A method for selecting and switching a cold and hot start temperature error compensation model of an optical gyroscope

Abstract

The application provides a kind of optical gyro cold and hot start temperature error compensation model selection and switching method, and the temperature space gradient between different positions of gyro at the moment of starting inertial navigation is obtained by multiple temperature sensors arranged in the gyro instrument area of inertial navigation.The "cold start" and "hot start" of inertial navigation are regarded as two groups, and the "Mahalanobis distance" of the sample set formed by the space gradient to the two groups is calculated.The application can judge the temperature state of the inertial navigation in a very short time after starting, and select the correct error model to compensate the gyro output, and at the same time, it can complete the compensation model transition switching of the gyro from the cold start stage to the hot state long time stable working stage, and is compatible with the different working states of cold start and hot start.The method has small calculation amount, is simple and easy to implement, and is convenient for engineering implementation.

Description

Technical Field

[0001] This application relates to the field of inertial navigation technology, specifically to a method for selecting and switching an optical gyroscope cold and hot start temperature error compensation model. Background Technology

[0002] Optical gyroscopes mainly include fiber optic gyroscopes and laser gyroscopes. Compared with classic rotor gyroscopes, they have significant advantages such as a wider angular velocity measurement range, resistance to overload shocks, and shorter startup time, making them particularly suitable for strapdown inertial navigation systems (INS). As the core instrument of an INS, the performance of the gyroscope directly determines the navigation accuracy of the INS. However, the output of optical gyroscopes is affected by temperature, and the resulting instrument temperature error severely restricts the improvement of system alignment, navigation accuracy, and rapid response performance. The commonly used solution is temperature error compensation. Temperature error compensation is essentially a mathematical modeling method. It uses temperature testing to excite the gyroscope and separates the temperature error from the gyroscope output. It then uses the correlation between the error and temperature to establish a model of the two and subtracts it from the system output. However, in actual implementation, the temperature error behavior of the gyroscope often differs greatly between two conditions: starting from a cold environment (the gyroscope has been de-powered and cooled for a long time before startup, and is approximately in thermal equilibrium with the environment) and starting from a hot environment (the gyroscope has been powered on for a long time before startup, and its internal temperature is higher than the ambient temperature).

[0003] The gyroscope temperature error model obtained through long-term, full-temperature-range testing is primarily a hot-state model and cannot accurately describe the error pattern during the cold-state startup phase. Therefore, it is necessary to specifically model this brief cold-state startup process. Furthermore, the gyroscope temperature error models established under both conditions are not interchangeable. Thus, it is crucial to accurately determine the inertial navigation system's temperature state within the extremely short startup time to select the correct error model for gyroscope output compensation and to facilitate a smooth transition from the cold-state startup phase to the long-term stable hot-state operating phase.

[0004] Currently, most publicly available research focuses on gyroscope temperature error compensation during long-term operation of inertial navigation systems, with little attention paid to gyroscope cold start-up, and almost no research on the selection and switching methods of cold and hot start-up models for different operating stages. Summary of the Invention

[0005] To address the aforementioned technical problems, this invention provides a method for selecting and switching an optical gyroscope's cold / hot start temperature error compensation model. The technical solution adopted by this invention is as follows:

[0006] A method for selecting and switching an optical gyroscope cold / hot start temperature error compensation model, the method comprising the following steps:

[0007] Step 1: Place multiple temperature sensors at different positions on different gyroscopes of the inertial navigation system, and let the total number of sensors be n.

[0008] Step 2: Calculate the spatial temperature gradient using the absolute temperature difference measured by temperature sensors at different locations;

[0009] Step 3: Conduct n1 groups of inertial navigation cold-state startup tests and n2 groups of inertial navigation hot-state startup tests as training datasets. Treat the cold and hot environments as two populations, denoted as G1 and G2; calculate the estimated population mean of the temperature gradient for each population. and covariance estimation

[0010] Step 4: Based on the Mahalanobis distance overall discrimination criterion, determine whether the startup is in a cold or hot state based on the discrimination threshold TSH;

[0011] Step 5: If the system is determined to be a hot start in Step 4, then the hot start temperature error compensation model Mod2 is directly selected to compensate the gyroscope output. Mod2 is the error compensation amount calculated in real time by the pre-established hot start temperature error model, which can be expressed as: Where T g (t)=[T g1 (t)T g2 (t)...T gn (t)] T The real-time temperature is collected by the gyroscope temperature sensor. This is a function for compensating for hot-state temperature error based on the correlation between error and temperature.

