An unmanned aerial vehicle group completely distributed network optimal control method and system

Abstract

The application discloses a kind of unmanned aerial vehicle group completely distributed network optimal control method and system, the method includes: with unmanned aerial vehicle group in non-matching interference under the goal of completing group control, based on unmanned aerial vehicle group error model constructs unmanned aerial vehicle group optimal control model;Unmanned aerial vehicle group optimal control model is solved using HJB equation, and the ideal optimal control signal of unmanned aerial vehicle group is obtained;Based on ACI framework, the parameter matrix in the ideal optimal control signal of unmanned aerial vehicle group is updated, and the actual optimal control signal of unmanned aerial vehicle group is obtained.The application can overcome the limitations of traditional control method, improve the adaptability and control accuracy of unmanned aerial vehicle group in air base station transmission limited and disturbance environment.

Description

Technical Field

[0001] This invention relates to an optimal control method and system for a fully distributed network of unmanned aerial vehicles (UAVs), belonging to the field of UAV communication. Background Technology

[0002] With the continuous advancement of technology, airborne base stations are playing an increasingly important role in UAV swarms. However, achieving communication transmission and emergency control among UAV groups in mission environments still faces many challenges. Traditional UAV swarms, when transmitting and receiving communication relay signals, are prone to incomplete signal reception due to asynchronous movement speeds of airborne base stations. Furthermore, some parameters in the transmitted signal structure are unknown, resulting in incomplete and difficult-to-observe data received by other UAVs. Simultaneously, UAV swarms are susceptible to external disturbances such as multi-angle, multi-modal gusts, and internal disturbances such as complex electromagnetic interference during group movement, leading to formation trajectory divergence and collision risks. Traditional motion control methods are still unable to effectively address internal and external mismatched interference in the inner and outer control signals of UAVs, making it difficult to ensure the system completes the target mission within the desired optimal motion state.

[0003] To address these issues, existing technologies employ robust control strategies for fully distributed UAV networks with mismatched interference. While interference can be suppressed through robust control methods, the sources and structures of interference received by the inner and outer loops differ, and state coupling exists between the position and velocity loops. Therefore, traditional robust control methods still have limitations in handling heterogeneous mismatched interference. Existing technologies also employ strategies that estimate unknown terms of the transmitted signal by designing state observers and constructing strict feedback systems based on the linearized structure of the transmitted signal to approximate the trajectory of the restricted signal. However, the structured parameters of some airborne base station signals are unknown, making it impossible to obtain the complete signal based on a single state approximation. Therefore, such methods have limitations in handling restricted signal transmission from UAV networks.

[0004] Therefore, there is an urgent need for a fully distributed network optimal control method and system for unmanned aerial vehicle (UAV) groups that can effectively cope with internal and external heterogeneous mismatched interference and accurately process restricted signal transmission, thereby ensuring stable communication and precise motion control of UAV groups. Summary of the Invention

[0005] The purpose of this invention is to overcome the shortcomings of the prior art and provide an optimal control method and system for a fully distributed network of unmanned aerial vehicle (UAV) swarms. This method can effectively overcome the limitations of traditional control methods, improve the adaptability and control accuracy of UAV swarms in environments with limited transmission from airborne base stations and disturbances, and provide strong technical support for the reliability of UAV swarm mission execution.

[0006] To achieve the above objectives, the present invention is implemented using the following technical solution:

[0007] On one hand, the present invention provides a fully distributed network optimal control method for unmanned aerial vehicle (UAV) groups, comprising:

[0008] Construct an error model for unmanned aerial vehicle (UAV) groups;

[0009] With the goal of enabling unmanned aerial vehicle (UAV) groups to complete unit control under unmatched disturbances, an optimal control model for UAV groups is constructed based on the UAV group error model.

[0010] The optimal control model of the UAV group is solved using the HJB equation to obtain the ideal optimal control signal of the UAV group;

[0011] Based on the ACI framework, the parameter matrix in the ideal optimal control signal of the UAV group is updated to obtain the actual optimal control signal of the UAV group.

[0012] Optionally, the process of obtaining the unmanned aerial vehicle (UAV) group error model includes:

[0013] Construct a communication network topology model for unmanned aerial vehicle (UAV) groups, including airborne base stations;

[0014] Establish a distance model between the airborne base station and the UAV group, a state model of the airborne base station, and a state model of the UAV group;

[0015] Construct an unmatched interference auxiliary function, and based on the unmatched interference auxiliary function and the UAV group state model, construct an UAV group state model under unmatched interference;

[0016] Based on the UAV group communication network topology model, the distance model between the airborne base station and the UAV group, the airborne base station state model, and the UAV group state model under mismatched interference, an UAV group error model is constructed.

[0017] On the other hand, the present invention provides a fully distributed network optimal control system for unmanned aerial vehicle (UAV) groups, comprising:

[0018] The data acquisition module is used to build an error model for the unmanned aerial vehicle (UAV) group.

[0019] The optimal control model construction module is used to construct the optimal control model of the UAV group based on the UAV group error model, with the goal of completing the group control under unmatched disturbances.

