A navigation error compensation method for a micro-inertial satellite integrated navigation system

Abstract

The present invention provides a navigation error compensation method for a micro-inertial satellite integrated navigation system, comprising: collecting gyro and meter outputs, as well as navigation information, from the micro-inertial satellite integrated navigation system, and preprocessing the gyro and meter outputs to obtain gyro angular rate increments and meter acceleration increments; setting model training inputs and outputs; using a deep learning LSTM model for model training to obtain an inertial navigation error training model; performing real-time navigation calculation based on the real-time measured gyro angular rate and meter specific force to obtain real-time angular rate increments and acceleration increments; if satellite receiver data is invalid and the current time is greater than or equal to l×ΔT, obtaining a real-time predicted inertial navigation error based on the real-time angular rate increments and acceleration increments based on the inertial navigation error training model, and completing navigation error compensation based on the real-time predicted inertial navigation error. The present invention can solve the technical problem of insufficient navigation accuracy of micro-inertial satellite integrated navigation systems in the prior art.

Description

Technical Field

[0001] The present invention relates to the technical field of micro-inertial satellite integrated navigation systems, and in particular to a navigation error compensation method for a micro-inertial satellite integrated navigation system. Background Art

[0002] Compared with traditional mechanical and optical inertial navigation systems, MEMS inertial navigation systems (MEMS) offer advantages such as small size, light weight, low power consumption, strong overload resistance, and high cost-performance. With technological advancements, MEMS inertial navigation systems are finding widespread application in areas such as small unmanned aerial vehicle (UAV) navigation and control, land vehicle stabilization and targeting, smart bomb guidance, seeker stabilization, short-range tactical missile mid-range guidance, torpedo attitude and heading systems, and individual soldier navigation and positioning. Currently, MEMS is the optimal choice for achieving the "small size, light weight, and low cost" navigation system while meeting precision requirements.

[0003] Compared to traditional optical inertial navigation systems, micro-electromechanical inertial navigation systems (MEMSs) are less accurate and typically require integration with satellite navigation systems to output high-precision combined navigation position, velocity, and attitude information. However, in scenarios such as plateaus, mountains, jungles, and urban roads, where satellite signals are subject to severe obstruction and interference, the positioning performance of the micro-inertial / satellite combined navigation system is insufficient to meet the navigation accuracy requirements of equipment. Therefore, research is needed to improve the accuracy of micro-inertial combined navigation in satellite-denied environments. Summary of the Invention

[0004] The present invention aims to solve at least one of the technical problems existing in the prior art.

[0005] The present invention provides a navigation error compensation method for a micro-inertial satellite integrated navigation system. The navigation error compensation method comprises the following steps: collecting a gyro and an add-table output of the micro-inertial satellite integrated navigation system, as well as navigation information, and pre-processing the gyro and the add-table output to obtain a gyro angular rate increment and an add-table acceleration increment; performing feature extraction based on the gyro angular rate increment, the add-table acceleration increment, and the navigation information, and setting a model training input and a model training output; performing model training using a deep learning LSTM model based on the model training input and the model training output to obtain an inertial navigation error training model; and performing feature extraction based on the inertial navigation system. The system uses the real-time measured gyro angular rate and gage force to perform real-time navigation calculations to obtain real-time angular rate increments and acceleration increments. It determines whether the satellite receiver data is valid. If the satellite receiver data is valid, the Kalman filter method is used for combined filtering to correct the navigation error in real time. If the satellite receiver data is invalid and the current time is greater than or equal to l×ΔT, where l is the lookback value of the inertial navigation error training model and ΔT is the interval period, the real-time predicted inertial navigation error is obtained based on the real-time angular rate increments and acceleration increments based on the inertial navigation error training model. The navigation error compensation of the micro-inertial satellite integrated navigation system is completed based on the real-time predicted inertial navigation error.

[0006] Further, according to Preprocessing obtains the gyro angular rate increment, where is the gyro angular rate increment during the ΔT period, is the angular velocity measured by the gyroscope, and T is the IMU sampling period.

[0007] Further, according to Preprocessing to obtain the acceleration increment, where: is the acceleration increment in the ΔT time period, is the specific force measured by the meter, and T is the IMU sampling period.

