Estimation method for installation error of visual inertial visual downward-looking integrated navigation of unmanned aerial vehicle

Abstract

The invention provides a method for estimating an installation error of visual inertial visual downward-looking combined navigation of an unmanned aerial vehicle, and relates to the technical field of inertial visual combined navigation, and the method comprises the steps: obtaining a current moment observation quantity, a current moment observation matrix and a historical state variable of a visual inertial visual downward-looking combination of the unmanned aerial vehicle in real time; wherein the historical state variable is a state variable at a previous moment of the current moment; determining an adaptive state estimation result and an adaptive state estimation variance according to the current moment observation quantity, the current moment observation matrix and the historical state variable; and performing navigation installation error convergence judgment on the self-adaptive state estimation result and the self-adaptive state estimation variance according to a navigation installation error convergence condition, and determining a navigation installation error convergence judgment result. According to the technical scheme, the error influence caused by re-assembly and disassembly of the integrated navigation system does not need to be considered, the navigation installation error is accurately estimated on line, and the ergodicity and the practical precision of the visual inertial visual downward-looking integrated navigation system of the unmanned aerial vehicle are improved.

Description

Technical Field

[0001] This invention relates to the field of inertial vision integrated navigation technology, and in particular to a method for estimating the installation error of UAV visual-inertial-visual downward-looking integrated navigation. Background Technology

[0002] Inertial-visual-look-down (INS) integrated navigation combines the autonomy, continuity, and real-time capabilities of inertial navigation with the high precision of visual navigation, making it an effective means for UAVs and other aircraft to achieve high-precision, highly autonomous navigation and positioning. During INS ...

[0003] Currently, existing methods for estimating and compensating installation errors involve pre-processing, which is cumbersome and has many limitations. Furthermore, once the integrated navigation system is disassembled and reassembled, the original installation errors are no longer applicable, thus limiting the flexible application of inertial vision integrated navigation systems.

[0004] Therefore, there is an urgent need for a method to estimate the installation error of UAV visual-inertial-visual-downward integrated navigation, which can use real-time information corresponding to the downward integration to estimate the navigation installation error online and accurately, thereby improving the ergonomics and practical accuracy of UAV visual-inertial-visual-downward integrated navigation system. Summary of the Invention

[0005] This invention provides a method for estimating the installation error of a UAV visual-inertial-visual-look-down integrated navigation system. It addresses the shortcomings of existing methods, which involve pre-processing for installation error estimation and compensation, resulting in cumbersome procedures, numerous limitations, and the inapplicability of existing installation errors once the integrated navigation system is disassembled and reassembled. This method achieves the following: by acquiring the current-moment observations, current-moment observation matrix, and historical state variables of the UAV visual-inertial-visual-look-down integrated navigation system in real time, the method determines the adaptive state estimation result and the adaptive state estimation variance. Based on the navigation installation error convergence condition, the method performs a navigation installation error convergence judgment on the adaptive state estimation result and the adaptive state estimation variance, determining the navigation installation error convergence judgment result. This eliminates the need to consider the error impact caused by the disassembly and reassembly of the integrated navigation system, enabling online and accurate estimation of navigation installation errors and improving the ergodicity and practical accuracy of the UAV visual-inertial-visual-look-down integrated navigation system.

[0006] This invention provides a method for estimating the installation error of a UAV's visual-inertial-visual-look-down integrated navigation system, comprising the following steps.

[0007] The system acquires in real time the current observations, current observation matrix, and historical state variables of the UAV's visual-inertial-visual-downward combination; among which, the historical state variables are the state variables of the previous moment.

[0008] The adaptive state estimation result and the adaptive state estimation variance are determined based on the current observations, the current observation matrix, and the historical state variables.

[0009] Based on the convergence condition of navigation installation error, the adaptive state estimation results and the adaptive state estimation variance are used to determine the convergence result of navigation installation error.

[0010] According to the present invention, a method for estimating the installation error of a UAV visual-inertial-visual-look-down integrated navigation system determines the adaptive state estimation result and the adaptive state estimation variance based on the current observations, the current observation matrix, and historical state variables, including: Determine the state prediction results based on historical state variables; Determine the adaptive filter gain based on the observation matrix at the current time. The adaptive state estimation result is determined based on the state prediction result, the adaptive filter gain, the current time observations, and the current time observation matrix. The adaptive state estimation variance is determined based on the adaptive filter gain and the observation matrix at the current time.

[0011] The present invention provides a method for estimating the installation error of a UAV visual-inertial-visual-look-down integrated navigation system, which determines the state prediction result based on historical state variables, including: Obtain the integrated navigation state transition matrix of the UAV based on the historical inertial navigation error patterns; The state prediction result is determined based on the integrated navigation state transition matrix and historical state variables.

[0012] According to the present invention, a method for estimating the installation error of a UAV visual-inertial-visual-look-down integrated navigation system determines the adaptive filtering gain based on the current observation matrix, including: Obtain the current observation noise matrix and determine the variance of the UAV state prediction; The adaptive filtering gain is determined based on the UAV state prediction variance, the current observation noise matrix, and the current observation matrix.

