Fault Detection Method, Device and Medium Based on Data-Driven of Unmanned Systems

Abstract

The present application discloses a fault detection method, device and medium based on data-driven of an unmanned system. The method includes connecting an unknown unmanned system to a controller, collecting the state data of the unknown unmanned system offline, and constructing a state data matrix and an evolutionary state data matrix based on the collected state data; constructing a fault detector based on the state data matrix and the evolutionary state data matrix, connecting the fault detector to the unknown unmanned system, and performing fault detection based on the state data of the unmanned system through the fault detector. By collecting the closed-loop noisy state data of the unknown unmanned system offline, without prior physical process identification, then constructing a fault detector based on the collected state data, and performing fault detection through the fault detector, the present application can trigger an alarm in time when the unmanned system appears in an abnormal state, realizing fault detection for any unmanned system operating stably.

Description

Technical Field

[0001] This application relates to the technical field of unmanned system failures, and particularly relates to a fault detection method, device, and medium based on data-driven of an unmanned system. Background Art

[0002] In the current era of artificial intelligence leading scientific and technological innovation, unmanned systems represented by driverless vehicles, robots, drones, unmanned boats, etc. have become a large stage for the display of artificial intelligence technology. At the same time, with the development of big data and computing technology, data-driven is applied to unmanned systems to improve the stability and optimization of unmanned systems through data-driven controllers.

[0003] In the actual operation process, in addition to stability and basic optimality, it is also necessary to limit the input and state of the system to ensure the safety of system operation. For example, in a fleet of driverless vehicles, if the distance between any two driverless vehicles cannot guarantee a certain lower limit, then collisions will occur between the driverless vehicles. However, when an abnormal state occurs, such as one of the driverless vehicles suddenly fails and becomes out of control, the safety distance between this vehicle and other vehicles cannot be guaranteed, which may lead to collisions and cause a series of serious subsequent consequences. Therefore, it is necessary to detect the abnormal state of the unmanned system. However, currently, the fault detection models of unmanned systems generally detect specific types of inorganic system faults and cannot be applied to the fault detection of unknown unmanned systems.

[0004] Therefore, the existing technology still needs to be improved and enhanced. Summary of the Invention

[0005] The technical problem to be solved by this application is to provide a fault detection method, device, and medium based on data-driven of an unmanned system in view of the deficiencies of the existing technology.

[0006] To solve the above technical problem, the first aspect of this application provides a fault detection method based on data-driven of an unmanned system. Specifically, the fault detection method based on data-driven of an unmanned system includes:

[0007] Connect the unknown unmanned system to the controller, offline collect the state data at a certain number of collection times, and construct a state data matrix and an evolutionary state data matrix based on the collected state data;

[0008] Construct a fault detector based on the state data matrix and the evolutionary state data matrix;

[0009] Connect the fault detector to the unknown unmanned system, and perform fault detection based on the state data of the unmanned system through the fault detector.

[0010] The described fault detection method based on data-driven of unmanned systems, wherein the offline collection of state data at a certain number of collection moments specifically includes:

[0011] Set initial state data for the unknown unmanned system and transmit the initial state data to the controller;

[0012] Generate a control input through the controller and apply the control input to the unknown unmanned system;

[0013] Collect the state data of the unknown unmanned system at a new moment formed based on the control input;

[0014] Use the state data at the new moment as the initial state data, and re-execute the step of transmitting the initial state data to the controller until a certain number of state data at collection moments are collected.

[0015] The described fault detection method based on data-driven of unmanned systems, wherein constructing a fault detector based on the state data matrix and the evolved state data matrix specifically includes:

[0016] Construct a first semidefinite programming problem based on the state data matrix and the evolved state data matrix, and solve the first semidefinite programming problem to determine the optimal solution of the first semidefinite programming problem;

[0017] Construct a second semidefinite programming problem based on the optimal solution of the first semidefinite programming problem, and solve the second semidefinite programming problem to construct a fault detector.