[0008] Furthermore, set the model training input to

[0009] in, Represents the gyro angular rate increment Components in the x, y, and z directions; Respectively represent the acceleration increment Components in the x, y, and z directions; L k-1 ,λ k-1 、h k-1 Respectively represent the latitude, longitude, and altitude values ​​at the last interval period; Vn k-1 、Vu k-1 、Ve k-1 Respectively represent the north, sky, and east speeds at the last interval period; γ k-1 ,θ k-1 , ψ k-1 They respectively represent the roll angle, pitch angle, and heading angle at the last interval period.

[0010] Furthermore, set the model training output to

[0011] Y OUT =(ΔSn, ΔSu, ΔSe, ΔVn, ΔVu, ΔVe, Δγ, Δθ, Δψ), where ΔSn, ΔSu, and ΔSe are the components of the position error in the east direction of the north sky calculated by the inertial navigation system in the current interval period; ΔVn, ΔVu, and ΔVe are the components of the velocity error in the east direction of the north sky calculated by the inertial navigation system in the current interval period; Δγ, Δθ, and Δψ are the roll angle error, pitch angle error, and heading angle error, respectively, calculated by the inertial navigation system in the current interval period.

[0012] Further, according to Perform closed-loop correction of navigation error, where V, S, γ, θ, and ψ are the compensated velocity, position, roll angle, pitch angle, and heading angle, respectively. V INS 、S INS , γ INS,θ INS , ψ INS are the speed, position, roll angle, pitch angle and heading angle obtained by real-time navigation solution, respectively. ΔV′, ΔS′, Δγ′, Δθ′ and Δψ′ are the speed error, position error, roll angle error, pitch angle error and heading angle error obtained by real-time prediction based on the inertial navigation error training model, respectively.

[0013] The technical solution of the present invention provides a method for compensating navigation errors in a micro-inertial satellite integrated navigation system. This method uses a deep learning LSTM model (Long Short-Term Memory Network) to train an inertial navigation error training model based on collected integrated navigation data. When satellites lose lock, the method uses the inertial navigation error training model to obtain real-time predicted inertial navigation errors, thereby completing navigation error compensation for the micro-inertial satellite integrated navigation system. Compared with the prior art, the technical solution of the present invention can solve the technical problem of insufficient navigation accuracy in the prior art micro-inertial satellite integrated navigation system. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] The accompanying drawings are included to provide a further understanding of the embodiments of the present invention, constitute a part of the specification, illustrate the embodiments of the present invention, and together with the description, explain the principles of the present invention. Obviously, the drawings described below are only some embodiments of the present invention, and those skilled in the art can derive other drawings based on these drawings without inventive effort.

[0015] Figure 1 The figure shows a flow chart of a navigation error compensation method for a micro-inertial satellite integrated navigation system provided according to a specific embodiment of the present invention. DETAILED DESCRIPTION

[0016] It should be noted that, in the absence of conflict, the embodiments in this application and the features in the embodiments can be combined with each other. The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings 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. The following description of at least one exemplary embodiment is actually only illustrative and is in no way intended to limit the present invention and its application or use. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0017] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present application. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that when the terms "comprise" and / or "include" are used in this specification, they indicate the presence of features, steps, operations, devices, components and / or combinations thereof.

[0018] Unless otherwise specifically stated, the relative arrangement of the parts and steps, numerical expressions and numerical values ​​set forth in these embodiments do not limit the scope of the present invention. At the same time, it should be understood that, for ease of description, the sizes of the various parts shown in the accompanying drawings are not drawn according to actual proportional relationships. The technology, methods and equipment known to those of ordinary skill in the relevant art may not be discussed in detail, but in appropriate cases, the technology, methods and equipment should be considered as a part of the specification. In all examples shown and discussed here, any specific value should be interpreted as being merely exemplary, rather than as a limitation. Therefore, other examples of the exemplary embodiments can have different values. It should be noted that similar numbers and letters represent similar items in the following drawings, and therefore, once an item is defined in an accompanying drawing, it does not need to be further discussed in subsequent drawings.

