reconnaissance equipment, such as the working angle of the camera, the drone group can complete the task of taking all-round three-dimensional photos of the target in a shorter time.

[0060] like Figure 1 As shown, a distributed UAV formation method for resisting complex wind field interference includes the following steps:

[0061] S101: obtaining a composite wind speed, inputting the composite wind speed into a pre-established single quad-rotor UAV dynamic model, and outputting a UAV model with wind interference based on an enhanced adaptive control method, wherein the composite wind speed includes a constant wind speed and a wind shear wind speed;

[0062] The specific process is as follows:

[0063] Step 1: Consider the constant wind To build a model, Here, constant wind force refers to the wind force that remains stable and uniform throughout the process, representing the basic size of the wind force with stable characteristics. We denote the constant wind speed as and the constant wind direction angle is Therefore, the basic wind speed and direction can be expressed as .

[0064] Step 2: Establish a wind shear model. We use the microburst numerical model to fit the wind shear engineering model of the flight simulation. In this case, the wind shear field is regarded as an incompressible, inviscid and non-rotating potential energy flow field, and temperature changes are not considered. This wind shear model is expressed as ,in, is the maximum outflow velocity, and By changing the size of these three parameters for the corresponding radius and height, tiny downdrafts of different intensities and ranges can be generated. , , x represents the actual radius distance. The applicable conditions of the model are and .

[0065] Step 3: Consider the composite wind speed and determine the effect of this wind speed on the UAV formation as .

[0066] Step 4: Build a dynamic model of a single quadrotor drone with the following assumptions: a) The quadrotor drone has a rigid and symmetrical structure that remains unchanged during flight. b) The center point of the quadrotor drone is precisely aligned with its center of gravity and remains unchanged throughout the flight. c) The gravity field and air density in the quadrotor drone’s operating environment remain unchanged.

[0067] Step 5: Based on the dynamic model of a single quad-rotor drone, the wind speed is incorporated into the quad-rotor drone dynamic model as additional wind force, thereby establishing the following drone model formula with wind interference:

[0068]

[0069]

[0070] in, Represents the flight position information of the drone in the inertial coordinate system. Represents the nonlinear speed information of the drone in the inertial coordinate system. The three-axis speeds are are the yaw angle, roll angle and pitch angle of the drone respectively. Represents the wind field resistance coefficient. Represents the coefficient of air resistance. represents gravity, represents the wind speed, Represents the quality of the drone. are the speeds of the four rotors, represents the inertia moment of a single motor, Represents the residual speed, Represents the elements of the inertia matrix of the quadrotor drone. This paper assumes that the quadrotor drone is a rigid symmetrical structure as a whole. Represents the thrust control amount. represent the control variables of roll, pitch and yaw respectively, represents the lift coefficient of the rotor (the magnitude depends on the material and diameter of the propeller, etc.), represents the anti-torque coefficient of the rotor, Represents the distance from the center of mass to the center of the rotor.

[0071] Step 6: Based on the above model construction, first consider the motion model of the quadrotor nonlinear UAV system on a two-dimensional plane. Unmanned aerial vehicles, the leader marked , N followers are marked as According to the above conditions, the motion model of the quadrotor nonlinear UAV system on a two-dimensional plane is:

[0072]

[0073] Step 7: Order , then we can get the following second-order integral formula , where the input of the drone control is expressed as , the unknown nonlinear part is expressed as .

[0074] Get the result of the second-order integrator model of the UAV Step 8: The result of the second-order integrator model of the UAV is shown in the above formula. In order to uniformly represent the position and velocity state of the UAV, let For the indivual status, , For the indivual The control input only uses the local information of neighboring drones. The nonlinear part is Then, step 7 is transformed into a general nonlinear system as follows: .

[0075] Step 9: Because the UAV formation flight under wind field interference needs to consider not only the influence of wind speed but also the influence of wind field noise and unknown interference in the environment, based on step 8, the nonlinear system is converted into the following formula ,in , other interferences are as follows . is an approximate small residual of the unknown disturbance in the wind field, which is assumed to be bounded ,in is an unknown positive constant. represents the noise received by the drone in the wind field, It is also an independent standard white noise with zero mean, and the variance may be affected by wind speed and some possible unknown random factors.