[0018] The described fault detection method based on data-driven of unmanned systems, wherein the first semidefinite programming problem is expressed as:

[0019] ,

[0020] ,

[0021] ,

[0022] ,

[0023] wherein, represents solving the value of the determinant, represents the dimension is a real positive definite symmetric matrix to be solved, represents a real number to be solved, represents the state data matrix, represents the evolved state data matrix, represents the first intermediate matrix, represents the second intermediate matrix, Denotes the identity matrix of dimension , Denotes the all-zero matrix of dimension , Denotes the all-zero matrix of dimension , Denotes a constant between 0 and 1, Denotes the transpose of the matrix , Denotes the transpose of the matrix , Denotes a known constant greater than or equal to zero, Denotes the number of time steps collected offline.

[0024] The described fault detection method based on data-driven of unmanned systems, wherein the second semi-definite programming problem is formulated as:

[0025] ,

[0026] ,

[0027] ,

[0028] ,

[0029] ,

[0030] ,

[0031] ,

[0032] ,

[0033] ,

[0034] Wherein, Denotes solving the value of the determinant, Denotes taking the maximum value in the curly brackets, Denotes the dimension to be solved of the matrix, is the scalar to be solved, Denotes the optimal solution of the first semi-definite programming problem, Denotes the inverse matrix of the optimal solution of the first semi-definite programming problem, Denotes the third intermediate matrix, Denotes the fourth intermediate matrix, Denotes the all-zero matrix of dimension , Denotes a constant, Denotes the dimension of The identity matrix, represents a zero matrix with dimensions of ; represents a known constant greater than or equal to zero, represents the number of time steps collected offline, represents the upper limit of the initial value during the online operation of the unknown unmanned system;

[0035] The fault detector is expressed as:

[0036] ,

[0037] wherein, represents the fault detector, represents the optimal solution of the second semi - definite programming problem.

[0038] In the described fault detection method based on data - driven of unmanned systems, wherein the connection of the fault detector to the unknown unmanned system is specifically:

[0039] Connect the fault detector to the output items of the unknown unmanned system, so that the fault detector is connected to the closed - loop system formed by the unknown unmanned system and the controller, where the fault detector is parallel to the controller.

[0040] In the described fault detection method based on data - driven of unmanned systems, wherein the fault detection based on the state data of the unmanned system by the fault detector specifically includes:

[0041] Obtain the state data of the unknown unmanned system at each collection moment through the fault detector;

[0042] Determine the state of the fault detector based on the state data, and form an alarm message when the state of the fault detector does not meet the preset requirements;

[0043] The state of the fault detector is:

[0044] When , ;

[0045] When at that time, is expressed as:

[0046] ,

[0047] wherein, represents the state of the fault detector at time represents the state data at time represents The state data at a moment, represents the state data matrix, represents the evolved state data matrix, represents the matrix right generalized inverse matrix of;

[0048] The preset requirement is:

[0049] ,

[0050] wherein, represents transpose of, represents the fault detector.

[0051] The second aspect of the present application provides a fault detection device based on data-driven of an unmanned system. Among them, the fault detection device based on data-driven of the unmanned system specifically includes:

[0052] An offline collection module, configured to connect an unknown unmanned system to a controller, offline collect state data at a certain number of collection moments, and construct a state data matrix and an evolved state data matrix based on the collected state data;

[0053] A construction module, configured to construct a fault detector based on the state data matrix and the evolved state data matrix;

[0054] A fault detection module, configured to connect the fault detector to the unknown unmanned system, and perform fault detection based on the state data of the unmanned system through the fault detector.

[0055] 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 any one of the above-mentioned data-driven fault detection methods based on an unmanned system.

[0056] The fourth aspect of the present application provides a terminal device, which includes: a processor and a memory;

[0057] The memory stores a computer-readable program executable by the processor;

[0058] When the processor executes the computer-readable program, it implements the steps in any one of the above-mentioned data-driven fault detection methods based on an unmanned system.

