Failure diagnosis system, failure diagnosis method, failure diagnosis program, and moving body

Abstract

To autonomously diagnose failures of a plurality of rudders.SOLUTION: A failure diagnosis system 40 applied to an underwater vehicle that controls an amount of control related to a posture of a body of the underwater vehicle by means of a plurality of rudders the number of which is larger than the number of amounts of control, includes: an estimation unit 41 that estimates, for each rudder, an amount of operation of another rudder, based on a command value of the rudder and an amount of state of a moving body; and a determination unit 42 that performs a failure determination of the plurality of rudders by using the plurality of estimated amounts of operation. The determination unit 42 performs: a first failure determination of determining whether or not a failure has occurred in any of the rudders, by using the plurality of amounts of operation corresponding to the respective rudders; and a second failure determination of determining whether or not a failure has occurred in any of the rudders other than the failed rudder, by using, as target amounts of operation, a plurality of amounts of operation excluding the amount of operation estimated based on the command value of the failed rudder determined to have the failure, when it is determined that the failure has occurred in any of the rudders.SELECTED DRAWING: Figure 5

Description

[Technical Field]

[0001] The present disclosure relates to a fault diagnosis system, a fault diagnosis method, a fault diagnosis program, and a mobile object. [Background technology]

[0002] In a moving body such as an underwater vehicle or a flying body, the attitude of the body is controlled by an operating means. For example, in an underwater vehicle, the attitude of the body is controlled by controlling a rudder provided on the body.

[0003] During navigation, for example, if the aircraft collides with a reef, mechanical damage (such as sticking or deformation) may occur to a part of the rudder. If one of the multiple rudders fails, it may cause problems in controlling the attitude of the aircraft. In such a case, it may become impossible to navigate along the target trajectory, and in the case of an unmanned aircraft, there is a possibility that the aircraft may be lost. For this reason, for example, Patent Document 1 proposes a technology that autonomously detects failures in the operating means. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Patent Publication No. 2021-77015 Summary of the Invention [Problem to be solved by the invention]

[0005] However, in the failure diagnosis disclosed in Patent Document 1, there is a risk that failures may not be detected when failures occur in multiple rudders.

[0006] The present disclosure has been made in consideration of the above circumstances, and aims to provide a fault diagnosis system, a fault diagnosis method, a fault diagnosis program, and a mobile body that can autonomously diagnose multiple rudder faults. [Means for solving the problem]

[0007] One aspect of the present disclosure is a fault diagnosis system applied to a mobile body in which control variables related to the attitude of the body are controlled by a plurality of operating means, the number of which is greater than the number of the control variables, and the fault diagnosis system includes: an estimation unit that, for each operating means, estimates the operating variables of the other operating means based on a command value of the operating means and a state variable of the mobile body; and a determination unit that performs a fault determination for the plurality of operating means using the estimated plurality of operating variables, wherein the determination unit performs a first fault determination that determines whether or not a fault has occurred in any of the operating means using the plurality of operating variables corresponding to each of the operating means; and, if it is determined that a fault has occurred in any of the operating means, performs a second fault determination that determines whether or not a fault has occurred in any of the operating means other than the faulty operating means, using the plurality of operating variables as target operating variables, excluding the operating variable estimated based on the command value of the faulty operating means determined to be faulty.

[0008] One aspect of the present disclosure is a moving body including a plurality of operating means for controlling the attitude of the body and the fault diagnosis system described above.

[0009] One aspect of the present disclosure is a fault diagnosis method applied to a mobile body in which control variables related to the attitude of the body are controlled by a plurality of operating means, the number of which is greater than the number of the control variables, in which a computer executes, for each of the operating means, an estimation process that estimates the operating variables of the other operating means based on a command value of the operating means and a state variable of the mobile body, and a judgment process that performs a fault judgment on the plurality of operating means using the estimated plurality of operating variables, and the judgment process includes: a first fault judgment process that determines whether a fault has occurred in any of the operating means using the plurality of operating variables corresponding to each of the operating means; and a second fault judgment process that, if it is determined that a fault has occurred in any of the operating means, uses the plurality of operating variables excluding the operating variable estimated based on the command value of the faulty operating means that has been judged to be faulty, as target operating variables, to determine whether a fault has occurred in any of the operating means other than the faulty operating means.

[0010] One aspect of the present disclosure is a program for causing a computer to function as the fault diagnosis system. [Effects of the Invention]

[0011] According to the present disclosure, it is possible to autonomously diagnose multiple rudders for faults. [Brief explanation of the drawings]

[0012] [Figure 1] 1 is a diagram illustrating an underwater vehicle according to an embodiment of the present disclosure. FIG. [Figure 2] FIG. 1 illustrates an X-rudder according to an embodiment of the present disclosure. [Figure 3] FIG. 2 is a diagram illustrating an example of a hardware configuration of a control device according to an embodiment of the present disclosure. [Figure 4] FIG. 2 is a functional block diagram illustrating functions of a control device according to an embodiment of the present disclosure. [Figure 5] FIG. 2 is a diagram illustrating a configuration example of a fault diagnosis unit according to an embodiment of the present disclosure. [Figure 6] FIG. 2 is a diagram illustrating a configuration example of a first fault candidate identifying unit according to an embodiment of the present disclosure. [Figure 7] 10 is a flowchart illustrating an example of a procedure of a determination process in a determination unit according to an embodiment of the present disclosure. [Figure 8] 10 is a flowchart showing an example of a processing procedure of a failure candidate identification process related to the No. 1 rudder that is performed in the first failure determination process according to an embodiment of the present disclosure. [Figure 9] 10 is a flowchart showing an example of a processing procedure of a failure candidate identification process related to the No. 1 rudder that is performed in the second failure determination process according to an embodiment of the present disclosure. [Figure 10] 10 is a flowchart showing an example of a processing procedure of a failure candidate identification process related to the No. 1 rudder that is performed in the second failure determination process according to an embodiment of the present disclosure. [Figure 11] FIG. 10 is a diagram showing an example of a simulation result of aircraft control when a conventional fault diagnosis unit is used. [Figure 12]FIG. 10 is a diagram illustrating an example of a simulation result of aircraft control when a fault diagnosis unit according to an embodiment of the present disclosure is used. [Figure 13] FIG. 10 is a diagram illustrating a configuration example of a Kalman filter according to a modified example of the present disclosure. DETAILED DESCRIPTION OF THE INVENTION

[0013] A fault diagnosis system, a fault diagnosis method, a fault diagnosis program, and a mobile body according to an embodiment of the present disclosure will be described below with reference to the drawings. In this embodiment, an underwater vehicle that navigates underwater will be described as an example of the mobile body, but the mobile body is not limited to this. For example, the mobile body may be any mobile body that is controlled by redundant operating means (e.g., rudders), and may be manned or unmanned. Other examples of mobile bodies include aircraft, rockets, drones, etc. Furthermore, the fault diagnosis system can be widely applied to the above-mentioned mobile bodies.

[0014] In the following description, the angle of rotation around the x-axis in the absolute coordinate system (Earth coordinate system) is defined as roll angle φ, the angle of rotation around the y-axis as pitch angle θ, and the angle of rotation around the z-axis as yaw angle ψ. The roll angle φ, pitch angle θ, and yaw angle ψ indicate the attitude of the underwater vehicle 1. That is, in the underwater vehicle 1, the roll angle φ, pitch angle θ, and yaw angle ψ related to the attitude of the vehicle are controlled as control objects (control variables), and attitude control is performed by an operating means (e.g., a rudder) described later. Note that the depth Z (z-axis direction) can be changed by controlling the pitch angle θ as attitude, and therefore the pitch angle θ and depth Z are correlated with each other.