[0076] Step 10: The wind field noise is expressed as ,in Represents the change of wind speed over time. It means that the noise intensity is a positive definite diagonal matrix. is a Gaussian white noise process with a mean of zero, representing the influence of other possible random factors on the noise. Assume that its boundary is ,in is an unknown positive constant.

[0077] S102: Solve the UAV model with wind interference based on the Lur'e nonlinear estimation method, output the UAV formation estimation interference result, and realize time-varying formation control based on the UAV formation estimation interference result, wherein non-smooth functions and environmental interference data are introduced in the solution process for auxiliary solution. The nonlinearity of UAVs and wind fields is abstracted into Lur'e nonlinearity for estimation and control, and non-smooth functions are introduced to apply adaptive gains to adaptively adjust and control unknown interferences generated by other environments. At the same time, an adaptive law is introduced for wind field noise to achieve a more accurate and more adaptive estimation and control. Compared with the neural network-based method, this method is closer to the essence of noise interference, especially emphasizing the influence of wind speed on wind field noise, so as to achieve more accurate interference estimation.

[0078] The specific process is as follows:

[0079] Step 11: Based on Lur'e nonlinearity, the nonlinear system is designed as ,in:

[0080]

[0081]

[0082]

[0083] It can be seen that the nonlinear system of the UAV conforms to Lur'e nonlinearity and is suitable for the characteristics of Lur'e nonlinearity.

[0084] Consider using Lur'e nonlinearity to abstract the nonlinear terms, which conforms to the nonlinear form of UAV and wind field, and use the formula To achieve nonlinear control, is a given symmetric positive definite matrix.

[0085] Step 12: The communication network between the drones consists of an undirected graph Description, where . is an edge set, It is a picture The adjacency matrix of . Indicates indivual and indivual Assume that is simple, that is, for The Laplacian matrix is ​​defined as follows ,in is a diagonal matrix with diagonal elements ,in , if there is a path between any two points in an undirected graph, the graph is called a strongly connected graph.

[0086] Step 13: Based on the above steps and considering time-varying formation control, the control protocol of the UAV is designed as follows

[0087]

[0088] in, Represents a constant, , , , is the feedback gain matrix, P is a positive definite matrix, It is estimated The upper limit of the adaptive gain, is the adaptive gain vector used to estimate the unknown constant vector Representative The formation control compensation vector of the UAVs, and , is the expected state of the ith and jth UAVs, is the actual state of the i-th UAV and the j-th UAV, is a non-smooth function.

[0089] Consider using adaptive law and adaptive gain to estimate and control noise interference and external unknown interference. Consider the external unknown interference state as:

[0090]

[0091] Step 14: Non-smooth functions are derived from sliding mode functions , defined as

[0092] ,

[0093] in

[0094] Step 15: Consider the desired formation configuration

[0095] Step 16: Consider a nonlinear system consisting of multiple drones whose formations vary over time. Assume that there is a given positive scalar , and there exists a positive definite matrix with appropriate dimension , there exists an identity matrix I that satisfies the following LMI: , the system is asymptotically stable, which means that the multi-UAV system forms an ideal formation and realizes time-varying formation control.

[0096] The new LMI matrix is ​​considered to achieve stable flight of UAV formation, and the Lyapunov stability proof is applied to prove the feasibility, thereby realizing the adaptive stable flight of UAV formation under complex wind field interference.