[0020] The first solution module is used to solve the optimal control model of the UAV group using the HJB equation to obtain the ideal optimal control signal of the UAV group;

[0021] The second solution module is used to update the parameter matrix in the ideal optimal control signal of the UAV group based on the ACI framework, so as to obtain the actual optimal control signal of the UAV group.

[0022] Compared with the prior art, the beneficial effects achieved by the present invention are as follows:

[0023] This invention proposes a fully distributed network optimal control method and system for unmanned aerial vehicle (UAV) groups. The method aims to achieve UAV group control under unmatched disturbances. It constructs an optimal control model for the UAV group based on the UAV group error model and combines the HJB (Hamilton-Jacobi-Bellman) equations and ACI (Actor-Critic-Identifier) ​​equations. The optimal control model of the UAV group is solved using the actuator-commentator-identifier (ACI) framework to obtain the actual optimal control signal of the UAV group. The UAV group error model is constructed based on the distance model between the airborne base station and the UAV group, the airborne base station state model, and the UAV group state model under mismatched interference. The UAV group state model under mismatched interference is constructed by handling different interferences in the UAV system's position and velocity loops using linear system theory. A linear decoupling method is used to decompose the communication transmission constraints of the airborne base station into system-determinable parameters, thereby constructing the airborne base station state model. Finally, the optimal control model of the UAV group is solved using the HJB and ACI algorithms. The actual optimal control signal enables optimal cooperation between the UAV groups, allowing the formation to suppress the effects of mismatched interference during movement. This enables the UAV group to complete precise movements in complex environments under conditions of limited information transmission, while also possessing interference suppression performance. Attached Figure Description

[0024] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly described below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort, wherein:

[0025] Figure 1 The flowchart shown is a fully distributed network optimal control method for unmanned aerial vehicle (UAV) groups in an embodiment of the present invention.

[0026] Figure 2 The flowchart shown is a process for obtaining the unmanned aerial vehicle (UAV) group error model in an embodiment of the present invention.

[0027] Figure 3 The diagram shown is a two-dimensional planar motion trajectory diagram of the unmanned aerial vehicle (UAV) group in this embodiment of the invention.

[0028] Figure 4 The diagram shows the trajectory of the unmanned aerial vehicle (UAV) group in the X direction in this embodiment of the invention.

[0029] Figure 5 The diagram shows the trajectory of the unmanned aerial vehicle (UAV) group in the Y direction in this embodiment of the invention.

[0030] Figure 6 The diagram shown is a schematic representation of the error between the UAV's trajectory in the X direction and the desired trajectory in the X direction under unmatched interference in this embodiment of the invention.

[0031] Figure 7 The diagram shown is a schematic representation of the error between the UAV's trajectory in the Y direction and the desired trajectory in the Y direction under unmatched interference in this embodiment of the invention.

[0032] Figure 8 The diagram shown is a schematic representation of the error between the total trajectory and the desired trajectory of the UAV under unmatched interference in an embodiment of the present invention.

[0033] Figure 9 The diagram shown is a comparison of the spacing between each pair of drones in the drone group in this embodiment of the invention. Detailed Implementation

[0034] The technical solutions of the embodiments of this disclosure will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this disclosure, and not all embodiments. The following description of at least one exemplary embodiment is merely illustrative and is in no way intended to limit this disclosure or its application or use. All other embodiments obtained by those skilled in the art based on the embodiments of this disclosure without creative effort are within the scope of protection of this disclosure.

[0035] Unless otherwise specifically stated, the relative arrangement, numerical expressions, and values ​​of the components and steps set forth in these embodiments do not limit the scope of this disclosure.

[0036] Techniques, methods, and equipment known to those skilled in the art may not be discussed in detail, but where appropriate, such techniques, methods, and equipment should be considered part of the specification.

[0037] In all examples shown and discussed herein, any specific values ​​should be interpreted as merely exemplary and not as limitations. Therefore, other examples of exemplary embodiments may have different values.

[0038] The application principle of the present invention will be described in detail below with reference to the accompanying drawings.

[0039] Example 1

[0040] like Figure 1 As shown, this embodiment of the invention provides a fully distributed network optimal control method for unmanned aerial vehicle (UAV) groups, comprising the following steps:

[0041] S01: Construct an error model for unmanned aerial vehicle (UAV) groups;

[0042] S02: With the goal of completing the unit control of the UAV group under unmatched interference, an optimal control model of the UAV group is constructed based on the UAV group error model;

[0043] S03: Solve the optimal control model of the UAV group using the HJB equation to obtain the ideal optimal control signal of the UAV group;

[0044] S04: Based on the ACI framework, update the parameter matrix in the ideal optimal control signal of the UAV group to obtain the actual optimal control signal of the UAV group.

[0045] Furthermore, the process of obtaining the unmanned aerial vehicle (UAV) group error model, such as... Figure 2 As shown, it includes:

[0046] S1: Construct a communication network topology model for unmanned aerial vehicles (UAVs) that includes airborne base stations;

[0047] S2: Establish the spacing model between the airborne base station and the UAV group, the airborne base station state model, and the UAV group state model;

[0048] S3: Construct an unmatched interference auxiliary function, and based on the unmatched interference auxiliary function and the UAV group state model, construct an UAV group state model under unmatched interference;

[0049] S4: Based on the UAV group communication network topology model, the distance model between the airborne base station and the UAV group, the airborne base station state model, and the UAV group state model under non-matching interference, construct the UAV group error model.

