Data-driven collaborative output adjustment method and device for heterogeneous unmanned system, and medium

Abstract

The invention discloses a heterogeneous unmanned system data-driven collaborative output adjustment method and device and a medium, and the method comprises the steps: collecting an adjustment input sequence, an adjustment output sequence and an internal model state sequence of each unmanned system monomer in an off-line manner, and carrying out the data-driven collaborative output adjustment of each unmanned system monomer based on the collected adjustment input sequence, adjustment output sequence and internal model state sequence; determining a control gain matrix of the unmanned system monomer; and constructing an output feedback controller for the unmanned system monomers based on the control gain matrix to obtain a distributed output feedback controller, and performing coordinated output adjustment on the cluster unmanned system through the distributed output feedback controller. According to the method, the distributed output feedback controller is constructed based on the input data and the output data by using the input data and the output data which are subjected to noise interference and are collected offline, so that dependence on an accurate system model is avoided, and collaborative output adjustment of an unknown cluster system can be realized only by using the data.

Description

Technical Field

[0001] The present application relates to the technical field of collaborative control of clustered unmanned systems, and in particular to a data-driven collaborative output adjustment method, device and medium for heterogeneous unmanned systems. Background Art

[0002] With the rapid development of artificial intelligence, by promoting the deep integration of artificial intelligence and scientific research, accelerating the paradigm change and capability improvement of scientific research, and promoting artificial intelligence to a new stage of high-quality application. Especially in the field of unmanned systems, unmanned systems represented by drones, unmanned vehicles, unmanned ships, and intelligent robots have become an important carrier of artificial intelligence research, showing the deep integration of artificial intelligence with cutting-edge disciplines such as big data, optimized decision-making, and collaborative control. Cluster unmanned systems are composed of a large number of unmanned system monomers. Through mutual interaction and collaboration, they complete complex tasks that a single unmanned system cannot complete independently, becoming a strategic focus of global scientific and technological competition.

[0003] In response to the requirements of trajectory tracking tasks in interference and confrontation environments, the problem of coordinated output regulation of clustered unmanned systems has attracted widespread attention. Existing solutions mainly rely on accurate models of the system. It is necessary to build a state space model based on mechanism or identification methods to determine the distributed control strategy and perform coordinated regulation based on the distributed control strategy. However, due to the characteristics of clustered unmanned systems such as large-scale distribution, heterogeneous node dynamics, and high model complexity, it is extremely difficult to build an accurate model, which limits the coordinated regulation through distributed control strategies in complex environments.

[0004] Therefore the prior art still needs to be improved and enhanced. Summary of the invention

[0005] The technical problem to be solved by the present application is to provide a data-driven collaborative output adjustment method, device and medium for heterogeneous unmanned systems in response to the deficiencies in the prior art.

[0006] In order to solve the above technical problems, the first aspect of the present application provides a data-driven collaborative output adjustment method for heterogeneous unmanned systems, wherein the data-driven collaborative output adjustment method for heterogeneous unmanned systems specifically includes: Constructing an internal model system for each unmanned system monomer in the swarm unmanned system, and connecting the unmanned system monomer with its corresponding internal model system, wherein the internal model system is constructed based on the evolution matrix of the external reference system of the swarm unmanned system; Offline collecting the adjustment input sequence, the adjustment output sequence and the internal model state sequence of the internal model system corresponding to each unmanned system unit in the cluster unmanned system, and determining the control gain matrix of the unmanned system unit based on the collected adjustment input sequence, the adjustment output sequence and the internal model state sequence; An output feedback controller is constructed for each unmanned system unit based on the control gain matrix of each unmanned system unit to obtain a distributed output feedback controller corresponding to the cluster unmanned system, and the cluster unmanned system is coordinated and output adjusted through the distributed output feedback controller.

[0007] The heterogeneous unmanned system data-driven collaborative output adjustment method, wherein the dynamic models of each unmanned system monomer in the cluster unmanned system are heterogeneous.

[0008] The heterogeneous unmanned system data-driven collaborative output adjustment method, wherein the internal model system is constructed for each unmanned system monomer in the cluster unmanned system specifically includes: Constructing a submatrix based on the minimum polynomial of the evolution matrix of the external reference system of the swarm unmanned system, and constructing a block diagonal matrix with the submatrix as a submatrix on the diagonal to form a first internal model matrix; Constructing a column vector having the same column dimension as the column-direction dimension of the submatrix, and constructing a block diagonal matrix with the column vector as a submatrix on the diagonal to form a second internal mode matrix, wherein the product of the submatrix and the column vector is controllable; Based on the first internal model matrix, the second internal model matrix and the virtual tracking difference of each unmanned system unit, an internal model system is constructed for each unmanned system unit.