[0097] Specifically, the present invention is further described below by way of embodiments:

[0098] The simulation results are as follows Figures 2 to 8 The detailed presentation focuses on Figure 2 , which vividly depicts the entire process of the drone formation from its initial state to its final stable configuration. Figure 2At the beginning of the simulation (t=0), the UAV formation presents a chaotic state. This is because the UAV group has not yet completed the communication synchronization and position calibration with each other at the moment of startup, and the slight differences in the initial conditions (such as speed, direction, etc.) make their distribution in space appear scattered and disordered. However, as the simulation time progresses (t=1 to t=2), it can be clearly observed that the UAV formation begins to show a certain degree of organization and coordination. The UAVs continuously exchange information such as position and speed through the built-in communication protocol and control algorithm, and adjust their flight status according to this information. In this process, the UAV formation gradually transitions from disorder to order, and the relative positions between the formation members begin to tend to be reasonably distributed. At t=3, the UAV formation has formed a clear parallelogram configuration. In the parallelogram configuration, the UAVs maintain sufficient distance to avoid collisions with each other, and can support each other to jointly deal with external interference. At this time, although the wind field interference still exists, the UAV formation has achieved stable configuration maintenance through the collaborative control strategy. Further observation of the formation at t=5 shows that the formation configuration has hardly changed compared to t=3. This shows that the UAV formation can not only form a stable configuration in a short period of time, but also maintain the stability of the configuration under continuous wind field interference. This stability is one of the important features of UAV formation technology and the key to its ability to complete complex tasks. In summary, Figure 2 Through the display of five key time points, the whole process of the UAV formation from initial chaos to the final formation of a stable parallelogram configuration is described in detail. This process fully demonstrates the coordination, adaptability and stability of UAV formation technology, and provides strong support for the widespread application of UAVs in complex environments.

[0099] Due to the influence of the wind field, the flight speed of the drones had some vibrations at first, but gradually became consistent during the flight and finally reached stability. It can be seen that the four-drone formation has a better effect, and they can transmit information to each other and maintain stability. Figure 3 and Figure 4 The speed changes are described along the x-axis and y-axis respectively. Under the influence of the complex and changeable wind field, the flight speed of the four drones did experience a significant vibration period at the beginning of takeoff. This vibration is mainly caused by the unevenness and changes in the wind field, which makes each drone receive different degrees of thrust or resistance during the flight, thus affecting the stability of its speed. However, as the flight time goes on, each drone in the drone formation shows good adaptability and coordination. Through the built-in communication system and control protocol, they transmit flight status information to each other in real time, including key parameters such as speed, position, and attitude. The sharing of this information enables the drones to establish close connections and jointly cope with the challenges brought by the wind field. Figure 3In the figure, we can clearly see the speed change along the x-axis. At first, due to the large and unstable component of the wind field in the x-axis direction, the speed curves of each drone showed obvious fluctuations. But as time goes by, these fluctuations gradually weakened, the speed curves tended to be smooth, and finally reached a relatively stable value. This shows that the drone formation has successfully minimized the impact of the wind field in the x-axis direction through collaborative control. Similarly, in Figure 4 In the figure, the speed change along the y-axis direction also shows a similar trend. Although the wind field in the y-axis direction may have different characteristics and intensities, the UAV formation can still maintain a high degree of coordination. Through information exchange and adjustment between each other, the speed of each UAV in the y-axis direction gradually converges and eventually reaches a stable state. In summary, the four-UAV formation showed excellent stability when facing the influence of the wind field. Through real-time information sharing and precise control adjustments, they successfully overcame the challenges brought by the wind field and achieved a smooth transition and final stabilization of the flight speed.

[0100] The control protocol showed a high degree of flexibility and adaptability in dealing with external disturbances suffered by the four drones. Figure 5 and Figure 6 The control protocol is a highly complex and variable interference mechanism that is used to control the flight status of each drone. The control protocol integrates an improved adaptive algorithm to monitor the flight status of each drone and the intensity and characteristics of external interference in real time. Based on this information, the control protocol calculates the optimal control instructions, aiming to minimize the impact of interference on the drone and maintain the overall stability and consistency of the formation. With the precise execution of control instructions and real-time information sharing between drones, the flight status of each drone begins to converge gradually. In this process, the control protocol continuously adjusts and optimizes control parameters to ensure that each drone can participate in formation flight in the best state. At the same time, through an effective collaborative control strategy, the relative position and speed between drones are precisely controlled, allowing the entire formation to maintain a compact and orderly structure. Finally, after a period of adaptation and adjustment, the control error of each drone tends to 0, that is, their actual flight status is almost completely consistent with the expected status. This indicates that the control protocol has successfully overcome the influence of external interference and achieved high-precision control of the drone's position and speed.