[0013] According to the present invention, a method for estimating the installation error of a UAV visual-inertial-visual-look-down integrated navigation system is provided, which determines the variance of UAV state prediction, including: Obtain the historical error covariance and historical system noise matrix; The UAV state prediction variance is determined based on the integrated navigation state transition matrix, historical error covariance, and historical system noise matrix.

[0014] According to the present invention, a method for estimating the installation error of a UAV visual-inertial-visual-look-down integrated navigation system determines the adaptive filtering gain based on the current observation matrix, including: The adaptive filtering gain is determined based on the UAV state prediction variance, the current observation matrix, and the current observation noise matrix.

[0015] According to the present invention, a method for estimating the installation error of a UAV visual-inertial-visual-look-down integrated navigation system determines the adaptive state estimation variance based on the adaptive filter gain and the current observation matrix, including: The adaptive state estimation variance is determined based on the adaptive filter gain, the current observation matrix, the UAV state prediction variance, and the current observation noise matrix.

[0016] According to the present invention, a method for estimating the installation error of a UAV visual-inertial-visual-downward integrated navigation system is provided. The convergence conditions for the navigation installation error include an effective filtering number threshold, an adaptive state estimation variance threshold, an adaptive state estimation variance absolute value threshold, an adaptive state estimation result peak-to-peak value threshold, and an adaptive state result absolute value threshold.

[0017] According to the present invention, a method for estimating the installation error of a UAV's visual-inertial-visual-look-down integrated navigation system is provided. Based on the navigation installation error convergence condition, the adaptive state estimation result and the adaptive state estimation variance are used to determine the navigation installation error convergence result, including: Obtain the current number of filters to determine the adaptive state estimation result and the adaptive state estimation variance; If the current number of filters is greater than or equal to the effective number of filters threshold, the adaptive state estimation variance is less than or equal to the adaptive state estimation variance threshold, the adaptive state estimation variance is less than or equal to the absolute value threshold of the adaptive state estimation variance, and the adaptive state estimation result is less than or equal to the peak-to-peak value threshold and the adaptive state estimation result is less than or equal to the absolute value threshold of the adaptive state result, then the navigation installation error convergence judgment result is determined to be navigation installation error convergence; otherwise, the navigation installation error convergence judgment result is determined to be navigation installation error non-convergence.

[0018] According to the present invention, a method for estimating the installation error of a UAV visual-inertial-visual-look-down integrated navigation system, after determining the navigation installation error convergence result by performing a navigation installation error convergence judgment on the adaptive state estimation result and the adaptive state estimation variance based on the navigation installation error convergence condition, further includes: If the navigation installation error convergence judgment result is determined to be converged, the navigation installation error convergence judgment result is adaptively corrected online to determine the error correction result.

[0019] According to the present invention, a method for estimating the installation error of a UAV visual-inertial-visual-look-down integrated navigation system, after determining the navigation installation error convergence result by performing a navigation installation error convergence judgment on the adaptive state estimation result and the adaptive state estimation variance based on the navigation installation error convergence condition, further includes: If the navigation installation error convergence judgment result is that the navigation installation error does not converge, continue to execute the steps of real-time acquisition of the current observations, current observation matrix and historical state variables of the UAV visual-inertial-visual downward combination.

[0020] The present invention also provides an estimation device for the installation error of a UAV visual-inertial-visual-look-down integrated navigation system, comprising the following modules: The variable acquisition module is used to acquire the current observations, current observation matrix, and historical state variables of the UAV's visual-inertial-visual-downward combination in real time; among which, the historical state variables are the state variables of the previous moment of the current moment. The variance determination module is used to determine the adaptive state estimation result and the adaptive state estimation variance based on the current observations, the current observation matrix, and the historical state variables. The result determination module is used to determine the convergence of navigation installation error based on the adaptive state estimation results and the adaptive state estimation variance according to the navigation installation error convergence conditions, and to determine the navigation installation error convergence judgment result.

[0021] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the estimation method for installation error of UAV visual-inertial-visual-downward integrated navigation as described above.

[0022] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements an estimation method for installation error of any of the above-described UAV visual-inertial-visual-downward integrated navigation systems.

[0023] The present invention also provides a computer program product, including a computer program that, when executed by a processor, implements an estimation method for installation error of any of the above-described UAV visual-inertial-visual-downward integrated navigation systems.