[0015] FIG. 1 is a diagram illustrating an underwater vehicle 1 according to an embodiment of the present disclosure. As shown in FIG. 1, the attitude of the underwater vehicle 1 is defined by six axes (six degrees of freedom) of a hull coordinate system consisting of three mutually orthogonal linear axes: the stern axis (hereinafter referred to as the "x-axis"), the lateral axis (hereinafter referred to as the "y-axis"), and the vertical axis (hereinafter referred to as the "z-axis"), and three rotational axes relative to these linear axes. The velocity in the x-axis direction is defined as velocity u, the velocity in the y-axis direction as velocity v, and the velocity in the z-axis direction as velocity w. The rotational angular velocity around the x-axis is defined as angular velocity p, the rotational angular velocity around the y-axis as angular velocity q, and the rotational angular velocity around the z-axis as angular velocity r. In other words, the velocities u, v, w, angular velocity p, q, and r are state quantities of the vehicle.

[0016] In this embodiment, the state quantities used are velocity u, velocity v, velocity w, angular velocity p, angular velocity q, and angular velocity r. Furthermore, the state quantities used are roll angle φ, pitch angle θ, yaw angle ψ, and depth Z. The state quantities of the underwater vehicle 1 are values ​​that indicate the state (attitude, etc.) of the underwater vehicle 1, and are detected, for example, by sensors provided on the underwater vehicle 1. Each detected state quantity is used in the control device 20, which will be described later. The sensor is, for example, an inertial navigation system provided inside the vehicle. By providing each sensor inside the vehicle, it is possible to suppress the occurrence of underwater pressure resistance and an increase in fluid resistance, etc. Note that the method of acquiring each state quantity is not limited to using sensors, and other methods, such as estimating and acquiring using a simulation model, can also be used. As long as the state quantities of the vehicle are acquired, there are no limitations on the method of specifying (calculating) each state quantity.

[0017] The control device 20 controls the roll angle φ, pitch angle θ, and yaw angle ψ of the underwater vehicle 1 (attitude control) by controlling a plurality of operating means provided on the vehicle. The operating means are means for controlling the attitude of the underwater vehicle 1, and rudders are an example. The operating means is not limited to rudders as long as it can control the attitude of the vehicle and is driven by an actuator or the like. For example, thrusters or the like may be used. In this embodiment, the operating means will be described as using four rudders (X rudders) R, namely, a first rudder R1, a second rudder R2, a third rudder R3, and a fourth rudders R4, as shown in FIG. 2. Note that FIG. 2 is a view from the stern, and also shows a propeller 2 for generating propulsive force. The propulsive force from the propeller 2 generates force in the x-axis direction. The control device 20 controls the rudders R1 to R4 of the X rudders to control the roll angle φ, pitch angle θ, and yaw angle ψ, which are control targets related to the vehicle's attitude. In this way, the number of control objects related to the attitude of the vehicle (three in this embodiment) is greater than the number of operation means (four in this embodiment), so the underwater vehicle 1 of this embodiment is a redundant system. The control objects may be controlled by four or more operation means provided on the vehicle, with the three elements of attitude being the pitch angle θ, yaw angle ψ, and roll angle φ of the vehicle. Any mobile vehicle in which the control objects related to the attitude of the vehicle are controlled by a plurality of operation means that are greater than the number of control objects can be similarly adapted as a redundant system.

[0018] Specifically, the control device 20 controls the attitude of the vehicle by giving control command values ​​to the first rudder actuator, the second rudder actuator, the third rudder actuator, and the fourth rudder actuator provided on the underwater vehicle 1. By driving each actuator, the angle (rudder angle) of each rudder is changed, thereby controlling the attitude of the vehicle.

[0019] FIG. 3 is a diagram showing an example of the hardware configuration of a controller 20 according to this embodiment. As shown in FIG. 3, the controller 20 includes, for example, a CPU (Central Processing Unit: processor) 11, a main memory 12, and a secondary storage 13. The controller 20 may also include an external interface 14 for connecting an external device, a communication interface 15 for communicating with the external device, and the like. The controller 20 may also include input devices such as a keyboard and a mouse, and a display unit such as a liquid crystal display device for displaying data. The components constituting the controller 20 are connected to each other directly or indirectly via a bus and cooperate with each other to perform various processes.

[0020] The CPU 11 controls the entire control device 20 using, for example, an OS (Operating System) stored in a secondary storage device 13 connected via a bus, and executes various processes by executing various programs stored in the secondary storage device 13. One or more CPUs 11 may be provided, and they may work together to realize processes. The main storage device 12 is composed of writable memory such as cache memory and RAM (Random Access Memory), and is used as a working area for reading out programs executed by the CPU 11 and writing data processed by the programs. The secondary storage device 13 is a non-transitory computer-readable storage medium. Examples of the secondary storage device 13 include a magnetic disk, a magneto-optical disk, a CD-ROM, a DVD-ROM, and a semiconductor memory. Examples of the secondary storage device 13 include a read-only memory (ROM), a hard disk drive (HDD), and a solid-state drive (SSD) flash memory. The secondary storage device 13 stores, for example, an operating system (OS) for controlling the entire control device 20, such as Windows (registered trademark), iOS (registered trademark), or Android (registered trademark), a basic input / output system (BIOS), various device drivers for operating peripheral devices, various application software, and various data and files. The secondary storage device 13 also stores programs for implementing various processes and various data required for implementing the various processes. A plurality of secondary storage devices 13 may be provided, and the programs and data described above may be stored separately in each secondary storage device 13.

[0021] 4 is a functional block diagram showing an example of functions of the control device 20. For example, a series of processes for realizing the functions of the control device 20 is stored in the secondary storage device 13 or the like in the form of a program. The CPU (processor) 11 reads this program into the main storage device 12 and executes information processing and arithmetic processing to realize various functions. The program may be pre-installed in the secondary storage device 13, provided in a state stored in another non-transitory computer-readable storage medium, or distributed via wired or wireless communication means. Examples of non-transitory computer-readable storage media include magnetic disks, magneto-optical disks, CD-ROMs, DVD-ROMs, and semiconductor memories.

[0022] 4, the control device 20 includes a steering angle command calculation unit 30 and a fault diagnosis unit (fault diagnosis system) 40. The control device 20 may also include a degeneration command unit 50, a fault-tolerant control unit 60, etc.

[0023] (Configuration of steering angle command calculation unit) The rudder angle command calculation unit 30 calculates a command value for each rudder in the X rudder based on the target value. The rudder angle command calculation unit 30 includes, for example, a virtual rudder angle calculation unit 31 and a distribution unit 32. The target value is a target value for a control variable related to the attitude of the aircraft. The control variable is a quantity that is desired to be controlled to a target value among quantities belonging to the control object. The target values ​​are target values ​​corresponding to the roll angle φ, pitch angle θ, and yaw angle ψ. Since the pitch angle θ is related to the depth Z of the aircraft, in the virtual rudder angle calculation unit 31 according to this embodiment, a case will be described in which the depth controller 33 calculates the target value of the pitch angle θ using the target value of the depth Z. However, since a control variable related to the attitude of the aircraft can be used as the target value, target values ​​corresponding to the roll angle φ, pitch angle θ, and yaw angle ψ may also be input to the virtual rudder angle calculation unit 31. The rudder angle command calculation unit 30 is not limited to the configuration shown in FIG. 4, as long as it can calculate command values ​​corresponding to each rudder based on target values ​​for control variables related to the attitude of the aircraft (roll angle φ, pitch angle θ, and yaw angle ψ).

[0024] The virtual rudder angle calculation unit 31 calculates command values ​​corresponding to the virtual rudder (lateral rudder, vertical rudder, and roll rudder) based on target values ​​of control variables related to the attitude of the aircraft. The target values ​​are the depth target value Z*, the yaw angle target value ψ*, and the roll angle target value φ*. The depth target value Z* is used to calculate the pitch angle target value θ*, so the command value of the virtual rudder is calculated based on the roll angle target value φ*, the pitch angle target value θ*, and the yaw angle target value ψ*. The virtual rudder is a virtually set rudder, specifically the horizontal rudder, vertical rudder, and roll rudder.