[0012] If step 4 determines it to be a cold start, then during the inertial navigation system's operating time t < t G The internal system uses a cold-start temperature error compensation model Mod1 to compensate for the gyroscope output. Mod1 is the error compensation amount calculated in real time by the pre-established cold-start temperature error model, which can be expressed as: This represents the cold-state temperature error compensation function established based on the correlation between error and temperature, t G The applicable duration for the cold start model; in t G ≤t≤t T Internal arrangements for the transition process, t T To schedule the model transition time, a transition function is used to smoothly and asymptotically switch the compensation amount output by the cold-start temperature error compensation model to the compensation amount output by the hot-start model; when t>t T In this case, the Mod2 hot-start temperature error compensation model is used to compensate the gyroscope output.

[0013] Furthermore, in step 1 above, the placement of multiple temperature sensors at different positions on different gyroscopes of the inertial navigation system includes:

[0014] One temperature sensor is placed on each of the gyroscopes in the X, Y, and Z directions, meaning n = 3; or

[0015] For laser gyroscopes, a temperature sensor is placed at both the anode and the casing of each gyroscope, i.e., n = 6; or

[0016] For fiber optic gyroscopes, a temperature sensor is placed near the fiber and close to the surface of the housing on each gyroscope, i.e., n=6.

[0017] Furthermore, in step 2, the calculation of the temperature spatial gradient includes:

[0018] Let the real-time temperature measurement value of n temperature sensors be T. gi (t), i=1,…,n, then the temperature gradient between the two temperature sensors is expressed as:

[0019]

[0020] Using the above calculation formula, the maximum value that can be calculated is... A temperature spatial gradient.

[0021] Furthermore, in step 2, after calculating the temperature spatial gradient, the method further includes:

[0022] Based on the actual situation, the calculation yields... Choose k from the temperature spatial gradients. Typical gradient composition of measurement sample ΔT g (t), the measured sample ΔT g (t) is represented as:

[0023]

[0024] Let the inertial navigation system start at t=0, and the temperature gradient ΔT be considered. g (t) is smoothed to obtain the smoothed temperature gradient value ΔT. gs Let the smoothing time be Δt, which is expressed as:

[0025]

[0026] Since the absolute temperature noise is very small, the value is taken as Δt = 1s.

[0027] Furthermore, the estimation of the population mean of the temperature gradient of the two populations is calculated separately. and include:

[0028] Based on the experimental results and the temperature gradient smoothing value ΔT gs The calculation formula, under the two total conditions, expresses the temperature gradient elements belonging to the two totalities as follows:

[0029]

[0030]

[0031] Estimation of the population mean of the temperature gradient of two populations and Calculated using the following formula:

[0032]

[0033]

[0034] Furthermore, the covariance estimation Calculated using the following formula:

[0035]

[0036]

[0037]

[0038] Furthermore, the discrimination threshold TSH is expressed as:

[0039]

[0040] in,

[0041] Furthermore, in step 4, cold and hot startup are determined based on the discrimination threshold TSH, including:

[0042] This discrimination threshold is used to analyze the real-time temperature spatial gradient ΔT obtained at the inertial navigation system startup time. gs Perform online discrimination; if the discrimination function ω(ΔT) gs ):

[0043] ω(ΔT gs ) = a T ·ΔT gs ≥TSH

[0044] If the inertial navigation system is in a cold start state, it is determined that the inertial navigation system is in a hot start state; otherwise, it is determined that the inertial navigation system is in a hot start state.