[0024] This invention provides a method for estimating the installation error of a UAV's visual-inertial-visual-look-down integrated navigation system. The method involves acquiring the current-moment observations, current-moment observation matrix, and historical state variables of the UAV's visual-inertial-visual-look-down integrated navigation system in real time; wherein the historical state variables are the state variables from the previous moment. Based on the current-moment observations, current-moment observation matrix, and historical state variables, an adaptive state estimation result and an adaptive state estimation variance are determined. Finally, based on the navigation installation error convergence condition, a navigation installation error convergence judgment is made on the adaptive state estimation result and the adaptive state estimation variance to determine the navigation installation error convergence judgment result. The technical solution of this invention addresses the shortcomings of existing methods for estimating and compensating installation errors, which involve pre-processing, cumbersome procedures, numerous limitations, and the fact that the original installation errors become inapplicable once the integrated navigation system is disassembled and reassembled, thus restricting the flexible application of inertial vision integrated navigation systems. This invention achieves the determination of adaptive state estimation results and adaptive state estimation variance by real-time acquisition of the current observations, current observation matrix, and historical state variables of the UAV's visual-inertial-visual-downward-looking integrated navigation system. Based on the navigation installation error convergence condition, the adaptive state estimation results and adaptive state estimation variance are used to determine the navigation installation error convergence result, eliminating the need to consider the error impact caused by the disassembly and reassembly of the integrated navigation system. This allows for online and accurate estimation of navigation installation errors, improving the ergodicity and practical accuracy of the UAV visual-inertial-visual-downward-looking integrated navigation system. Attached Figure Description

[0025] To more clearly illustrate the technical solutions in this invention or the prior art, 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 invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0026] Figure 1 This is a flowchart illustrating the method for estimating the installation error of the UAV visual-inertial-visual-downward integrated navigation system provided by the present invention.

[0027] Figure 2 This is a schematic diagram of the structure of the device for estimating the installation error of the UAV visual-inertial-visual-downward integrated navigation system provided by the present invention.

[0028] Figure 3 This is a schematic diagram of the structure of the electronic device provided by the present invention. Detailed Implementation

[0029] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0030] The following is combined with Figure 1 This invention describes a method for estimating the installation error of a UAV visual-inertial-visual-look-down integrated navigation system. This method is applicable to situations involving the estimation of installation errors in UAV visual-inertial-visual-look-down integrated navigation systems. The execution subject of this method can be an electronic device or a device for estimating the installation error of a UAV visual-inertial-visual-look-down integrated navigation system installed in that electronic device. This device can be implemented through software, hardware, or a combination of both. Figure 1 This is a flowchart illustrating the method for estimating installation errors of the UAV visual-inertial-visual-look-down integrated navigation system provided by the present invention. Figure 1 As shown, the method includes the following steps: 101, 102 and 103.

[0031] Step 101: Real-time acquisition of the current observations, current observation matrix, and historical state variables of the UAV visual-inertial-visual downward-looking combination.

[0032] In this step, the historical state variable is the state variable of the previous time step.

[0033] Specifically, the system acquires in real time the current observations, current observation matrix, and historical state variables of the UAV's visual-inertial-visual-downward combination.

[0034] Current observation The calculation is shown in formula (1).

[0035] (1) In formula (1), This indicates the latitude calculated by the inertial navigation system of the UAV. This indicates the longitude calculated by the inertial navigation system of the UAV. This indicates the latitude calculated for the UAV's visual navigation. This indicates the longitude calculated by the drone's visual navigation.

[0036] The current observation matrix is ​​calculated as shown in formula (2).

[0037] (2) In formula (2), Indicates the misalignment angle of the drone. This indicates the positional error of the drone. This indicates the installation error and misalignment angle of the drone. The calculation is shown in formula (3), installation error The calculation is shown in formula (4).

[0038] (3) (4) In formulas (3) and (4), Indicates the drone's flight altitude. This represents the slant distance from the ground along the optical axis of the drone's visual sensor. This represents the radius of the Earth's meridian. Indicates the radius of the Earth's circumpolar orbit; The transformation matrix, representing the transformation from the camera coordinate system to the vehicle coordinate system, is 3. 3 matrices Represents the first in the transformation matrix Line number Column elements, .

[0039] Step 102: Determine the adaptive state estimation result and the adaptive state estimation variance based on the current observations, the current observation matrix, and the historical state variables.

[0040] In this step, historical state variables The state variable is the state variable of the previous time step. Indicates the current moment. This refers to the time before the current time, but this embodiment does not limit this.

[0041] Specifically, the state variables of the inertial / look-down integrated navigation filter, which include the installation error between the inertial navigation and vision sensors, are constructed. It is a 17-dimensional state variable, as shown in formula (5).

[0042] (5) In formula (5), This indicates the northward misalignment angle of the drone's inertial navigation. This indicates the yaw misalignment angle of the drone's inertial navigation. This indicates the eastward misalignment angle of the drone's inertial navigation. This represents the northbound velocity error of the UAV's inertial navigation. This represents the azimuth velocity error of the drone's inertial navigation. This represents the eastward velocity error of the UAV's inertial navigation. This represents the latitude error of the drone's inertial navigation. This indicates the altitude error of the drone's inertial navigation. This indicates the longitude error of the drone's inertial navigation. , and These represent the coordinate systems of the UAV's inertial navigation carrier ( , and The gyroscope on the screen drifts. , and These represent the coordinate systems of the UAV's inertial navigation carrier ( , and The acceleration on the surface is zero bias. and This indicates the installation error between the inertial navigation and visual sensors of the drone.