[0059] Beneficial effects: Compared with the prior art, the present application provides a fault detection method, device, and medium based on data-driven of an unmanned system. The method includes connecting an unknown unmanned system to a controller, offline collecting state data at a certain number of collection moments, and constructing a state data matrix and an evolution state data matrix based on the collected state data; constructing a fault detector based on the state data matrix and the evolution state data matrix; connecting the fault detector to the unknown unmanned system, and performing fault detection based on the state data of the unmanned system through the fault detector. The present application collects the closed-loop noisy state data of the unknown unmanned system offline, without prior physical process identification, then constructs a fault detector based on the collected state data, and performs fault detection through the fault detector, so as to trigger an alarm in time when the unmanned system is in an abnormal state, realizing fault detection for any unmanned system operating stably. Description of the Drawings

[0060] To more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0061] Figure 1 It is a flowchart of the fault detection method based on data-driven of an unmanned system provided by the embodiment of the present application.

[0062] Figure 2 It is a principle block diagram of a detection system formed by a physical process, a controller, and a fault detector.

[0063] Figure 3 It is a state trajectory evolution diagram of an unmanned system in a specific embodiment.

[0064] Figure 4 It is a trajectory diagram of the detection state of the fault detector in this specific embodiment.

[0065] Figure 5 It is a principle block diagram of the fault detection device based on data-driven of an unmanned system provided by the embodiment of the present application.

[0066] Figure 6 It is a principle block diagram of the terminal device provided by the embodiment of the present application. Detailed Embodiments

[0067] An embodiment of the present application provides a fault detection method, device and medium based on data-driven of an unmanned system. To make the purpose, technical solution and effect of the present application clearer and more definite, the following further elaborates on the present application with reference to the accompanying drawings and by way of 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.

[0068] Those skilled in the art of this technology can understand that unless specifically stated, the singular forms "a", "an", "the" and "said" used herein may also include the plural forms. It should be further understood that the term "comprising" used in the specification of the present application means the presence of the described 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 their groups. It should be understood that when we say an element is "connected" or "coupled" to another element, it can be directly connected or coupled to other elements, or there may also be intermediate elements. In addition, the "connection" or "coupling" used herein may include wireless connection or wireless coupling. The phrase "and / or" used herein includes all or any unit and all combinations of one or more related listed items.

[0069] Those skilled in the art of this technology can understand that unless otherwise defined, all terms (including technical terms and scientific terms) used herein have the same meaning as the general understanding of those of ordinary skill in the art to which the present application belongs. It should also be understood that terms such as those defined in a general dictionary should be understood to have a meaning consistent with the meaning in the context of the prior art, and will not be interpreted with an idealized or overly formal meaning unless specifically defined as here.

[0070] It should be understood that the sequence numbers and magnitudes 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 to the implementation process of the embodiments of the present application.

[0071] Next, with reference to the accompanying drawings, the content of the application will be further described through the description of the embodiments.

[0072] This embodiment provides a fault detection method based on data-driven of an unmanned system, as Figure 1 shown, the method includes:

[0073] S10. Connect an unknown unmanned system to a controller, offline collect state data at a certain number of collection moments, and construct a state data matrix and an evolutionary state data matrix based on the collected state data.

[0074] Specifically, connecting the unknown unmanned system to the controller means connecting the physical process of the unknown unmanned system to the controller to form a closed-loop system. Among them, the physical process is the physical movement process of the unmanned aerial vehicle system, and the dynamic equation of the physical process can be expressed as:

[0075] ,

[0076] Among them, represents t the state data of the unknown unmanned system at time ; represents t the unknown disturbance received by the unknown unmanned system at time , and it satisfies that for all times there is , represents a known constant greater than or equal to zero, represents the two-norm of the unknown disturbance ; the matrix represents dimensional unknown real matrix, the control matrix represents dimensional unknown real matrix; the control gain matrix represents dimensional matrix that can ensure the Schur stability of the matrix , and this control gain matrix can be determined based on data-driven.