[0050] This embodiment proposes a fully distributed network optimal control method for unmanned aerial vehicle (UAV) groups. It uses linear system theory to handle different mismatched disturbances in the UAV system's position and velocity loops, constructs an auxiliary function for mismatched disturbances, and updates the UAV group state model based on this function to obtain the UAV group state model under mismatched disturbances. Combining the UAV group state model under mismatched disturbances, the distance model between the airborne base station and the UAV group, and the airborne base station state model, an UAV group error model is constructed. Finally, with the goal of achieving UAV group control under mismatched disturbances, an optimal control model for the UAV group is constructed based on the UAV group error model. By combining the optimal control model solved using HJB and ACI, optimal cooperation between UAV groups is achieved through actual optimal control signals. This enables the formation to suppress the effects of mismatched disturbances during movement, thereby enabling the UAV group to complete precise movements in complex environments with limited information transmission, while also possessing interference suppression performance.

[0051] Specifically, the construction of the UAV group communication network topology model in step S1 includes:

[0052] S11. Define the vertex set of the drone group. , This indicates the total number of drones in the drone group. Represents the vertex of the first drone representation; Represents the vertex of the nth drone; This represents the vertex of the second drone.

[0053] S12, Define the edge set of the drone group , This indicates the communication connection between drones;

[0054] S13. Define the adjacency matrix. , Indicates drone With drones Communication connection between them, when the drone With drones When there is a communication connection ,otherwise, ;

[0055] S14, Define the vertex of the air base station , The vertex representing the aerial base station;

[0056] Communication matrix between drones and airborne base stations Where, diag is a diagonal matrix. This represents the communication connection between the airborne base station and the nth drone; This represents the communication connection between the airborne base station and the i-th drone; This represents the communication connection between the airborne base station and the second drone; it is assumed that at least one drone is connected to the airborne base station, i.e. ; This represents the communication connection between the airborne base station and the i-th drone;

[0057] S15. Define an undirected connected graph for the drone group. Degree matrix of unmanned aerial vehicle (UAV) groups , This represents the communication connection between the nth drone and the jth drone; This represents the communication connection between the first drone and the j-th drone;

[0058] Specifically, step S2 establishes the spacing model between the airborne base station and the UAV group, the airborne base station state model, and the UAV group state model, including:

[0059] S21. Construct the unmanned aerial vehicle (UAV) group state model:

[0060] S211. First, construct the dynamic state equations of the unmanned aerial vehicle group:

[0061] ;

[0062] In the formula, , , They are respectively The first derivative with respect to time These represent the i-th drone along... Positional component of direction; Let their masses and lift be represented by , respectively, of the i-th UAV. Represents gravitational acceleration; These represent the i-th drone along... Aerodynamic damping coefficient in the direction; Let represent the pitch angle, roll angle, and yaw angle of the i-th UAV, respectively.

[0063] S212. Based on the dynamic state equations of the unmanned aerial vehicle (UAV) group, construct the UAV group state model:

[0064] ;

[0065] in, Represents the position vector of the i-th UAV; These represent the i-th drone along... Positional component of direction; for The first derivative with respect to time; Represents the velocity vector of the i-th UAV; These represent the i-th drone along... velocity in the direction; for The first derivative with respect to time; and All represent intermediate matrices, and the formulas are as follows:

[0066] ;

[0067] ;

[0068] This represents the control signal for the i-th UAV; These represent the i-th drone along... The directional control signal component is expressed by the following formula:

[0069] ;

[0070] This embodiment constructs the state and dynamic characteristics of the UAV (UAV group state model) in detail through the above formula, and reflects the relative positional relationship between the UAV and the airborne base station through kinematic equations, thereby quantifying the error in the UAV formation process. Combined with the constructed performance indicators, it provides a foundation for the subsequent completion of UAV group control under unmatched interference.

[0071] S22: Constructing an aerial base station state model:

[0072] ;

[0073] in, This represents the location vector of the airborne base station, and T represents the transpose operation; for The first derivative with respect to time; These represent the airborne base station along Positional component of direction; Represents the velocity vector of the airborne base station; These represent the airborne base station along The velocity component in the direction; for The first derivative with respect to time; This represents the known motion transmission signals of the airborne base station; among which, In the formula Both represent coefficient matrices. The norm is bounded and the eigenvalues ​​are positive.

[0074] S23: Construct a model of the distance between the aerial base station and the drone group:

[0075] ;

[0076] In the formula, for The first derivative with respect to time; The vector represents the distance between the i-th UAV and the airborne base station; T represents the transpose operation. Represent the i-th drone and the airborne base station along the path respectively. Positional spacing components in the direction; This represents the velocity distance vector between the i-th UAV and the airborne base station; for The first derivative with respect to time; This represents a known bounded input signal. These represent known bounded input signals along... The signal component in the direction;

[0077] This embodiment lays the foundation for the position and velocity loop errors in the subsequent group movement process by constructing the above model.