[0043] Constructing filter state variables Subsequently, based on the filter state variables The current observations, the current observation matrix, and the historical state variables determine the adaptive state estimation result and the adaptive state estimation variance. That is Filter state at the current moment .

[0044] In one specific embodiment, determining the adaptive state estimation result and the adaptive state estimation variance based on the current observations, the current observation matrix, and the historical state variables includes: determining the state prediction result based on the historical state variables; determining the adaptive filter gain based on the current observation matrix; determining the adaptive state estimation result based on the state prediction result, the adaptive filter gain, the current observations, and the current observation matrix; and determining the adaptive state estimation variance based on the adaptive filter gain and the current observation matrix.

[0045] In one specific embodiment, determining the state prediction result based on historical state variables includes: obtaining the integrated navigation state transition matrix established by the UAV based on historical inertial navigation error patterns; and determining the state prediction result based on the integrated navigation state transition matrix and historical state variables.

[0046] Specifically, the adaptive state estimation results The calculation is shown in formula (6).

[0047] (6) In formula (6), This indicates the state prediction result. express Adaptive filter gain at time step express The current time observation at time , express Current observation matrix at time step 1, state prediction result The calculation is shown in formula (7), and the adaptive filter gain is... The calculation is shown in formula (8).

[0048] (7) In formula (7), This represents the integrated navigation state transition matrix established by the UAV based on the laws governing inertial navigation errors. express Historical state variables at any given time.

[0049] In one specific embodiment, determining the adaptive filtering gain based on the current observation matrix includes: obtaining the current observation noise matrix and determining the UAV state prediction variance; and determining the adaptive filtering gain based on the UAV state prediction variance, the current observation noise matrix, and the current observation matrix.

[0050] In one specific embodiment, determining the adaptive filtering gain based on the current observation matrix includes: determining the adaptive filtering gain based on the UAV state prediction variance, the current observation matrix, and the current observation noise matrix.

[0051] Specifically, adaptive filter gain The calculation is shown in formula (8).

[0052] (8) In formula (8), This represents the variance of the drone state prediction. express The transpose of the current observation matrix at time t. express Current observation noise matrix at time step 1, and UAV state prediction variance. The calculation is shown in formula (9).

[0053] In one specific embodiment, determining the UAV state prediction variance includes: obtaining the historical error covariance and the historical system noise matrix; and determining the UAV state prediction variance based on the integrated navigation state transition matrix, the historical error covariance, and the historical system noise matrix.

[0054] Specifically, the variance of UAV state prediction The calculation is shown in formula (9).

[0055] (9) In formula (9), This represents the integrated navigation state transition matrix established by the UAV based on the laws governing inertial navigation errors. express Historical error covariance at time point This represents the transpose of the integrated navigation state transition matrix. express The historical system noise matrix at any given time.

[0056] In one specific embodiment, determining the adaptive state estimation variance based on the adaptive filter gain and the current observation matrix includes: determining the adaptive state estimation variance based on the adaptive filter gain, the current observation matrix, the UAV state prediction variance, and the current observation noise matrix.

[0057] Specifically, adaptive state estimation variance The calculation is shown in formula (10).

[0058] (10) In formula (10), Indicates adaptive filter gain The transpose of .

[0059] Step 103: Based on the navigation installation error convergence condition, determine the navigation installation error convergence of the adaptive state estimation result and the adaptive state estimation variance, and determine the navigation installation error convergence result.

[0060] Specifically, after determining the adaptive state estimation result and the adaptive state estimation variance, the navigation installation error convergence judgment is performed on the adaptive state estimation result and the adaptive state estimation variance according to the navigation installation error convergence condition, and the navigation installation error convergence judgment result is determined.

[0061] The advantage of this setup is that it utilizes the current-time observations and the current-time observation matrix obtained through downward matching to determine the adaptive state estimation result and the adaptive state estimation variance. Based on the navigation installation error convergence condition, it performs a navigation installation error convergence judgment on the adaptive state estimation result and the adaptive state estimation variance, thereby determining the navigation installation error convergence judgment result and realizing real-time online estimation of navigation installation error. Compared with traditional methods that pre-calibrate or offline estimate installation errors, this significantly simplifies the system usage process and reduces the difficulty of ensuring accuracy.

[0062] In one specific embodiment, the navigation installation error convergence condition includes an effective filtering number threshold, an adaptive state estimation variance threshold, an adaptive state estimation variance absolute value threshold, an adaptive state estimation result peak-to-peak value threshold, and an adaptive state result absolute value threshold.

