Aerospace measurement and control external measurement data real-time adaptive filtering system based on prior information

Abstract

The invention relates to a spaceflight measurement and control external measurement data real-time adaptive filtering system based on prior information, and belongs to the technical field of spaceflight measurement and control data processing. Comprising a filtering initial value calculation module, a maneuvering detection module, an adaptive filtering module and a parameter configuration module, and the filtering initial value calculation module is used for accurately calculating a filtering initial value; the maneuvering detection module is used for detecting target maneuvering; and the adaptive filtering module modifies the filter parameter configuration module according to the target maneuver detected by the maneuver detection module so as to realize accurate tracking of the target. The filtering initial value calculation module comprises a priori information calculation filtering initial value or an external measurement actual measurement track data multi-point smooth value calculation filtering initial value. The prior information includes telemetry trajectory data and theoretical trajectory data. According to the method, the real-time filtering effect of the external measurement data of the sea-based mobile aerospace observation station is improved, and the oscillation amplitude of the speed curve of the external measurement trajectory data filtering result during low-elevation measurement is reduced.

Description

Technical Field

[0001] The invention relates to a real-time adaptive filtering system for aerospace measurement and control external measurement data based on prior information, belonging to the technical field of aerospace measurement and control data processing. Background Art

[0002] Real-time processing of external measurement data is an important part of aerospace measurement and control, and is one of the main responsibilities of sea-based mobile aerospace stations. When a sea-based mobile aerospace station tracks an aircraft, the original measurement of the external measurement data is the azimuth, pitch, and slant range of the aircraft relative to the sea-based mobile platform. Combined with the real-time position data and real-time attitude data of the sea-based mobile platform, a series of coordinate transformations are performed to obtain the position of the aircraft in the geocentric rectangular coordinate system. The position is filtered and smoothed to obtain the position, velocity, and other trajectory information of the aircraft in the geocentric rectangular coordinate system. The most commonly used data filtering and smoothing method in the real-time processing of external measurement data of sea-based mobile platforms is the optimal linear filter of polynomial plus white noise. At present, the commonly used N-point data center smoothing in the external measurement data processing model of sea-based mobile stations is polynomial filtering. When the elevation angle of the measurement data is high, the speed curve of the processed target trajectory data is smooth and the effect is good; but when the elevation angle of the measurement data is low, the speed curve of the processed trajectory data shows obvious fluctuations. Summary of the invention

[0003] The technical problem to be solved by the present invention is to provide a real-time adaptive filtering system for aerospace measurement and control external measurement data based on prior information in response to the above-mentioned existing technology, so as to improve the real-time filtering effect of external measurement data of sea-based mobile aerospace measurement stations and reduce the oscillation amplitude of the velocity curve of the external measurement data filtering result during low elevation angle measurement.

[0004] The technical solution adopted by the present invention to solve the above-mentioned problem is: a real-time adaptive filtering system for aerospace measurement and control external measurement data based on prior information, including a filter initial value calculation module, a maneuver detection module, an adaptive filtering module and a parameter configuration module, the filter initial value calculation module is used to accurately calculate the filter initial value; the maneuver detection module is used to detect the target maneuver; the adaptive filtering module modifies the filter parameter configuration module according to the target maneuver detected by the maneuver detection module to achieve accurate tracking of the target.

[0005] The filter initial value calculation module includes calculating the filter initial value using prior information or calculating the filter initial value using multi-point smoothing values ​​of external measured trajectory data.

[0006] The prior information includes telemetry trajectory data and theoretical trajectory data.

[0007] The maneuver detection module includes a new information offset detection method and a new information distance weighted square detection method. The new information offset detection method: under normal circumstances, the new information mean oscillates around zero; when the target maneuvers, the new information deviates from the zero mean; the new information within a period of time is counted, and when the new information continuously exceeds a given new information threshold within the statistical period, it is determined that the target maneuver occurs, otherwise the target maneuver is eliminated.

[0008] The new information threshold is a positive value or 0.

