Multi-source information anomaly detection and processing method and device for MEMS inertial integrated navigation system

Abstract

This invention discloses a method and apparatus for detecting and processing multi-source information anomalies in a MEMS inertial navigation system. The method first establishes a Cartesian square value calculation function for the inertial navigation system; secondly, it determines filter parameters and the range of observed jumps, and simulates and calculates the Cartesian square threshold value T during the jump. D The external reference information is detected again based on the k-square threshold; finally, multi-source information anomaly processing is performed. This invention can monitor multi-source fusion information of MEMS inertial navigation systems in real time and process abnormal information.

Description

TECHNICAL FIELD

[0001] The application belongs to the field of MEMS inertial navigation, and relates to a multi-source information anomaly detection and processing method and device for a micro inertial integrated navigation system. BACKGROUND

[0002] MEMS inertial devices have the characteristics of small volume, light weight and low power consumption, and are a development direction of future inertial devices, while the precision of MEMS inertial devices is relatively poor, and needs to be fused with external reference information to obtain higher navigation precision.

[0003] Multi-source information fusion of inertial navigation is mostly performed by using a kalman filter to fuse external information and inertial navigation results to obtain optimal results, and the fusion results not only retain the advantages of high-frequency output, strong autonomy and high reliability of inertial navigation, but also retain the advantage of non-divergence of the precision of external information.

[0004] In actual use, the external reference signal will appear distorted, that is, there is a deviation between the actual observation value and the theoretically inferred value, and fusion with severely distorted reference data will affect the navigation precision, and therefore a multi-source information detection and processing method for a MEMS inertial integrated navigation system needs to be proposed. SUMMARY

[0005] The application aims to provide a multi-source information anomaly detection and processing method and device for a MEMS inertial integrated navigation system, and timely detect and process abnormal information.

[0006] To achieve the object of the application, the multi-source information anomaly detection and processing method for a MEMS inertial integrated navigation system provided by the application adopts the technical scheme comprising the following steps:

[0007] Step 1. For an inertial navigation system, a chi-square value calculation function is established;

[0008] Step 2. Filter parameters and observation jump ranges are determined, and a chi-square threshold value T D is simulated and calculated;

[0009] Step 3. Whether the external reference is abnormal is detected based on the chi-square threshold value T D ;

[0010] When the chi-square value λ κ >T D , it is detected that the multi-source information is abnormal;

[0011] When the chi-square value λ κ ≤T D , it is detected that the multi-source information is fault-free.

[0012] Step 4. Multi-source information anomaly processing

[0013] When the single chi-square value exceeds the chi-square threshold T D , the filter only performs time update and does not perform measurement update; the threshold number N is set according to the system characteristics, that is, when the chi-square value exceeds the chi-square threshold T D for N times in succession, it is considered that the external information reference is faulty, and the filter is restarted.

[0014] Further, the setting method of the chi-square threshold T D is as follows: the observation noise R k and the observation Z k of the combined navigation filter parameter are determined, the maximum value of the allowable jump is determined, the position and speed jump limit are determined, and the chi-square threshold T D is determined through simulation calculation. The position includes latitude, altitude and longitude, and the observation includes latitude, altitude, longitude, north speed, celestial speed and east speed.

[0015] Further, the chi-square value is determined according to the inertial navigation accuracy, the characteristics of the external information source and the use requirement, and the method is as follows: when the inertial navigation system accuracy is low or the external information source is more trusted, the chi-square value threshold is appropriately relaxed; when the inertial navigation system accuracy is high or the inertial navigation system is more trusted, the chi-square value threshold is appropriately tightened.

[0016] According to another aspect of the present application, a MEMS inertial combined navigation system multi-source information anomaly detection and processing device is provided, which comprises a chi-square value calculation module, a chi-square threshold setting module, a fault detection module and a fault processing module.

[0017] The chi-square value calculation module establishes a chi-square value calculation function for the inertial navigation system and carries out chi-square value calculation.