[0025] The deviation between each target value and the state quantity corresponding to each target value is input to the virtual steering angle calculation unit 31. Specifically, the deviation between the depth target value Z* and the state quantity of the depth, the deviation between the yaw angle target value ψ* and the state quantity of the yaw angle ψ, and the deviation between the roll angle target value φ* and the state quantity of the roll angle φ are input to the virtual steering angle calculation unit 31. The state quantity of the pitch angle θ is also input to the virtual steering angle calculation unit 31.

[0026] The virtual rudder angle calculation unit 31 includes a depth controller 33, an attitude angle controller 36, a heading controller 34, and a roll controller 35. The depth controller 33 calculates a pitch angle target value θ* based on the deviation between the depth target value Z* and the state quantity of the depth. Then, the deviation between the pitch angle target value θ* and the state quantity of the pitch angle θ is input to the attitude angle controller 36. The attitude angle controller 36 calculates a rudder command value based on the deviation between the pitch angle target value θ* and the state quantity of the pitch angle θ. In other words, the command value of the rudder, which is a virtual rudder, is calculated based on the pitch angle target value θ*.

[0027] The heading controller 34 calculates a command value for the vertical rudder based on the deviation between the target yaw angle value ψ* and the state quantity of the yaw angle ψ. That is, the command value for the vertical rudder, which is a virtual rudder, is calculated based on the target yaw angle value ψ*.

[0028] The roll controller 35 calculates a command value for the roll rudder based on the deviation between the roll angle target value φ* and the state quantity of the roll angle φ. In other words, the command value for the roll rudder, which is a virtual rudder, is calculated based on the roll angle target value φ*.

[0029] Once the command values ​​for the lateral rudder, vertical rudder, and roll rudder are calculated, they are output to the distribution unit 32. A nullifying unit 37 is provided between the virtual rudder angle calculation unit 31 and the distribution unit 32. The nullifying unit 37 is controlled by a degeneration command unit 50, which will be described later. Specifically, when there is no degeneration command from the degeneration command unit 50, the nullifying unit 37 outputs the output value from the virtual rudder angle calculation unit 31 to the distribution unit 32 as is. When there is a degeneration command from the degeneration command unit 50, the nullifying unit 37 nullifies the output value (specific output) from the virtual rudder angle calculation unit 31. In this embodiment, since a low priority is set for the roll angle φ, as a specific configuration, the nullifying unit 37 is provided on the output side of the roll controller 35. That is, when there is a degeneration command from the degeneration command unit 50, which will be described later, the output of the roll controller 35 is nullified, and 0 is input to the distribution unit 32 as the roll rudder command value. It should be noted that, depending on the setting of the priority regarding degeneration, it is possible to adopt a configuration other than the case where the invalidation unit 37 is provided on the output side of the roll controller 35.

[0030] The distribution unit 32 calculates command values ​​for the X rudder (No. 1 rudder R1, No. 2 rudder R2, No. 3 rudder R3, and No. 4 rudder R4), which are actual rudder, based on the command values ​​of the lateral rudder, vertical rudder, and roll rudder, which are virtual rudder. For example, the command value for the X rudder is calculated by numerically converting (for example, multiplying by a constant) each command value for the lateral rudder, vertical rudder, and roll rudder, and adding or subtracting the numerically converted values. For example, the command value for the No. 1 rudder R1 is calculated by adding or subtracting a value based on the command value for the lateral rudder, a value based on the command value for the vertical rudder, and a value based on the command value for the roll rudder. Note that the method of calculating the command value for the X rudder based on the virtual rudder is not limited to the above.

[0031] The command value to the X rudder calculated by the distribution unit 32 is output to the actuator of each rudder in the X rudder, and the rudder is controlled.

[0032] (Configuration of the fault diagnosis unit) The fault diagnosis unit 40 performs fault diagnosis on the X rudder. That is, the fault diagnosis unit 40 identifies the faulty rudder, which is the rudder in which a fault has occurred. As shown in FIG. 5, the fault diagnosis unit 40 includes an estimation unit 41 and a determination unit 42. The fault diagnosis unit 40 receives, as state quantities of the underwater vehicle 1, inputs such as the speed u, speed v, speed w, angular velocity p, angular velocity q, angular velocity r, pitch angle θ, and roll angle φ. The state quantities are current values ​​(present values) of predetermined parameters of the underwater vehicle 1. Note that the state quantities used by the fault diagnosis unit 40 are not limited to those described above, and are set depending on the configuration of the observer described below. For example, the state quantities can be at least one of the speed u, speed v, speed w, angular velocity p, angular velocity q, angular velocity r, pitch angle θ, yaw angle ψ, and roll angle φ. The state quantities of the vehicle are measured, for example, by sensors or the like provided on the vehicle and output to the estimation unit 41.

[0033] The estimation unit 41 estimates the operation amount of each of the other rudders based on the rudder command value and the state quantity of the underwater vehicle 1 for each rudder. Specifically, for the first rudder R1, the estimation unit 41 estimates the operation amount of each of the other rudders (i.e., the second rudder R2, the third rudder R3, and the fourth rudder R4) based on the command value of the first rudder R1 and the state quantity of the underwater vehicle 1 (estimation process). The operation amount is the angle (rudder angle) of the rudder when driven based on the command value. The estimation unit 41 also performs the above estimation process for each of the second rudder R2, the third rudder R3, and the fourth rudder R4.

[0034] The estimation unit 41 performs estimation processing using an observer, for example, as shown in Fig. 5. An observer is provided for each rudder. Fig. 5 specifically shows the configuration of only the observer OS1, and the configurations of the other observers (observer OS2, observer OS3, observer OS4) are shown in simplified form, but they have the same configuration as observer OS1.

[0035] If the state (state variable) is x and the input is the amount of operation for each rudder, the linear approximation model (state equation) is given by the following equation (1). The amount of operation for rudder No. 1 R1 is δr1, the amount of operation for rudder No. 2 R2 is δr2, the amount of operation for rudder No. 3 R3 is δr3, and the amount of operation for rudder No. 4 R4 is δr4.

[0036]

number

[0037] In equation (1), A is the state matrix (A matrix), and B (a matrix with elements b1, b2, b3, and b4) is the input matrix (B matrix). The B matrix is ​​the rudder influence coefficient, and is also called the lift coefficient or sensitivity coefficient. Equation (1) shows the relationship between the operation amount for each rudder and the state variables. The state variables are matrices of each state variable (speed u, speed v, speed w, angular velocity p, angular velocity q, angular velocity r, pitch angle θ, and roll angle φ) input to the estimation unit 41. In equation (1), for example, if the first rudder R1 is considered normal, equation (1) can be transformed into the following equation (2) using the B matrices for the second rudder R2, the third rudder R3, and the fourth rudder R4.

[0038]

number

[0039] Equation (2) can be transformed into equation (3) using a pseudo-inverse matrix.

[0040]

number

[0041] That is, assuming that one rudder (first rudder R1 in equation (3)) is normal, in other words, assuming that the operation amount and command value are equal, the operation amount of the other rudder is calculated by using each state quantity. Note that each element of matrix A and matrix B is set in advance based on the characteristics of the underwater vehicle 1.

[0042] The estimation unit 41 estimates the rudder operation amount by an observer that uses the calculation principle of equation (3). Specifically, as shown in Fig. 5, the observer OS1 multiplies the state quantity by the A matrix and also multiplies the state quantity by s. Multiplying the state quantity by s corresponds to differentiating the state quantity in the time domain. Then, assuming that no failure has occurred in the first rudder R1 (i.e., assuming that the operation amount and the command value are equal), the command value of the first rudder R1 is multiplied by the component (b1) of the B matrix that corresponds to the first rudder R1. Then, a value is calculated by subtracting the value obtained by multiplying the state quantity x by the A matrix and the value obtained by multiplying the command value of the first rudder R1 by b1 from the value obtained by differentiating the state quantity x. The calculated value corresponds to the left side of equation (2). Then, the pseudo-inverse matrix B of the B matrix components that correspond to the second rudder R2, the third rudder R3, and the fourth rudder R4 is calculated. + 234 By multiplying by , the right side of equation (3) is calculated, and the estimated values ​​dr of the rudder angles of the second rudder R2, the third rudder R3, and the fourth rudder R4 are obtained. 21 ,dr 31 ,dr 41 Calculate.