[0078] In this embodiment, step S3 constructs an unmatched interference auxiliary function, and based on the unmatched interference auxiliary function and the UAV group state model, constructs a UAV group state model under unmatched interference. The specific steps are as follows:

[0079] S31: Construct an unmatched interference auxiliary function, which includes a multimodal gust interference function and an electromagnetic interference function;

[0080] The state equation for the multimodal gust disturbance function is shown below:

[0081] ;

[0082] In the formula, This represents a given constant value for the o-th gust interference; This represents the preset minimum value for the o-th gust interference; for The first derivative with respect to time; This represents the state variable for the o-th gust disturbance; They represent along The state component of the o-th gust interference in the direction; Each represents a constant matrix representing the 0th gust of wind interference; Let represent the set of state variables representing the gust interference of the i-th drone; They represent along The state variable of the i-th drone in the direction of gust interference; This represents the total number of gust interferences experienced by the i-th drone;

[0083] The state equation for the electromagnetic interference function is shown below:

[0084] ;

[0085] In the formula, This represents a given constant value for the k-th electromagnetic interference. for The first derivative with respect to time; This represents the state variable of the k-th electromagnetic interference. They represent along The state component of the k-th electromagnetic interference in the direction; This represents the saturation value of the k-th electromagnetic interference. The constant matrix of electromagnetic interference is represented by both. Let i represent the set of state variables representing the electromagnetic interference of the i-th UAV; They represent along The state variables of electromagnetic interference in the direction of the i-th UAV; This represents the total number of electromagnetic interferences from the i-th drone;

[0086] S32: The state model of the UAV group under unmatched disturbance is constructed based on the unmatched disturbance auxiliary function as follows:

[0087] ;

[0088] In the formula, This represents the position vector of the i-th UAV under non-matching interference; for The first derivative with respect to time; This represents the velocity vector of the i-th UAV under unmatched interference. for The first derivative with respect to time; Indicates the first Uncertain control signals generated by a drone after being subjected to electromagnetic interference; They represent the first drones along Uncertain control signal components generated after the direction is affected by electromagnetic interference.

[0089] In this embodiment, mismatched interference is introduced. Because the interference is estimated and compensated for in real time, the steady-state values ​​of the UAV's state errors (such as position and velocity errors) can be significantly reduced, resulting in more precise UAV control and more accurate formation maintenance in the later stages. Furthermore, in swarm collaboration, mismatched interference can affect different UAVs to varying degrees, disrupting formation consistency. By independently estimating and compensating for the interference of each UAV, the impact of interference on the overall swarm collaboration can be effectively isolated, ensuring that the formation remains stable even under disturbances.

[0090] In this embodiment, step S4: Based on the UAV group communication network topology model, the distance model between the airborne base station and the UAV group, the airborne base station state model, and the UAV group state model under mismatched interference, a UAV group error model is constructed. The specific steps are as follows:

[0091] S41. Construct the position and velocity transformation equations as follows:

[0092] ;

[0093] In the formula: They represent the first Position and velocity errors between the drone and the airborne base station; This represents the position vector of the i-th UAV under non-matching interference; This represents the velocity vector of the i-th UAV under unmatched interference. This represents the positional distance vector between the i-th UAV and the airborne base station; Let represent the velocity distance vector between the i-th UAV and the airborne base station.

[0094] S42. Based on the position and velocity transformation equations, construct the UAV group error model, the formula of which is:

[0095] :

[0096] In the formula: Indicates the first The relative positional error vector between the drone and the other drones in the drone group; Indicates the first The relative speed error vector between the drone and the rest of the drone group; n is the total number of drones in the drone group; Let $\mathbf{j}$ represent the position error and velocity error between the j-th UAV and the airborne base station, respectively.

[0097] In this embodiment, step S02 aims to achieve drone group control under unmatched disturbances. Based on the drone group error model, an optimal control model for the drone group is constructed, including:

[0098] S021. Construct the performance index function, the formula is:

[0099] ;

[0100]

[0101] In the formula, Representing the Performance metrics of the drone; and Both represent symmetric positive definite constant matrices; represents the given normal coefficient; t represents the current time; s represents the integration variable; Describes the integrand over the interval from t to ∞. Integrate with respect to s; Indicates the intermediate vector; express transpose; express transpose; express transpose; express The transpose of .

[0102] S022. Constructing an optimal control model for unmanned aerial vehicle (UAV) groups based on performance index functions:

[0103] ;

[0104] In the formula: This represents the objective function of the optimal control model for unmanned aerial vehicle (UAV) groups.