[0009] The weighted square innovation distance detection method is based on the weighted square deviation of the innovation, and the innovation distance function is:

[0010]

[0011] v k is the new information at time k, v k The transpose of

[0012] S k is the new interest variance, For S k The inverse matrix of

[0013] D k Subject to χ with m degrees of freedom 2 distributed;

[0014] Detection of the occurrence and elimination of maneuvers: P[D k >M]=a

[0015] Where: a is the allowed false alarm probability; M is the threshold, and P is the cumulative probability;

[0016] When D k When >M, maneuvers occur, increasing the noise variance;

[0017] When D k When ≤M, the maneuver is eliminated and the noise variance is reduced.

[0018] The false alarm probability a is 0.03-0.08.

[0019] The adaptive filtering module adjusts the filter output parameters: the filter output data is adjusted to the output data output one second before the current filter input data. When the filter input data is an integer second, the filter output data of the previous integer second is sent for communication. When the target maneuvers, the filter parameters of the input data within the last two seconds can be adjusted and then re-filtered, thereby optimizing the delay characteristics of the adaptive filtering.

[0020] Compared with the prior art, the advantages of the present invention are: a real-time adaptive filtering system for aerospace measurement and control external measurement data based on prior information, which effectively improves the real-time filtering effect of external measurement data of sea-based mobile aerospace measurement stations and reduces the oscillation amplitude of the speed curve of the filtering result of external measurement trajectory data during low elevation angle measurement. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 This is a flowchart of initial filter value calculation in a real-time adaptive filtering system for aerospace measurement and control external data based on prior information according to an embodiment of the present invention;

[0022] Figure 2 This is a flow chart of the new information offset maneuver detection in an embodiment of the present invention;

[0023] Figure 3 This is a flow chart of weighted square maneuver detection in an embodiment of the present invention;

[0024] Figure 4 This is a flow chart of a time delay optimized maneuver detection adaptive filter in an embodiment of the present invention;

[0025] Figure 5 It is a comparison diagram of the speed-height curve of the trajectory data of the adaptive filtering result in the embodiment of the present invention;

[0026] Figure 6 for Figure 5 A magnified image of

[0027] Figure 7 It is a comparison diagram of the trajectory data component curves of the adaptive filtering result in the embodiment of the present invention;

[0028] Figure 8 for Figure 7 Magnified image of . DETAILED DESCRIPTION

[0029] The present invention is further described in detail below with reference to the accompanying drawings.

[0030] In this embodiment, a real-time adaptive filtering system for aerospace measurement and control external data based on prior information includes four parts: a filter initial value calculation module, a maneuver detection module, an adaptive filtering module, and a parameter configuration module. The filter initial value calculation module is used to accurately calculate the filter initial value; the maneuver detection module is used to detect the target maneuver; the adaptive filtering module modifies the filter parameter configuration module according to the target maneuver detected by the maneuver detection module to achieve accurate tracking of the target.

[0031] Filter initial value calculation module: Accurate filter initial value can effectively improve the filter convergence speed and improve the filtering effect. When the sea-based mobile station performs aerospace measurement and control tasks, it generally undertakes both aircraft telemetry and external measurement tasks, and the telemetry stable tracking time is earlier than the external measurement tracking time. Figure 1As shown in the figure, the initial filter value calculation process is as follows: 1. In the case of telemetry data, high-precision telemetry trajectory data (such as aircraft GNSS position and speed data) is used to calculate the initial filter value of the external data. 2. In the case of no telemetry trajectory data, if the previous station tracking shows that the actual flight trajectory of the aircraft is consistent with the theoretical trajectory of the target during aircraft relay tracking, the theoretical trajectory data is used to calculate the initial filter value of the external data. 3. In the absence of telemetry trajectory data and the previous station tracking shows that the actual flight trajectory of the aircraft is significantly different from the theoretical value, the multi-point smoothed value of the external measured trajectory data is used as the initial filter value. When the external measurement equipment captures the tracking target, the target elevation angle is generally low. At this time, the error of the external measurement data is large. If single-point external measurement data is used as the initial filter value, the filter convergence time is long and the filtering result oscillates greatly. Therefore, the multi-point smoothed value of the external measured trajectory data is used as the initial filter value.

