Novel four-axis redundant inertial measurement unit structure and fault detection and isolation method thereof
By installing a three-axis accelerometer on the periphery of the four-axis RIMU system and using the Kalman filtering algorithm, efficient detection and isolation of gyroscope faults in the four-axis RIMU system is achieved, solving the problem that traditional methods cannot isolate the faulty gyroscope, improving the The system's information utilization and reliability are achieved.
Patent Information
- Application Number
- CN202411211288.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-30
- Publication Date
- 2024-11-29
AI Technical Summary
Traditional fault detection and isolation methods cannot effectively isolate the fault of the gyroscope in the four-axis redundant inertial measurement system, limiting the application of this system.
A three-axis accelerometer is installed on the periphery of the four-axis obliquely positioned optimal redundant configuration. The extension system is used to obtain the carrier's reference angular velocity information, and combined with the generalized likelihood ratio fault detection algorithm improved by Kalman filtering to detect gyroscope failure. Direct comparison method to locate faulty gyroscope.
In the case of small size, weight and cost, the information utilization rate was improved, and the faulty gyroscope in the four-axis RIMU system was successfully isolated, solving the problem that traditional methods could not locate the faulty gyroscope.
Smart Images

Figure CN119043374A_ABST
Abstract
Description
Technical Field
[0001] The present invention designs a novel redundant inertial measurement unit structure and designs a fault detection algorithm for the proposed structure. Background Art
[0002] Inertial navigation is an important part of modern navigation technology. It has good autonomy, does not require external signals or send any signals, and has the ability to work around the clock. In some important tasks, the reliability of the inertial measurement unit (IMU) is related to the success or failure of the mission, among which the reliability of the gyroscope with a complex structure that can provide the carrier angular velocity is the most important. The current solution is to design the IMU redundantly, including a system-level redundant solution with multiple sets of IMUs and a device-level redundant solution with multiple inertial measurement sensors configured on a single IMU. Considering that the volume and weight of high-precision IMUs are large, the volume and weight of the system-level redundant solution will increase exponentially, which is not suitable for some scenarios with limited volume and load, such as spacecraft and submarines. Therefore, a single-machine multi-meter redundant inertial measurement unit (RIMU) has a better application prospect. Among them, the four-axis RIMU system composed of a single machine with eight meters (four gyroscopes and four accelerometers) has the lowest volume cost and the highest cost performance. However, due to insufficient information, traditional fault detection and isolation methods cannot isolate the faulty gyroscope in the four-axis system, thus limiting the application of the four-axis system. Therefore, it is necessary to invent a method that can effectively isolate a faulty gyroscope in a four-axis RIMU system. Summary of the invention
[0003] The object of the present invention is achieved in that:
[0004] Step 1: Add a three-axis accelerometer to the periphery of the four-axis tilted optimal redundant configuration to form an extended system;
[0005] Step 2: Using all accelerometer sensors in the extended system to obtain reference angular velocity information of the carrier;
[0006] Step 3: Use the four-axis RIMU to obtain high-precision inertial information of the measurement carrier, and use the generalized likelihood ratio fault detection algorithm improved by Kalman filtering to detect whether the gyroscope is faulty;
[0007] Step 4: Use the acquired angular velocity reference information to convert the theoretical output information of the four gyroscopes, and use the direct comparison method to locate the faulty gyroscope;
[0008] Step 5: Simulate and verify the proposed method.
[0009] Beneficial effects:
[0010] (1) By introducing an additional triaxial accelerometer at a non-center of mass, the mutual conversion of two heterogeneous information types, specific force and angular velocity, within the system is realized, thereby improving the information utilization rate without significantly increasing the volume, weight, and cost.
[0011] (2) It solves the problem that traditional fault detection and isolation methods cannot locate faulty gyroscopes in a four-axis redundant inertial measurement system. BRIEF DESCRIPTION OF THE DRAWINGS
[0012] Figure 1 To expand the four-axis RIMU in the system;
[0013] Figure 2 Schematic diagram of the installation of the triaxial accelerometer;
[0014] Figure 3 It is the optimal configuration for the four-axis RIMU;
[0015] Figure 4 It is a four-axis change diagram;
[0016] Figure 5 This is the schematic diagram of the system after the change;
[0017] Figure 6 System workflow diagram. DETAILED DESCRIPTION
[0018] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0019] In the description of the present invention, it should be understood that the terms "upper", "middle", "outer", "inner" and the like indicating directions or positional relationships are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the components or elements referred to must have a specific direction, be constructed and operated in a specific direction, and therefore should not be understood as limiting the present invention.