[0019] like Figure 1 As shown, according to a specific embodiment of the present invention, a method for compensating navigation errors of a micro-inertial satellite integrated navigation system is provided. The method for compensating navigation errors of a micro-inertial satellite integrated navigation system includes:

[0020] Collect the gyro and table outputs and navigation information of the micro-inertial satellite integrated navigation system, and pre-process the gyro and table outputs to obtain the gyro angular rate increment and table acceleration increment;

[0021] Perform feature extraction based on the gyro angular rate increment, the gimbal acceleration increment, and navigation information, and set the model training input and output;

[0022] Based on the model training input and model training output, a deep learning LSTM model is used to train the model and obtain the inertial guidance error training model.

[0023] Perform real-time navigation calculation based on the gyro angular rate and acceleration ratio measured by the inertial navigation system to obtain real-time angular rate increment and acceleration increment;

[0024] Determine whether the satellite receiver data is valid. If the satellite receiver data is valid, the Kalman filter method is used for combined filtering to correct the navigation error in real time. If the satellite receiver data is invalid and the current time is greater than or equal to l×ΔT, based on the inertial navigation error training model, the real-time predicted inertial navigation error is obtained according to the real-time angular rate increment and acceleration increment, and the navigation error compensation of the micro-inertial satellite integrated navigation system is completed based on the real-time predicted inertial navigation error.

[0025] This configuration provides a method for compensating navigation errors in a micro-inertial satellite integrated navigation system. This method uses a deep learning LSTM model for model training based on collected integrated navigation data to obtain an inertial navigation error training model. When satellites lose lock, the method uses the inertial navigation error training model to obtain real-time predicted inertial navigation errors, completing navigation error compensation for the micro-inertial satellite integrated navigation system. Compared to existing technologies, this technical solution can address the technical problem of insufficient navigation accuracy in existing micro-inertial satellite integrated navigation systems.

[0026] Furthermore, in the present invention, in order to realize navigation error compensation of the micro-inertial satellite integrated navigation system, the gyroscope and summing table outputs and navigation information of the micro-inertial satellite integrated navigation system are first collected, and the gyroscope and summing table outputs are preprocessed to obtain the gyroscope angular rate increment and the summing table acceleration increment.

[0027] As a specific embodiment of the present invention, Preprocessing obtains the gyro angular rate increment, where is the gyro angular rate increment within the ΔT time period, ΔT is the interval period, is the angular velocity measured by the gyroscope, and T is the IMU sampling period.

[0028] According to Preprocessing to obtain the acceleration increment, where: is the acceleration increment within the ΔT time period, ΔT is the interval period, is the specific force measured by the meter, and T is the IMU sampling period.

[0029] Furthermore, in the present invention, after pre-processing the gyro and adder table outputs, feature extraction is performed based on the gyro angular rate increment and adder table acceleration increment, as well as navigation information, and model training input and output are set.

[0030] As a specific embodiment of the present invention, the model training input is set to

[0031] in, Represents the gyro angular rate increment Components in the x, y, and z directions; Respectively represent the acceleration increment Components in the x, y, and z directions; L k-1 ,λ k-1 、h k-1 Respectively represent the latitude, longitude, and altitude values ​​at the last interval period; Vn k-1 、Vu k-1 、Ve k-1 Respectively represent the north, sky, and east speeds at the last interval period; γ k-1 ,θ k-1 , ψ k-1 They respectively represent the roll angle, pitch angle, and heading angle at the last interval period.

[0032] Set the model training output to Y OUT =(ΔSn, ΔSu, ΔSe, ΔVn, ΔVu, ΔVe, Δγ, Δθ, Δψ), where ΔSn, ΔSu, and ΔSe are the components of the position error in the east direction of the north sky calculated by the inertial navigation system in the current interval period; ΔVn, ΔVu, and ΔVe are the components of the velocity error in the east direction of the north sky calculated by the inertial navigation system in the current interval period; Δγ, Δθ, and Δψ are the roll angle error, pitch angle error, and heading angle error, respectively, calculated by the inertial navigation system in the current interval period.

[0033] Furthermore, in the present invention, after setting the model training input and output, a deep learning LSTM model is used to perform model training according to the model training input and output, and the inertial navigation error based on the lateral zero speed is obtained as a specific embodiment of the present invention. The inertial navigation error training model review value is set to l, and the inertial navigation error training model is F(p1, p2, p3…pn), where p1, p2, p3…pn are the parameters obtained by training, and n is an integer.