[0063] In this step, the effective filtering threshold can be, for example, Second-rate, It is a positive integer greater than or equal to 0; the adaptive state estimation variance threshold can be, for example, a pre-set threshold. The absolute value threshold for the adaptive state estimation variance can be, for example, a pre-set threshold. The peak-to-peak threshold of the adaptive state estimation result can be, for example, a pre-set threshold. The absolute value threshold for the adaptive state result can be, for example, a pre-set threshold. This embodiment does not limit this aspect.

[0064] Specifically, by combining the UAV's flight scenario and downward-looking matching status, navigation installation error convergence conditions are designed to determine the convergence of installation errors and thus the installation error convergence status is determined. The value of is related to the flight scenario and trajectory; for example, it can be between 200 and 500, which is the adaptive state estimation variance threshold. Adaptive state estimation absolute variance threshold Peak-to-peak threshold of adaptive state estimation results and the absolute value threshold of the adaptive state result The settings can be adjusted according to the actual situation; this embodiment does not impose any limitations on them.

[0065] In one specific embodiment, the navigation installation error convergence judgment is performed on the adaptive state estimation result and the adaptive state estimation variance according to the navigation installation error convergence condition, and the navigation installation error convergence judgment result is determined. This includes: obtaining the current number of filtering iterations for determining the adaptive state estimation result and the adaptive state estimation variance; if the current number of filtering iterations is greater than or equal to the effective number of filtering iterations threshold, the adaptive state estimation variance is less than or equal to the adaptive state estimation variance threshold, the adaptive state estimation variance is less than or equal to the absolute value threshold of the adaptive state estimation variance, and the adaptive state estimation result is less than or equal to the peak-to-peak value threshold and the adaptive state estimation result is less than or equal to the absolute value threshold of the adaptive state result, the navigation installation error convergence judgment result is determined to be navigation installation error convergence; otherwise, the navigation installation error convergence judgment result is determined to be navigation installation error non-convergence.

[0066] Specifically, the current number of filtering iterations is obtained to determine the adaptive state estimation result and the adaptive state estimation variance. If the current number of filtering iterations is greater than or equal to the effective number of filtering iterations threshold, the adaptive state estimation variance is less than or equal to the adaptive state estimation variance threshold, the adaptive state estimation variance is less than or equal to the absolute value threshold of the adaptive state estimation variance, and the adaptive state estimation result is less than or equal to the peak-to-peak value threshold and the adaptive state estimation result is less than or equal to the absolute value threshold of the adaptive state result, the navigation installation error convergence judgment result is determined to be navigation installation error convergence; otherwise, the navigation installation error convergence judgment result is determined to be navigation installation error non-convergence.

[0067] In one specific embodiment, after determining the navigation installation error convergence result by judging the adaptive state estimation result and the adaptive state estimation variance based on the navigation installation error convergence condition, the method further includes: if the navigation installation error convergence result is determined to be navigation installation error convergence, performing online adaptive correction on the navigation installation error convergence result to determine the error correction result.

[0068] Specifically, after determining the convergence of the navigation installation error by judging the adaptive state estimation result and the adaptive state estimation variance based on the navigation installation error convergence condition, and then determining the navigation installation error convergence result, online adaptive correction is performed on the navigation installation error convergence result to determine the error correction result.

[0069] For example, the online adaptive correction method may refer to the inertial navigation misalignment angle for correction, but this embodiment does not limit it.

[0070] The advantage of this setup is that it allows for online adaptive correction of installation errors during UAV flight missions, significantly reducing the adverse effects of installation errors on observation accuracy and improving the accuracy of inertial vision integrated navigation.

[0071] In one specific embodiment, after determining the navigation installation error convergence result by judging the adaptive state estimation result and the adaptive state estimation variance based on the navigation installation error convergence condition, the method further includes: if the navigation installation error convergence result is determined to be that the navigation installation error does not converge, the method continues to return to the step of real-time acquisition of the current observations, the current observation matrix, and the historical state variables of the UAV visual-inertial-visual downward-looking combination.

[0072] Specifically, after determining the convergence of the navigation installation error by judging the adaptive state estimation result and the adaptive state estimation variance based on the navigation installation error convergence condition, and determining the navigation installation error convergence result, if the navigation installation error convergence result is determined to be that the navigation installation error does not converge, the process continues to return to the step of real-time acquisition of the current observations, current observation matrix and historical state variables of the UAV visual-inertial-visual downward-looking combination.