[0018] The chi-square threshold setting module determines the filter parameter and the observation jump range, and simulates and calculates the chi-square threshold T D when jumping;

[0019] The fault detection module detects whether the multi-source information is abnormal based on the chi-square threshold T D when jumping; when the chi-square value λ κ >T D , it is detected that the source information is abnormal; when the chi-square value λ κ ≤T D , it is detected that the multi-source information is not faulty;

[0020] The fault processing module is used for multi-source information anomaly processing, and when the single chi-square value exceeds the chi-square threshold T DWhen the filter is only updated in time, no measurement update is performed; according to the system characteristics, the threshold number N is set, i.e. when the square value exceeds the square threshold T for N times in succession, it is considered that the external information reference is faulty, and the filter is restarted. D

[0021] Further, the square threshold value T in the square threshold setting module is set according to the allowable jump maximum value of the position and velocity, and the position and velocity jump limit is determined through simulation calculation. D The setting method is to determine the combined navigation filter parameter observation noise R k and the observation Z k The square threshold value T is set according to the allowable jump maximum value of the position and velocity, and the position and velocity jump limit is determined through simulation calculation. D The position includes latitude, altitude and longitude, and the observation includes latitude, altitude, longitude, north speed, celestial speed and east speed.

[0022] Further, the square value is determined according to the inertial navigation accuracy, the characteristics of the external information source and the use requirement, and the method is as follows: when the inertial navigation system accuracy is low, or the external information source is more trusted, the square value threshold is appropriately relaxed; when the inertial navigation system accuracy is high, or the inertial navigation system is more trusted, the square value threshold is appropriately tightened.

[0023] Compared with the prior art, the beneficial effects of the present application are as follows:

[0024] The MEMS inertial combined navigation system multi-source information anomaly detection and processing method and device provided by the present application can monitor the external reference information of the micro inertial combined navigation system in real time and eliminate abnormal information, thereby improving the navigation accuracy. BRIEF DESCRIPTION OF DRAWINGS

[0025] The accompanying drawings included to provide a further understanding of the embodiments of the present application, constitute a part of the specification and serve to explain the principles of the present application together with the text. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0026] Figure 1 A flowchart of the MEMS inertial combined navigation system multi-source information anomaly detection and processing method provided by the specific embodiment of the present application is shown;

[0027] Figure 2 A flowchart of the square threshold value setting provided by the specific embodiment of the present application is shown;

[0028] Figure 3 A schematic diagram of the principle of the MEMS inertial combined navigation system multi-source information anomaly detection and processing device provided by the specific embodiment of the present application is shown. ​DETAILED DESCRIPTION

[0029] It should be noted that the embodiments in the present application and the features in the embodiments can be combined with each other in the case of no conflict. The technical solutions in the embodiments of the present application will be described clearly and completely in combination with the drawings of the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. The description of the at least one exemplary embodiment is actually only illustrative, but not as any limitation on the present application and its application or use. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the protection scope of the present application.

[0030] The present application adopts the chi-square test as the method of multi-source information anomaly detection, and selects different processing methods according to different situations. The chi-square test is a statistical analysis tool for comparing the consistency of data and the uncertainty of reference value. Specifically, as shown in the figure, Figure 1 The method provided by the present application comprises the following steps:

[0031] Step 1. For the inertial navigation system, a chi-square value calculation function is established

[0032] The chi-square value calculation function is:

[0033]

[0034] Wherein:

[0035] λ k obeys chi-square distribution; 2

[0036] r k is the residual of Kalman filter, which is zero-mean white noise, and the calculation formula of the residual is as follows:

[0037]

[0038] Wherein, Z k is the observation, H k is the measurement matrix, is the one-step prediction estimated state vector;

[0039] α is the variance, and the calculation formula is as follows:

[0040]

[0041] Wherein, R k is the observation noise; P k,k-1 is the one-step prediction covariance matrix.

[0042] ​Step 2. Determine filter parameters and observation jump index, and simulate and calculate the threshold value T of the chi-square value D .

[0043] From the formula of the chi-square value calculation function, the chi-square value is affected by the observation Z k , one-step prediction state vector value, one-step prediction P matrix, and observation noise R k . In the normal filtering process, the one-step prediction is basically stable, so the chi-square value is greatly affected by the observation Z k , observation noise R k . Therefore, the setting of the chi-square threshold value T D mainly includes determining the combined navigation filter parameters (mainly the observation noise R k ) and the jump index of the observation Z k , and further simulating and verifying to determine the chi-square threshold value, as shown in Figure 2 .

[0044] Step 3. Detect external information reference abnormality based on the chi-square threshold value T D

[0045] When the chi-square value λ κ > T D , it is detected that the multi-source information is abnormal.

[0046] When the chi-square value λ κ ≤ T D , it is detected that the multi-source information is fault-free.

[0047] Step 4. Multi-source information abnormality processing

[0048] When the single chi-square value exceeds the chi-square threshold value T D , the filter only performs time update and does not perform measurement update; according to the system characteristics, the threshold number N is set, that is, when the chi-square value exceeds the chi-square threshold value T D for N times, it is considered that the external information reference is faulty, and the filter is restarted.