[0101] The estimated results of wind field noise are as follows: Figure 7 As shown. Due to the randomness of noise, It also experiences significant fluctuations at the beginning of the estimation, but eventually stabilizes. This indicates that the noise is eventually well estimated. Figure 7 In the presented wind farm noise estimation results, the dynamic changes of noise characteristics and its estimation process can be observed in detail. In the initial stage, due to the randomness and unpredictability of the wind farm noise itself, the estimated value showed significant fluctuations. This fluctuation reflects the instability of the noise source and the complexity of the environment, which makes the initial estimation process face certain challenges. However, with the gradual adjustment of the adaptive law, the estimation system begins to gradually adapt to and learn the characteristics of the noise, and gradually reduces the deviation between the estimated value and the actual noise. Finally, after a period of optimization, the estimated value of the wind farm noise tends to stabilize and the fluctuation amplitude is significantly reduced. This result not only shows that the estimation system effectively captures the noise characteristics, but also reflects the stability and accuracy of the algorithm in complex environments. At this point, it can be considered that the noise has been well estimated, and its estimated value can more realistically reflect the actual situation of the wind farm noise. From Figure 8 It can be seen that the adaptive gain gradually converges to a stable value, which means that the unknown disturbance has also been controlled. It can be clearly observed that the adaptive gain (a parameter used in the control system to dynamically adjust the system response to compensate for unknown environmental disturbances) has undergone a process of gradual adjustment and finally stabilized at a specific value. In the initial stage, the system may be faced with the influence of external unknown disturbances, resulting in instability of system performance or increased difficulty in prediction. In order to meet these challenges, the adaptive gain begins to play its role and automatically adjusts its value according to the real-time performance of the system. This process is dynamic and aims to find the gain value that can most effectively offset the influence of unknown disturbances through continuous trial and error and optimization. Over time, the adjustment of the adaptive gain gradually becomes flat, indicating that the system has begun to adapt to the current environment and conditions. At this stage, small changes in the gain value no longer cause significant fluctuations in system performance, but gradually converge to a relatively stable level. This stable value reflects the system's best compensation ability for unknown disturbances in the current state. Finally, when the adaptive gain fully converges to a stable value, it can be considered that the system has successfully controlled the unknown disturbance. This means that the system performance has reached a relatively stable state and can more effectively respond to changes in the external environment while maintaining high operating efficiency and accuracy.

[0102] The simulation focuses on studying the stable flight of UAVs in a composite wind field. First, the performance of the UAV formation configuration before and after completion under coupled interference is compared. Second, an attempt is made to compare the speeds of four UAVs to achieve UAV formation flight. Finally, the interference changes are compared to achieve stable control of the interference.

[0103] Second, as Fig. 9As shown, in order to achieve the above-mentioned purpose, the present invention discloses a distributed UAV formation system for resisting composite wind field interference, comprising:

[0104] The model processing module 11 is used to obtain the composite wind speed, input the composite wind speed into a pre-established single quad-rotor UAV dynamic model, and output a UAV model with wind interference based on an enhanced adaptive control method, wherein the composite wind speed includes a constant wind speed and a wind shear speed;

[0105] The UAV formation module 12 is used to solve the UAV model with wind interference based on the Lur'e nonlinear estimation method, output the UAV formation estimation interference result, and realize time-varying formation control based on the UAV formation estimation interference result, wherein non-smooth functions and environmental interference data are introduced in the solution process for auxiliary solution.

[0106] Based on the same inventive concept, the present invention also provides a computer device, which includes: one or more processors, and a memory for storing one or more computer programs; the program includes program instructions, and the processor is used to execute the program instructions stored in the memory. The processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. It is the computing core and control core of the terminal, which is used to implement one or more instructions, specifically for loading and executing one or more instructions in a computer storage medium to implement the above method.