[0077] Exemplarily, the offline collection of the state data at a certain number of collection times specifically includes:

[0078] Set the initial state data for the unknown unmanned system and transmit the initial state data to the controller;

[0079] Generate a control input through the controller and apply the control input to the unknown unmanned system;

[0080] Collect the state data of the new moment formed by the unknown unmanned system based on the control input;

[0081] Take the state data of the new moment as the initial state data and re-execute the step of transmitting the initial state data to the controller until the state data at a certain number of collection times is collected.

[0082] Specifically, the initial state data is randomly generated, that is, an arbitrary initial value is applied to the closed-loop system formed by the unknown unmanned system and the controller. Of course, in practical applications, an initial state data can also be preset in advance, and then this preset initial state data is used as the initial state data of the unknown unmanned system when performing offline collection. When an arbitrary initial value is applied to the closed-loop system formed by the unknown unmanned system and the controller, during the operation of the physical process of the unknown unmanned system, the state data is transmitted to the controller in real time. The controller multiplies the state data by the control gain matrix K and then transmits it back to the physical process through the control matrix B. The physical process generates the state data at the next moment according to this control input, and then transmits this state data to the controller, and loops in turn until a certain number of state data at the collection moments are collected.

[0083] Furthermore, the certain number is preset, or it can be determined according to the preset requirements of the state data matrix. For example, the certain number is the minimum amount of state data that makes the state data matrix a full row rank matrix, etc. And, the certain number is equal to the number of running steps of the closed-loop system plus 1, that is , and satisfies . That is to say, for the number of running steps of the closed-loop system , the state of the physical process with a length of will be collected . . Use the collected certain number of state data to construct a state data matrix and an evolution state data matrix. Among them, the state data matrix and the evolution state data matrix are respectively expressed as:

[0084] ,

[0085] ,

[0086] where represents the state data matrix, represents the evolution state data matrix.

[0087] It should be noted that after constructing the state data matrix and the evolution state data matrix, it will be detected whether the state data matrix is row full rank. If it is row full rank, it means that the state data matrix meets the requirements. On the contrary, if it is not row full rank, it means that the state data matrix meets the requirements, and the above offline collection process needs to be continued until the constructed state data matrix is full row rank.

[0088] S20. Construct a fault detector based on the state data matrix and the evolution state data matrix.

[0089] Specifically, a fault detector is used to monitor the abnormal state of an unmanned system. Among them, the construction process of the fault detector can construct a semidefinite programming problem based on the state data matrix and the evolved state data matrix. However, the fault detector is constructed based on the optimal solution of the semidefinite programming problem. Among them, the semidefinite programming problem can include a first semidefinite programming problem and a second semidefinite programming problem, and the optimal solution of the semidefinite programming problem is the optimal solution of the second semidefinite programming problem. That is, the fault detector can be expressed as:

[0090] ,

[0091] wherein, represents the fault detector, represents the optimal solution of the semidefinite programming problem.

[0092] Exemplarily, the construction of the fault detector based on the state data matrix and the evolved state data matrix specifically includes:

[0093] Construct a first semidefinite programming problem based on the state data matrix and the evolved state data matrix, and solve the first semidefinite programming problem to determine the optimal solution of the first semidefinite programming problem;

[0094] Construct a second semidefinite programming problem based on the optimal solution of the first semidefinite programming problem, and solve the second semidefinite programming problem to construct the fault detector.

[0095] Specifically, the first semidefinite programming problem is expressed as:

[0096] ,

[0097] ,

[0098] ,

[0099] ,

[0100] wherein, represents the value of solving the determinant, represents the dimension of the real positive definite symmetric matrix to be solved, represents the real number to be solved, represents the state data matrix, represents the evolved state data matrix, represents the first intermediate matrix, represents the second intermediate matrix, represents the identity matrix of dimension , represents the all-zero matrix of dimension . represents a zero matrix of dimension , is a constant between 0 and 1, represents the matrix transpose, represents the matrix transpose, represents a known constant greater than or equal to zero, represents the number of time steps collected offline.