[0073] This invention provides a method for estimating the installation error of a UAV's visual-inertial-visual-look-down integrated navigation system. The method involves acquiring the current-moment observations, current-moment observation matrix, and historical state variables of the UAV's visual-inertial-visual-look-down integrated navigation system in real time; wherein the historical state variables are the state variables from the previous moment. Based on the current-moment observations, current-moment observation matrix, and historical state variables, an adaptive state estimation result and an adaptive state estimation variance are determined. Finally, based on the navigation installation error convergence condition, a navigation installation error convergence judgment is made on the adaptive state estimation result and the adaptive state estimation variance to determine the navigation installation error convergence judgment result. In the above embodiments, the technical solution of the present invention addresses the shortcomings of existing methods for estimating and compensating installation errors, which involve pre-processing, cumbersome procedures, numerous limitations, and the fact that the original installation errors become inapplicable once the integrated navigation system is disassembled and reassembled, thus restricting the flexible application of the inertial vision integrated navigation system. The present invention achieves the determination of adaptive state estimation results and adaptive state estimation variance by real-time acquisition of the current observations, current observation matrix, and historical state variables of the UAV visual-inertial-visual-downward combined navigation system; and determines the navigation installation error convergence result by judging the adaptive state estimation results and adaptive state estimation variance based on the navigation installation error convergence condition. This eliminates the need to consider the error impact caused by the disassembly and reassembly of the integrated navigation system, enabling online and accurate estimation of navigation installation errors and improving the ergodicity and practical accuracy of the UAV visual-inertial-visual-downward combined navigation system.

[0074] The following describes the device for estimating the installation error of UAV visual-inertial-visual-downward integrated navigation provided by the present invention. The device for estimating the installation error of UAV visual-inertial-visual-downward integrated navigation described below can be referred to in correspondence with the method for estimating the installation error of UAV visual-inertial-visual-downward integrated navigation described above.

[0075] Figure 2 This is a schematic diagram of the structure of the device for estimating the installation error of the UAV visual-inertial-visual-look-down integrated navigation system provided by the present invention, with reference to... Figure 2 As shown, the device 200 for estimating the installation error of a UAV's visual-inertial-visual-look-down integrated navigation system includes: a variable acquisition module 201, a variance determination module 202, and a result determination module 203; wherein, The variable acquisition module 201 is used to acquire the current observations, current observation matrix and historical state variables of the UAV visual-inertial-visual-downward combination in real time; wherein, the historical state variables are the state variables of the previous moment of the current moment.

[0076] The variance determination module 202 is used to determine the adaptive state estimation result and the adaptive state estimation variance based on the current observations, the current observation matrix and the historical state variables.

[0077] The result determination module 203 is used to determine the navigation installation error convergence result by judging the adaptive state estimation result and the adaptive state estimation variance based on the navigation installation error convergence condition.

[0078] In one example embodiment, the variance determination module 202 is specifically used to: determine the state prediction result based on historical state variables; determine the adaptive filter gain based on the current observation matrix; determine the adaptive state estimation result based on the state prediction result, the adaptive filter gain, the current observations, and the current observation matrix; and determine the adaptive state estimation variance based on the adaptive filter gain and the current observation matrix.

[0079] In one example embodiment, the variance determination module 202 determines the state prediction result based on historical state variables, specifically for: obtaining the integrated navigation state transition matrix established by the UAV based on the historical inertial navigation error rules; and determining the state prediction result based on the integrated navigation state transition matrix and historical state variables.

[0080] In one example embodiment, the variance determination module 202 determines the adaptive filtering gain based on the current observation matrix, specifically for: obtaining the current observation noise matrix and determining the UAV state prediction variance; and determining the adaptive filtering gain based on the UAV state prediction variance, the current observation noise matrix, and the current observation matrix.

[0081] In one example embodiment, the variance determination module 202 determines the UAV state prediction variance, specifically for: obtaining the historical error covariance and the historical system noise matrix; and determining the UAV state prediction variance based on the integrated navigation state transition matrix, the historical error covariance, and the historical system noise matrix.

[0082] In one example embodiment, the variance determination module 202 determines the adaptive filtering gain based on the current observation matrix, specifically for: determining the adaptive filtering gain based on the UAV state prediction variance, the current observation matrix, and the current observation noise matrix.

[0083] In one example embodiment, the variance determination module 202 determines the adaptive state estimation variance based on the adaptive filter gain and the current observation matrix, specifically: determining the adaptive state estimation variance based on the adaptive filter gain, the current observation matrix, the UAV state prediction variance, and the current observation noise matrix.

[0084] In one example embodiment, the navigation installation error convergence condition includes an effective filtering number threshold, an adaptive state estimation variance threshold, an adaptive state estimation variance absolute value threshold, an adaptive state estimation result peak-to-peak value threshold, and an adaptive state result absolute value threshold.

[0085] In one example embodiment, the result determination module 203 is specifically used to: obtain the current number of filtering iterations for determining the adaptive state estimation result and the adaptive state estimation variance; if the current number of filtering iterations is greater than or equal to the effective number of filtering iterations threshold, the adaptive state estimation variance is less than or equal to the adaptive state estimation variance threshold, the adaptive state estimation variance is less than or equal to the absolute value threshold of the adaptive state estimation variance, and the adaptive state estimation result is less than or equal to the peak-to-peak value threshold of the adaptive state estimation result and the adaptive state estimation result is less than or equal to the absolute value threshold of the adaptive state result, the navigation installation error convergence judgment result is determined to be navigation installation error convergence; otherwise, the navigation installation error convergence judgment result is determined to be navigation installation error non-convergence.