[0107] It should be further explained that, based on the same inventive concept, the present invention also provides a computer storage medium, on which a computer program is stored, and the computer program is executed by the processor to execute the above method. The storage medium can adopt any combination of one or more computer-readable media. The computer-readable medium can be a computer-readable signal medium or a computer-readable storage medium. The computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electrical, magnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples (non-exhaustive list) of computer-readable storage media include: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present invention, a computer-readable storage medium can be any tangible medium containing or storing a program, which can be used by an instruction execution system, device or device or used in combination with it.

[0108] In the description of this specification, the description with reference to the terms "one embodiment", "example", "specific example", etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present disclosure. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.

[0109] The above shows and describes the basic principles, main features and advantages of the present disclosure. Those skilled in the art should understand that the present disclosure is not limited by the above embodiments, and the above embodiments and descriptions are only for explaining the principles of the present disclosure. Without departing from the spirit and scope of the present disclosure, the present disclosure may have various changes and improvements, and these changes and improvements fall within the scope of the present disclosure to be protected.

Claims

1. A distributed UAV formation method for resisting complex wind field interference, characterized in that: The method comprises the following steps: Obtaining a composite wind speed, inputting the composite wind speed into a pre-established single quad-rotor UAV dynamic model, and outputting a UAV model with wind disturbance based on an enhanced adaptive control method, wherein the composite wind speed includes a constant wind speed and a wind shear wind speed; The UAV model with wind interference is as follows: U4=K M (Ω1 2 -Ω2 2 +Ω3 2 -Ω4 2 ). Among them, x i ,y i ,z i Represents the flight position information of the drone in the inertial coordinate system. It represents the nonlinear speed information of the UAV in the inertial coordinate system. The three-axis speeds are ψ, φ, θ are the yaw angle, roll angle and pitch angle of the UAV respectively, K w Represents the wind field resistance coefficient, K D represents the air resistance coefficient, g represents gravity, V b represents the wind speed, m represents the mass of the drone, Ω i , where i = 1, 2, 3, 4, are the speeds of the four rotors, I r represents the inertia moment of a single motor, Ω=-Ω1+Ω2-Ω3+Ω4 represents the residual speed, I x ,I y ,I z represents the element of the inertia matrix of the quadrotor drone, U1 represents the thrust control variable, U2, U3, U4 represent the control variables of roll, pitch and yaw respectively, and K T represents the lift coefficient of the rotor, K M represents the anti-torque coefficient of the rotor, l represents the distance from the center of mass to the center of the rotor; The UAV model with wind interference first considers the motion model of the quadrotor nonlinear UAV system on a two-dimensional plane. There are N+1 unmanned aerial vehicles, the leader is marked as 0, and the N followers are marked as i=1, 2, ..., N. According to the above conditions, the motion model of the quadrotor nonlinear UAV system on a two-dimensional plane is Let p i (t) = [x i ,y i ] T , The following second-order integral formula is obtained The input of the drone control is expressed as The unknown nonlinear part is expressed as The results of the UAV second-order integrator model were obtained; Assume ε i (t)∈R n is the state of the i-th UAV, ε i =[p i T ,v i T ] T ,u i (t)∈R m is the control input of the i-th UAV, and the nonlinear part is represented by f i (t)∈R m Then the results of the UAV second-order integrator model are transformed into a nonlinear system as shown below The nonlinear system is transformed into the following formula where A∈R n×n ,B∈R n×m , other interferences are as follows i (t)∈R m ,d i (t)∈R m , d i (t) is the approximate small residual of the unknown disturbance in the wind field, which is bounded by ||d i (t)||<τ, where τ is an unknown positive constant, w i (t) represents the noise received by the UAV in the wind field, w i (t) is also an independent standard white noise with a mean of zero; The noise received by the drone in the wind farm Where V w =[V b ,V b ] T represents the change of wind speed over time, ξ i (t)∈R,i=1,2,...,N indicates that the noise intensity is a positive definite diagonal matrix, ∈(t) is a Gaussian white noise process with a mean of zero, and the boundary is set to ||∈(t)||<γ, where γ is an unknown positive constant; Communication graph content: The communication network between N+1 drones is composed of an undirected graph G(A)=(V,E,A), where V=[v,i=1,2,3,...,n], is an edge set, A=[a ij ]∈R n×n is the adjacency matrix of graph G, a ij ≠0 indicates the connection between the i-th UAV and the j-th UAV, for i∈V,a ii =0, the Laplace matrix is ​​as follows L = C R -A, where C R is a diagonal matrix with diagonal elements Where i = 1, 2, ..., N; The UAV model with wind interference is solved based on the Lur'e nonlinear estimation method, and the UAV formation estimation interference result is output, and the time-varying formation control is realized based on the UAV formation estimation interference result, wherein non-smooth functions and environmental interference data are introduced in the solution process for auxiliary solution; The nonlinear system design based on Lur'e nonlinearity is in The control protocol of the drone is designed as follows: in, α represents a constant, α>0,K∈R m×n , K=-B T P is the feedback gain matrix, P is a positive definite matrix, r i (t) is the estimated d i (t) Upper limit adaptive gain, is the adaptive gain vector, ξ(t),η i (t)∈R m represents the formation control compensation vector of the i-th UAV, and h i (t), h j (t) is the expected state of the i-th UAV and the j-th UAV, ε i (t), ε j (t) is the actual state of the i-th UAV and the j-th UAV, is a non-smooth function; The non-smooth function is derived from the sliding mode function Defined as in Consider the desired formation configuration Suppose there is a given positive scalar α, and there exists a positive definite matrix P with set dimension, there exists an identity matrix I, and the matrix satisfies the following LMI: The system is stable.