[0020] Step 1: Add a three-axis accelerometer to the periphery of the four-axis tilted optimal redundant configuration to form an extended system;
[0021] First, install the four-axis RIMU and the three-axis accelerometer on the same rigid base. The installation diagram is shown in the figure. Figure 1 The four-axis RIMU is installed at the center of mass of the carrier, and its coordinate system is O b X b Y b Z bThe additional triaxial accelerometer is installed diagonally, and its coordinate system is O p X p Y p Z p .
[0022] The internal structure of the four-axis RIMU adopts a four-axis oblique symmetrical configuration. Figure 2 As shown, the included angles between the sensitive axis h and the three coordinate axes are all β=54.73°.
[0023] At this time, the installation matrix of RIMU is:
[0024]
[0025] The arm between the additional accelerometer and the RIMU is denoted by L, which is a constant value, and is set to L = [1 1 1] T .
[0026] Step 2: Using all accelerometer sensors in the extended system to obtain reference angular velocity information of the carrier;
[0027] When the carrier moves, let the vector from the center of the earth to the center of mass of the carrier be R, and the vector from a point P on the carrier that is not the center of mass to the center of mass be L. Then the vector R' from the center of the earth to the point P that is not the center of mass is:
[0028] R'=R+L (2)
[0029] Then the specific force equation at the non-center of mass is:
[0030]
[0031] Among them: A b is the specific force at the center of mass of the carrier, is the angular velocity of the carrier system relative to the navigation system, is the Earth's rotation angular velocity, is the angular velocity of the navigation system relative to the Earth system, L is the lever arm, is the accelerometer bias.
[0032] It can be seen that the specific force information from the non-center of mass contains the angular velocity information of the carrier, and the angular velocity can be obtained by using a reasonable method.
[0033] According to the structure of the six-accelerometer gyro-free strapdown inertial navigation system, the output information of the three-axis accelerometer needs to be projected onto the coordinate axis respectively. The calculation formula is as follows:
[0034]
[0035] Where L x =[0;1;0],L y =[0;0;1],L z=[1;0;0], ΔL x =LL x ,ΔL y =LL y ,ΔL z =LL z At this point the system becomes Figure 3 As shown, it is equivalent to using three single-axis accelerometers a, b, c installed on the coordinate axis. The specific force information d, e, f is the carrier center of mass specific force information calculated by the accelerometer measurement information in the RIMU, so a, b, c, d, e, f form a six-axis GFSINS system.
[0036] Incorporating the installation position and sensitive direction of the accelerometer into equation (3), the specific force equation can be transformed into:
[0037]
[0038] Where: P is the lever-arm effect matrix, determined by the expansion of the specific force equation; L b is the installation position of the accelerometer; θ is the sensitive direction of the accelerometer; is the accelerometer bias.
[0039] Taking accelerometer a as an example, the specific force at a is:
[0040]
[0041] The specific force at the center of mass of the carrier is The installation position of accelerometer a is The sensitive direction is θ a =
[001] T , then the output of accelerometer a is:
[0042]
[0043] Where P 12 is the element in the first row and second column of matrix P.
[0044] Similarly, the outputs of all six accelerometers can be obtained as follows:
[0045]
[0046] Solving the equation group (10), we can solve the differential equation of the carrier angular velocity:
[0047]
[0048] The fourth-order Runge-Kutta method is used to calculate the reference angular velocity of the current cycle, which is represented by X, Y, and Z respectively. Simplify equation (11) into the following system of equations:
[0049]
[0050] Then the fourth-order Runge-Kutta method can be expressed as follows:
[0051]
[0052] in:
[0053]
[0054] In summary, the reference angular velocity information of the carrier can be obtained by using the specific force information of the non-center of mass.
[0055] Step 3: Use the four-axis RIMU to obtain high-precision inertial information of the measurement carrier, and use the generalized likelihood ratio fault detection algorithm improved by Kalman filtering to detect whether the gyroscope is faulty;
[0056] First, define the measurement model of the redundant system. The measurement equation of the redundant system without fault is:
[0057] Z=(I+H se )[(H+H me )X+b1+ε1] (14)
[0058] Where: Z is the n-dimensional gyroscope measurement vector, H se is the n-order scale coefficient error matrix, H me is the n×3 dimensional installation error matrix, X is the 3D carrier angular velocity, b1 is the constant bias, and ε1 is Gaussian white noise.