[0105] In this embodiment, the time derivative of the objective function of the UAV group's optimal control model is calculated, and then the HJB equation is introduced for derivation to solve the objective function and obtain the ideal optimal control signal. This ensures that the UAV group achieves fully distributed control under unmatched interference. That is, step S03 uses the HJB equation to solve the optimal control model of the UAV group to obtain the ideal optimal control signal of the UAV group. The steps are as follows:

[0106] S031: Assume the ideal optimal control signal for the UAV group, and adjust the objective function of the optimal control model of the UAV group based on the assumed ideal optimal control signal;

[0107] Specifically, assuming the ideal optimal control signal for the unmanned aerial vehicle (UAV) group is: Then the objective function of the optimal control model for the unmanned aerial vehicle (UAV) group is adjusted to: ;in, Let these represent the ideal optimal control signals for the i-th, 1st, and nth UAVs in the UAV group, respectively; and let these represent... transpose; express transpose; express transpose;

[0108] S032: Differentiate the objective function of the adjusted optimal control model of the unmanned aerial vehicle group and construct the HJB equation;

[0109] Specifically, the HJB equation is expressed as follows:

[0110] ;

[0111] In the formula, The equation is HJB; This represents partial differential calculation; express 2-norm;

[0112] , ; Represent and 2-norm; Let represent the set of state variables representing the gust interference of the j-th UAV; express The derivative;

[0113] S033: Solve the HJB equation;

[0114] Specifically, let ,get Thus, the ideal optimal control signal can be solved. ;

[0115] In the formula, ;

[0116] ;

[0117] in, All are parameters to be designed; For the communication connection between the i-th UAV and the airborne base station and the other UAVs; Let be the constant upper bound matrix representing the unmatched interference of the i-th UAV; The state variables to be estimated; The adaptive structure representing the matrix to be estimated in the unknown signal from an airborne base station is as follows:

[0118] ;

[0119] In the formula, They are respectively column vectors, All are adaptive laws, expressed as follows:

[0120] ;

[0121] In the formula, All are given, known constant values; for The first derivative with respect to time; for The first derivative with respect to time;

[0122] in, and It is continuous and exists. and Make:

[0123] ;

[0124] in, and Both represent parameter matrices; and They represent and transpose; and Both represent basis function vectors; and Both represent approximation errors, and and Each satisfies and ; and All are constants.

[0125] In this embodiment, the HJB equation is used to solve for the ideal optimal control signal of the UAV, which can obtain the theoretically global optimal solution, thereby achieving the theoretical limit of comprehensive performance in tasks such as trajectory tracking and formation coordination. This method naturally generates a state feedback control law, giving the UAV inherent robustness and enabling it to proactively adapt to uncertainties such as wind disturbances and model biases. Furthermore, it can uniformly handle complex constraints such as nonlinear dynamics and input saturation, and by combining with adaptive dynamic programming (ACI), it endows the UAV with advanced intelligence for online learning and continuous optimization, ultimately achieving stable, efficient, and adaptive autonomous control of the unmanned system.

[0126] Based on the above solution steps, the ideal optimal control signal is divided into multiple parts, and the influence of the design adaptive parameters and ACI basis functions is considered separately to further ensure the accuracy and robustness of the control signal. For example, in step S04: based on the ACI framework, the parameter matrix in the ideal optimal control signal of the UAV group is updated to obtain the actual optimal control signal of the UAV group, as follows:

[0127] S041: Based on the known motion transmission signals of the airborne base station and the unmanned aerial vehicle group state model under unmanned aerial vehicle group conditions and mismatched interference, design the identifier, commentator, actor in the ACI framework, as well as the update rules of the identifier, commentator and actor;

[0128] S042: Based on the designed identifiers, commenters, and actors, and the update rules of identifiers, commenters, and actors, perform iterative updates over time until the parameter matrix represented by the identifier is the same as the parameter matrix represented by the actor, i.e. The parameter matrix in the ideal optimal control signal of the UAV group and The actual optimal control signal for the UAV group is obtained by replacing the parameter matrix represented by the commentator and the parameter matrix represented by the identifier with the updated parameter matrix.

[0129] The design identifier is:

[0130] ;

[0131] in, Indicates the output of the identifier; The parameter matrix representing the identifier;

[0132] The update rule for identifiers is as follows:

[0133] ;

[0134] In the formula, The rate of change of the parameter matrix represented by the identifier; Indicate the design parameters of the i-th UAV;

[0135] The design reviewers used the following formula to evaluate control performance:

[0136] ;

[0137] In the formula, This represents the parameter matrix as characterized by the commentator; and They represent and transpose; express The estimation results;

[0138] The commenters' update pattern is as follows:

[0139] ;

[0140] In the formula, This represents the rate of change of the parameter matrix as represented by the commentator; Design parameters for commentators; express transpose;

[0141] The actor is designed to complete the control signal design, as expressed in the following formula:

[0142] ;

[0143] in, The parameter matrix representing the actor's characterization; express transpose; Represents a known constant vector; express The estimation results;

[0144] The updating pattern of design actors is as follows:

[0145] ;

[0146] in, Represents the rate of change of the parameter matrix represented by the actor; Design parameters for the actors.

[0147] By adopting the above scheme, a network adaptive update law for the ideal optimal control signal is designed to update the parameter matrix and ensure the robustness of the control signal.