[0009] The heterogeneous unmanned system data-driven collaborative output adjustment method, wherein the step of determining the control gain matrix of the unmanned system monomer based on the collected adjustment input sequence, adjustment output sequence and internal model state sequence specifically includes: For each unmanned system unit, determining an observability index of the unmanned system unit; Constructing a linear matrix inequality based on the observability index, the adjustment input sequence, the adjustment output sequence and the internal model state sequence, and solving the linear matrix inequality to obtain a first gain parameter and a second gain parameter; Based on the first gain parameter and the second gain parameter, a control gain matrix of the unmanned system unit is determined.

[0010] The heterogeneous unmanned system data-driven collaborative output adjustment method, wherein the linear matrix inequality is: , , , , , , , , , , in, represents the first gain parameter; represents a second gain parameter; , , , , Both represent the identity matrix; represents the first data matrix; represents the second data matrix; , , , , , , , , All are known matrices; Indicates unmanned system unit The observability index; represents transpose; Indicates unmanned system unit The upper limit of noise; , , , , , , , , , All represent all-zero matrices; , , All represent positive integers.

[0011] The heterogeneous unmanned system data-driven collaborative output adjustment method, wherein the expression of the control gain matrix is: , , , in, represents the first gain parameter, represents the second gain parameter, represents the control gain matrix, express No. eigenvalues, The Laplacian matrix representing the graph structure of the communication network between each unmanned system in the cluster unmanned system and the external reference signal, Indicates is a diagonal matrix with principal diagonal elements, represents the intermediate matrix, Indicates unmanned unit The connectivity relationship with the external reference system.

[0012] The heterogeneous unmanned system data-driven collaborative output regulation method, wherein the control gain matrix of each unmanned system unit is used to construct an output feedback controller for each unmanned system unit, specifically comprising: For each unmanned system monomer, construct a control difference of the unmanned system monomer, and construct monomer data of the unmanned system monomer based on the control difference and the internal model state of the unmanned system monomer; Constructing an output feedback controller for the unmanned system monomer based on the monomer data of the unmanned system monomer and the control gain matrix; The expression of the control difference is: , , , in, Indicates control differences; The adjacency matrix of the graph structure representing the communication network between each unmanned system in the cluster unmanned system and the external reference signal is Row, No. Elements of a column; Indicates unmanned unit Connectivity with external reference systems; Indicates unmanned system unit The observability index; and All represent intermediate variables; express The single unit output data of the unmanned system at the moment; express The system output of the external reference system at the moment; express Input data of the unmanned system unit at the moment; Indicates transpose.

[0013] A second aspect of the present application provides a data-driven collaborative output adjustment device for heterogeneous unmanned systems, wherein the data-driven collaborative output adjustment device for heterogeneous unmanned systems specifically includes: An internal model system construction module, used to construct an internal model system for each unmanned system monomer in the swarm unmanned system, and connect the unmanned system monomer with its corresponding internal model system, wherein the internal model system is constructed based on the evolution matrix of the external reference system of the swarm unmanned system; An offline collection module is used to collect offline the adjustment input sequence, the adjustment output sequence and the internal model state sequence of the corresponding internal model system of each unmanned system unit in the cluster unmanned system, and determine the control gain matrix of the unmanned system unit based on the collected adjustment input sequence, adjustment output sequence and internal model state sequence; A controller construction module is used to construct an output feedback controller for each unmanned system unit based on the control gain matrix of each unmanned system unit to obtain a distributed output feedback controller corresponding to the cluster unmanned system, and an adjustment module is used to coordinate the output adjustment of the cluster unmanned system through the distributed output feedback controller.

[0014] The third aspect of the present application provides a computer-readable storage medium, which stores one or more programs, and the one or more programs can be executed by one or more processors to implement the steps in the heterogeneous unmanned system data-driven collaborative output adjustment method as described above.

[0015] A fourth aspect of the present application provides a server, which includes: a processor and a memory; The memory stores a computer-readable program executable by the processor; When the processor executes the computer-readable program, it implements the steps in any of the heterogeneous unmanned system data-driven collaborative output adjustment methods described above.

[0016] Beneficial effects: 1. Only the input data and output data collected offline and affected by noise are used, and a distributed output feedback controller is built based on the input data and output data, avoiding the dependence on the precise model of the system. The coordinated output regulation of the unknown cluster system can be achieved only by using data.

[0017] 2. It avoids the use of single state data of the unmanned system, and only uses input data and output data to achieve collaborative output adjustment, effectively solving the problem of unpredictable state data caused by privacy protection requirements when performing tasks.