[0086] In one example embodiment, the device further includes an error correction module. The error correction module is configured to: after determining the navigation installation error convergence result by performing a navigation installation error convergence judgment on the adaptive state estimation result and the adaptive state estimation variance based on the navigation installation error convergence condition, and then, if the navigation installation error convergence judgment result is determined to be navigation installation error convergence, perform online adaptive correction on the navigation installation error convergence judgment result to determine the error correction result.

[0087] In one example embodiment, the device further includes a feedback execution module. The feedback execution module is configured to: after determining the navigation installation error convergence result by performing a navigation installation error convergence judgment on the adaptive state estimation result and the adaptive state estimation variance based on the navigation installation error convergence condition, and determining that the navigation installation error convergence judgment result is that the navigation installation error does not converge, continue to return to the step of real-time acquisition of the current-moment observations, the current-moment observation matrix, and the historical state variables of the UAV visual-inertial-visual downward-looking combination.

[0088] The apparatus of this embodiment can be used to execute any of the methods in the side embodiment of the method for estimating the installation error of UAV visual-inertial-visual-downward integrated navigation. Its specific implementation process and technical effects are similar to those in the side embodiment of the method for estimating the installation error of UAV visual-inertial-visual-downward integrated navigation. For details, please refer to the detailed description in the side embodiment of the method for estimating the installation error of UAV visual-inertial-visual-downward integrated navigation, which will not be repeated here.

[0089] Figure 3 This is a schematic diagram of the structure of the electronic device provided by the present invention, such as... Figure 3 As shown, the electronic device may include a processor 310, a communication interface 320, a memory 330, and a communication bus 340. The processor 310, communication interface 320, and memory 330 communicate with each other via the communication bus 340. The processor 310 can call logical instructions in the memory 330 to execute a method for estimating the installation error of the UAV's visual-inertial-visual-downward integrated navigation system. This method includes: real-time acquisition of the current-moment observations, the current-moment observation matrix, and historical state variables of the UAV's visual-inertial-visual-downward integrated navigation system; wherein the historical state variables are the state variables of the previous moment; determining the adaptive state estimation result and the adaptive state estimation variance based on the current-moment observations, the current-moment observation matrix, and the historical state variables; and performing a navigation installation error convergence judgment on the adaptive state estimation result and the adaptive state estimation variance according to the navigation installation error convergence condition, thereby determining the navigation installation error convergence judgment result.

[0090] Furthermore, the logical instructions in the aforementioned memory 330 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0091] On the other hand, the present invention also provides a computer program product, which includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the estimation method for the installation error of the UAV visual-inertial-visual-downward integrated navigation provided by the above methods. The method includes: acquiring in real time the current observations, the current observation matrix, and historical state variables of the UAV visual-inertial-visual-downward integrated navigation; wherein the historical state variables are the state variables of the previous moment of the current moment; determining the adaptive state estimation result and the adaptive state estimation variance based on the current observations, the current observation matrix, and the historical state variables; and performing a navigation installation error convergence judgment on the adaptive state estimation result and the adaptive state estimation variance based on the navigation installation error convergence condition, and determining the navigation installation error convergence judgment result.

[0092] In another aspect, the present invention also provides a non-transitory computer-readable storage medium storing a computer program thereon. When executed by a processor, the computer program implements a method for estimating the installation error of a UAV visual-inertial-visual-downward integrated navigation system provided by the methods described above. The method includes: acquiring in real time the current-moment observations, the current-moment observation matrix, and historical state variables of the UAV visual-inertial-visual-downward integrated navigation system; wherein the historical state variables are the state variables of the previous moment; determining an adaptive state estimation result and an adaptive state estimation variance based on the current-moment observations, the current-moment observation matrix, and the historical state variables; and performing a navigation installation error convergence judgment on the adaptive state estimation result and the adaptive state estimation variance based on the navigation installation error convergence condition, thereby determining the navigation installation error convergence judgment result.

[0093] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0094] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0095] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for estimating the installation error of a UAV's visual-inertial-visual-look-down integrated navigation system, characterized in that, include: The current observations, current observation matrix, and historical state variables of the UAV's visual-inertial-visual-downward combination are acquired in real time; wherein, the historical state variables are the state variables of the previous moment of the current moment. The adaptive state estimation result and the adaptive state estimation variance are determined based on the current observations, the current observation matrix, and the historical state variables. Based on the navigation installation error convergence condition, the adaptive state estimation result and the adaptive state estimation variance are used to determine the navigation installation error convergence result.

2. The method for estimating the installation error of a UAV visual-inertial-visual-look-down integrated navigation system according to claim 1, characterized in that, The step of determining the adaptive state estimation result and the adaptive state estimation variance based on the current-time observations, the current-time observation matrix, and the historical state variables includes: The state prediction result is determined based on the historical state variables; The adaptive filter gain is determined based on the observation matrix at the current time. The adaptive state estimation result is determined based on the state prediction result, the adaptive filter gain, the current time observations, and the current time observation matrix; The adaptive state estimation variance is determined based on the adaptive filter gain and the observation matrix at the current time.