[0045] Furthermore, the step of smoothly and asymptotically switching the compensation amount output by the cold-state start-up temperature error compensation model to the compensation amount output by the hot-state start-up model using a transition function includes:

[0046] In t G ≤t≤t T During the time period, the real-time compensation values ​​ΔM1(t) and ΔM2(t) of the cold-state model Mod1 and the hot-state model Mod2 are calculated simultaneously. The compensation value output by the cold-state start-up temperature error compensation model is smoothly and asymptotically switched to the compensation value output by the hot-state start-up model using the transition function F(t), expressed as:

[0047]

[0048] ΔM(t)=[1-F(t)]ΔM1(t)+F(t)ΔM2(t)t G ≤t≤t T

[0049] Where F(t) is the transition function, and ΔM(t) is the temperature error compensation amount of the transition process that integrates the cold and hot models.

[0050] Furthermore, in step 5, t G With t T The selection should be based on the cold start modeling dataset duration of the inertial navigation system and the temperature characteristics of the inertial navigation system.

[0051] The following technical effects can be achieved through the embodiments of this application:

[0052] (1) This invention can accurately distinguish between cold and hot start-up of the inertial navigation system in a very short time and select the correct error model to compensate the gyroscope output. At the same time, it can smoothly switch the compensation model from cold start-up to hot stable working stage, and is compatible with different working states of cold start-up and hot start-up.

[0053] (2) The present invention uses the spatial gradient of the temperature field distribution of the gyroscope as the distinguishing mark between hot and cold states. There is no need to calculate the temperature change rate over time, thus avoiding the influence of noise introduced by the calculation of the first derivative of temperature on the accuracy of the judgment and the start-up delay introduced by smoothing differential noise. Attached Figure Description

[0054] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0055] Figure 1 This is a schematic diagram of the temperature error compensation model selection and switching process of the present invention. Detailed Implementation

[0056] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0057] Figure 1 This is a schematic diagram illustrating the selection and switching process of the temperature error compensation model in this invention. The technical solution of this invention involves obtaining the spatial temperature gradient between different positions of the gyroscope at the start-up time of the inertial navigation system (INS) using multiple temperature sensors placed in the gyroscope instrument area. The INS' "cold start" and "hot start" are considered as two populations, and the "Mahavior distance" between the sample set formed by the spatial gradient and the two populations is calculated. A distance discrimination threshold is generated through multiple pre-trained cold and hot start temperature experiments, serving as the discrimination criterion for cold and hot starts. This criterion is used to perform online discrimination of the spatial temperature gradient obtained at the start-up time of the INS. If it is determined to be a hot start, the hot start temperature error compensation model is directly selected to compensate the gyroscope output; if it is determined to be a cold start, the temperature error compensation model is used to compensate the gyroscope output when the INS operating time t < t G The internal cold-start temperature error compensation model is used to compensate the gyroscope output (t). G (This refers to the applicable duration of the cold start model). Afterwards, at t... G ≤t≤t T Internal arrangement of transition process (t) T (For the planned model transition time), a transition function is used to smoothly and asymptotically switch the compensation amount output by the cold-start temperature error compensation model to the compensation amount output by the hot-start model. When t>t T In this case, a hot-state start-up temperature error compensation model is used to compensate for the gyroscope output. The optical gyroscope cold / hot start-up temperature error compensation model selection and switching method based on spatial gradient discrimination utilizes the Mahalanobis distance calculated from the temperature spatial gradients measured by different gyroscope temperature sensors to accurately distinguish between cold and hot start-ups of the inertial navigation system, and arranges a transition process to achieve a natural switching between the cold and hot state models, such as... Figure 1 As shown.

[0058] The method for selecting and switching the temperature error compensation model of the present invention includes the following steps:

[0059] Step 1: Place multiple temperature sensors at different positions on different gyroscopes of the inertial navigation system. Let the total number of sensors be n.

[0060] In step 1 above, the placement of multiple temperature sensors at different positions on different inertial navigation gyroscopes includes:

[0061] One temperature sensor is placed on each of the gyroscopes in the X, Y, and Z directions, so n = 3.