[0101] Furthermore, the second semidefinite programming problem is expressed as:

[0102] ,

[0103] ,

[0104] ,

[0105] ,

[0106] ,

[0107] ,

[0108] ,

[0109] ,

[0110] ,

[0111] where represents solving the value of the determinant, represents taking the maximum value in the curly brackets, represents the dimension to be solved matrix, is the scalar to be solved, represents the optimal solution of the first semidefinite programming problem, represents the inverse matrix of the optimal solution of the first semidefinite programming problem, represents the third intermediate matrix, represents the fourth intermediate matrix, represents a zero matrix of dimension , represents a constant, represents a known constant greater than or equal to zero, represents a zero matrix of dimension identity matrix, represents a zero matrix of dimension , represents a known constant greater than or equal to zero, represents the number of time steps collected offline, represents the upper limit value of the initial value when the unknown unmanned system is operating online.

[0112] It should be noted that the solution processes of the first semidefinite programming problem and the second semidefinite programming problem can both be carried out by existing processes, and will not be specifically described here. At the same time, when obtaining the fault detector constructed based on the state data matrix and the evolution state data matrix, other methods can also be used to construct it. For example, by fusing the state data matrix and the evolution state data matrix, etc., as long as a method can construct a fault detector that can perform fault detection on the unmanned system is acceptable.

[0113] S30. Connect the fault detector to the unknown unmanned system, and perform fault detection based on the state data of the unmanned system through the fault detector.

[0114] Specifically, after obtaining the fault detector, during the operation stage of the unmanned system, as Figure 2 shown, the physical process, the controller, and the fault detector can be connected. The fault detector performs anomaly detection on the state data at each moment, and when an anomaly in the state data is detected, a warning is given to inform the unmanned system of the anomaly. Among them, when connecting the physical process, the controller, and the fault detector, the fault detector is connected to the output item of the unknown unmanned system, so that the fault detector is connected to the closed-loop system formed by the unknown unmanned system and the controller, where the fault detector is parallel to the controller.

[0115] Furthermore, when performing fault detection based on the state data of the unmanned system through the fault detector, the fault detector can obtain the state data of the unmanned system, then determine the fault detector state based on this state data, and finally determine whether the state data is abnormal based on the fault detector state. Based on this, the performing fault detection based on the state data of the unmanned system through the fault detector specifically includes:

[0116] Obtaining the state data of the unknown unmanned system at each collection moment through the fault detector;

[0117] Determining the fault detector state based on the state data, and forming an alarm message when the fault detector state does not meet the preset requirements.

[0118] Specifically, the fault detector state is used to reflect the fault state of the state data. Among them, the fault detector state can be calculated based on the state time, and the fault detector can be expressed as:

[0119] When , ;

[0120] When time, It is expressed as:

[0121] ,

[0122] Among them, represents the state of the fault detector at time represents the state data at time represents the state data at time represents the state data matrix, represents the evolved state data matrix, represents the matrix right contra-inverse matrix of.

[0123] Furthermore, the prediction requirement is pre-set as the judgment basis for determining whether the state data is abnormal. That is to say, when the fault detector state meets the preset requirement, it indicates that the state data is normal, and when the fault detector state does not meet the preset requirement, it indicates that the state data is abnormal. Among them, the preset requirement is:

[0124] ,

[0125] Among them, represents transpose of, represents the fault detector.