3. The method for estimating the installation error of a UAV visual-inertial-visual-look-down integrated navigation system according to claim 2, characterized in that, Determining the state prediction result based on the historical state variables includes: Obtain the integrated navigation state transition matrix of the UAV based on the historical inertial navigation error patterns; The state prediction result is determined based on the integrated navigation state transition matrix and the historical state variables.

4. The method for estimating the installation error of the UAV visual-inertial-visual-look-down integrated navigation system according to claim 3, characterized in that, The step of determining the adaptive filtering gain based on the current observation matrix includes: Obtain the current observation noise matrix and determine the variance of the UAV state prediction; The adaptive filtering gain is determined based on the UAV state prediction variance, the current time observation noise matrix, and the current time observation matrix.

5. The method for estimating the installation error of a UAV visual-inertial-visual-look-down integrated navigation system according to claim 4, characterized in that, The determination of the UAV state prediction variance includes: Obtain the historical error covariance and historical system noise matrix; The UAV state prediction variance is determined based on the integrated navigation state transition matrix, the historical error covariance, and the historical system noise matrix.

6. The method for estimating the installation error of a UAV visual-inertial-visual-look-down integrated navigation system according to claim 5, characterized in that, The step of determining the adaptive filtering gain based on the current observation matrix includes: The adaptive filtering gain is determined based on the UAV state prediction variance, the current time observation matrix, and the current time observation noise matrix.

7. The method for estimating the installation error of a UAV visual-inertial-visual-look-down integrated navigation system according to claim 6, characterized in that, The step of determining the adaptive state estimation variance based on the adaptive filter gain and the current time observation matrix includes: The adaptive state estimation variance is determined based on the adaptive filter gain, the current observation matrix, the UAV state prediction variance, and the current observation noise matrix.

8. The method for estimating the installation error of a UAV visual-inertial-visual-look-down integrated navigation system according to claim 1, characterized in that, The convergence conditions for navigation installation error include an effective filtering number threshold, an adaptive state estimation variance threshold, an adaptive state estimation variance absolute value threshold, an adaptive state estimation result peak-to-peak value threshold, and an adaptive state result absolute value threshold.

9. The method for estimating the installation error of a UAV visual-inertial-visual-look-down integrated navigation system according to claim 8, characterized in that, The step of determining the navigation installation error convergence result by judging the adaptive state estimation result and the adaptive state estimation variance based on the navigation installation error convergence condition includes: Obtain the current number of filtering iterations to determine the adaptive state estimation result and the adaptive state estimation variance; If the current filtering count is greater than or equal to the effective filtering count threshold, the adaptive state estimation variance is less than or equal to the adaptive state estimation variance threshold, the adaptive state estimation variance is less than or equal to the absolute value threshold of the adaptive state estimation variance, and the adaptive state estimation result is less than or equal to the peak-to-peak value threshold of the adaptive state estimation result and the adaptive state estimation result is less than or equal to the absolute value threshold of the adaptive state result, then the navigation installation error convergence judgment result is determined to be navigation installation error convergence; otherwise, the navigation installation error convergence judgment result is determined to be navigation installation error non-convergence.

10. The method for estimating the installation error of a UAV visual-inertial-visual-look-down integrated navigation system according to any one of claims 1-9, characterized in that, After determining the navigation installation error convergence result by performing a navigation installation error convergence judgment on the adaptive state estimation result and the adaptive state estimation variance based on the navigation installation error convergence condition, the method further includes: If the navigation installation error convergence judgment result is determined to be that the navigation installation error has converged, the navigation installation error convergence judgment result is adaptively corrected online to determine the error correction result.

11. The method for estimating the installation error of a UAV visual-inertial-visual-downward integrated navigation system according to claim 10, characterized in that, After determining the navigation installation error convergence result by performing a navigation installation error convergence judgment on the adaptive state estimation result and the adaptive state estimation variance based on the navigation installation error convergence condition, the method further includes: If the navigation installation error convergence judgment result is that the navigation installation error does not converge, the process continues to return to the step of real-time acquisition of the current observation, current observation matrix and historical state variables of the UAV visual-inertial-visual downward combination.

12. A device for estimating the installation error of a UAV's visual-inertial-visual-look-down integrated navigation system, characterized in that, include: The variable acquisition module is used to acquire in real time the current observations, current observation matrix, and historical state variables of the UAV's visual-inertial-visual-downward combination; wherein, the historical state variables are the state variables of the previous moment of the current moment; The variance determination module is used to determine the adaptive state estimation result and the adaptive state estimation variance based on the current time observations, the current time observation matrix, and the historical state variables. The result determination module is used to determine the navigation installation error convergence result by judging the adaptive state estimation result and the adaptive state estimation variance based on the navigation installation error convergence condition.

Images