[0018] 3. The construction of a distributed output feedback controller can be realized by only solving a low-complexity linear matrix inequality offline, which saves computing resources and energy consumption; and the constructed controller has good robustness, which effectively solves the zero-deviation tracking of the output of each unmanned system unit and the external reference system when the data is interfered. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0020] Figure 1 A flowchart of a data-driven collaborative output adjustment method for heterogeneous unmanned systems provided in an embodiment of the present application.

[0021] Figure 2 This is an effect diagram of the first dimension output of a specific embodiment.

[0022] Figure 3 This is an effect diagram of the second dimension output of this specific embodiment.

[0023] Figure 4 A principle block diagram of a heterogeneous unmanned system data-driven collaborative output regulation device provided in an embodiment of the present application.

[0024] Figure 5 A block diagram of the server provided in accordance with an embodiment of the present application. DETAILED DESCRIPTION

[0025] The embodiments of the present application provide a method, device and medium for coordinated output adjustment driven by data of heterogeneous unmanned systems. In order to make the purpose, technical solution and effect of the present application clearer and more specific, the present application is further described in detail with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0026] It will be understood by those skilled in the art that, unless expressly stated, the singular forms "one", "said", and "the" used herein may also include plural forms. It should be further understood that the term "comprising" used in the specification of the present application refers to the presence of the features, integers, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or groups thereof. It should be understood that when we refer to an element as being "connected" or "coupled" to another element, it may be directly connected or coupled to the other element, or there may be an intermediate element. In addition, the "connection" or "coupling" used herein may include wireless connection or wireless coupling. The term "and / or" used herein includes all or any unit and all combinations of one or more associated listed items.

[0027] It will be understood by those skilled in the art that, unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as those generally understood by those skilled in the art to which this application belongs. It should also be understood that terms such as those defined in general dictionaries should be understood to have meanings consistent with those in the context of the prior art, and will not be interpreted with idealized or overly formal meanings unless specifically defined as here.

[0028] It should be understood that the sequence numbers and sizes of the steps in this embodiment do not mean the order of execution. The execution order of each process is determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiment of the present application.

[0029] After research, it was found that with the rapid development of artificial intelligence, by promoting the deep integration of artificial intelligence and scientific research, accelerating the paradigm change and capability improvement of scientific research, and promoting artificial intelligence to a new stage of high-quality application. Especially in the field of unmanned systems, unmanned systems represented by drones, unmanned vehicles, unmanned ships, and intelligent robots have become an important carrier of artificial intelligence research, showing the deep integration of artificial intelligence with cutting-edge disciplines such as big data, optimized decision-making, and collaborative control. Cluster unmanned systems are composed of a large number of unmanned system monomers. Through mutual interaction and collaboration, they complete complex tasks that a single unmanned system cannot complete independently, becoming a strategic focus of global scientific and technological competition.

[0030] In response to the requirements of trajectory tracking tasks in interference and confrontation environments, the problem of coordinated output regulation of clustered unmanned systems has attracted widespread attention. Existing solutions mainly rely on accurate models of the system. It is necessary to build a state space model based on mechanism or identification methods to determine the distributed control strategy and perform coordinated regulation based on the distributed control strategy. However, due to the characteristics of clustered unmanned systems such as large-scale distribution, heterogeneous node dynamics, and high model complexity, it is extremely difficult to build an accurate model, which limits the coordinated regulation through distributed control strategies in complex environments.

[0031] In order to solve the above problems, in an embodiment of the present application, an internal model system is constructed for each unmanned system monomer in the cluster unmanned system, and the unmanned system monomer is connected to its corresponding internal model system, and the adjustment input sequence, adjustment output sequence and internal model state sequence of each unmanned system monomer in the cluster unmanned system are collected offline, and the control gain matrix of the unmanned system monomer is determined based on the collected adjustment input sequence, adjustment output sequence and internal model state sequence; an output feedback controller is constructed for each unmanned system monomer based on the control gain matrix of each unmanned system monomer to obtain a distributed output feedback controller corresponding to the cluster unmanned system, and the cluster unmanned system is coordinated and output regulated by the distributed output feedback controller. The present application avoids dependence on the precise model of the system by utilizing the input data and output data collected offline and interfered by noise, and constructs a distributed output feedback controller based on the input data and output data, so as to achieve coordinated output regulation of unknown cluster systems using only data.

[0032] The application content is further explained below through the description of embodiments in conjunction with the accompanying drawings.

[0033] This embodiment provides a data-driven collaborative output adjustment method for heterogeneous unmanned systems, such as Figure 1 As shown, the method includes: S10. Construct an internal model system for each unmanned system unit in the cluster unmanned system, and connect the unmanned system unit with its corresponding internal model system.