[0049] In an embodiment of the present application, the chi-square threshold value T D in step 2 is obtained as follows:

[0050] Step 2.1 Determine filter parameters

[0051] According to the performance index, the measured results of the external information source of the system, and the combined filtering requirements, the parameters (mainly the observation noise R k ) of the combined navigation filter are determined.

[0052] Step 2.2 Determine the observation jump range

[0053] ​According to the performance index of the external information source, the measured result, the inertial navigation system precision and the combined navigation precision requirement, the maximum value of the jump of each observation quantity is determined, and when the jump of the external information source is greater than the value, the Kalman filter detection fails. The observation quantity includes latitude, altitude, longitude, north speed, sky speed and east speed.

[0054] When the general inertial navigation precision is high, the influence of the failure of the external information source for a few times / short time is small, the requirement for the data jump can be increased, otherwise the requirement for the data jump can be appropriately relaxed.

[0055] Step 2.3: Simulation to determine the Kalman filter threshold value

[0056] By using the actual vehicle-mounted, airborne or theoretical trajectory data, the position (latitude, altitude, longitude) jump limit and the speed (north speed, sky speed and east speed) jump limit are superimposed, the Kalman filter value when jumping is calculated through simulation, and the result is listed as shown in Table 2.

[0057] Table 2: Kalman filter result record table when jumping

[0058]

[0059] In some embodiments of the application, the Kalman filter value is further determined according to the inertial navigation precision, the use requirement and the like.

[0060] According to the simulation result, the inertial precision of the system, the characteristics of the external information source, the use requirement and the like, whether the Kalman filter threshold value needs to be relaxed or increased is determined. When the inertial navigation system precision is low, or the external information source is more trusted, the Kalman filter value threshold can be appropriately relaxed; when the inertial navigation system precision is high, or the inertial navigation system is more trusted, the Kalman filter value threshold can be appropriately increased.

[0061] Taking a combined navigation system as an example, the setting of the Kalman filter threshold value of the application is further described.

[0062] (1) Determine the filter parameter

[0063] The setting of the R matrix of the combined navigation system is as follows:

[0064] Noise matrix Wherein, σ x , σ h , σ v and respectively represent the position error and the speed error of the navigation system, and are set as:

[0065] σ x = (3m / R e ) rad

[0066] σ h = 3m

[0067] σ vn= 0.06 m / s

[0068] σ vu = 0.15 m / s

[0069] σ ve = 0.06 m / s

[0070] (2) Determine the observation jump range

[0071] The external information source (satellite receiver) of a certain integrated navigation system measured results / indicators are as follows:

[0072] Navigation positioning accuracy: horizontal ≤ 30 m (3σ), elevation ≤ 45 m (3σ);

[0073] Navigation speed accuracy: horizontal ≤ 0.6 m / s (3σ), vertical ≤ 0.9 m / s (3σ).

[0074] (3) Simulate the determination of the chi-square threshold value

[0075] Simulate using actual racing cars, airborne data or theoretical trajectories. After initializing the filter, superimpose the jump simulation on the original data of the satellite receiver to calculate the chi-square value. The effective filtering number is greater than 6, and the chi-square test is used to test the effectiveness of the filter. The results of the chi-square value when the observation jumps are shown in Table 3.

[0076] Table 3 Chi-square result record table when jumping

[0077] Serial number Latitude jump Altitude jump Longitude jump North speed jump Sky speed jump East speed jump Kappa value 1 30 101 2 45 219 3 30 169 4 0.6 93 5 0.9 38 6 0.6 90 7 30 45 30 0.6 0.9 0.6 703

[0078] According to the previous calculation formula, when the position and speed jump simultaneously in multiple directions, the corresponding chi-square value of the jump can be calculated by scaling and superimposing in proportion.

[0079] (3) Determine the chi-square value according to the inertial navigation accuracy, usage requirements, etc.

[0080] In this example, the gyro accuracy of the integrated navigation system is about 1° / h-10° / h (1σ), the pure inertial navigation accuracy is poor, the satellite receiver usage environment is less blocked, and the jump is generally small, so the chi-square threshold value is relaxed, and the chi-square value of 700 is selected after comprehensive consideration.

[0081] (4) Determine the abnormal processing method according to the inertial navigation accuracy, usage requirements, etc.

[0082] When the single chi-square value exceeds the chi-square threshold value of 700, the filter only performs time update and does not perform measurement update; according to the system characteristics, set the threshold number N, in this example, N is 3, that is, when the chi-square value exceeds the chi-square threshold value of 700 for 3 consecutive times, it is considered that the external information reference is faulty, and the filter is restarted.