[0043] Similar calculations are performed for the other observers. Note that the pseudo-inverse matrices of the component (b1) of the B matrix corresponding to the first rudder R1 in the observer OS1, and the B matrix components corresponding to the second rudder R2, the third rudder R3, and the fourth rudder R4 are set for each observer.

[0044] The estimation unit 41 performs the estimation process corresponding to the first rudder R1 as described above for the other rudders. That is, by performing the estimation process assuming that the second rudder R2 is normal, the estimated values ​​dr of the rudder angles of the first rudder R1, the third rudder R3, and the fourth rudder R4 are calculated. 12 ,dr 32 ,dr 42 (Observer OS2) By performing estimation processing assuming that the third rudder R3 is normal, the estimated values ​​dr of the rudder angles of the first rudder R1, the second rudder R2, and the fourth rudder R4 are calculated. 13 ,dr 23 ,dr 43(Observer OS3) By performing estimation processing assuming that the fourth rudder R4 is normal, the estimated values ​​dr of the rudder angles of the first rudder R1, the second rudder R2, and the third rudder R3 are calculated. 14 ,dr 24 ,dr 34 (observer OS4). In this way, three rudder angle estimates corresponding to each rudder are calculated (12 in total) by each estimation process.

[0045] The estimated value (operation amount) of the rudder angle calculated in this way is output to the determination unit 42. The determination unit 42 uses the estimated values ​​of the plurality of rudder angles to perform failure determination for the plurality of rudders R1 to R4. Specifically, the judgment unit 42 performs a first failure judgment and a second failure judgment. In the first failure judgment, a plurality of operation amounts (estimated values ​​of rudder angles) corresponding to each rudder is used to judge whether or not a failure has occurred in any of the rudder. In the second failure judgment, if it is judged in the first failure judgment that a failure has occurred in any of the rudder, a plurality of operation amounts excluding the operation amount (estimated value of rudder angle) estimated based on the command value of the faulty rudder judged to have failed is used as target operation amounts to further judge whether or not a failure has occurred in any of the rudder other than the faulty rudder.

[0046] For example, the determination unit 42 includes a failure candidate identification unit 43 and a failed rudder identification unit 44. The fault candidate identification unit 43 identifies fault candidates for each rudder using a plurality of comparison parameters calculated based on the estimated value and command value of the rudder angle, and outputs a fault candidate flag corresponding to each identified fault candidate.

[0047] More specifically, the fault candidate identification unit 43 includes a first fault candidate identification unit 43a corresponding to the first rudder R1, a second fault candidate identification unit 43b corresponding to the second rudder R2, a third fault candidate identification unit 43c corresponding to the third rudder R3, and a fourth fault candidate identification unit 43d corresponding to the fourth rudder R4.

[0048] The first fault candidate identifying unit 43a includes the respective rudder angles dr estimated for the first rudder R1. 12 ,dr. 13 ,dr. 14and the command value δr1 of the first rudder R1. The second fault candidate identifying unit 43b receives the respective rudder angles dr estimated for the second rudder R2. 21 ,dr 23 ,dr 24 and the command value δr2 of the second rudder R2 are input. The third fault candidate identifying unit 43c receives the respective rudder angles dr estimated for the third rudder R3. 31 ,dr 32 ,dr 34 and the command value δr3 of the third rudder R3 are input. The fourth fault candidate identifying unit 43d receives the estimated rudder angles dr 41 ,dr 42 ,dr 43 and the command value δr4 for the fourth rudder R4 is input.

[0049] Fig. 6 is a functional configuration diagram showing the functions of the first fault candidate identifying unit 43a. As shown in Fig. 6, the first fault candidate identifying unit 43a includes a parameter calculation unit 46 and an identifying unit 47. The parameter calculation unit 46 calculates the rudder angles dr estimated for the first rudder R1. 12 ,dr. 13 ,dr. 14 and the command value δr1 of the first rudder R1, the comparison parameters are calculated. 12 is the amount of operation estimated based on the command value of the second rudder R2, dr 13 is the estimated operation amount based on the command value of the third rudder R3, dr 14 indicates the amount of operation estimated based on the command value of the fourth rudder R4. For example, the parameter calculation unit 46 calculates the steering angle dr 12 ,dr. 13 ,dr. 14 The deviation between the command value δr1 and the reference value δr2 is calculated, and the absolute value of the deviation is used as the comparison parameter e 12 , e 13 , e 14 Output as

[0050] The identification unit 47 uses the plurality of comparison parameters e calculated by the parameter calculation unit 46. 12 , e 13 , e 14and outputs a fault candidate flag FLc indicating the identified fault candidate rudder. For example, the identification unit 47 uses a plurality of comparison parameters e 12 , e 13 , e 14 If there is a comparison parameter among these that has a value greater than the other comparison parameters by a predetermined threshold value ε1 or more, that comparison parameter is identified as a faulty parameter.

[0051] If the parameter is greater than the other comparison parameters by a predetermined value or more, it is assumed that there was an error in the estimation process. Therefore, the rudder that was assumed to be normal in the rudder angle estimation process used to calculate the fault parameter is identified as a candidate for the faulty rudder. For example, the specification unit 47 may use the comparison parameter e 12 is identified as a fault parameter, the second rudder R2 is identified as a faulty candidate rudder, and a fault candidate flag FLc=2 indicating this second rudder R2 is output. Note that, if the differences of all comparison parameters are within the threshold, it is determined that there is no faulty rudder, and a fault candidate flag FLc=0 is output.

[0052] Then, the second fault candidate identifying section 43b, the third fault candidate identifying section 43c, and the fourth fault candidate identifying section 43d perform similar processing, thereby outputting a fault candidate flag FLc, respectively. As a result, the fault candidate identifying section 43 outputs, for example, four fault candidate flags FLc to the faulty rudder identifying section 44.

[0053] The faulty rudder identifying unit 44 identifies the faulty rudder based on the multiple fault candidate flags output from the fault candidate identifying unit 43. For example, when all the fault candidate flags FLc indicate the same rudder, the faulty rudder identifying unit 44 identifies that rudder as a faulty rudder and outputs the fault flag FL corresponding to the faulty rudder. Note that the identification of the faulty rudder based on the fault candidate flags FLc is not limited to this method. For example, it may be determined by majority vote, or other methods may be used. Also, the faulty rudder may be identified by analyzing the information output from the fault candidate identifying unit 43. For the analysis, a statistical method may be used, or machine learning such as AI may be used. When a fault flag FL is output from the faulty rudder identifying section 44, the fault flag FL is fed back to the fault candidate identifying section 43.

[0054] In this way, when the failure flag FL is fed back, the judgment unit 42 makes a second failure judgment (second failure judgment). In the second failure judgment, the estimation unit 41 uses, as target operation amounts, the operation amounts (specifically, comparison parameters) estimated based on the command value of the rudder judged to be faulty, in other words, a plurality of operation amounts excluding the rudder angle estimated by the estimation process assuming that the rudder judged to be the faulty rudder is normal, to judge whether or not a failure has occurred in another rudder.

[0055] The second failure determination is the same as the first failure determination described above, except that the rudder angle estimated by the estimation process assuming that the rudder determined as the faulty rudder is normal is excluded. For example, if the second rudder R2 is determined to be abnormal in the first failure determination, the rudder angle dr 12 ,dr. 32 ,dr. 42 The fault candidate is identified using the rudder angle (target operation amount) excluding the rudder angle. Specifically, the comparison parameter e for the second rudder R2 is 12 , e 32 , e 42 Fault candidates are identified using parameters other than the above. In this way, by performing a second failure judgment, when multiple rudders have failed, it is possible to identify the failed rudder without overlooking the detection of the second failed rudder.