[0034] Specifically, the swarm unmanned system includes several unmanned system units, wherein the dynamic models of at least some of the several unmanned system units are heterogeneous, and in particular, the dynamic models of each unmanned system unit in the swarm unmanned system are heterogeneous. For example, the swarm unmanned system is a drone swarm, which includes several quadcopter drones with heterogeneous dynamic models. The mathematical model of each unmanned system unit in the swarm unmanned system is unknown, and its dynamic equation can be expressed as: , in, express Moment The state data of an unmanned system unit, whose dimensions are ; express Moment The input data of an unmanned system unit has the dimension ; express Moment The single output data of an unmanned system single unit has the dimension of ; , and All are positive integers; express unknown real matrix of dimension; yes unknown real matrix of dimension; express unknown real matrix of dimension; express unknown real matrix of dimension; The dimension is The external reference signal, its dynamic model is: , in, express The system output of the external reference system at time , whose dimension is ; express dimension unknown real matrix, represents the evolution matrix, whose dimension is , and has no eigenvalues ​​with negative real parts.

[0035] Each unmanned system in the cluster unmanned system communicates with the external reference system, and the communication network between each unmanned system in the cluster unmanned system and the external reference system is a directed spanning tree with the external reference signal as the root node. Here, the communication network is recorded as .

[0036] To this end, at a given regulated output When , the closed-loop system formed by the cluster unmanned system and the external reference system can be expressed as: .

[0037] The goal of collaborative output regulation in the present embodiment is to construct input data for each unmanned system unit. In the absence of an external reference signal (i.e. The closed-loop system achieves exponential stabilization when there is an external reference signal (i.e. is not zero), for any system initial state data , can be achieved over time Going to infinity, tracking deviation Tends to zero.

[0038] Based on this, when performing collaborative output adjustment on the cluster unmanned system, an internal model system will first be constructed for each unmanned system unit in the collaborative output adjustment. The internal model system is used to perform internal model compensation on the unmanned system unit. The internal model system is constructed based on the evolution matrix of the external reference system, and the external reference system is used to provide an external reference signal for the cluster unmanned system.

[0039] Exemplarily, the step of constructing an internal model system for each unmanned system in the swarm unmanned system specifically includes: Constructing a submatrix based on the minimum polynomial of the evolution matrix of the external reference system of the swarm unmanned system, and constructing a block diagonal matrix with the submatrix as a submatrix on the diagonal to form a first internal model matrix; Constructing a column vector having the same column dimension as the column-direction dimension of the submatrix, and constructing a block diagonal matrix with the column vector as a submatrix on the diagonal to form a second internal mode matrix, wherein the product of the submatrix and the column vector is controllable; Based on the first internal model matrix, the second internal model matrix and the virtual tracking difference of each unmanned system unit, an internal model system is constructed for each unmanned system unit.

[0040] Specifically, after obtaining the evolution matrix of the external reference system, first determine the evolution matrix The minimum polynomial of , and then construct a constant square matrix whose characteristic polynomial is the same as the minimum polynomial , and finally with a constant square matrix A block diagonal matrix is ​​constructed for the sub-matrices on the diagonal to form a first internal mode matrix, wherein the first internal mode matrix can be expressed as: , in, represents the first internal model matrix; A constant square matrix representing the characteristic polynomial and the minimal polynomial, whose dimension is , represents a positive integer; Indicates A block diagonal matrix constructed for the submatrices on the diagonal; This means that the block diagonal matrix has sub-matrices.

[0041] After obtaining the first internal model matrix, a second internal model matrix is ​​constructed based on the first internal model matrix, and the product of the submatrix of the first internal model matrix and the column vector of the second internal model matrix is ​​controllable, wherein the second internal model matrix can be expressed as: , in, represents the second internal model matrix; represents an arbitrarily chosen constant column vector with dimension ,and It is controllable; Indicates A block diagonal matrix constructed for column vectors on the diagonal; This means that the block diagonal matrix has A column vector.

[0042] It should be noted that, in practical applications, other methods may also be used to construct the first internal model matrix and the second internal model matrix, for example, directly using the evolution matrix The coefficients of the minimum polynomial are used to construct the first internal model matrix and the second internal model matrix, and so on.

[0043] After determining the first internal model matrix and the second internal model matrix, an internal model system is constructed based on the first internal model matrix, the second internal model matrix and the virtual tracking difference, wherein the internal model system can be expressed as: , in, express The internal model state at the moment, express The internal model state at the moment, represents the first internal mode matrix, represents the second internal mode matrix, Represents the tracking error.

[0044] Further, the tracking error may include a first output difference between the monomer output data of each unmanned system monomer, and may also include a second output difference between the monomer output data of the unmanned system monomer and the system output of the external signal system. In the embodiment of the present application, the tracking error includes the first output difference between the monomer output data of each unmanned system monomer and the second output difference between the monomer output data of the unmanned system monomer and the system output of the external signal system. Accordingly, the tracking error can be expressed as: , in, represents the tracking error, Indicates the communication network The adjacency matrix of No. Row, No. The elements of the column, Indicates unmanned system unit The connectivity relationship with the external reference system, Indicates unmanned system unit The single output data of Indicates unmanned system unit The single output data of represents the system output of the external reference system, Indicates the number of unmanned system units.