[0126] In summary, this embodiment provides a fault detection method, device and medium based on data-driven of an unmanned system. The method includes connecting an unknown unmanned system to a controller, offline collecting state data at a certain number of collection times, and constructing a state data matrix and an evolved state data matrix based on the collected state data; constructing a fault detector based on the state data matrix and the evolved state data matrix; connecting the fault detector to the unknown unmanned system, and performing fault detection based on the state data of the unmanned system through the fault detector. This application collects the closed-loop noisy state data of the unknown unmanned system offline, without prior physical process identification, then constructs a fault detector based on the collected state data, and performs fault detection through the fault detector to trigger an alarm in time when the unmanned system is in an abnormal state, realizing fault detection for any unmanned system running stably.

[0127] To illustrate the effect of the fault detection method based on data-driven of an unmanned system adopted in this embodiment of the application, a specific example is given below.

[0128] As Figure 3 and Figure 4 shown is the effect diagram of a robot embodiment operating with the safety controller of the present invention for 50 unit times. The system matrix of the dynamic equation of the physical process of the robot is:

[0129] , ,

[0130] The number of running steps collected offline , , , and by solving the first semi-definite programming problem, we get:

[0131] ,

[0132] The running range of the given system initial value is , and we get , and by solving the second semi-definite programming problem, we get the fault detector:

[0133] ,

[0134] By installing the fault detector for the unknown unmanned system and running for 50 unit times, we get the state trajectory evolution diagram as shown in Figure 3 and the trajectory diagram of the fault detector detection state as shown in Figure 4 .

[0135] From Figure 3 , it can be seen that the fault occurs after the 25th step length, and the fault detector alarms. From Figure 4 , it can be seen that when the state of the fault detector does not meet the preset requirements, the fault alarm sounds, thus indicating the effectiveness of the fault detector.

[0136] Based on the above fault detection method based on unmanned system data-driven, an embodiment of the present application provides a fault detection device based on unmanned system data-driven, as shown in Figure 5 , the fault detection device based on unmanned system data-driven specifically includes:

[0137] An offline collection module 100, which is used to connect the unknown unmanned system to the controller, offline collect the state data at a certain number of collection moments, and construct a state data matrix and an evolution state data matrix based on the collected state data;

[0138] A construction module 200, which is used to construct a fault detector based on the state data matrix and the evolution state data matrix;

[0139] A fault detection module 300 is configured to connect the fault detector to the unknown unmanned system and perform fault detection based on the status data of the unmanned system through the fault detector.

[0140] Based on the above-mentioned fault detection method driven by unmanned system data, this embodiment provides a computer-readable storage medium storing one or more programs, which can be executed by one or more processors to implement the steps in the fault detection method driven by unmanned system data as described in the above embodiment.

[0141] Based on the above-mentioned fault detection method driven by unmanned system data, this application also provides a terminal device, as Figure 6 shown, which includes at least one processor 20, a display screen 21, and a memory 22. It may also include a communication interface 23 and a bus 24. Among them, the processor 20, the display screen 21, the memory 22, and the communication interface 23 can communicate with each other through the bus 24. The display screen 21 is set to display a user guidance interface preset in the initial setting mode. The communication interface 23 can transmit information. The processor 20 can call the logic instructions in the memory 22 to execute the method in the above embodiment.

[0142] In addition, when the logic instructions in the above-mentioned memory 22 are implemented in the form of a software functional unit and sold or used as an independent product, they can be stored in a computer-readable storage medium.

[0143] The memory 22, as a computer-readable storage medium, can be set to store software programs and computer-executable programs, such as the program instructions or modules corresponding to the method in the embodiment 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, implements the method in the above embodiment.

[0144] The memory 22 may include a program storage area and a data storage area. Among them, the program storage area can store an operating system and application programs required for at least one function; the data storage area can store data created according to the use of the terminal device. In addition, the memory 22 may include a high-speed random access memory and may also include a non-volatile memory. For example, various media such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disc that can store program codes can also be a transient storage medium.

[0145] In addition, the specific processes loaded and executed by the multi-instruction processor in the above storage medium and terminal device have been described in detail in the above method, and will not be repeated here one by one.