[0148] In summary, this embodiment introduces a fully distributed network optimal control method for unmanned aerial vehicle (UAV) groups. The combination of the HJB equations and the ACI framework provides a perfect theoretical and practical bridge for solving the actual optimal control signals for UAVs. This method utilizes the HJB equations to reveal the theoretical structure and global performance upper bound of optimal control, guiding the learning direction of the commentator network for the value function and the actuator network for the control strategy within the ACI framework. By estimating uncertainties such as mismatched disturbances online using identifiers, the system can maintain the accuracy of the learning process even in environments where the model is unknown. The commentator approximates the solution of the HJB equations through time-series difference errors, thereby dynamically correcting the saturation control strategy of the actuators. This fusion mechanism enables UAVs to autonomously evolve intelligent control behaviors that combine optimality, robustness, and adaptability under complex disturbances and constraints.

[0149] Example 2

[0150] This embodiment verifies the optimal control method for a fully distributed network of unmanned aerial vehicle (UAV) groups provided by the present invention, in conjunction with specific embodiments:

[0151] This embodiment constructs a formation (drone group) consisting of one airborne base station and two drones. It then uses steps S01-S04 from Embodiment 1 to solve for the actual optimal control signal of the formation (drone group) and employs this signal to control the formation. The parameters for unmatched interference are set as follows:

[0152] ;

[0153] ;

[0154] ;

[0155] ;

[0156] ;

[0157] Then, based on the formation control results, simulation verification was performed on the trajectory, trajectory error, and spacing error. The simulation results are analyzed as follows:

[0158] like Figure 3 As shown, the blue, red, and green trajectory lines are the two-dimensional planar motion trajectories of the drone group (drone 1, drone 2, and drone 3 equipped with an airborne base station), respectively. It can be seen that the drone group can effectively complete group movement under unmatched interference.

[0159] like Figure 4and Figure 5 As shown in the trajectory diagrams of the UAV group in the X and Y directions, it can be seen that the UAV group maintained an ideal spacing during its forward movement, such as a spacing of 1m in the X direction and a spacing of 1.7m in the Y direction.

[0160] like Figure 6 , Figure 7 and Figure 8 As shown, the errors of the UAV's trajectory in the X direction, Y direction, and total trajectory (two-dimensional plane) compared to the desired trajectory under unmatched interference are illustrated; where UAV1 is UAV 1, UAV2 is UAV 2, and UAV3 is UAV 3. It can be seen that the errors in the X, Y directions, and the total trajectory all remain within 10. -3 Within the unit m.

[0161] like Figure 9 The diagram shows a comparison of the spacing between each pair of drones in the drone group, revealing that even under unmatched interference, the spacing error between the drones is within 10 units. -3 Within the unit m.

[0162] In this embodiment, the X and Y directions represent the x and y directions in Embodiment 1.

[0163] In summary, the fully distributed network optimal control method for unmanned aerial vehicle (UAV) groups introduced in Example 1 enables UAVs to achieve precise intelligent control behavior under complex disturbances (mismatched interference) and constraints.

[0164] Example 3

[0165] This invention provides a fully distributed network-optimal control system for unmanned aerial vehicle (UAV) groups, comprising:

[0166] The data acquisition module is used to build an error model for the unmanned aerial vehicle (UAV) group.

[0167] The optimal control model construction module is used to construct the optimal control model of the UAV group based on the UAV group error model, with the goal of completing the group control under unmatched disturbances.

[0168] The first solution module is used to solve the optimal control model of the UAV group using the HJB equation to obtain the ideal optimal control signal of the UAV group;

[0169] The second solution module is used to update the parameter matrix in the ideal optimal control signal of the UAV group based on the ACI framework, so as to obtain the actual optimal control signal of the UAV group.

[0170] The specific functions of each module described above are explained in the relevant content of the method in Embodiment 1, and will not be repeated here.

[0171] Example 4

[0172] This embodiment also introduces a readable storage medium on which a computer program is stored. When the computer program is executed by a processor, it implements the steps of the fully distributed network optimal control method for unmanned aerial vehicle groups in Embodiment 1.

[0173] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0174] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0175] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0176] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0177] The embodiments of the present invention have been described above with reference to the accompanying drawings. However, the present invention is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of the present invention without departing from the spirit and scope of the claims. All of these forms are within the protection scope of the present invention.

Claims

1. A fully distributed network optimal control method for unmanned aerial vehicle (UAV) groups, characterized in that, include: Construct an error model for unmanned aerial vehicle (UAV) groups; With the goal of enabling unmanned aerial vehicle (UAV) groups to complete unit control under unmatched disturbances, an optimal control model for UAV groups is constructed based on the UAV group error model. The optimal control model of the UAV group is solved using the HJB equation to obtain the ideal optimal control signal of the UAV group; Based on the ACI framework, the parameter matrix in the ideal optimal control signal of the UAV group is updated to obtain the actual optimal control signal of the UAV group. The process of obtaining the unmanned aerial vehicle (UAV) group error model includes: Construct a communication network topology model for unmanned aerial vehicle (UAV) groups, including airborne base stations; Establish a distance model between the airborne base station and the UAV group, a state model of the airborne base station, and a state model of the UAV group; Construct an unmatched interference auxiliary function, and based on the unmatched interference auxiliary function and the UAV group state model, construct an UAV group state model under unmatched interference; Based on the UAV group communication network topology model, the distance model between the airborne base station and the UAV group, the airborne base station state model, and the UAV group state model under mismatched interference, an UAV group error model is constructed.