[0045] Furthermore, after constructing the internal model system for each unmanned system monomer, each unmanned system monomer is connected to the corresponding internal model system, and the operation system connected to the internal model system is operated in an open loop, wherein the operation system can be expressed as: , in, Indicates the input noise present on the input side, Indicates the output noise present on the output side, Indicates the measured unmanned system unit affected by noise The disturbed regulated input, Indicates the measured unmanned system unit affected by noise The disturbed regulated output, represents the measured endomembrane system affected by noise exist The internal model state at the moment, represents the measured endomembrane system affected by noise exist The internal model state at the moment.

[0046] S20. Offline collect the adjustment input sequence, adjustment output sequence and internal model state sequence of the corresponding internal model system of each unmanned system unit in the cluster unmanned system, and determine the control gain matrix of the unmanned system unit based on the collected adjustment input sequence, adjustment output sequence and internal model state sequence.

[0047] Specifically, after running the above operating system, collect 0 to Moment , , , To obtain the adjustment input sequence, adjustment output sequence and internal model state sequence, wherein the collection process can be specifically: for any period of time arrive , random input generation in real time To unknown unmanned system , get the unmanned system monomer at the corresponding time The output sequence of the regulation , randomly generate noise sequence and , calculate the single unit of unmanned system under noise interference The adjustment input sequence and adjust output data , and record the input sequence at the corresponding time and adjust the output sequence ; will adjust the output sequence Send it to the inner membrane system, record the state data of the inner membrane system at the corresponding time to obtain the inner model state sequence

[0048] Further, after obtaining the adjustment input sequence, the adjustment output sequence and the internal model state sequence, the control gain matrix of the unmanned system unit is determined based on the adjustment input sequence, the adjustment output sequence and the internal model state sequence, wherein the process of determining the control gain matrix specifically includes: For each unmanned system unit, determining an observability index of the unmanned system unit; Constructing a linear matrix inequality based on the observability index, the adjustment input sequence, the adjustment output sequence and the internal model state sequence, and solving the linear matrix inequality to obtain a first gain parameter and a second gain parameter; Based on the first gain parameter and the second gain parameter, a control gain matrix of the unmanned system unit is determined.

[0049] Specifically, the observability index is the matrix pair in the dynamics model When it is considerable, it makes and The rank of the target matrix reaches the unmanned system monomer Status Dimension The smallest integer of , where the target matrix is: , in, represents the target matrix, Indicates transpose.

[0050] Further, when constructing the linear matrix inequality, two intermediate data matrices can be constructed based on the observability index, the adjustment input sequence, the adjustment output sequence and the internal model state sequence, which are respectively recorded as the first data matrix and the second data matrix, wherein the matrix dimensions of the first data matrix and the second data matrix are the same, and the number of columns of the two is = , the number of rows = , the first data matrix and the second data matrix can be expressed as: , , in, Represents the observability index.

[0051] After determining the first data matrix and the second data matrix, set the given matrix As an unmanned system unit in the data collection phase The noise upper limit of the introduced noise includes the input measurement noise, the adjusted output measurement noise and the external reference system in the data collection phase. Then, a known matrix for intermediate data is constructed based on the first data matrix, the second data matrix and the noise upper limit. , and ,in, , and Respectively expressed as: , , , in, represents the first data matrix, represents the second data matrix, Indicates unmanned system unit The upper limit of noise, Indicates unmanned system unit The upper limit of noise, is a known matrix used as intermediate data, which can be expressed as: , , in, For a known matrix as intermediate data, , All represent all-zero matrices, and All represent positive integers, , denotes the identity matrix, , , All represent positive integers.

[0052] In getting , and After, based on , and Construct a linear matrix inequality, where the linear matrix inequality is: , , , , , in, represents the first gain parameter, represents a second gain parameter; , , Both represent the identity matrix; , , , All are known matrices; Indicates unmanned system unit The observability index; represents transpose; , , , , , , , , All represent all-zero matrices; , All represent positive integers.

[0053] After obtaining the linear matrix inequality, the first gain parameter and the second gain parameter can be determined by solving the linear matrix inequality, wherein the solution method of the linear matrix inequality can adopt the existing method, which will not be repeated here. In addition, after obtaining the first gain parameter and the second gain parameter, the control gain matrix is ​​determined based on the first gain parameter and the second gain parameter, and the control gain matrix can be expressed as: , , , in, represents the first gain parameter, represents the second gain parameter, represents the control gain matrix, express No. eigenvalues, The Laplacian matrix representing the graph structure of the communication network between each unmanned system in the cluster unmanned system and the external reference signal, Indicates is a diagonal matrix with principal diagonal elements, represents the intermediate matrix, Indicates unmanned unit The connectivity relationship with the external reference system.