[0032] Maneuver detection module: When the target is maneuvering, the original filter model will deteriorate, causing the target state to deviate from the true state and the filter residual characteristics to change. The target maneuver is detected by the innovation offset detection method and the innovation distance square detection method.

[0033] 1) Innovation offset detection method

[0034] The mean of the new information is zero. Under normal circumstances, the new information oscillates around the 0 mean. When the target maneuvers, the new information sequence will deviate from the 0 mean. Statistics are collected for a period of time. When the new information exceeds the given threshold continuously during the statistical period, the target maneuver is determined to have occurred, otherwise the target maneuver is canceled. The new information deviation maneuver detection process is as follows: Figure 2 As shown in the figure, NewsVal is the given new information threshold, which is a positive value or 0; Cnt1 is the count of new information continuously greater than the new information threshold, and Cnt2 is the count of new information continuously less than the negative new information threshold; CntVal is the number of times the new information continuously deviates, which can be 2N (the measurement data is sampled at N points per second for 2 seconds). When the maneuver occurs, the noise variance is increased; when the maneuver is released, the noise variance is reduced.

[0035] 2) Weighted square detection method of innovation distance

[0036] The weighted squared innovation distance detection method is based on the weighted squared deviation of the innovation, and defines the innovation distance function as:

[0037]

[0038] Among them, v k is the new information at time k, v k The transpose of v is the same as v for a 1×1 vector. k Consistent. k is the new interest variance, For S kThe inverse matrix of . From the statistical properties of the new information sequence, we know that D k Subject to χ with m degrees of freedom 2 If the target maneuvers, the new information v k It will no longer be a Gaussian white noise process with a mean of zero. k will increase, so the following method can be used to detect the occurrence and elimination of maneuvers: Take D k The probability of being greater than a certain distance threshold M is a, that is,

[0039] P[D k >M]=a

[0040] Where: a is the allowed false alarm probability; M is the distance threshold, and P is the cumulative probability.

[0041] When D k When >M, maneuvers occur, increasing the noise variance to ensure continuous tracking; generally 1.2 times the noise variance.

[0042] When D k When ≤M, maneuvering is eliminated to reduce the noise variance and ensure tracking accuracy, which is generally 0.8 times the noise variance.

[0043] For χ with m degrees of freedom 2 The variable of the distribution, χ 2 The larger the value is, the smaller the false alarm probability a is, and the higher the credibility is. Usually the false alarm probability is 0.05, which means 95% credibility. a and M are one-to-one corresponding. Once a is set, M is naturally set.

[0044] The innovation distance weighted square maneuver detection process is as follows Figure 3 As shown in the figure, M1 and M2 are the detection distance thresholds. For filtering each component of the geocentric trajectory data, M1 can take a value of 3.84 and M2 can take a value of 1.07. Cnt1 is the count of the number of times the new information distance is continuously greater than the M1 detection distance threshold or continuously less than the M2 detection distance threshold; CntVal is the number of times the new information distance is continuously offset, which can be 2N (the measurement data is sampled at N points per second for 2 seconds).

[0045] Adaptive filtering module: Kalman filtering and particle filtering are performed on the external trajectory data. After detecting the target maneuver, the filter parameters are modified to achieve accurate tracking of the target.

[0046] The shortcoming of detecting adaptive filtering is that there is a time delay between the occurrence of maneuvers and the detection moment, which has a lag. The time stamp of the output data of the particle filter and the Kalman filter is the time stamp of the filter input data, that is, the filter output data has no time delay. The data sampling frequency of the external measurement equipment of the sea-based mobile aerospace station is generally much higher than the frequency of the communication trajectory data sent to the mission center. Adjust the filter output parameters: adjust the filter output data to the output data output one second before the current filter input data. When the filter input data is an integer second, the filter output data of the previous integer second is sent for communication. When the target maneuvers, the filter parameters can be adjusted for the input data in the last two seconds (recalculate the initial filter value, adjust the variance of the new information, and restart the filter) and re-filtered, thereby optimizing the delay characteristics of the adaptive filter. Such as Figure 4 As shown, the flow chart of the adaptive filtering module. Figure 5 , 6 It is the speed curve of the trajectory data of the adaptive filtering module. Figure 7 , 8 It is a curve diagram of the component of the adaptive filtering trajectory data. In the figure, “polynomial filtering” is the filtering effect of the existing method of the sea-based mobile aerospace station, and “GNSS” is the telemetry prior trajectory data.