2. The fully distributed network optimal control method for unmanned aerial vehicle (UAV) groups according to claim 1, characterized in that, The unmanned aerial vehicle (UAV) group communication network topology model includes: ； ； ； ； ； ， ； ； in, V represents an undirected connected graph of the UAV group; V represents the vertex set of the UAV group. This represents the vertex of the first drone. This represents the vertex of the second drone. Represents the vertex of the nth drone; W represents the edge set of the UAV group; W represents the adjacency matrix of the UAV group. This represents the communication connection between the i-th drone and the j-th drone; Represents the vertex of an aerial base station; represents the vertex of the airborne base station; Q represents the communication matrix between the UAV and the airborne base station; diag represents the diagonal matrix form; This represents the communication connection between the airborne base station and the nth drone; This indicates the communication connection between the airborne base station and the first drone; This indicates the communication connection between the airborne base station and the second drone; D represents the communication connection between the airborne base station and the i-th UAV; D represents the degree matrix of the UAV group; This represents the communication connection between the nth drone and the jth drone; This represents the communication connection between the first drone and the j-th drone.

3. The fully distributed network optimal control method for unmanned aerial vehicle (UAV) groups according to claim 2, characterized in that, The process involves establishing a distance model between the airborne base station and the UAV group, a state model for the airborne base station, and a state model for the UAV group. The distance model between the airborne base station and the UAV group is represented as follows: ； In the formula, for The first derivative with respect to time; The vector represents the distance between the i-th UAV and the airborne base station; T represents the transpose operation. Represent the i-th drone and the airborne base station along the path respectively. Positional spacing components in the direction; This represents the velocity distance vector between the i-th UAV and the airborne base station; for The first derivative with respect to time; This represents the velocity distance vector between the i-th UAV and the airborne base station; This represents a known bounded input signal. These represent known bounded input signals along... The signal component in the direction; The state model of the airborne base station is represented as follows: ； ； In the formula, for The first derivative with respect to time; This represents the location vector of the airborne base station. These represent the airborne base station along Positional component of direction; Represents the velocity vector of the airborne base station; for The first derivative with respect to time; This indicates known motion transmission signals from the airborne base station. These represent the known edges of the airborne base station. The direction of motion transmission signal components; Both represent coefficient matrices. The norm is bounded and the eigenvalues ​​are positive; The unmanned aerial vehicle (UAV) group state model is represented as follows: ； ； ； ； ； ； ； ； In the formula, Represents the position vector of the i-th UAV; These represent the i-th drone along... Positional component of direction; for The first derivative with respect to time; Represents the velocity vector of the i-th UAV; These represent the i-th drone along... velocity in the direction; for The first derivative with respect to time; and Both represent intermediate matrices; This represents the control signal for the i-th UAV; These represent the i-th drone along... Directional control signal components; Let these represent the mass and lift of the i-th UAV, respectively. Represents gravitational acceleration; These represent the i-th drone along... Aerodynamic damping coefficient in the direction; Let represent the pitch angle, roll angle, and yaw angle of the i-th UAV, respectively. , , They are respectively The first derivative with respect to time.

4. The fully distributed network optimal control method for unmanned aerial vehicle (UAV) groups according to claim 3, characterized in that, The construction of the unmatched interference auxiliary function, and the construction of the UAV group state model under unmatched interference based on the unmatched interference auxiliary function and the UAV group state model, include: An unmatched interference auxiliary function is constructed, which includes a multimodal gust interference function and an electromagnetic interference function. The state equation of the multimodal gust interference function is shown below: ； In the formula, This represents a given constant value for the 0th gust of wind interference; This represents the preset minimum value for the 0th gust interference; for The first derivative with respect to time; This represents the state variable for the o-th gust disturbance; They represent along The state component of the o-th gust interference in the direction; Each represents a constant matrix representing the 0th gust interference; Let represent the set of state variables representing the gust interference of the i-th drone; They represent along The state variable of the i-th drone in the direction of gust interference; This represents the total number of gust interferences experienced by the i-th drone; The state equation for the electromagnetic interference function is shown below: ； In the formula, This represents a given constant value for the k-th electromagnetic interference. for The first derivative with respect to time; This represents the state variable of the k-th electromagnetic interference. They represent along The state component of the k-th electromagnetic interference in the direction; This represents the saturation value of the k-th electromagnetic interference. The constant matrix of electromagnetic interference is represented by both. Let i represent the set of state variables representing the electromagnetic interference of the i-th UAV; They represent along The state variables of electromagnetic interference in the direction of the i-th UAV; This represents the total number of electromagnetic interferences from the i-th drone; The state model of the UAV group under unmatched disturbance is represented as follows: ； In the formula, This represents the position vector of the i-th UAV under non-matching interference; for The first derivative with respect to time; This represents the velocity vector of the i-th UAV under unmatched interference. for The first derivative with respect to time; Indicates the first Uncertain control signals generated by a drone after being subjected to electromagnetic interference; They represent the first drones along Uncertain control signal components generated after the direction is affected by electromagnetic interference.