[0059] Expanding formula (14) yields:
[0060] Z=(H+H me +H se H+H se H me )X+(I+H se )(b1+ε1) (15)
[0061] Considering that the scale factor error and installation error are both small in magnitude, it can be considered that H se H me ≈0, then:
[0062] Z=(H+H me +H se H)X+(I+H se )(b1+ε1) (16)
[0063] make
[0064]
[0065] Formula (16) can be written as:
[0066] Z=(H+H m )X+b+ε (18)
[0067] The parity vector in the absence of faults is:
[0068] P=VZ=VH m X+Vb+Vε (19)
[0069] P=AX+b'+Vε (20)
[0070] Where: A = VH m , b'=Vb.
[0071] The quantities to be estimated are the elements of the (n-3)×3-dimensional matrix A and the (n-3) elements of the b' vector. Expand the matrix A into a column vector as follows:
[0072] a=[a 11 a 12 a 13 a 21 ... a l1 ] T (twenty one)
[0073] where a ij is the element in the i-th row and j-th column of matrix A, l=n-3, and the first term of formula (20) can be written as:
[0074] AX=Ca (22)
[0075] Where C is a l×3l matrix, defined as:
[0076]
[0077] Then the parity vector can be expressed as:
[0078] P=Ca+b'+Vε (24)
[0079] The above equation is used as the measurement equation, where the matrix C is related to the true state X. Since the true value cannot be obtained, the estimated value of the state is used here To express:
[0080]
[0081] Substituting the expression for C yields an estimate of C At this time, the measurement equation is expressed as:
[0082]
[0083] The estimated quantities a and b' are taken as state quantities and synthesized into an augmented vector e, then the measurement equation is:
[0084]
[0085] in: ε'=Vε.
[0086] In order to estimate the state quantity e, a state equation is also required. Considering that the installation error and zero bias are constants, assuming that they satisfy the discrete Markov process, the state equation is:
[0087] e k =e k-1 +w k-1 (28)
[0088] Where: w k is white noise with variance Q k .
[0089] According to the measurement equation of formula (27) and the state equation of formula (28), the Kalman filter algorithm can be used to estimate e k , and then compensate the parity vector.
[0090]
[0091] It can be seen that the parity vector P * After compensation, only the measurement noise is included and it is independent of the state X. At this time, the fault detection function and fault isolation function become:
[0092]
[0093] Step 4: Use the acquired angular velocity reference information to convert the theoretical output information of the four gyroscopes, and use the direct comparison method to locate the faulty gyroscope.
[0094] After the fault is detected in step 3, the reference acceleration obtained in step 2 is used to isolate the faulty instrument. The work flow diagram is as follows: Figure 4 shown.
[0095] Use the installation matrix to get reference information for each gyroscope:
[0096]
[0097] In the formula is the reference output of the corresponding gyroscope.
[0098] Subtract the reference information from the actual output to obtain the residual
[0099]
[0100] The gyroscope with the largest residual value is considered to be the faulty gyroscope. The fault can be isolated by setting all the corresponding row elements in the installation matrix H to zero, and the remaining three non-coplanar gyroscopes continue to work.
[0101] Step 5: Simulate and verify the proposed method.
[0102] First, the motion trajectory and the ideal inertia information of the carrier are generated.
[0103] The initial parameter settings of the extended system sensor are as follows:
[0104] Initial simulation parameters
[0105]
[0106] Sensor sampling period ts = 0.01s, fault detection threshold T D =13.28.
[0107] The ideal output of the sensor is obtained using the installation matrix H. Zero bias and noise are added to simulate the actual usage scenario. At the same time, a fault value of 10° / h is randomly added to the gyroscope. The gyroscope output is as follows.
[0108] In the description of this specification, the description with reference to the terms "one embodiment", "example", "specific example", etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.
[0109] The preferred embodiments of the present invention disclosed above are only used to help illustrate the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to the specific implementation methods described. Obviously, many modifications and changes can be made according to the content of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present invention, so that those skilled in the art can understand and use the present invention well. The present invention is limited only by the claims and their full scope and equivalents.
Claims
1. A novel four-axis redundant inertial group structure and its fault detection and isolation method, characterized by: The steps include: Step 1: Add a three-axis accelerometer to the periphery of the four-axis tilted optimal redundant configuration to form an extended system; Step 2: Using all accelerometer sensors in the extended system to obtain reference angular velocity information of the carrier; Step 3: Use the four-axis RIMU to obtain high-precision inertial information of the measurement carrier, and use the generalized likelihood ratio fault detection algorithm improved by Kalman filtering to detect whether the gyroscope is faulty; Step 4: Use the acquired angular velocity reference information to convert the theoretical output information of the four gyroscopes, and use the direct comparison method to locate the faulty gyroscope; Step 5: Simulate and verify the proposed method.