2. The distributed UAV formation method for resisting complex wind field interference according to claim 1 is characterized in that: The constant wind speed V d Calculation: represents the wind direction angle, the constant wind speed is recorded as V0 and the constant wind direction angle is The constant wind speed and direction is denoted by V d =V0, Wind shear speed V q It is expressed as: Among them, v max is the maximum outflow velocity, r m and h m For the corresponding maximum radius and height, changing the size of these three parameters will produce tiny downdrafts of different intensities and ranges. q =x-r0, r0=2km, x represents the actual radius distance, h q Indicates the height of a small downdraft; Composite wind speed V b =V d +V q .

3. The distributed UAV formation method for resisting complex wind field interference according to claim 1 is characterized in that: The setting conditions of the pre-established single quadrotor UAV dynamic model are: A quadrotor drone has a rigid and symmetrical structure that remains unchanged during flight; the center point of the quadrotor drone is precisely aligned with its center of gravity and remains unchanged throughout the flight; the gravity field and air density in the quadrotor drone's operating environment remain unchanged.

4. The distributed UAV formation method for resisting complex wind field interference according to claim 1 is characterized in that: The nonlinear formula based on Lur'e Φ T (σ(t))(Φ(σ(t))-K1σ(t))≤0 to achieve nonlinear control, where K1∈R m×m is a given symmetric positive definite matrix.