[0083] Based on the same concept, according to another aspect of the present application, a MEMS inertial integrated navigation system multi-source information anomaly detection and processing device is provided.

[0084] The MEMS inertial integrated navigation system multi-source information anomaly detection and processing device provided by the present application is described below, and the MEMS inertial integrated navigation system multi-source information anomaly detection and processing device described below can be referred to each other corresponding to the MEMS inertial integrated navigation system multi-source information anomaly detection and processing method described above.

[0085] In an exemplary embodiment of the present application, the MEMS inertial integrated navigation system multi-source information anomaly detection and processing device includes a skew value calculation module, a skew threshold setting module, a fault detection module and a fault processing module, as shown in Figure 3 .

[0086] The skew value calculation module establishes a skew value calculation function for the inertial navigation system and carries out skew value calculation.

[0087] The skew threshold setting module determines the filter parameters and the observation jump range, and simulates and calculates the skew threshold value T D .

[0088] The fault detection module is used to detect whether the multi-source information is abnormal. When the skew value λ κ > T D , it is determined that the multi-source information is abnormal; when the skew value λ κ ≤ T D , it is determined that the multi-source information is fault-free.

[0089] The fault processing module is used to process the multi-source information abnormality. When the skew value exceeds the skew threshold value T D each time, the filter only performs time update and does not perform measurement update; according to the system characteristics, the threshold number N is set, that is, when the skew value exceeds the skew threshold value T D each time for N times in succession, it is considered that the external information reference is faulty, and the filter is restarted.

[0090] Further,

[0091] The skew value calculation function is:

[0092]

[0093] Wherein:

[0094] λ k obeys χ 2 distribution.

[0095] r kis the residual of the Kalman filter, is zero-mean white noise, and the calculation formula of the residual is as follows:

[0096]

[0097] In the formula, Z k is an observation, H k is a measurement matrix, is a one-step prediction estimated state vector;

[0098] is a variance, and the calculation formula is as follows:

[0099]

[0100] In the formula, R k is observation noise; P k,k-1 is a one-step prediction covariance matrix.

[0101] The Kalman threshold value T D in the Kalman threshold setting module is set as follows: the combined navigation filter parameter observation noise R k and the observation Z k are determined, and the maximum value of the allowable jump is determined, and the position and speed jump limit allowed are superimposed, and the Kalman threshold value T D is determined through simulation calculation. The position includes latitude, altitude, and longitude, and the observation includes latitude, altitude, longitude, north speed, sky speed, and east speed.

[0102] The Kalman value is further determined according to the inertial navigation accuracy, the characteristics of the external information source, and the use requirement, and the method is as follows: when the inertial navigation system accuracy is low, or the external information source is more trusted, the Kalman value threshold is appropriately relaxed; when the inertial navigation system accuracy is high, or the inertial navigation system is more trusted, the Kalman value threshold is appropriately tightened.

[0103] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, and not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: it can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement for part of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for multi-source information anomaly detection and processing of a MEMS inertial integrated navigation system, characterized in that, The method comprises the following steps: Step 1. Establishing a Kalman value calculation function for an inertial navigation system; Step 2. Determine filter parameters and observation jump range, simulate and calculate jump time threshold value ; Step 3. Thresholding based on carbazole detecting whether the external information reference is abnormal; Performing when the value of the karnaugh If the detection is multi-source information anomaly; Step 4. Abnormal processing of multi-source information: When the single chi-square value exceeds the chi-square threshold value The filter only performs time update and does not perform measurement update; the threshold number N is set according to the system characteristics, that is, when the chi-square value exceeds the chi-square threshold value for N times in succession, it is considered that the external information reference fault occurs, and the filter is restarted; the chi-square threshold value in step 2 is obtained as follows: Step 2.1 Determining filter parameters According to the performance index, measured results and combination filtering requirements of the external information source of the system, the parameters of the combination navigation filter are determined; Step 2.2 Determining the jump range of the observation According to the performance index, measured results, inertial navigation system accuracy and combination navigation accuracy requirements of the external information source, the maximum value of the jump of each observation that can be tolerated is determined, and when the jump of the external information source is greater than the value, the Kalman detection will not pass, and the observation includes latitude, altitude, longitude, north speed, sky speed and east speed; Step 2.3 Simulating to determine the Kalman threshold value Actual vehicle, airborne or theoretical trajectory data are superimposed with the allowable position and speed jump limit, and the Kalman value when jumping is calculated through simulation. 2.The MEMS integrated navigation system multi-source information anomaly detection and processing method according to claim 1, characterized in that, The Kalman value calculation function is: wherein: subject to distribution; is the residual of the Kalman filter, which is zero-mean white noise, and the calculation formula of the residual is as follows: ， wherein is the observation, is the measurement matrix, is the one-step prediction estimate of the state vector; is the variance, calculated as follows: wherein is the observation noise; is the one-step prediction covariance matrix.