[0062] For laser gyroscopes, a temperature sensor is placed between the anode of each gyroscope and the shell, i.e., n = 6.

[0063] For fiber optic gyroscopes, a temperature sensor is placed at the position of each gyroscope near the fiber and at the position of each gyroscope near the surface of the housing, that is, n=6;

[0064] Step 2: Calculate the spatial temperature gradient using the absolute temperature difference measured by temperature sensors at different locations;

[0065] In step 2, the calculation of the temperature spatial gradient includes:

[0066] Let the real-time temperature measurement value of n temperature sensors be T. gi (t), i=1,…,n, then the temperature gradient between the two temperature sensors is expressed as:

[0067]

[0068] Using the above calculation formula, the maximum value that can be calculated is... A temperature spatial gradient;

[0069] In step 2, after calculating the temperature spatial gradient, the following is also included:

[0070] Based on the actual situation, the calculation yields... Choose k from the temperature spatial gradients. Typical gradient composition of measurement sample ΔT g (t), the measured sample ΔT g (t) is represented as:

[0071]

[0072] Let the inertial navigation system start at t=0, and the temperature gradient ΔT be considered. g (t) is smoothed to obtain the smoothed temperature gradient value ΔT. gs Let the smoothing time be Δt, which is expressed as:

[0073]

[0074] Since the absolute temperature noise is very small, the value is taken as Δt = 1s;

[0075] Step 3: Conduct n1 groups of inertial navigation cold-state startup tests and n2 groups of inertial navigation hot-state startup tests as training datasets. Treat the cold and hot environments as two populations, denoted as G1 and G2; calculate the estimated population mean of the temperature gradient for each population. and covariance estimation

[0076] The estimation of the population mean of the temperature gradient of the two populations is calculated separately. and include:

[0077] Based on the experimental results and the temperature gradient smoothing value ΔT gs The calculation formula, under the two total conditions, expresses the temperature gradient elements belonging to the two totalities as follows:

[0078]

[0079]

[0080] Estimation of the population mean of the temperature gradient of two populations and Calculated using the following formula:

[0081]

[0082]

[0083] The covariance estimation Calculated using the following formula:

[0084]

[0085]

[0086]

[0087] Step 4: Based on the Mahalanobis distance overall discrimination criterion, determine whether the startup is in a cold or hot state based on the discrimination threshold TSH;

[0088] The discrimination threshold TSH is expressed as:

[0089]

[0090] in,

[0091] In step 4, cold and hot start-up are determined based on the discrimination threshold TSH, including:

[0092] This discrimination threshold is used to analyze the real-time temperature spatial gradient ΔT obtained at the inertial navigation system startup time. gs Perform online discrimination; if the discrimination function ω(ΔT) gs ):

[0093] ω(ΔT gs ) = a T ·ΔT gs≥TSH

[0094] If the inertial navigation system is in a cold start state, it is determined that the inertial navigation system is in a hot start state; otherwise, it is determined that the inertial navigation system is in a hot start state.

[0095] Step 5: If the system is determined to be a hot start in Step 4, then the hot start temperature error compensation model Mod2 is directly selected to compensate the gyroscope output. Mod2 is the error compensation amount calculated in real time by the pre-established hot start temperature error model, which can be expressed as: Where T g (t)=[T g1 (t)T g2 (t)...T gn (t)] T The real-time temperature is collected by the gyroscope temperature sensor. The hot-state temperature error compensation function, established based on the correlation between error and temperature, can be modeled using a polynomial. For example, if a single temperature sensor is used, It can be represented as (a0, a1, a2 are model coefficients);

[0096] If step 4 determines it to be a cold start, then during the inertial navigation system's operating time t < t G The internal system uses a cold-start temperature error compensation model to compensate for the gyroscope output. G The applicable duration for the cold start model; in t G ≤t≤t T Internal arrangements for the transition process, t T To schedule the model transition time, a transition function is used to smoothly and asymptotically switch the compensation amount output by the cold-start temperature error compensation model to the compensation amount output by the hot-start model; when t>t T In this case, a hot-start temperature error compensation model is used to compensate for the gyroscope output.