[0056] Next, the determination process executed by the above-mentioned determination unit 42 will be described with reference to Fig. 7. Fig. 7 is a flowchart showing an example of the processing procedure of the determination process executed by the determination unit 42. In the various processes described below, steps S1 to S6 correspond to a first fault determination process, and steps S7 to S10 correspond to a second fault determination process.

[0057] First, in the first fault determination process, various data required for fault determination, such as the rudder angle estimated by the estimation unit 41, are acquired (S1). Next, a comparison parameter is calculated for each rudder (S2). As a result, for the first rudder R1, the comparison parameter e 12 , e 13 , e 14 However, for the second rudder R2, the comparison parameter e 21 , e 23 , e 24 However, for the third rudder R3, the comparison parameter e 31 , e 32 , e 34 However, for the fourth rudder R4, the comparison parameter e 41 , e 42 , e 43 are calculated respectively.

[0058] Next, a fault candidate identification process is performed for each rudder, and a fault candidate flag FLc is output (S3). Details of this process will be described later. Next, a determination is made as to whether a rudder is at fault based on the fault candidate flag FLc output for each rudder (S4). Specifically, it is determined whether the rudder indicated by the four fault candidate flags FLc output from the first fault candidate identifying unit 43a to the fourth fault candidate identifying unit 43d is the same (S5). As a result, if the fault candidate flags FLc are not the same, it is determined that there is no rudder in which a fault has occurred (S5: NO), and the process is terminated.

[0059] On the other hand, if all the failure candidate flags FLc indicate the same rudder, it is determined that there is a malfunctioning rudder (S5: YES), and a failure flag FL based on the malfunctioning rudder is output (S6). In this way, when a faulty rudder is identified in the first fault determination process, the second fault determination process, which is the second determination process, is subsequently executed.

[0060] In the second fault determination process, first, as in the first fault determination process described above, a fault candidate identification process is performed for each rudder, and a fault candidate flag is output (S7). However, in the fault candidate identification process of the second fault determination process, the comparison parameter calculated using the rudder angle estimated using the command value of the rudder identified as the faulty rudder is excluded from the comparison. For example, if the third rudder R3 is identified as the faulty rudder, the comparison parameter e 13 , e 23 , e 43 are excluded from the comparison.

[0061] Next, the malfunction candidate flag FLc output for each rudder is used to determine whether the rudder is malfunctioning (S8). As a result, if there is no malfunctioning rudder (S9: NO), the process ends. On the other hand, if a malfunctioning rudder is identified (S9: YES), the malfunction flag FL indicating the malfunctioning rudder is output (S10), and the process ends.

[0062] Next, the fault candidate identification process carried out in the first fault determination process will be described with reference to Fig. 8. The fault candidate identification process is carried out for each rudder, but since each process is carried out in the same procedure, the processing procedure in the first fault candidate identification unit 43a corresponding to the No. 1 rudder R1 will be described here as a representative example. Fig. 8 is a flowchart showing an example of the processing procedure of the fault candidate identification process for the No. 1 rudder R1 carried out in the first fault determination process.

[0063] First, the comparison parameter e 12 , e 13 , e 14 (SA1) Then, the comparison parameter e 12 is the other comparison parameter e 13 , e 14 If the result is a positive determination (SA2: YES), the comparison parameter e 12 The second rudder R2 associated with this is identified as a fault candidate, and a fault candidate flag based on the identified second rudder R2 is output (SA3). As a result, a fault candidate flag in which FLc is set to "2" (FLc=2) is output from the first fault candidate identifying unit 43a.

[0064] On the other hand, if the determination in step SA2 is negative (SA2: NO), the comparison parameter e 13 is the other comparison parameter e 12 , e 14 If the result is a positive determination (SA4: YES), the comparison parameter e 13 The third rudder R3 associated with this is identified as a fault candidate, and a fault candidate flag based on the identified third rudder R3 is output (SA5). As a result, a fault candidate flag in which FLc is set to "3" (FLc=3) is output from the first fault candidate identifying unit 43a.

[0065] On the other hand, if the determination in step SA4 is negative (SA4: NO), the comparison parameter e 14 is the other comparison parameter e 12 , e 13 If the result is a positive determination (SA6: YES), the comparison parameter e 14 The fourth rudder R4 associated with this is identified as a fault candidate, and a fault candidate flag based on the identified fourth rudder R4 is output (SA7). As a result, a fault candidate flag in which FLc is set to "4" (FLc=4) is output from the first fault candidate identifying unit 43a. If the determination in step SA6 is negative (SA6: NO), it is determined that there is no rudder that is a fault candidate, and the fault candidate flag is set to "0" and output (FLc=0) (SA8), and the process ends.

[0066] Next, the fault candidate identification process performed in the second fault determination process will be explained with reference to Figures 9 and 10. As with the first fault determination process, the fault candidate identification process is executed for each rudder. Since the fault candidate identification process is the same for each rudder, the processing procedure in the first fault candidate identification unit 43a corresponding to the No. 1 rudder R1 will be explained here as a representative example. Figures 9 and 10 are flowcharts showing an example of the processing procedure of the fault candidate identification process for the No. 1 rudder R1 performed in the second fault determination process.

[0067] First, it is determined whether the failure flag FL=1 (SB1). That is, it is determined whether the rudder itself has been identified as the failed rudder. As a result, if the determination is affirmative (SB1: YES), the process is terminated. On the other hand, if FL=1 is not true (SB1: NO), it is determined whether FL=2 (SB2). That is, it is determined whether the second rudder R2 has been identified as the failed rudder. As a result, if the determination is affirmative (SB2: YES), the comparison parameter e for the second rudder R2 is set to 0. 12 Excluding and comparing parameters e 13 , e 14 Specifically, the comparison parameter e 13 is the other comparison parameter e 14 If the result is a positive determination (SB3: YES), the comparison parameter e 13 The third rudder R3 associated with this is identified as a fault candidate, and a fault candidate flag based on the identified third rudder R3 is output (SB4). As a result, a fault candidate flag in which FLc is set to "3" (FLc=3) is output from the first fault candidate identifying unit 43a.

[0068] On the other hand, if the determination in step SB3 is negative (SB3: NO), the comparison parameter e 14 is the other comparison parameter e 13 If the result is a positive determination (SB5: YES), the comparison parameter e 14 The fourth rudder R4 associated with this is identified as a fault candidate, and a fault candidate flag based on the identified fourth rudder R4 is output (SB6). As a result, a fault candidate flag in which FLc is set to "4" (FLc=4) is output from the first fault candidate identifying unit 43a.

[0069] Furthermore, if the determination is negative in step SB5 (SB5: NO), it is determined that there is no rudder that is a failure candidate, and the failure candidate flag is set to "0" and output (FLc=0) (SB7), and the process ends.

[0070] If the determination in step SB2 is negative (SB2: NO), it is determined whether FL=3 (SB8). That is, it is determined whether the third rudder R3 has been identified as the faulty rudder. As a result, if the determination is positive (SB8: YES), the comparison parameter e for the third rudder R3 is 13 Excluding and comparing parameters e 12 , e 14 Specifically, the comparison parameter e 12 is the other comparison parameter e 14 If the result is a positive determination (SB9: YES), the comparison parameter e 12 The second rudder R2 associated with this is identified as a fault candidate, and a fault candidate flag based on the identified second rudder R2 is output (SB10). As a result, a fault candidate flag in which FLc is set to "2" (FLc=2) is output from the first fault candidate identifying unit 43a.

[0071] On the other hand, if the determination in step SB9 is negative (SB9: NO), the comparison parameter e 14 is the other comparison parameter e 12 If the result is a positive determination (SB11: YES), the comparison parameter e 14 The fourth rudder R4 associated with this is identified as a fault candidate, and a fault candidate flag based on the identified fourth rudder R4 is output (SB12). As a result, a fault candidate flag in which FLc is set to "4" (FLc=4) is output from the first fault candidate identifying unit 43a.