[0054] S30. Construct an output feedback controller for each unmanned system unit based on the control gain matrix of each unmanned system unit to obtain a distributed output feedback controller corresponding to the cluster unmanned system, and coordinate the output adjustment of the cluster unmanned system through the distributed output feedback controller.

[0055] Specifically, the control gain matrix is ​​used to construct an output feedback controller, and the output of each unmanned system unit in the cluster unmanned system is controlled by the output feedback controller corresponding to each unmanned system unit, so as to achieve exponential stabilization of the unknown cluster unmanned system when the external disturbance and the external signal are 0, and when the external disturbance and the external signal are not 0, for any system initial state, the system output can tend to infinity with time, and the tracking deviation tends to zero.

[0056] The output feedback controller is used to generate control inputs based on the input data of the unmanned system monomer, monomer output data, system output and internal model state, and send the control inputs back to the unknown unmanned system monomer to achieve the coordinated output adjustment of the cluster unmanned system of the unknown model. Among them, when generating control inputs based on the input data, monomer output data, system output and internal model state of the unmanned system monomer, the control gain matrix can be multiplied with the monomer data of the unmanned system monomer determined according to the input data, monomer output data, system output and internal model state. In other words, during the online operation stage, the unmanned system monomer of the unknown model At each moment, the current monomer data is sent to the corresponding output feedback controller, and the inner membrane system sends the current inner membrane state to the output feedback controller. , the output feedback controller generates the control input and pass the generated control input Send back an unknown unmanned system unit , to achieve the collaborative output regulation of the cluster unmanned system of the unknown model.

[0057] To this end, the output feedback controller can be expressed as: , in, Indicates the internal model state. represents the control difference, represents the control gain matrix, Represents the control input generated by the output feedback controller.

[0058] Exemplarily, constructing an output feedback controller for each unmanned system unit based on the control gain matrix of each unmanned system unit specifically includes: For each unmanned system monomer, construct a control difference of the unmanned system monomer, and construct monomer data of the unmanned system monomer based on the control difference and the internal model state of the unmanned system monomer; An output feedback controller is constructed for the unmanned system unit based on the unit data of the unmanned system unit and the control gain matrix.

[0059] Specifically, the control difference includes a first output difference between the monomer output data of each unmanned system monomer and a second output difference between the monomer output data of the unmanned system monomer and the system output. The control difference can be expressed as: , , , in, represents the control difference, The adjacency matrix of the graph structure representing the communication network between each unmanned system in the cluster unmanned system and the external reference signal is Row, No. The elements of the column, Indicates unmanned unit The connectivity relationship with the external reference system, Indicates unmanned system unit The observability index, and All represent intermediate variables. express The single output data of the unmanned system at the moment, express The system output of the external reference system at time instant, express The input data of the unmanned system at the moment, Indicates transpose.

[0060] Of course, in practical applications, the control difference can take other forms, for example, it can include taking the average of the output differences between the individual output data of each unmanned system unit as the first output difference, or weighting the output differences between the individual output data of each unmanned system unit to determine the first output difference, etc.

[0061] In summary, this embodiment provides a data-driven collaborative output regulation method for heterogeneous unmanned systems, the method constructs an internal model system for each unmanned system monomer in a cluster unmanned system, connects the unmanned system monomer to its corresponding internal model system, collects the adjustment input sequence, adjustment output sequence and the internal model state sequence of the corresponding internal model system of each unmanned system monomer in the cluster unmanned system offline, and determines the control gain matrix of the unmanned system monomer based on the collected adjustment input sequence, adjustment output sequence and internal model state sequence; constructs an output feedback controller for each unmanned system monomer based on the control gain matrix of each unmanned system monomer to obtain a distributed output feedback controller corresponding to the cluster unmanned system, and coordinates the output regulation of the cluster unmanned system through the distributed output feedback controller. This application avoids dependence on the precise model of the system by utilizing the input data and output data collected offline and interfered by noise, and constructs a distributed output feedback controller based on the input data and output data, so as to achieve collaborative output regulation of unknown cluster systems using only data.

[0062] In order to further illustrate the effectiveness of the data-driven collaborative output adjustment method for heterogeneous unmanned systems provided in the embodiment of the present application, a specific simulation example is given below to illustrate it.

[0063] like Figure 2 and Figure 3 The figure shows the effect of running the method of the present invention for 80 seconds in a discrete-time heterogeneous multi-agent system. The system matrix in this embodiment is: , , , , , , in, , , .