[0146] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, and are not intended to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A fault detection method based on data-driven of unmanned systems, characterized in that, The described fault detection method based on data-driven of unmanned systems specifically includes: Connect the unknown unmanned system to the controller, offline collect the state data at a certain number of collection moments, and construct a state data matrix and an evolutionary state data matrix based on the collected state data; Construct a first semidefinite programming problem based on the state data matrix and the evolutionary state data matrix, and solve the first semidefinite programming problem to determine the optimal solution of the first semidefinite programming problem; Construct a second semidefinite programming problem based on the optimal solution of the first semidefinite programming problem, and solve the second semidefinite programming problem to construct a fault detector; Connect the fault detector to the unknown unmanned system, and perform fault detection based on the state data of the unmanned system through the fault detector; The first semidefinite programming problem is expressed as: ， ， ， ， Among them, represents the value of solving the determinant, represents the dimension, of the real positive definite symmetric matrix to be solved, represents the real number to be solved, represents the state data matrix, represents the evolved state data matrix, represents the first intermediate matrix, represents the second intermediate matrix, represents the dimension of the identity matrix, represents the dimension of the all-zero matrix, represents the dimension of the all-zero matrix, represents a constant between 0 and 1, represents the matrix transpose of, represents the matrix transpose of, represents a known constant greater than or equal to zero, represents the number of time steps collected offline, represents the dimension of the state data of the unknown unmanned system; The second semidefinite programming problem is expressed as: ， ， ， ， ， ， ， ， ， Among them, represents the value of solving the determinant, represents taking the maximum value in the curly brackets, represents the dimension to be solved of the matrix, is the scalar to be solved, represents the optimal solution of the first semidefinite programming problem, represents the inverse matrix of the optimal solution of the first semidefinite programming problem, represents the third intermediate matrix, represents the fourth intermediate matrix, represents the dimension of the all-zero matrix, represents a constant, represents the dimension of the identity matrix, represents the dimension of the all-zero matrix, represents a known constant greater than or equal to zero, represents the number of time steps collected offline, represents the upper limit of the initial value when the unknown unmanned system is running online; The fault detector is expressed as: ， Among them, represents a fault detector, represents the optimal solution of the second semi - definite programming problem.

2. The fault detection method based on unmanned system data-driven according to claim 1, characterized in that The specific process of offline collecting the state data at a certain number of collection moments includes: Set the initial state data for the unknown unmanned system and transmit the initial state data to the controller; Generate a control input through the controller and apply the control input to the unknown unmanned system; Collect the state data of the new moment formed by the unknown unmanned system based on the control input; Take the state data of the new moment as the initial state data, and re-execute the step of transmitting the initial state data to the controller until the state data at a certain number of collection moments is collected.

3. The fault detection method based on unmanned system data-driven according to claim 1, characterized in that The specific way of connecting the fault detector to the unknown unmanned system is: Connect the fault detector to the output item of the unknown unmanned system, so that the fault detector is connected to the closed-loop system formed by the unknown unmanned system and the controller, where the fault detector is parallel to the controller.

4. The fault detection method based on unmanned system data-driven according to claim 1 or 3, characterized in that, The specific process of performing fault detection based on the state data of the unmanned system through the fault detector includes: Obtain the state data of the unknown unmanned system at each collection moment through the fault detector; Determine the state of the fault detector based on the state data, and form an alarm message when the state of the fault detector does not meet the preset requirements; The state of the fault detector is: When then ; When time It is expressed as: ， Among them, represents the state of the fault detector at time represents the state data at time represents the state data at time represents the state data matrix represents the evolved state data matrix represents the matrix right generalized inverse matrix of The preset requirements are: ， Among them, denotes the transpose of denotes a fault detector.

5. 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 fault detection method based on data-driven of unmanned systems according to any one of claims 1-4.

6. A terminal device, characterized in that, It 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 the fault detection method based on data-driven of unmanned systems according to any one of claims 1-4.

Images