[0072] In addition, in the case of a negative determination in step SB8 (SB8: NO), FL=4, that is, the fourth rudder R4 is considered to have been identified as the faulty rudder, and therefore the comparison parameter e 14 Excluding and comparing parameters e 12 , e 13 Specifically, the comparison parameter e 12 is the other comparison parameter e 13If the result is a positive determination (SB13: YES), the comparison parameter e 12 The second rudder R2 associated with this is identified as a fault candidate, and a fault candidate flag based on the identified second rudder R2 is output (SB14). As a result, a fault candidate flag in which FLc is set to "2" (FLc=2) is output from the first fault candidate identifying unit 43a.

[0073] On the other hand, if the determination in step SB13 is negative (SB13: NO), the comparison parameter e 13 is the other comparison parameter e 12 If the result is a positive determination (SB15: YES), the comparison parameter e 13 The third rudder R3 associated with this is identified as a fault candidate, and a fault candidate flag based on the identified third rudder R3 is output (SB16). As a result, a fault candidate flag in which FLc is set to "3" (FLc=3) is output from the first fault candidate identifying unit 43a. Also, in the case of a negative determination in step SB15 and step SB11 of Fig. 9, it is determined that there is no rudder that is a fault candidate, and the fault candidate flag is set to "0" and output (FLc=0) (SB7), and the processing ends.

[0074] In addition, when a rudder failure is detected by the above-mentioned failure diagnosis unit 40, the rudder angle estimated for the rudder identified as the failed rudder may be used to calculate the failed rudder angle. For example, when the first rudder R1 is identified as the failed rudder (FL=1), the rudder angle dr estimated for the first rudder R1 is 12 ,dr. 13 ,dr. 14 The failed rudder angle of the first rudder R1 may be calculated by statistically processing the above. As an example of the statistical processing, an average value or a median value may be used.

[0075] (Effect of fault diagnosis processing by the fault diagnosis unit) Next, the effect of the fault diagnosis processing by the fault diagnosis unit 40 described above will be described with reference to Figs. 11 and 12. Figs. 11 and 12 are diagrams showing an example of changes in the attitude of the aircraft when a fault occurs in the fourth rudder R4 at time t1, and then a fault occurs in the third rudder R3 at time t2, a predetermined period after the fault occurred in the fourth rudder R4. Fig. 11 is a diagram illustrating changes in the attitude of the aircraft when conventional fault diagnosis is applied, and Fig. 12 is a diagram illustrating changes in the attitude of the aircraft when the fault diagnosis unit 40 according to this embodiment is applied. Figs. 11 and 12 show, from top to bottom, changes over time in the position on the X axis, the position on the Y axis, the depth, the roll angle, the pitch angle, and the direction. In each graph, the solid line indicates the change in the aircraft, and the dotted line indicates the target value.

[0076] In the conventional fault diagnosis shown in Fig. 11, since it is not possible to detect the failure of the third rudder R3 after the failure of the fourth rudder R4 has occurred, it can be seen that azimuth control has become impossible from time t2 onwards. In contrast, when the fault diagnosis unit 40 according to this embodiment is adopted, as shown in Fig. 12, it can be seen that the azimuth angle can be controlled in accordance with the target value even after the failure of the third rudder R3 has occurred at time t2.

[0077] (Configuration of the degeneration command unit) The degeneration command unit 50 outputs a command to the invalidation unit 37 in the rudder angle command calculation unit 30 to degenerate the control system in the rudder angle command calculation unit 30. Specifically, if the number of normally controllable operating means is equal to or greater than the number of control variables, the degeneration command unit 50 calculates a command value for each operating means based on a target value corresponding to each control variable. If the number of normally controllable operating means is less than the number of control variables, the degeneration command unit 50 degenerates the number of control variables used to calculate each command value and calculates a command value for each operating means based on at least one of the target values. Note that the number of faulty rudder is determined using the results of the fault diagnosis unit 40. Note that the degeneration command unit 50 is a known technique, so a detailed description thereof will be omitted here. For example, the degeneration command unit disclosed in Patent Document 1 can be employed.

[0078] (Configuration of fault-tolerant control unit) The fault-tolerant control unit 60 calculates command values ​​for each rudder so as to decouple the influence of the faulty rudder. For example, if one of the X rudder fails and becomes stuck at a certain rudder angle, the faulty rudder will exert a force on the aircraft according to the stuck rudder angle. Such a force exerted by the faulty rudder will affect the attitude control of the aircraft. For this reason, the fault-tolerant control unit 60 controls the other rudder (the rudder that operates normally) so as to suppress the influence of the faulty rudder. Note that the fault-tolerant control unit 60 is a well-known technique, so a detailed description thereof will be omitted here. For example, it is possible to adopt the fault-tolerant control unit disclosed in Patent Document 1.

[0079] As described above, according to this embodiment, there is provided an estimation unit 41 that estimates, for each rudder, the operation amount of each of the other rudder based on the command value of that rudder and the state amount of the underwater vehicle 1, and a determination unit 42 that performs a failure determination for the multiple rudder using the multiple estimated operation amounts. The determination unit 42 performs a first failure determination that determines whether or not a failure has occurred in any of the rudder using the multiple operation amounts corresponding to each rudder, and if it is determined that a failure has occurred in any of the rudder, performs a second failure determination that determines whether or not a failure has occurred in any of the rudder other than the failed rudder using the multiple operation amounts excluding the operation amount estimated based on the command value of the failed rudder that has been determined to be failed as target operation amounts.

[0080] In this way, if it is determined in the first failure determination that a failure has occurred in any of the rudder, a second failure determination is performed. In the second failure determination, a plurality of operation amounts excluding the operation amount estimated based on the command value of the failed rudder determined to be failed is used as a target to determine whether or not a failure has occurred in other rudder(s). This makes it possible to detect abnormalities in the plurality of failed rudder(s) without overlooking the detection of the second failed rudder when multiple rudder(s) have failed. In particular, if the influence of the second failure is smaller than that of the first failure, the influence of the second failure is overshadowed by the influence of the first failure and is unlikely to appear in the value of the comparison parameter. Even in such a case, according to this embodiment, the second failure determination is performed by excluding the operation amount estimated using the command value of the failed rudder detected in the first failure determination, so that the influences of the first rudder failure and the second rudder failure can be separated. This improves the accuracy of detecting the second rudder failure.

[0081] In this embodiment, the underwater vehicle 1 has four rudders, so the failure determination is performed twice, but the number of times the failure determination is performed is not limited to this. For example, if there are a large number of operating means, the failure determination may be performed three or more times. For the second and subsequent failure determinations, as described above, the estimated values ​​for the operating means determined to be faulty can be excluded from the comparison targets.

[0082] The fault diagnosis system, fault diagnosis method, fault diagnosis program, and mobile body of the present disclosure have been described above using embodiments, but the fault diagnosis system, fault diagnosis method, fault diagnosis program, and mobile body of the present disclosure are not limited to the above-described embodiments, and various modifications or improvements can be made to the above embodiments without departing from the gist of the present disclosure, and forms incorporating such modifications or improvements are also included in the technical scope of the present disclosure. Furthermore, the processing flow described in the above embodiment is just one example, and unnecessary steps can be deleted, new steps can be added, or the processing order can be rearranged without departing from the spirit of the present disclosure.

[0083] For example, in the above-described embodiment, a case has been described in which the fault diagnosis unit (fault diagnosis system) 40 is incorporated into the control device 20 of the underwater vehicle (mobile body) 1, but the present invention is not limited to this configuration. For example, the fault diagnosis unit 40 and the control device 20 may be provided separately, and a system configuration may be adopted in which information can be exchanged between the fault diagnosis unit 40 and the control device 20. In this case, the fault diagnosis unit (fault diagnosis system) may be installed in a location different from the mobile body, and may perform fault diagnosis remotely by exchanging information with a control device mounted on the mobile body via communication.