[0064] Accordingly, the inner membrane matrix and for: , , .

[0065] Depend on Figure 2 The output of the four unmanned system units shown and the system output of the external reference system The first dimension output and Figure 3 The output of the four unmanned system units shown and the system output of the external reference system It can be seen from the rendering of the second-dimensional output that the output of each unmanned system unit can track the output of the external reference system without deviation, that is, the collaborative output regulation of the multi-agent system is realized, which proves the effectiveness of the data-driven collaborative output regulation method of the heterogeneous unmanned system.

[0066] Based on the above-mentioned heterogeneous unmanned system data-driven collaborative output adjustment method, this embodiment provides a heterogeneous unmanned system data-driven collaborative output adjustment device, such as Figure 4 As shown, the heterogeneous unmanned system data-driven collaborative output adjustment device specifically includes: The internal model system construction module 100 is used to construct an internal model system for each unmanned system monomer in the swarm unmanned system, and connect the unmanned system monomer with its corresponding internal model system, wherein the internal model system is constructed based on the evolution matrix of the external reference system of the swarm unmanned system; The offline collection module 200 is used to collect the adjustment input sequence, the adjustment output sequence and the internal model state sequence of the corresponding internal model system of each unmanned system unit in the cluster unmanned system offline, and determine the control gain matrix of the unmanned system unit based on the collected adjustment input sequence, adjustment output sequence and internal model state sequence; The controller construction module 300 is used to construct an output feedback controller for each unmanned system unit based on the control gain matrix of each unmanned system unit to obtain a distributed output feedback controller corresponding to the cluster unmanned system, and the adjustment module is used to coordinate the output adjustment of the cluster unmanned system through the distributed output feedback controller.

[0067] Based on the above-mentioned heterogeneous unmanned system data-driven collaborative output adjustment method, this embodiment provides a computer-readable storage medium, which stores one or more programs. The one or more programs can be executed by one or more processors to implement the steps in the heterogeneous unmanned system data-driven collaborative output adjustment method as described in the above-mentioned embodiment.

[0068] Based on the above heterogeneous unmanned system data-driven collaborative output adjustment method, the present application also provides a server, such as Figure 5As shown, it includes at least one processor 20; a display screen 21; and a memory 22, and may also include a communications interface 23 and a bus 24. The processor 20, the display screen 21, the memory 22, and the communications interface 23 may communicate with each other through the bus 24. The display screen 21 is configured to display a preset user guide interface in the initial setting mode. The communications interface 23 may transmit information. The processor 20 may call the logic instructions in the memory 22 to execute the method in the above embodiment.

[0069] In addition, the logic instructions in the memory 22 can be implemented in the form of software functional units and can be stored in a computer-readable storage medium when sold or used as an independent product.

[0070] The memory 22 is a computer-readable storage medium that can be configured to store software programs, computer executable programs, such as program instructions or modules corresponding to the methods in the embodiments of the present disclosure. The processor 20 executes functional applications and data processing by running the software programs, instructions or modules stored in the memory 22, that is, implementing the methods in the above embodiments.

[0071] The memory 22 may include a program storage area and a data storage area, wherein the program storage area may store an operating system and at least one application required for a function; the data storage area may store data created according to the use of the server, etc. In addition, the memory 22 may include a high-speed random access memory and may also include a non-volatile memory. For example, a variety of media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a disk or an optical disk, may also be a transient storage medium.

[0072] In addition, the specific process of loading and executing the multiple instruction processors in the above-mentioned storage medium and the server has been described in detail in the above-mentioned method, and will not be described here one by one.

[0073] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A data-driven collaborative output adjustment method for heterogeneous unmanned systems, characterized in that: The heterogeneous unmanned system data-driven collaborative output adjustment method specifically includes: Constructing an internal model system for each unmanned system monomer in the swarm unmanned system, and connecting the unmanned system monomer with its corresponding internal model system, wherein the internal model system is constructed based on the evolution matrix of the external reference system of the swarm unmanned system; Offline collecting the adjustment input sequence, the adjustment output sequence and the internal model state sequence of the internal model system corresponding to each unmanned system unit in the cluster unmanned system, and determining the control gain matrix of the unmanned system unit based on the collected adjustment input sequence, the adjustment output sequence and the internal model state sequence; An output feedback controller is constructed for each unmanned system unit based on the control gain matrix of each unmanned system unit to obtain a distributed output feedback controller corresponding to the cluster unmanned system, and the cluster unmanned system is coordinated and output adjusted through the distributed output feedback controller.

2. The data-driven collaborative output adjustment method for heterogeneous unmanned systems according to claim 1 is characterized in that: The dynamic models of each unmanned system in the cluster unmanned system are heterogeneous.