[0034] Furthermore, in the present invention, after obtaining the inertial navigation error training model, real-time navigation solution is performed based on the gyro angular rate and the gimbal force measured in real time by the inertial navigation system to obtain the real-time angular rate increment and the acceleration increment; whether the satellite receiver data is valid is judged; if the satellite receiver data is valid, the Kalman filter method is used for combined filtering to correct the navigation error in real time; if the satellite receiver data is invalid and the current time is greater than or equal to l×ΔT, based on the inertial navigation error training model, the real-time predicted inertial navigation error is obtained according to the real-time angular rate increment and the acceleration increment, and the navigation error compensation of the micro-inertial satellite integrated navigation system is completed according to the real-time predicted inertial navigation error.

[0035] As a specific embodiment of the present invention, if the satellite receiver data is invalid and the current time is greater than or equal to l×ΔT, based on the inertial navigation error training model, the real-time predicted inertial navigation errors ΔV′, ΔS′, Δγ′, Δθ′, Δψ′ are obtained according to the real-time angular rate increment and acceleration increment, and Perform closed-loop correction of navigation error, where V, S, γ, θ, and ψ are the compensated velocity, position, roll angle, pitch angle, and heading angle, respectively. V INS 、S INS , γ INS ,θ INS , ψ INS are the speed, position, roll angle, pitch angle and heading angle obtained by real-time navigation solution, respectively. ΔV′, ΔS′, Δγ′, Δθ′ and Δψ′ are the speed error, position error, roll angle error, pitch angle error and heading angle error obtained by real-time prediction based on the inertial navigation error training model, respectively.

[0036] In addition, the interval period of the real-time predicted inertial navigation error obtained based on the inertial navigation error training model is ΔT.

[0037] The present invention provides a navigation error compensation method for a micro-inertial satellite integrated navigation system. The method utilizes a deep learning method based on collected normal integrated navigation data to establish a deep learning model. The method uses the micro-inertial system's gyroscope, gating data, and navigation speed and attitude data as inputs, and uses navigation speed error, attitude error, and position error as outputs. The method then performs model training to obtain an inertial navigation error training model. When a satellite fails, the trained inertial navigation error training model is used to input gyroscope, gating data, speed, and attitude information. The model then outputs the current inertial navigation speed error, attitude error, and position error in real time, and then performs closed-loop compensation to obtain a final navigation result, effectively improving the integrated navigation measurement accuracy.

[0038] In order to have a further understanding of the present invention, the following Figure 1 The navigation error compensation method of the micro-inertial satellite integrated navigation system of the present invention is described in detail.

[0039] like Figure 1 As shown, according to a specific embodiment of the present invention, a method for compensating navigation errors of a micro-inertial satellite integrated navigation system is provided, which includes the following steps.

[0040] Step 1: collect the gyro and table output of the micro-inertial satellite integrated navigation system, as well as navigation information; Preprocessing to obtain gyro angular rate increment; according to Preprocessing obtains the acceleration increment of the addition table.

[0041] Step 2: Set the model training input to The model training output is Y OUT =(ΔSn, ΔSu, ΔSe, ΔVn, ΔVu, ΔVe, Δγ, Δθ, Δψ).

[0042] Step 3: Use the deep learning LSTM model to perform model training based on the model training input and output to obtain the inertial error training model F(p1, p2, p3…pn).

[0043] Step 4: Perform real-time navigation calculation based on the gyro angular rate and acceleration ratio measured in real time by the inertial navigation system to obtain the real-time angular rate increment and acceleration increment.

[0044] Step 5: Determine whether the satellite receiver data is valid. If the satellite receiver data is valid, the Kalman filter method is used for combined filtering to correct the navigation error in real time. If the satellite receiver data is invalid and the current time is greater than or equal to l×ΔT, the inertial navigation error training model is used to obtain the real-time predicted inertial navigation error according to the real-time angular rate increment and acceleration increment. Perform closed-loop correction of navigation error to complete navigation error compensation of micro-inertial satellite integrated navigation system.

[0045] In summary, the present invention provides a method for compensating navigation errors in a micro-inertial satellite integrated navigation system. This method uses a deep learning LSTM model for model training based on collected integrated navigation data to obtain an inertial navigation error training model. When satellites lose lock, the method uses the inertial navigation error training model to obtain real-time predicted inertial navigation errors, thereby completing navigation error compensation for the micro-inertial satellite integrated navigation system. Compared with the prior art, the technical solution of the present invention can solve the technical problem of insufficient navigation accuracy in prior art micro-inertial satellite integrated navigation systems.