[0084] Furthermore, in the above embodiment, the first fault determination is not limited to the above-described method, and other known methods may be used to determine a rudder fault. For example, the first fault determination may be performed using the method disclosed in Patent Document 1.

[0085] Furthermore, in the above embodiment, if a rudder failure occurs, the control device 20 of the underwater vehicle 1 may reduce the vehicle's travel speed. For example, the vehicle may travel at approximately half the normal speed. The higher the speed, the stronger the steering force. Therefore, if a rudder failure is detected, reducing the speed can reduce the impact of the rudder failure on the vehicle's navigation. This increases the chances of reaching the destination.

[0086] Furthermore, in the above-described embodiment, the estimator 41 estimates each rudder angle using an observer, but this is not limiting. For example, instead of an observer, a Kalman filter as shown in FIG. 13 may be used to estimate the rudder angle. A Kalman filter does not require parameter design that takes into account the aircraft's bandwidth, making it easier to design than an observer. Furthermore, a Kalman filter is robust against measurement noise, and is expected to improve the detection rate.

[0087] In the above-described embodiment, the control device 20 of the underwater vehicle 1 may forcibly change the direction and depth during navigation and actively move the rudder R. For example, rudder failures include multiple failure patterns, such as sticking, floating, hard-over, missing, and offset. Sticking refers to a failure in which the rudder surface becomes stuck in a certain position and no longer responds to subsequent commands. Floating refers to a failure in which the rudder is not properly fixed, making it impossible to control the rudder angle to the desired angle. Hard-over refers to a situation in which the rudder surface moves to an upper or lower limit regardless of the command value. Missing refers to a situation in which a missing rudder occurs, causing the fluid force and fluid moment generated by the rudder to be smaller than the nominal state. Offset refers to a situation in which the actual angle of the rudder surface is offset from the command value, for example, due to a twist in the attachment between the rudder and the link.

[0088] Among these various failure patterns, detection of float or missing float faults becomes impossible or extremely difficult during straight sailing without rudder movement. Therefore, by actively changing the heading and depth and actively moving the rudder R to change the state quantities, it becomes possible to detect failure patterns such as float or missing float faults. For example, the control device 20 has a failure detection navigation mode that actively performs failure detection, and by executing this failure detection navigation mode at a predetermined timing, for example, at predetermined time intervals, it becomes possible to detect failure modes that are difficult to detect during straight sailing. In this failure detection navigation mode, for example, navigation is performed in which at least one of the heading and depth is forcibly changed.

[0089] Furthermore, as described above, there are failure modes that are likely to be affected by changes in the amount of operation, and failure modes that are unlikely to be affected by changes in the amount of operation. Therefore, the threshold value ε1 used when identifying a rudder that is a candidate for failure may be adjusted depending on the failure mode to be diagnosed. For example, when detecting a float or missing failure pattern, the value of the threshold value ε1 may be reduced to increase the sensitivity of failure determination.

[0090] The fault diagnosis system, fault diagnosis method, fault diagnosis program, and mobile body described in the above-described embodiments can be understood, for example, as follows.

[0091] A fault diagnosis system according to a first aspect of the present disclosure is a fault diagnosis system (40) applied to a moving body (1) that controls control variables (θ, ψ, φ) related to the attitude of the body by a plurality of operating means (R) that is greater than the number of the control variables, and includes: an estimation unit (41) that, for each operating means, estimates the operating variables of the other operating means based on a command value of the operating means and a state variable of the moving body; and a determination unit (42) that performs fault determination for the plurality of operating means using the estimated plurality of operating variables, wherein the determination unit performs a first fault determination for determining whether or not a fault has occurred in any of the operating means using the plurality of operating variables corresponding to each of the operating means; and, if it is determined that a fault has occurred in any of the operating means, performs a second fault determination for determining whether or not a fault has occurred in any of the operating means other than the faulty operating means, using the plurality of operating variables as target operating variables, excluding the operating variable estimated based on the command value of the faulty operating means determined to be faulty.

[0092] According to the above fault diagnosis system, when a fault occurs in a plurality of operating means, the faults in the plurality of operating means can be detected without overlooking the fault in the second operating means. Furthermore, in the second fault determination, the fault determination of the operating means is performed by excluding the operation amount estimated using the command value of the operating means detected in the first fault determination, so that the influence of the fault in the first operating means and the influence of the fault in the second operating means can be separated. This improves the accuracy of detecting the fault in the second operating means.

[0093] In a fault diagnosis system (40) according to a second aspect of the present disclosure, in the first aspect, the determination unit (42) performs the second fault determination using a plurality of comparison parameters calculated based on each of the target operation amounts and command values ​​for each of the operation means other than the faulty operation means.

[0094] For example, if an abnormality occurs in the operating means, the value of the operation amount estimated by the estimation unit will indicate an abnormal value. In this case, it is difficult to determine how much the operation amount deviates from the normal value (command value) by using a method of comparing each operation amount. In contrast, according to the fault diagnosis system, comparison parameters calculated using the target operation amount and the command value are used for fault determination, making it possible to more accurately identify the target operation amount indicating an abnormal value.

[0095] In a fault diagnosis system (40) according to a third aspect of the present disclosure, in the second aspect, if there is a comparison parameter for each of the operation means excluding the faulty operation means that has a value greater than the other comparison parameters by a predetermined threshold or more, the determination unit identifies a faulty candidate operation means based on the comparison parameter in the second fault determination, and determines whether a fault has occurred in any of the operation means other than the faulty operation means based on the plurality of faulty candidate operation means identified for each of the operation means.

[0096] According to the fault diagnosis system, for each operation means excluding the faulty operation means, a faulty candidate operation means is identified based on the comparison parameters. Then, based on the multiple faulty candidate operation means identified for each operation means, it is determined whether or not a fault has occurred in any operation means other than the faulty operation means. This makes it possible to easily detect a second fault.

[0097] In a fault diagnosis system (40) according to a fourth aspect of the present disclosure, in the third aspect, the determination unit identifies the operating means as the faulty operating means when the multiple fault candidate operating means identified for each operating means excluding the faulty operating means are all the same operating means.

[0098] According to the fault diagnosis system, whether or not a fault has occurred in an operating means is determined by comprehensively using the multiple fault candidate operating means identified for each operating means, thereby reducing the probability of erroneous detection of a faulty operating means.

[0099] A fault diagnosis system (40) according to a fifth aspect of the present disclosure is the third or fourth aspect, wherein the determination unit adjusts the threshold value depending on a fault mode to be diagnosed.

[0100] According to the above-described fault diagnosis system, it is possible to adjust the sensitivity of the fault diagnosis depending on the ease of detecting the fault mode, thereby making it possible to detect even fault modes that are difficult to detect.

[0101] A fault diagnosis system (40) according to a sixth aspect of the present disclosure is any one of the first to fifth aspects, wherein the determination unit performs the first fault determination using a plurality of comparison parameters calculated for each of the operation means based on the target operation amount and a command value.

[0102] According to the above fault diagnosis system, the comparison parameters calculated using the target manipulated variable and the command value are used for fault determination, so that it is possible to more accurately identify the manipulated variable that shows an abnormal value.

[0103] A fault diagnosis system (40) according to a seventh aspect of the present disclosure is configured such that, in the sixth aspect, if there is a comparison parameter for each operating means that has a value greater than the other comparison parameters by a predetermined threshold or more in the first fault judgment, the determination unit identifies a fault-candidate operating means based on the comparison parameter, and determines whether a fault has occurred in any of the operating means based on the plurality of fault-candidate operating means identified for each operating means.

[0104] According to the fault diagnosis system, for each operating means, a fault candidate operating means is identified based on the comparison parameters. Then, based on the multiple fault candidate operating means identified for each operating means, it is determined whether or not a fault has occurred in the operating means. This makes it possible to easily detect a fault in the operating means.