[0097] The method of compensating the gyroscope output using a cold-start temperature error compensation model includes: compensating the gyroscope output using a cold-start temperature error compensation model Mod1, where Mod1 is the error compensation amount calculated in real time by a pre-established cold-start temperature error model, which can be expressed as: The cold-state temperature error compensation function, established based on the correlation between error and temperature, can also be modeled using a polynomial: if a single temperature sensor is used, It can be represented as (b0, b1, b2 are model coefficients).

[0098] The process of smoothly and asymptotically switching the compensation amount output by the cold-start temperature error compensation model to the compensation amount output by the hot-start model using a transition function includes:

[0099] In t G ≤t≤t T During the time period, the real-time compensation values ​​ΔM1(t) and ΔM2(t) of the cold-state model Mod1 and the hot-state model Mod2 are calculated simultaneously. The compensation value output by the cold-state start-up temperature error compensation model is smoothly and asymptotically switched to the compensation value output by the hot-state start-up model using the transition function F(t), expressed as:

[0100]

[0101] ΔM(t)=[1-F(t)]ΔM1(t)+F(t)ΔM2(t)t G ≤t≤t T

[0102] Where F(t) is the transition function, and ΔM(t) is the temperature error compensation amount for the transition process that integrates the cold-state model and the hot-state model;

[0103] The method of using a hot-start temperature error compensation model to compensate for the gyroscope output includes: using a hot-start temperature error compensation model Mod2 to compensate for the gyroscope output.

[0104] In the specific implementation process, t G With t T The selection should be based on the cold start modeling dataset duration of the inertial navigation system and the temperature characteristics of the inertial navigation system.

[0105] By adopting the above scheme, the present invention can accurately distinguish between cold and hot start-up of the inertial navigation system in a very short time and select the correct error model to compensate for the gyroscope output. At the same time, it can smoothly switch the compensation model from cold start-up to the stable hot working stage of the gyroscope, and is compatible with different working states of cold start-up and hot start-up.

[0106] The functions described above in this application can be performed at least in part by one or more hardware logic components. For example, without limitation, exemplary types of hardware logic components that can be used include: field programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload programmable logic devices (CPLDs), and so on.

[0107] Furthermore, although the operations are described in a specific order, this should be understood as requiring that such operations be performed in the specific order shown or in sequential order, or requiring that all illustrated operations be performed to achieve the desired result. In certain environments, multitasking and parallel processing may be advantageous. Similarly, although several specific implementation details are included in the above discussion, these should not be construed as limiting the scope of this disclosure. Certain features described in the context of individual embodiments may also be implemented in combination in a single implementation. Conversely, various features described in the context of a single implementation may also be implemented individually or in any suitable sub-combination in multiple implementations.

[0108] Although the subject matter has been described using language specific to structural features and / or device logic, it should be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or actions described above. Rather, the specific features and actions described above are merely illustrative examples of implementing the claims.