[0046] The foregoing description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Those skilled in the art will readily appreciate that various modifications and variations of the present invention are possible. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present invention are intended to be within the scope of protection of the present invention.

Claims

1. A navigation error compensation method for a micro-inertial satellite integrated navigation system, characterized in that: The navigation error compensation method of the micro-inertial satellite integrated navigation system comprises: Gyroscope and table outputs and navigation information of the micro-inertial satellite integrated navigation system are collected, and the gyroscope and table outputs are preprocessed to obtain gyroscope angular rate increments and table acceleration increments; Perform feature extraction based on the gyro angular rate increment and the gimbal acceleration increment, and the navigation information, and set a model training input and a model training output; Performing model training using a deep learning LSTM model based on the model training input and model training output to obtain an inertial guidance error training model; Perform real-time navigation calculation based on the gyro angular rate and acceleration ratio measured by the inertial navigation system to obtain real-time angular rate increment and acceleration increment; Whether the satellite receiver data is valid is determined. If the satellite receiver data is valid, a Kalman filter method is used for combined filtering to correct the navigation error in real time. If the satellite receiver data is invalid and the current time is greater than or equal to l×ΔT, where l is a look-back value of the inertial navigation error training model and ΔT is an interval period, a real-time predicted inertial navigation error is obtained based on the real-time angular rate increment and acceleration increment based on the inertial navigation error training model, and navigation error compensation of the micro-inertial satellite integrated navigation system is completed based on the real-time predicted inertial navigation error.

2. The navigation error compensation method of the micro-inertial satellite integrated navigation system according to claim 1, characterized in that: according to Preprocessing obtains the gyro angular rate increment, where is the gyro angular rate increment during the ΔT period, is the angular velocity measured by the gyroscope, and T is the IMU sampling period.

3. The navigation error compensation method of the micro-inertial satellite integrated navigation system according to claim 1, characterized in that: according to Preprocessing to obtain the acceleration increment, where: is the acceleration increment in the ΔT time period, is the specific force measured by the meter, and T is the IMU sampling period.

4. The navigation error compensation method of a micro-inertial satellite integrated navigation system according to any one of claims 1 to 3, characterized in that: Set the model training input to in, Represents the gyro angular rate increment Components in the x, y, and z directions; Respectively represent the acceleration increment Components in the x, y, and z directions; L k-1 ,λ k-1 、h k-1 Respectively represent the latitude, longitude, and altitude values ​​at the last interval period; Vn k-1 、Vu k-1 、Ve k-1 Respectively represent the north, sky, and east speeds at the last interval period; γ k-1 ,θ k-1 , ψ k-1 They respectively represent the roll angle, pitch angle, and heading angle at the last interval period.

5. The navigation error compensation method of a micro-inertial satellite integrated navigation system according to any one of claims 1 to 4, characterized in that: Set the model training output to Y OUT =(ΔSn, ΔSu, ΔSe, ΔVn, ΔVu, ΔVe, Δγ, Δθ, Δψ), where ΔSn, ΔSu, and ΔSe are the components of the position error in the east direction of the north sky calculated by the inertial navigation system in the current interval period; ΔVn, ΔVu, and ΔVe are the components of the velocity error in the east direction of the north sky calculated by the inertial navigation system in the current interval period; Δγ, Δθ, and Δψ are the roll angle error, pitch angle error, and heading angle error, respectively, calculated by the inertial navigation system in the current interval period.

6. The navigation error compensation method of the micro-inertial satellite integrated navigation system according to claim 1, characterized in that: according to Perform closed-loop correction of navigation error, where V, S, γ, θ, and ψ are the compensated velocity, position, roll angle, pitch angle, and heading angle, respectively. V INS 、S INS , γ INS ,θ INS , ψ INS are the speed, position, roll angle, pitch angle and heading angle obtained by real-time navigation solution, respectively; ΔV′, ΔS′, Δγ′, Δθ′, Δψ′ are the speed error, position error, roll angle error, pitch angle error and heading angle error obtained by real-time prediction based on the inertial navigation error training model, respectively.

Images