3. The method of claim 1, wherein the method further comprises: The threshold value of the Kalman filter The setting method is: determining the combined navigation filter parameter observation noise And the allowable maximum jump of the observation Superimposed allowable position and speed jump limit, and the threshold value of the Kalman filter is determined by simulation calculation .

4. The method of claim 3, wherein the method further comprises: The position includes latitude, altitude and longitude, and the observation includes latitude, altitude, longitude, north speed, sky speed and east speed.

5. The method of claim 3, wherein the method further comprises: According to the inertial navigation accuracy, external information source characteristics and use requirements, the Kalman value is further determined, and the method is as follows: when the inertial navigation system accuracy is low, or the external information source is more trusted, the Kalman value threshold is appropriately relaxed; when the inertial navigation system accuracy is high, or the inertial navigation is more trusted, the Kalman value threshold is appropriately tightened.

6. A multi-source information anomaly detection and processing device for a MEMS inertial integrated navigation system, characterized in that, It comprises a Kalman value calculation module, a Kalman threshold setting module, a fault detection module and a fault processing module, The Kalman value calculation module establishes a Kalman value calculation function for the inertial navigation system and carries out Kalman value calculation, The threshold setting module determines filter parameters and an observation jump range, and simulates and calculates the threshold value at the time of jumping The fault detection module detects whether the multi-source information is abnormal based on the threshold value at the time of jumping When the threshold value is less than the threshold value , it is detected that the source information is abnormal; and when the threshold value is greater than the threshold value , it is detected that the multi-source information is fault-free. The fault processing module is used for multi-source information abnormal situation processing, when the single chi-square value exceeds the chi-square threshold value , the filter only performs time update, and does not perform measurement update; the threshold number N is set according to the system characteristics, that is, when the chi-square value exceeds the chi-square threshold value for N times in succession, it is considered that the external information reference fault occurs, and the filter is restarted; the chi-square threshold value The acquisition method is as follows: Step 2.1 Determining filter parameters According to the performance index, measured results and combination filtering requirements of the external information source of the system, the parameters of the combination navigation filter are determined; Step 2.2 Determining the jump range of the observation According to the performance index, measured results, inertial navigation system accuracy and combination navigation accuracy requirements of the external information source, the maximum value of the jump of each observation that can be tolerated is determined, and when the jump of the external information source is greater than the value, the Kalman detection will not pass, and the observation includes latitude, altitude, longitude, north speed, sky speed and east speed; Step 2.3 Simulating to determine the Kalman threshold value Actual vehicle, airborne or theoretical trajectory data are superimposed with the allowable position and speed jump limit, and the Kalman value when jumping is calculated through simulation.

7. The MEMS inertial integrated navigation system multi-source information anomaly detection and processing apparatus according to claim 6, characterized in that, The Kalman value calculation function is: wherein: subject to distributed; is the residual of the Kalman filter, which is zero-mean white noise, and the calculation formula of the residual is as follows: wherein is the observation, is the measurement matrix, is the one-step prediction estimate of the state vector; For the variance, the formula is as follows: where is the observation noise; is the one-step prediction covariance matrix.

8. The MEMS inertial integrated navigation system multi-source information anomaly detection and processing apparatus according to claim 7, characterized in that, The threshold value of the threshold setting module The setting method is to determine the combined navigation filter parameter observation noise And the observation The maximum allowable jump value, the superimposed allowable position and speed jump limit, and the threshold value are determined by simulation calculation .

9. The MEMS Inertial Integrated Navigation System Multi-Source Information Anomaly Detection and Processing Device according to claim 8, characterized in that, The position includes latitude, altitude and longitude, and the observation includes latitude, altitude, longitude, north speed, sky speed and east speed.

10. The MEMS inertial integrated navigation system multi-source information anomaly detection and processing apparatus according to claim 8, characterized in that, According to the inertial navigation accuracy, external information source characteristics and use requirements, the Kalman value is further determined, and the method is as follows: when the inertial navigation system accuracy is low, or the external information source is more trusted, the Kalman value threshold is appropriately relaxed; when the inertial navigation system accuracy is high, or the inertial navigation is more trusted, the Kalman value threshold is appropriately tightened.

Images