Claims

1. A method for selecting and switching an optical gyroscope cold / hot start temperature error compensation model, characterized in that, The method includes the following steps: Step 1: Place multiple temperature sensors at different positions on different gyroscopes of the inertial navigation system, and let the total number of sensors be n. Step 2: Calculate the spatial temperature gradient using the absolute temperature difference measured by temperature sensors at different locations; Step 3: Perform separately Group inertial navigation cold environment start-up test and The inertial conduction hot-state start-up experiment was used as the training dataset, treating the cold and hot environments as two populations, denoted as . and ; Calculate the estimates of the population mean of the temperature gradient for each of the two populations. , and covariance estimation ; Step 4: Based on the overall discrimination criterion of Mahalanobis distance, and based on the discrimination threshold... To determine whether the startup is in a cold or hot state; Step 5: If the start-up is determined to be hot based on Step 4, then the hot-start temperature error compensation model Mod2 is directly selected to compensate the gyroscope output. Mod2 is the error compensation amount calculated in real time by the pre-established hot-start temperature error model, expressed as... ,in The real-time temperature is collected by the gyroscope temperature sensor. This is a function for compensating for hot-state temperature error based on the correlation between error and temperature. If step 4 determines it to be a cold start, then the inertial navigation system will operate for a certain period of time. The internal system uses a cold-start temperature error compensation model Mod1 to compensate for the gyroscope output. Mod1 is the error compensation amount obtained in real time from the pre-established cold-start temperature error model, expressed as: , This represents the cold-state temperature error compensation function established based on the correlation between error and temperature. The applicable duration for the cold start model; in Internal arrangements for the transition process, To schedule the model transition time, a transition function is used to smoothly and asymptotically switch the compensation amount output by the cold-start temperature error compensation model to the compensation amount output by the hot-start model; when t>t T In this case, the Mod2 hot-start temperature error compensation model is used to compensate the gyroscope output.

2. The method according to claim 1, characterized in that, In step 1 above, the placement of multiple temperature sensors at different positions on different inertial navigation gyroscopes includes: One temperature sensor is placed on each of the gyroscopes in the X, Y, and Z directions, meaning n=3; or For laser gyroscopes, a temperature sensor is placed at both the anode and the casing of each gyroscope, i.e., n=6; or For fiber optic gyroscopes, a temperature sensor is placed near the fiber and close to the surface of the housing on each gyroscope, i.e., n=6.

3. The method according to claim 1, characterized in that, In step 2, the calculation of the temperature spatial gradient includes: Let the real-time temperature measurement values ​​of n temperature sensors be... , The temperature gradient between the two temperature sensors is expressed as: Using the above calculation formula, the maximum value that can be calculated is... A temperature spatial gradient.

4. The method according to claim 3, characterized in that, In step 2, after calculating the temperature spatial gradient, the following is also included: Based on the actual situation, the calculation yields... A measurement sample is composed of k typical gradients selected from a given temperature space gradient. , The measurement sample Represented as: Record the start time of inertial navigation For temperature gradient Smoothing is performed to obtain the smoothed temperature gradient value. Let the smoothing time be... , is represented as: Since the absolute temperature noise is very small, the value is... .

5. The method according to claim 1, characterized in that, The estimation of the population mean of the temperature gradient of the two populations is calculated separately. and ,include: Based on the experimental results and the temperature gradient smoothing value... The calculation formula, in the overall and Below, the temperature gradient elements belonging to the two populations are respectively expressed as: Estimation of the population mean of the temperature gradient of two populations and It is calculated using the following formula: 。 6. The method according to claim 5, characterized in that, The covariance estimation It is calculated using the following formula: 。 7. The method according to claim 1, characterized in that, The discrimination threshold Represented as in, , .

8. The method according to claim 1 or 7, characterized in that, In step 4, based on the discrimination threshold To determine whether a startup is in a cold or hot state, the following methods are used: This discrimination threshold is used to analyze the real-time temperature spatial gradient obtained at the inertial navigation system startup time. Perform online discrimination; if the discrimination function... : If the inertial navigation system is in a cold start state, it is determined that the inertial navigation system is in a hot start state; otherwise, it is determined that the inertial navigation system is in a hot start state.

9. The method according to claim 1, characterized in that, The process of smoothly and asymptotically switching the compensation amount output by the cold-start temperature error compensation model to the compensation amount output by the hot-start model using a transition function includes: exist Simultaneously calculate the real-time compensation amounts of the cold-state model Mod1 and the hot-state model Mod2 within the time period. and Using transition functions The compensation amount output by the cold-start temperature error compensation model is smoothly and asymptotically switched to the compensation amount output by the hot-start model, expressed as: in, For transition function, This is a temperature error compensation amount for the transition process that integrates the cold and hot state models.

10. The method according to claim 9, characterized in that, In step 5, and The selection should be based on the cold start modeling dataset duration of the inertial navigation system and the temperature characteristics of the inertial navigation system.

Images