5. A distributed UAV formation system that resists complex wind field interference, characterized in that: A distributed UAV formation method for resisting composite wind field interference according to any one of claims 1 to 4 is adopted, comprising: A model processing module is used to obtain a composite wind speed, input the composite wind speed into a pre-established single quad-rotor UAV dynamic model, and output a UAV model with wind interference based on an enhanced adaptive control method, wherein the composite wind speed includes a constant wind speed and a wind shear speed; The UAV model with wind interference is as follows: U4=K M (Ω1 2 -Ω2 2 +Ω3 2 -Ω4 2 ). Among them, x i ,y i ,z i Represents the flight position information of the drone in the inertial coordinate system. It represents the nonlinear speed information of the UAV in the inertial coordinate system. The three-axis speeds are ψ, φ, θ are the yaw angle, roll angle and pitch angle of the UAV respectively, K w Represents the wind field resistance coefficient, K D represents the air resistance coefficient, g represents gravity, V b represents the wind speed, m represents the mass of the drone, Ω i , where i = 1, 2, 3, 4, are the speeds of the four rotors, I r represents the inertia moment of a single motor, Ω=-Ω1+Ω2-Ω3+Ω4 represents the residual speed, I x ,I y ,I z represents the element of the inertia matrix of the quadrotor drone, U1 represents the thrust control variable, U2, U3, U4 represent the control variables of roll, pitch and yaw respectively, and K T represents the lift coefficient of the rotor, K M represents the anti-torque coefficient of the rotor, l represents the distance from the center of mass to the center of the rotor; The UAV model with wind interference first considers the motion model of the quadrotor nonlinear UAV system on a two-dimensional plane. There are N+1 unmanned aerial vehicles, the leader is marked as 0, and the N followers are marked as i=1, 2, ..., N. According to the above conditions, the motion model of the quadrotor nonlinear UAV system on a two-dimensional plane is make The following second-order integral formula is obtained The input of the drone control is expressed as The unknown nonlinear part is expressed as The results of the UAV second-order integrator model were obtained; Assume ε i (t)∈R n is the state of the i-th UAV, ε i =[p i T ,v i T ] T ,u i (t)∈R m is the control input of the i-th UAV, and the nonlinear part is represented by f i (t)∈R m Then the results of the UAV second-order integrator model are transformed into a nonlinear system as shown below The nonlinear system is transformed into the following formula where A∈R n×n ,B∈R n×m , other interferences are as follows i (t)∈R m ,d i (t)∈R m , d i (t) is the approximate small residual of the unknown disturbance in the wind field, which is bounded by ||d i (t)||<τ, where τ is an unknown positive constant, w i (t) represents the noise received by the UAV in the wind field, w i (t) is also an independent standard white noise with a mean of zero; The noise w received by the UAV in the wind field i (t) = V w (t)ξ i (t)+ò(t), where V w =[V b ,V b ] T represents the change of wind speed over time, ξ i (t)∈R, i=1,2,...,N indicates that the noise intensity is a positive definite diagonal matrix, ∈(t) is a Gaussian white noise process with a mean of zero, and the boundary is set to ‖∈(t)‖<γ, where γ is an unknown positive constant; Communication graph content: The communication network between N+1 drones is composed of an undirected graph G(A)=(V,E,A), where V=[v,i=1,2,3,...,n], is an edge set, A=[a ij ]∈R n×n is the adjacency matrix of graph G, a ij ≠0 indicates the connection between the i-th UAV and the j-th UAV, for i∈V,a ii =0, the Laplace matrix is ​​as follows L = C R -A, where C R is a diagonal matrix with diagonal elements Where i = 1, 2, ..., N; A UAV formation module is used to solve the UAV model with wind interference based on the Lur'e nonlinear estimation method, output the UAV formation estimation interference result, and realize time-varying formation control based on the UAV formation estimation interference result, wherein non-smooth functions and environmental interference data are introduced in the solution process for auxiliary solution; The nonlinear system design based on Lur'e nonlinearity is in The control protocol of the drone is designed as follows: in, α represents a constant, α>0,K∈R m×n , K=-B T P is the feedback gain matrix, P is a positive definite matrix, r i (t) is the estimated d i (t) Upper limit adaptive gain, is the adaptive gain vector, ξ(t),η i (t)∈R m represents the formation control compensation vector of the i-th UAV, and h i (t), h j (t) is the expected state of the i-th UAV and the j-th UAV, ε i (t), ε j (t) is the actual state of the i-th UAV and the j-th UAV, is a non-smooth function; The non-smooth function is derived from the sliding mode function Defined as in Consider the desired formation configuration Suppose there is a given positive scalar α, and there exists a positive definite matrix P with set dimension, there exists an identity matrix I, and the matrix satisfies the following LMI: The system is stable.

Images