[0105] In a fault diagnosis system (40) according to an eighth aspect of the present disclosure, in the seventh aspect, the determination unit identifies an operating means as a faulty operating means when the multiple fault candidate operating means identified for each operating means are all the same operating means.

[0106] According to the fault diagnosis system, whether or not a fault has occurred in an operating means is determined by comprehensively using the multiple fault candidate operating means identified for each operating means, thereby reducing the probability of erroneous detection of a faulty operating means.

[0107] A fault diagnosis system (40) according to a ninth aspect of the present disclosure is the seventh or eighth aspect, wherein the determination unit adjusts the threshold value depending on a fault mode to be diagnosed.

[0108] According to the above-described fault diagnosis system, it is possible to adjust the sensitivity of the fault diagnosis depending on the ease of detecting the fault mode, thereby making it possible to detect even fault modes that are difficult to detect.

[0109] A moving body (1) according to a tenth aspect of the present disclosure includes a plurality of operating means (R) for controlling the attitude of the body, and a fault diagnosis system (40) according to any one of the first to ninth aspects.

[0110] The moving body (1) according to an eleventh aspect of the present disclosure reduces the moving speed when the failure diagnosis system (40) determines that a failure has occurred in the operating means (R).

[0111] According to the above-described moving body, when a malfunction of the operating means is detected, the speed is reduced, thereby suppressing the influence of the malfunction of the operating means on the navigation of the moving body, thereby increasing the possibility of navigating to the destination.

[0112] The moving body (1) according to a twelfth aspect of the present disclosure is the moving body (1) of the tenth or eleventh aspect, which has a fault determination navigation mode that forcibly changes at least one of the heading and the depth.

[0113] According to the above-described mobile body, it is possible to detect failure patterns that are difficult to detect when the mobile body is traveling in a straight line.

[0114] A fault diagnosis method according to a thirteenth aspect of the present disclosure is a fault diagnosis method applied to a moving body in which control variables related to the attitude of the body are controlled by a plurality of operating means, the number of which is greater than the number of the control variables, wherein a computer executes, for each of the operating means, an estimation process that estimates the operating variables of the other operating means based on a command value of the operating means and a state variable of the moving body, and a judgment process that performs a fault judgment on the plurality of operating means using the estimated plurality of operating variables, wherein the judgment process includes a first fault judgment process that uses the plurality of operating variables corresponding to each of the operating means to judge whether a fault has occurred in any of the operating means, and a second fault judgment process that, when it is judged that a fault has occurred in any of the operating means, uses the plurality of operating variables excluding the operating variable estimated based on the command value of the faulty operating means judged to be faulty as target operating variables to judge whether a fault has occurred in any of the operating means other than the faulty operating means.

[0115] A program according to a fourteenth aspect of the present disclosure is a program for causing a computer to function as the fault diagnosis system according to any one of the first to ninth aspects. [Explanation of symbols]

[0116] 1: Underwater vehicle 2: Propeller 11: CPU 12: Main memory 13:Secondary storage device 14: External interface 15: Communication interface 20: Control device 30: Steering angle command calculation unit 31: Virtual steering angle calculation unit 32:Distribution section 33: Depth controller 34: Orientation controller 35: Roll controller 36: Attitude angle controller 37: Invalidation section 40: Fault diagnosis unit (fault diagnosis system) 41:Estimation part 42: Judgment section 43: Failure candidate identification section 43a: 1st failure candidate identification section 43b:Second failure candidate identification section 43c: 3rd failure candidate identification section 43d: 4th failure candidate identification section 44:Faulty rudder identification section 46: Parameter calculation section 47: Specific part 50: Degeneration command unit 60: Fault-tolerant control unit FL: Fault flag FLc: Fault candidate flag R: Rudder (operation means) R1: Rudder 1 (operation means) R2: Rudder 2 (operation means) R3: Rudder 3 (operation means) R4: Rudder 4 (operation means) OS1: Observer OS2: Observer OS3: Observer OS4: Observer

Claims

1. 1. A fault diagnosis system applied to a moving body that controls control variables related to the attitude of the body by a plurality of operation means, the number of operation means being greater than the number of control variables, an estimation unit that estimates, for each of the operation means, an operation amount of each of the other operation means based on a command value of the operation means and a state amount of the moving object; a determination unit that determines whether or not the plurality of operation means have a malfunction using the plurality of estimated operation amounts; Equipped with The determination unit a first failure determination that determines whether or not a failure has occurred in any of the operation means by using the plurality of operation amounts corresponding to each of the operation means; When it is determined that a fault has occurred in any of the operating means, the fault diagnosis system performs a second fault determination in which a second fault determination is made to determine whether a fault has occurred in any of the operating means other than the faulty operating means, using a plurality of the operating amounts excluding the operating amount estimated based on the command value of the faulty operating means determined to be faulty, as target operating amounts.

2. 2. The fault diagnosis system according to claim 1, wherein the determination unit performs the second fault determination using a plurality of comparison parameters calculated based on each of the target operation amounts and command values ​​for each of the operation means other than the faulty operation means.

3. In the second fault determination, the determination unit If there is a comparison parameter for each of the operation means excluding the faulty operation means that has a value greater than the other comparison parameters by a predetermined threshold or more, a faulty candidate operation means is identified based on the comparison parameter; 3. The fault diagnosis system according to claim 2, wherein it is determined whether or not a fault has occurred in any of the operation means other than the faulty operation means based on the plurality of fault-candidate operation means identified for each of the operation means.

4. 4. The fault diagnosis system according to claim 3, wherein the determination unit identifies an operation means as the faulty operation means when the plurality of fault candidate operation means identified for each of the operation means excluding the faulty operation means are all the same operation means.

5. The fault diagnosis system according to claim 3 , wherein the determining unit adjusts the threshold value depending on a fault mode to be diagnosed.

6. 2. The fault diagnosis system according to claim 1, wherein the determination unit performs the first fault determination using a plurality of comparison parameters calculated for each of the operating means based on the target operating amount and a command value.

7. In the first fault determination, the determination unit If there is a comparison parameter for each of the operation means that has a value greater than the other comparison parameters by a predetermined threshold or more, a fault candidate operation means is identified based on the comparison parameter; 7. The fault diagnosis system according to claim 6, wherein it is determined whether or not a fault has occurred in any of the operating means based on the plurality of fault candidate operating means identified for each operating means.

8. 8. The fault diagnosis system according to claim 7, wherein the determination unit identifies an operating means as a faulty operating means when the plurality of fault candidate operating means identified for each operating means are all the same operating means.

9. The fault diagnosis system according to claim 7 , wherein the determining unit adjusts the threshold value depending on a fault mode to be diagnosed.

10. A plurality of operating means for controlling the attitude of the aircraft; A fault diagnosis system according to any one of claims 1 to 9; A mobile body comprising:

11. 11. The moving body according to claim 10, wherein the moving speed is reduced when the failure diagnosis system determines that a failure has occurred in the operating means.

12. 11. The mobile body according to claim 10, further comprising a navigation mode for fault detection that forcibly changes at least one of the heading and the depth.

13. 1. A fault diagnosis method applied to a moving body in which control variables relating to the attitude of the body are controlled by a plurality of operation means, the number of operation means being greater than the number of the control variables, The computer an estimation process for estimating, for each of the operation means, the operation amounts of the other operation means based on a command value of the operation means and a state amount of the moving body; a determination process for determining whether or not a malfunction has occurred in the plurality of operation means using the plurality of estimated operation amounts; The determination process includes: a first failure determination process for determining whether or not a failure has occurred in any of the operation means by using the plurality of operation amounts corresponding to each of the operation means; a second fault determination process for, when it is determined that a fault has occurred in any of the operating means, determining whether a fault has occurred in any of the operating means other than the faulty operating means by using, as target operating amounts, a plurality of the operating amounts excluding an operating amount estimated based on a command value of the faulty operating means determined to be faulty.

14. A program for causing a computer to function as the fault diagnosis system according to any one of claims 1 to 9.

Images