5. The fully distributed network optimal control method for unmanned aerial vehicle (UAV) groups according to claim 4, characterized in that, The unmanned aerial vehicle (UAV) group error model is expressed as follows: ： ； In the formula: Indicates the first The relative positional error vector between the drone and the other drones in the drone group; Indicates the first The relative speed error vector between the drone and the rest of the drone group; n is the total number of drones in the drone group; They represent the first Position and velocity errors between the drone and the airborne base station; Let $\mathbf{j}$ represent the position error and velocity error between the j-th UAV and the airborne base station, respectively.

6. The fully distributed network optimal control method for unmanned aerial vehicle (UAV) groups according to claim 5, characterized in that, The optimal control model for the unmanned aerial vehicle (UAV) group is expressed as follows: ； ； In the formula: This represents the objective function of the optimal control model for unmanned aerial vehicle (UAV) groups. and Both represent symmetric positive definite constant matrices; Represents the coefficient for a given normal value; Representing the The performance index values ​​of the drone; t represents the current time; s represents the integral variable; Describes the integrand over the interval from t to ∞. Integrate with respect to s; Indicates the intermediate vector; express transpose; express transpose; express transpose; express The transpose of .

7. The fully distributed network optimal control method for unmanned aerial vehicle (UAV) groups according to claim 6, characterized in that, The process of solving the optimal control model of the UAV group using the HJB equation to obtain the ideal optimal control signal for the UAV group includes: Assume the ideal optimal control signal for the unmanned aerial vehicle (UAV) group is Then the objective function of the optimal control model for the unmanned aerial vehicle (UAV) group is adjusted to: ;in, Let represent the ideal optimal control signals for the i-th, 1st, and nth UAVs in the UAV group, respectively; express transpose; express transpose; express transpose; Based on the adjusted objective function, construct the HJB equation: ； In the formula, The equation is HJB; , Let represent the set of state variables representing the gust interference of the j-th UAV; express The derivative; This represents partial differential calculation; express 2-norm; make ,get Thus, the ideal optimal control signal can be solved. ; In the formula, ; ； in, All are parameters to be designed; For the communication connection between the i-th UAV and the airborne base station and the other UAVs; Let be the constant upper bound matrix representing the unmatched interference of the i-th UAV; The state variables to be estimated; The adaptive structure representing the matrix to be estimated in the unknown signal from an airborne base station is as follows: ； In the formula, They are respectively column vectors, Both are adaptive laws, expressed as follows: ； In the formula, All are given, known constant values; for The first derivative with respect to time; for The first derivative with respect to time; in, and It is continuous and exists. and Make: ； in, and Both represent parameter matrices; and They represent and transpose; and Both represent basis function vectors; and Both represent approximation errors, and and Each satisfies and ; and All are constants.

8. The fully distributed network optimal control method for unmanned aerial vehicle (UAV) groups according to claim 7, characterized in that, The process of updating the parameter matrix in the ideal optimal control signal of the UAV group based on the ACI framework to obtain the actual optimal control signal of the UAV group includes: Based on the known motion transmission signals from the airborne base station and the unmanned aerial vehicle group state model under unmatched interference, we design the identifiers, commentators, and actors in the ACI framework, as well as the update rules for the identifiers, commentators, and actors. Based on the designed identifiers, commentators, and actors, and their update rules, iterative updates are performed over time until the parameter matrix represented by the identifier is the same as the parameter matrix represented by the actor. This process then refines the parameter matrix in the ideal optimal control signal for the UAV group. and The actual optimal control signal for the UAV group is obtained by replacing the parameter matrix represented by the commentator and the parameter matrix represented by the identifier with the updated parameter matrix. The design identifier is: ； Among them, represents the output of the identifier; represents the parameter matrix characterized by the identifier; The update rule for identifiers is as follows: ； In the formula, The rate of change of the parameter matrix represented by the identifier; Indicate the design parameters of the i-th UAV; The design reviewers used the following formula to evaluate control performance: ； In the formula, This represents the parameter matrix as characterized by the commentator; and They represent and transpose; express The estimation results; The commenters' update pattern is as follows: ； In the formula, This represents the rate of change of the parameter matrix as represented by the commentator; Design parameters for commentators; express transpose; The actor is designed to complete the control signal design, as expressed in the following formula: ； in, The parameter matrix representing the actor's characterization; express transpose; Represents a known constant vector; express The estimation results; The updating pattern of design actors is as follows: ； in, Represents the rate of change of the parameter matrix represented by the actor; Design parameters for the actor.

9. A fully distributed network optimal control system for unmanned aerial vehicle (UAV) groups, the system being used to implement the fully distributed network optimal control method for UAV groups as described in any one of claims 1-8, characterized in that, include: The data acquisition module is used to build an error model for the unmanned aerial vehicle (UAV) group. The optimal control model construction module is used to construct the optimal control model of the UAV group based on the UAV group error model, with the goal of completing the group control under unmatched disturbances. The first solution module is used to solve the optimal control model of the UAV group using the HJB equation to obtain the ideal optimal control signal of the UAV group; The second solution module is used to update the parameter matrix in the ideal optimal control signal of the UAV group based on the ACI framework, so as to obtain the actual optimal control signal of the UAV group.

Images