[0047] Parameter configuration module: The initial filter value calculation is set to calculate the initial filter value based on prior information or calculate the initial filter value based on multi-point smoothing values ​​of external measured trajectory data, and automatically optimize.

[0048] The initial value calculation parameter settings of the filter mainly include the number of points for external multi-point smoothing.

[0049] The mobile detection method sets the new information offset method and the new information distance weighted square distance detection method.

[0050] The parameters related to maneuver detection mainly refer to the number of consecutive offsets, the threshold of new information offset, the distance threshold and other parameters.

[0051] Through the parameter configuration module, the filter-related settings can be manually adjusted in real time, which is flexible to use.

[0052] The present application effectively improves the real-time filtering effect of the external measurement data of the sea-based mobile aerospace measurement station, and reduces the oscillation amplitude of the velocity curve of the filtering result of the external measurement trajectory data during low elevation angle measurement.

[0053] In addition to the above embodiments, the present invention also includes other implementation modes. Any technical solutions formed by equivalent transformation or equivalent replacement should fall within the protection scope of the claims of the present invention.

Claims

1. A real-time adaptive filtering system for aerospace measurement and control external data based on prior information, characterized in that: It includes a filter initial value calculation module, a maneuver detection module, an adaptive filter module and a parameter configuration module. The filter initial value calculation module is used to accurately calculate the filter initial value; the maneuver detection module is used to detect the target maneuver; the adaptive filter module modifies the filter parameter configuration module according to the target maneuver detected by the maneuver detection module to achieve accurate tracking of the target.

2. The real-time adaptive filtering system for aerospace measurement and control external data based on prior information according to claim 1 is characterized by: The filter initial value calculation module includes calculating the filter initial value using prior information or calculating the filter initial value using multi-point smoothing values ​​of external measured trajectory data.

3. The real-time adaptive filtering system for aerospace measurement and control external data based on prior information according to claim 2 is characterized in that: The prior information includes telemetry trajectory data and theoretical trajectory data.

4. The real-time adaptive filtering system for aerospace measurement and control external data based on prior information according to claim 1 is characterized by: The maneuver detection module includes an innovation offset detection method and an innovation distance weighted square detection method. The innovation offset detection method: under normal circumstances, the innovation mean oscillates around zero; when the target maneuvers, the innovation deviates from the zero mean; Statistics are collected for a period of time. When the number of new information exceeds a given new information threshold continuously within the statistical period, the target maneuver is determined to have occurred, otherwise the target maneuver is eliminated.

5. The real-time adaptive filtering system for aerospace measurement and control external data based on prior information according to claim 4 is characterized in that: The new information threshold is a positive value or 0.

6. The real-time adaptive filtering system for aerospace measurement and control external data based on prior information according to claim 1 is characterized by: The weighted square innovation distance detection method is based on the weighted square deviation of the innovation, and the innovation distance function is: v k is the new information at time k, v k The transpose of S k is the new interest variance, For S k The inverse matrix of D k Subject to χ with m degrees of freedom 2 distributed; Detection of the occurrence and elimination of maneuvers: P[D k >M]=a Where: a is the allowed false alarm probability; M is the threshold, and P is the cumulative probability; When D k When >M, maneuvers occur, increasing the noise variance; When D k When ≤M, the maneuver is eliminated and the noise variance is reduced.

7. The real-time adaptive filtering system for aerospace measurement and control external data based on prior information according to claim 6 is characterized by: The false alarm probability a is 0.03-0.

08.

8. The real-time adaptive filtering system for aerospace measurement and control external data based on prior information according to claim 6 is characterized by: The adaptive filtering module adjusts the filter output parameters: the filter output data is adjusted to the output data output one second before the current filter input data. When the filter input data is an integer second, the filter output data of the previous integer second is sent for communication. When the target maneuvers, the filter parameters of the input data within the last two seconds can be adjusted and then re-filtered, thereby optimizing the delay characteristics of the adaptive filtering.

Images