3. The data-driven collaborative output adjustment method for heterogeneous unmanned systems according to claim 1 is characterized in that: The construction of an internal model system for each unmanned system in the swarm unmanned system specifically includes: Constructing a submatrix based on the minimum polynomial of the evolution matrix of the external reference system of the swarm unmanned system, and constructing a block diagonal matrix with the submatrix as a submatrix on the diagonal to form a first internal model matrix; Constructing a column vector having the same column dimension as the column-direction dimension of the submatrix, and constructing a block diagonal matrix with the column vector as a submatrix on the diagonal to form a second internal mode matrix, wherein the product of the submatrix and the column vector is controllable; Based on the first internal model matrix, the second internal model matrix and the virtual tracking difference of each unmanned system unit, an internal model system is constructed for each unmanned system unit.

4. The data-driven collaborative output adjustment method for heterogeneous unmanned systems according to claim 1 is characterized in that: The step of determining the control gain matrix of the unmanned system unit based on the collected adjustment input sequence, adjustment output sequence and internal model state sequence specifically includes: For each unmanned system unit, determining an observability index of the unmanned system unit; Constructing a linear matrix inequality based on the observability index, the adjustment input sequence, the adjustment output sequence and the internal model state sequence, and solving the linear matrix inequality to obtain a first gain parameter and a second gain parameter; Based on the first gain parameter and the second gain parameter, a control gain matrix of the unmanned system unit is determined.

5. The data-driven collaborative output adjustment method for heterogeneous unmanned systems according to claim 4 is characterized in that: The linear matrix inequality is: ， ， ， ， ， ， ， ， ， in, represents a first gain parameter; represents a second gain parameter; , , , , Both represent the identity matrix; represents the first data matrix; represents the second data matrix; , , , , , , , , All are known matrices; Indicates unmanned system unit The observability index; represents transpose; Indicates unmanned system unit The upper limit of noise; , , , , , , , , , All represent all-zero matrices; , , All represent positive integers.

6. The data-driven collaborative output adjustment method for heterogeneous unmanned systems according to claim 4 is characterized in that: The expression of the control gain matrix is: ， ， ， in, represents the first gain parameter, represents the second gain parameter, represents the control gain matrix, express No. eigenvalues, The Laplacian matrix representing the graph structure of the communication network between each unmanned system in the cluster unmanned system and the external reference signal, Indicates is a diagonal matrix with principal diagonal elements, represents the intermediate matrix, Indicates unmanned unit The connectivity relationship with the external reference system.

7. The data-driven collaborative output adjustment method for heterogeneous unmanned systems according to claim 1 is characterized in that: The constructing of an output feedback controller for each unmanned system unit based on the control gain matrix of each unmanned system unit specifically includes: For each unmanned system monomer, construct a control difference of the unmanned system monomer, and construct monomer data of the unmanned system monomer based on the control difference and the internal model state of the unmanned system monomer; Constructing an output feedback controller for the unmanned system monomer based on the monomer data of the unmanned system monomer and the control gain matrix; The expression of the control difference is: ， ， ， in, Indicates control differences; The adjacency matrix of the graph structure representing the communication network between each unmanned system in the cluster unmanned system and the external reference signal is Row, No. Elements of a column; Indicates unmanned unit Connectivity with external reference systems; Indicates unmanned system unit The observability index; and All represent intermediate variables; express The single unit output data of the unmanned system at the moment; express The system output of the external reference system at the moment; express Input data of the unmanned system unit at the moment; Indicates transpose.

8. A data-driven collaborative output adjustment device for heterogeneous unmanned systems, characterized in that: The heterogeneous unmanned system data-driven collaborative output adjustment device specifically includes: An internal model system construction module, used to construct an internal model system for each unmanned system monomer in the swarm unmanned system, and connect the unmanned system monomer with its corresponding internal model system, wherein the internal model system is constructed based on the evolution matrix of the external reference system of the swarm unmanned system; An offline collection module is used to collect offline the adjustment input sequence, the adjustment output sequence and the internal model state sequence of the corresponding internal model system of each unmanned system unit in the cluster unmanned system, and determine the control gain matrix of the unmanned system unit based on the collected adjustment input sequence, adjustment output sequence and internal model state sequence; A controller construction module is used to construct an output feedback controller for each unmanned system unit based on the control gain matrix of each unmanned system unit to obtain a distributed output feedback controller corresponding to the cluster unmanned system, and an adjustment module is used to coordinate the output adjustment of the cluster unmanned system through the distributed output feedback controller.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores one or more programs, and the one or more programs can be executed by one or more processors to implement the steps in the heterogeneous unmanned system data-driven collaborative output adjustment method as described in any one of claims 1-7.

Images